Search results for: factorization of computation

Authors: Michel Soto Chalhoub

Abstract:

Building code-related literature provides recommendations on normalizing approaches to the calculation of the dynamic properties of structures. Most building codes make a distinction among types of structural systems, construction material, and configuration through a numerical coefficient in the expression for the fundamental period. The period is then used in normalized response spectra to compute base shear. The typical parameter used in simplified code formulas for the fundamental period is overall building height raised to a power determined from analytical and experimental results. However, reinforced concrete buildings which constitute the majority of built space in less developed countries pose additional challenges to the ones built with homogeneous material such as steel, or with concrete under stricter quality control. In the present paper, the particularities of reinforced concrete buildings are explored and related to current methods of equivalent static analysis. A comparative study is presented between the Uniform Building Code, commonly used for buildings within and outside the USA, and data from the Middle East used to model 151 reinforced concrete buildings of varying number of bays, number of floors, overall building height, and individual story height. The fundamental period was calculated using eigenvalue matrix computation. The results were also used in a separate regression analysis where the computed period serves as dependent variable, while five building properties serve as independent variables. The statistical analysis shed light on important parameters that simplified code formulas need to account for including individual story height, overall building height, floor plan, number of bays, and concrete properties. Such inclusions are important for reinforced concrete buildings of special conditions due to the level of concrete damage, aging, or materials quality control during construction. Overall results of the present analysis show that simplified code formulas for fundamental period and base shear may be applied but they require revisions to account for multiple parameters. The conclusion above is confirmed by the analytical model where fundamental periods were computed using numerical techniques and eigenvalue solutions. This recommendation is particularly relevant to code upgrades in less developed countries where it is customary to adopt, and mildly adapt international codes. We also note the necessity of further research using empirical data from buildings in Lebanon that were subjected to severe damage due to impulse loading or accelerated aging. However, we excluded this study from the present paper and left it for future research as it has its own peculiarities and requires a different type of analysis.

Keywords: seismic behaviour, reinforced concrete, simplified code formulas, equivalent static analysis, base shear, response spectra

Procedia PDF Downloads 232

Authors: Aline V. Arroteia, Gustavo G. Do Amaral, Simone Z. Kikuti, Norberto C. S. Moura, Silvio B. Melhado

Abstract:

Despite the significant advances made by the construction industry in recent years, the crystalized absence of integration between the design and construction phases is still an evident and costly problem in building construction. Globally, the construction industry has sought to adopt collaborative practices through new technologies to mitigate impacts of this fragmented process and to optimize its production. In this new technological business environment, professionals are required to develop new methodologies based on the notion of collaboration and integration of information throughout the building lifecycle. This scenario also represents the industry’s reality in developing nations, and the increasing need for overall efficiency has demanded new educational alternatives at the undergraduate and post-graduate levels. In countries like Brazil, it is the common understanding that Architecture, Engineering and Building Construction educational programs are being required to review the traditional design pedagogical processes to promote a comprehensive notion about integration and simultaneity between the phases of the project. In this context, the coherent inclusion of computation design to all segments of the educational programs of construction related professionals represents a significant research topic that, in fact, can affect the industry practice. Thus, the main objective of the present study was to comparatively measure the effectiveness of the Building Information Modeling courses offered by the University of Sao Paulo, the most important academic institution in Brazil, at the Schools of Architecture and Civil Engineering and the courses offered in well recognized BIM research institutions, such as the School of Design in the College of Architecture of the Georgia Institute of Technology, USA, to evaluate the dissemination of BIM knowledge amongst students in post graduate level. The qualitative research methodology was developed based on the analysis of the program and activities proposed by two BIM courses offered in each of the above-mentioned institutions, which were used as case studies. The data collection instruments were a student questionnaire, semi-structured interviews, participatory evaluation and pedagogical practices. The found results have detected a broad heterogeneity of the students regarding their professional experience, hours dedicated to training, and especially in relation to their general knowledge of BIM technology and its applications. The research observed that BIM is mostly understood as an operational tool and not as methodological project development approach, relevant to the whole building life cycle. The present research offers in its conclusion an assessment about the importance of the incorporation of BIM, with efficiency and in its totality, as a teaching method in undergraduate and graduate courses in the Brazilian architecture, engineering and building construction schools.

Keywords: building information modeling (BIM), BIM education, BIM process, design teaching

Procedia PDF Downloads 154

Authors: Baha Eddine Aissou, Aichouche Belhadj Aissa

Abstract:

Light detection and ranging (LiDAR) is an active remote sensing technology used for several applications. Airborne LiDAR is becoming an important technology for the acquisition of a highly accurate dense point cloud. A classification of airborne laser scanning (ALS) point cloud is a very important task that still remains a real challenge for many scientists. Support vector machine (SVM) is one of the most used statistical learning algorithms based on kernels. SVM is a non-parametric method, and it is recommended to be used in cases where the data distribution cannot be well modeled by a standard parametric probability density function. Using a kernel, it performs a robust non-linear classification of samples. Often, the data are rarely linearly separable. SVMs are able to map the data into a higher-dimensional space to become linearly separable, which allows performing all the computations in the original space. This is one of the main reasons that SVMs are well suited for high-dimensional classification problems. Only a few training samples, called support vectors, are required. SVM has also shown its potential to cope with uncertainty in data caused by noise and fluctuation, and it is computationally efficient as compared to several other methods. Such properties are particularly suited for remote sensing classification problems and explain their recent adoption. In this poster, the SVM classification of ALS LiDAR data is proposed. Firstly, connected component analysis is applied for clustering the point cloud. Secondly, the resulting clusters are incorporated in the SVM classifier. Radial basic function (RFB) kernel is used due to the few numbers of parameters (C and γ) that needs to be chosen, which decreases the computation time. In order to optimize the classification rates, the parameters selection is explored. It consists to find the parameters (C and γ) leading to the best overall accuracy using grid search and 5-fold cross-validation. The exploited LiDAR point cloud is provided by the German Society for Photogrammetry, Remote Sensing, and Geoinformation. The ALS data used is characterized by a low density (4-6 points/m²) and is covering an urban area located in residential parts of the city Vaihingen in southern Germany. The class ground and three other classes belonging to roof superstructures are considered, i.e., a total of 4 classes. The training and test sets are selected randomly several times. The obtained results demonstrated that a parameters selection can orient the selection in a restricted interval of (C and γ) that can be further explored but does not systematically lead to the optimal rates. The SVM classifier with hyper-parameters is compared with the most used classifiers in literature for LiDAR data, random forest, AdaBoost, and decision tree. The comparison showed the superiority of the SVM classifier using parameters selection for LiDAR data compared to other classifiers.

Keywords: classification, airborne LiDAR, parameters selection, support vector machine

Procedia PDF Downloads 147

Authors: Ana Paula Soares, Alexandrina Lages, Helena Oliveira, Francisco-Javier Gutiérrez-Domínguez, Marisa Lousada

Abstract:

From an early age, exposure to a spoken language allows us to implicitly capture the structure underlying the succession of the speech sounds in that language and to segment it into meaningful units (words). Statistical learning (SL), i.e., the ability to pick up patterns in the sensory environment even without intention or consciousness of doing it, is thus assumed to play a central role in the acquisition of the rule-governed aspects of language and possibly to lie behind the language difficulties exhibited by children with development language disorder (DLD). The research conducted so far has, however, led to inconsistent results, which might stem from the behavioral tasks used to test SL. In a classic SL experiment, participants are first exposed to a continuous stream (e.g., syllables) in which, unbeknownst to the participants, stimuli are grouped into triplets that always appear together in the stream (e.g., ‘tokibu’, ‘tipolu’), with no pauses between each other (e.g., ‘tokibutipolugopilatokibu’) and without any information regarding the task or the stimuli. Following exposure, SL is assessed by asking participants to discriminate between triplets previously presented (‘tokibu’) from new sequences never presented together during exposure (‘kipopi’), i.e., to perform a two-alternative-forced-choice (2-AFC) task. Despite the widespread use of the 2-AFC to test SL, it has come under increasing criticism as it is an offline post-learning task that only assesses the result of the learning that had occurred during the previous exposure phase and that might be affected by other factors beyond the computation of regularities embedded in the input, typically the likelihood two syllables occurring together, a statistic known as transitional probability (TP). One solution to overcome these limitations is to assess SL as exposure to the stream unfolds using online techniques such as event-related potentials (ERP) that is highly sensitive to the time-course of the learning in the brain. Here we collected ERPs to examine the neurofunctional correlates of SL in preschool children with DLD, and chronological-age typical language development (TLD) controls who were exposed to an auditory stream in which eight three-syllable nonsense words, four of which presenting high-TPs and the other four low-TPs, to further analyze whether the ability of DLD and TLD children to extract-word-like units from the steam was modulated by words’ predictability. Moreover, to ascertain if the previous knowledge of the to-be-learned-regularities affected the neural responses to high- and low-TP words, children performed the auditory SL task, firstly, under implicit, and, subsequently, under explicit conditions. Although behavioral evidence of SL was not obtained in either group, the neural responses elicited during the exposure phases of the SL tasks differentiated children with DLD from children with TLD. Specifically, the results indicated that only children from the TDL group showed neural evidence of SL, particularly in the SL task performed under explicit conditions, firstly, for the low-TP, and, subsequently, for the high-TP ‘words’. Taken together, these findings support the view that children with DLD showed deficits in the extraction of the regularities embedded in the auditory input which might underlie the language difficulties.

Keywords: development language disorder, statistical learning, transitional probabilities, word segmentation

Procedia PDF Downloads 188

Authors: Piyush Sharma, Laltu Chandra, P. S. Ghoshdastidar, Rajiv Shekhar

Abstract:

Metals processing operations such as melting and heat treatment of metals are energy-intensive, requiring temperatures greater than 500oC. The desired temperature in these industrial furnaces is attained by circulating electrically-heated air. In most of these furnaces, electricity produced from captive coal-based thermal power plants is used. Solar thermal energy could be a viable heat source in these furnaces. A retrofitted solar convective furnace (SCF) concept, which uses solar thermal generated hot air, has been proposed. Critical to the success of a SCF is the design of an open volumetric air receiver (OVAR), which can heat air in excess of 800oC. The OVAR is placed on top of a tower and receives concentrated solar radiation from a heliostat field. Absorbers, mixer assembly, and the return air flow chamber (RAFC) are the major components of an OVAR. The absorber is a porous structure that transfers heat from concentrated solar radiation to ambient air, referred to as primary air. The mixer ensures uniform air temperature at the receiver exit. Flow of the relatively cooler return air in the RAFC ensures that the absorbers do not fail by overheating. In an earlier publication, the detailed design basis, fabrication, and characterization of a 2 kWth open volumetric air receiver (OVAR) based laboratory solar air tower simulator was presented. Development of an experimentally-validated, CFD based mathematical model which can ultimately be used for the design and scale-up of an OVAR has been the major objective of this investigation. In contrast to the published literature, where flow and heat transfer have been modeled primarily in a single absorber module, the present study has modeled the entire receiver assembly, including the RAFC. Flow and heat transfer calculations have been carried out in ANSYS using the LTNE model. The complex return air flow pattern in the RAFC requires complicated meshes and is computational and time intensive. Hence a simple, realistic 1-D mathematical model, which circumvents the need for carrying out detailed flow and heat transfer calculations, has also been proposed. Several important results have emerged from this investigation. Circumferential electrical heating of absorbers can mimic frontal heating by concentrated solar radiation reasonably well in testing and characterizing the performance of an OVAR. Circumferential heating, therefore, obviates the need for expensive high solar concentration simulators. Predictions suggest that the ratio of power on aperture (POA) and mass flow rate of air (MFR) is a normalizing parameter for characterizing the thermal performance of an OVAR. Increasing POA/MFR increases the maximum temperature of air, but decreases the thermal efficiency of an OVAR. Predictions of the 1-D mathematical are within 5% of ANSYS predictions and computation time is reduced from ~ 5 hours to a few seconds.

Keywords: absorbers, mixer assembly, open volumetric air receiver, return air flow chamber, solar thermal energy

Procedia PDF Downloads 197

Authors: V. E. Messerle, O. A. Lavrichshev, A. B. Ustimenko

Abstract:

Currently and in the foreseeable future (up to 2100), the global economy is oriented to the use of organic fuel, mostly, solid fuels, the share of which constitutes 40% in the generation of electric power. Therefore, the development of technologies for their effective and environmentally friendly application represents a priority problem nowadays. This work presents the results of thermodynamic and experimental investigations of plasma technology for processing of low-grade coals. The use of this technology for producing target products (synthesis gas, hydrogen, technical carbon, and valuable components of mineral mass of coals) meets the modern environmental and economic requirements applied to basic industrial sectors. The plasma technology of coal processing for the production of synthesis gas from the coal organic mass (COM) and valuable components from coal mineral mass (CMM) is highly promising. Its essence is heating the coal dust by reducing electric arc plasma to the complete gasification temperature, when the COM converts into synthesis gas, free from particles of ash, nitrogen oxides and sulfur. At the same time, oxides of the CMM are reduced by the carbon residue, producing valuable components, such as technical silicon, ferrosilicon, aluminum and carbon silicon, as well as microelements of rare metals, such as uranium, molybdenum, vanadium, titanium. Thermodynamic analysis of the process was made using a versatile computation program TERRA. Calculations were carried out in the temperature range 300 - 4000 K and a pressure of 0.1 MPa. Bituminous coal with the ash content of 40% and the heating value 16,632 kJ/kg was taken for the investigation. The gaseous phase of coal processing products includes, basically, a synthesis gas with a concentration of up to 99 vol.% at 1500 K. CMM components completely converts from the condensed phase into the gaseous phase at a temperature above 2600 K. At temperatures above 3000 K, the gaseous phase includes, basically, Si, Al, Ca, Fe, Na, and compounds of SiO, SiH, AlH, and SiS. The latter compounds dissociate into relevant elements with increasing temperature. Complex coal conversion for the production of synthesis gas from COM and valuable components from CMM was investigated using a versatile experimental plant the main element of which was plug and flow plasma reactor. The material and thermal balances helped to find the integral indicators for the process. Plasma-steam gasification of the low-grade coal with CMM processing gave the synthesis gas yield 95.2%, the carbon gasification 92.3%, and coal desulfurization 95.2%. The reduced material of the CMM was found in the slag in the form of ferrosilicon as well as silicon and iron carbides. The maximum reduction of the CMM oxides was observed in the slag from the walls of the plasma reactor in the areas with maximum temperatures, reaching 47%. The thusly produced synthesis gas can be used for synthesis of methanol, or as a high-calorific reducing gas instead of blast-furnace coke as well as power gas for thermal power plants. Reduced material of CMM can be used in metallurgy.

Keywords: gasification, mineral mass, organic mass, plasma, processing, solid fuel, synthesis gas, valuable components

Procedia PDF Downloads 608

Authors: Maxwell Henderson, Tristan Cook, Justin Chan Jin Le, Mark Hodson, YoungJung Chang, John Novak, Daniel Padilha, Nishan Kulatilaka, Ansu Bagchi, Sanjoy Ray, John Kelly

Abstract:

We present a method of using adiabatic quantum optimization (AQO) to solve a mixed integer nonlinear programming (MINLP) problem instance. The MINLP problem is a general form of a set of NP-hard optimization problems that are critical to many business applications. It requires optimizing a set of discrete and continuous variables with nonlinear and potentially nonconvex constraints. Obtaining an exact, optimal solution for MINLP problem instances of non-trivial size using classical computation methods is currently intractable. Current leading algorithms leverage heuristic and divide-and-conquer methods to determine approximate solutions. Creating more accurate and efficient algorithms is an active area of research. Quantum computing (QC) has several theoretical benefits compared to classical computing, through which QC algorithms could obtain MINLP solutions that are superior to current algorithms. AQO is a particular form of QC that could offer more near-term benefits compared to other forms of QC, as hardware development is in a more mature state and devices are currently commercially available from D-Wave Systems Inc. It is also designed for optimization problems: it uses an effect called quantum tunneling to explore all lowest points of an energy landscape where classical approaches could become stuck in local minima. Our work used a novel algorithm formulated for AQO to solve a special type of MINLP problem. The research focused on determining: 1) if the problem is possible to solve using AQO, 2) if it can be solved by current hardware, 3) what the currently achievable performance is, 4) what the performance will be on projected future hardware, and 5) when AQO is likely to provide a benefit over classical computing methods. Two different methods, integer range and 1-hot encoding, were investigated for transforming the MINLP problem instance constraints into a mathematical structure that can be embedded directly onto the current D-Wave architecture. For testing and validation a D-Wave 2X device was used, as well as QxBranch’s QxLib software library, which includes a QC simulator based on simulated annealing. Our results indicate that it is mathematically possible to formulate the MINLP problem for AQO, but that currently available hardware is unable to solve problems of useful size. Classical general-purpose simulated annealing is currently able to solve larger problem sizes, but does not scale well and such methods would likely be outperformed in the future by improved AQO hardware with higher qubit connectivity and lower temperatures. If larger AQO devices are able to show improvements that trend in this direction, commercially viable solutions to the MINLP for particular applications could be implemented on hardware projected to be available in 5-10 years. Continued investigation into optimal AQO hardware architectures and novel methods for embedding MINLP problem constraints on to those architectures is needed to realize those commercial benefits.

Keywords: adiabatic quantum optimization, mixed integer nonlinear programming, quantum computing, NP-hard

Procedia PDF Downloads 525

Authors: B. Umaru Mohammed, Faris A. Al-Maziad, Mohammad Y. Numan

Abstract:

One of the most important factors affecting the energy performance within a building is a carefully and efficiently designed facade. The primary aim of this research was to identify and present the potentiality of utilising Double-Skin Facade (DSF) construction and critically examine its effect on the energy consumption of an office building located within a maritime desert climate as to the conventional single-skin curtain wall system. A comparative analysis of the effect on the overall energy consumption within an office building was investigated in which a combination of various Double-Skin Facade configurations, systems, and cavity depths, glazing types and orientations were utilised. A computer dynamic modelling was utilised in order to ensure accurate calculations and efficient simulations of the various DSF systems due to the complex nature of the various functions within the Facade cavity. Through the use of the dynamic thermal modelling simulations, the best cavity size glazed type and orientation were determined to lead to a detailed analysis of the efficiency of each respective combination of Double-Skin Facade construction. As such the optimal facade combination for use within an office building located in a maritime desert climate was identified. Results demonstrated that a multi-story Facade, depending on its configuration, save up to 5% on annual cooling loads respect to a Corridor Facade and while vented can save unto 12% when compared to the single skin façade, on annual cooling load in the maritime desert climate. The selected configuration of the DSF from SSF saves an overall annual cooling load of 32%.A comparative analysis of the effect on the overall energy consumption within an office building was investigated in which a combination of various Double-Skin Facade configurations, systems, and cavity depths, glazing types and orientations were utilized. A computer dynamic modelling was utilized in order to ensure accurate calculations and efficient simulations of the various DSF systems due to the complex nature of the various functions within the Facade cavity. Through the use of the dynamic thermal modelling simulations, the best cavity size glazed type and orientation were determined to lead to a detailed analysis of the efficiency of each respective combination of Double-Skin Facade construction. As such the optimal facade combination for use within an office building located in a maritime desert climate was identified. Results demonstrated that a multi-story Facade, depending on its configuration, save up to 5% on annual cooling loads respect to a Corridor Facade and while vented can save unto 12% when compared to the single skin facade, on annual cooling load in the maritime desert climate. The selected configuration of the DSF from SSF saves an overall annual cooling load of 32%.

Keywords: computer dynamics modelling, comparative analysis, energy computation, double skin facade, single skin curtain wall, maritime desert climate

Procedia PDF Downloads 342

Authors: Abderrhamane Jarou, Dominique Sauter, Christophe Aubrun

Abstract:

Over the past decade, the growing demand for energy efficiency in buildings has attracted the attention of the control community. Failures in HVAC (heating, ventilation and air conditioning) systems in buildings can have a significant impact on the desired and expected energy performance of buildings and on the user's comfort as well. FTC is a recent technology area that studies the adaptation of control algorithms to faulty operating conditions of a system. The application of Fault-Tolerant Control (FTC) in HVAC systems has gained attention in the last two decades. The objective is to maintain the variations in system performance due to faults within an acceptable range with respect to the desired nominal behavior. This paper considers the so-called active approach, which is based on fault and identification scheme combined with a control reconfiguration algorithm that consists in determining a new set of control parameters so that the reconfigured performance is "as close as possible, "in some sense, to the nominal performance. Thermal models of buildings and their HVAC systems are described by non-linear (usually bi-linear) equations. Most of the works carried out so far in FDI (fault diagnosis and isolation) or FTC consider a linearized model of the studied system. However, this model is only valid in a reduced range of variation. This study presents a new fault diagnosis (FD) algorithm based on a bilinear observer for the detection and accurate estimation of the magnitude of the HVAC system failure. The main contribution of the proposed FD algorithm is that instead of using specific linearized models, the algorithm inherits the structure of the actual bilinear model of the building thermal dynamics. As an immediate consequence, the algorithm is applicable to a wide range of unpredictable operating conditions, i.e., weather dynamics, outdoor air temperature, zone occupancy profile. A bilinear fault detection observer is proposed for a bilinear system with unknown inputs. The residual vector in the observer design is decoupled from the unknown inputs and, under certain conditions, is made sensitive to all faults. Sufficient conditions are given for the existence of the observer and results are given for the explicit computation of observer design matrices. Dedicated observer schemes (DOS) are considered for sensor FDI while unknown input bilinear observers are considered for actuator or system components FDI. The proposed strategy for FTC works as follows: At a first level, FDI algorithms are implemented, making it also possible to estimate the magnitude of the fault. Once the fault is detected, the fault estimation is then used to feed the second level and reconfigure the control low so that that expected performances are recovered. This paper is organized as follows. A general structure for fault-tolerant control of buildings is first presented and the building model under consideration is introduced. Then, the observer-based design for Fault Diagnosis of bilinear systems is studied. The FTC approach is developed in Section IV. Finally, a simulation example is given in Section V to illustrate the proposed method.

Keywords: bilinear systems, fault diagnosis, fault-tolerant control, multi-zones building

Procedia PDF Downloads 172

Authors: Jiaying Zhang, Xin Mu, Chang Ni, Jeff L. Zhang

Abstract:

Magnetic resonance elastography (MRE) non-invasively assesses tissue elastic properties, such as shear modulus, by measuring tissue’s displacement in response to mechanical waves. The estimated metrics on tissue elasticity or stiffness have been shown to be valuable for monitoring physiologic or pathophysiologic status of tissue, such as a tumor or fatty liver. To quantify tissue shear modulus from MRE-acquired displacements (essentially an inverse problem), multiple approaches have been proposed, including Local Frequency Estimation (LFE) and Direct Inversion (DI). However, one common problem with these methods is that the estimates are severely noise-sensitive due to either the inverse-problem nature or noise propagation in the pixel-by-pixel process. With the advent of deep learning (DL) and its promise in solving inverse problems, a few groups in the field of MRE have explored the feasibility of using DL methods for quantifying shear modulus from MRE data. Most of the groups chose to use real MRE data for DL model training and to cut training images into smaller patches, which enriches feature characteristics of training data but inevitably increases computation time and results in outcomes with patched patterns. In this study, simulated wave images generated by Finite Differential Time Domain (FDTD) simulation are used for network training, and U-Net is used to extract features from each training image without cutting it into patches. The use of simulated data for model training has the flexibility of customizing training datasets to match specific applications. The proposed method aimed to estimate tissue shear modulus from MRE data with high robustness to noise and high model-training efficiency. Specifically, a set of 3000 maps of shear modulus (with a range of 1 kPa to 15 kPa) containing randomly positioned objects were simulated, and their corresponding wave images were generated. The two types of data were fed into the training of a U-Net model as its output and input, respectively. For an independently simulated set of 1000 images, the performance of the proposed method against DI and LFE was compared by the relative errors (root mean square error or RMSE divided by averaged shear modulus) between the true shear modulus map and the estimated ones. The results showed that the estimated shear modulus by the proposed method achieved a relative error of 4.91%±0.66%, substantially lower than 78.20%±1.11% by LFE. Using simulated data, the proposed method significantly outperformed LFE and DI in resilience to increasing noise levels and in resolving fine changes of shear modulus. The feasibility of the proposed method was also tested on MRE data acquired from phantoms and from human calf muscles, resulting in maps of shear modulus with low noise. In future work, the method’s performance on phantom and its repeatability on human data will be tested in a more quantitative manner. In conclusion, the proposed method showed much promise in quantifying tissue shear modulus from MRE with high robustness and efficiency.

Keywords: deep learning, magnetic resonance elastography, magnetic resonance imaging, shear modulus estimation

Procedia PDF Downloads 68

Authors: Soha Saad, Jean Bigeon, Florence Ossart, Etienne Sourdille

Abstract:

Developing better solutions for train rescheduling problems has been drawing the attention of researchers for decades. Most researches in this field deal with minor incidents that affect a large number of trains due to cascading effects. They focus on timetables, rolling stock and crew duties, but do not take into account infrastructure limits. The present work addresses electric infrastructure incidents that limit the power available for train traction, and hence the transportation capacity of the railway system. Rescheduling is needed in order to optimally share the available power among the different trains. We propose a rescheduling process based on dynamic multiphysics railway simulations that include the mechanical and electrical properties of all the system components and calculate physical quantities such as the train speed profiles, voltage along the catenary lines, temperatures, etc. The optimization problem to solve has a large number of continuous and discrete variables, several output constraints due to physical limitations of the system, and a high computation cost. Our approach includes a phase of sensitivity analysis in order to analyze the behavior of the system and help the decision making process and/or more precise optimization. This approach is a quantitative method based on simulation statistics of the dynamic railway system, considering a predefined range of variation of the input parameters. Three important settings are defined. Factor prioritization detects the input variables that contribute the most to the outputs variation. Then, factor fixing allows calibrating the input variables which do not influence the outputs. Lastly, factor mapping is used to study which ranges of input values lead to model realizations that correspond to feasible solutions according to defined criteria or objectives. Generalized Sobol indexes are used for factor prioritization and factor fixing. The approach is tested in the case of a simple railway system, with a nominal traffic running on a single track line. The considered incident is the loss of a feeding power substation, which limits the power available and the train speed. Rescheduling is needed and the variables to be adjusted are the trains departure times, train speed reduction at a given position and the number of trains (cancellation of some trains if needed). The results show that the spacing between train departure times is the most critical variable, contributing to more than 50% of the variation of the model outputs. In addition, we identify the reduced range of variation of this variable which guarantees that the output constraints are respected. Optimal solutions are extracted, according to different potential objectives: minimizing the traveling time, the train delays, the traction energy, etc. Pareto front is also built.

Keywords: optimization, rescheduling, railway system, sensitivity analysis, train timetable

Procedia PDF Downloads 399

Authors: George Zhou, Yunchan Chen, Candace Chien

Abstract:

Kidney replacement therapy is the current standard of care for end-stage renal diseases. In-center or home hemodialysis remains an integral component of the therapeutic regimen. Arteriovenous fistulas (AVF) make up the vascular circuit through which blood is filtered and returned. Naturally, AVF patency determines whether adequate clearance and filtration can be achieved and directly influences clinical outcomes. Our aim was to build a deep learning model for automated AVF stenosis screening based on the sound of blood flow through the AVF. A total of 311 patients with AVF were enrolled in this study. Blood flow sounds were collected using a digital stethoscope. For each patient, blood flow sounds were collected at 6 different locations along the patient’s AVF. The 6 locations are artery, anastomosis, distal vein, middle vein, proximal vein, and venous arch. A total of 1866 sounds were collected. The blood flow sounds are labeled as “patent” (normal) or “stenotic” (abnormal). The labels are validated from concurrent ultrasound. Our dataset included 1527 “patent” and 339 “stenotic” sounds. We show that blood flow sounds vary significantly along the AVF. For example, the blood flow sound is loudest at the anastomosis site and softest at the cephalic arch. Contextualizing the sound with location metadata significantly improves classification performance. How to encode and incorporate categorical metadata is an active area of research1. Herein, we study ordinal (i.e., integer) encoding schemes. The numerical representation is concatenated to the flattened feature vector. We train a vision transformer (ViT) on spectrogram image representations of the sound and demonstrate that using scalar multiples of our integer encodings improves classification performance. Models are evaluated using a 10-fold cross-validation procedure. The baseline performance of our ViT without any location metadata achieves an AuROC and AuPRC of 0.68 ± 0.05 and 0.28 ± 0.09, respectively. Using the following encodings of Artery:0; Arch: 1; Proximal: 2; Middle: 3; Distal 4: Anastomosis: 5, the ViT achieves an AuROC and AuPRC of 0.69 ± 0.06 and 0.30 ± 0.10, respectively. Using the following encodings of Artery:0; Arch: 10; Proximal: 20; Middle: 30; Distal 40: Anastomosis: 50, the ViT achieves an AuROC and AuPRC of 0.74 ± 0.06 and 0.38 ± 0.10, respectively. Using the following encodings of Artery:0; Arch: 100; Proximal: 200; Middle: 300; Distal 400: Anastomosis: 500, the ViT achieves an AuROC and AuPRC of 0.78 ± 0.06 and 0.43 ± 0.11. respectively. Interestingly, we see that using increasing scalar multiples of our integer encoding scheme (i.e., encoding “venous arch” as 1,10,100) results in progressively improved performance. In theory, the integer values do not matter since we are optimizing the same loss function; the model can learn to increase or decrease the weights associated with location encodings and converge on the same solution. However, in the setting of limited data and computation resources, increasing the importance at initialization either leads to faster convergence or helps the model escape a local minimum.

Keywords: arteriovenous fistula, blood flow sounds, metadata encoding, deep learning

Procedia PDF Downloads 88

Authors: Pawel Martynowicz

Abstract:

Vibrations are a crucial problem for slender structures such as towers, masts, chimneys, wind turbines, bridges, high buildings, etc., that is why most of them are equipped with vibration attenuation or fatigue reduction solutions. In this work, a slender structure (i.e., wind turbine tower-nacelle model) equipped with nonlinear, semiactive tuned vibration absorber(s) is analyzed. For this study purposes, magnetorheological (MR) dampers are used as semiactive actuators. Several optimal-based approaches to structural vibration attenuation are investigated against the standard ‘ground-hook’ law and passive tuned vibration absorber(s) implementations. The common approach to optimal control of nonlinear systems is offline computation of the optimal solution, however, so determined open loop control suffers from lack of robustness to uncertainties (e.g., unmodelled dynamics, perturbations of external forces or initial conditions), and thus perturbation control techniques are often used. However, proper linearization may be an issue for highly nonlinear systems with implicit relations between state, co-state, and control. The main contribution of the author is the development as well as numerical and experimental verification of the Pontriagin maximum-principle-based vibration control concepts that produce directly actuator control input (not the demanded force), thus force tracking algorithm that results in control inaccuracy is entirely omitted. These concepts, including one-step optimal control, quasi-optimal control, and optimal-based modified ‘ground-hook’ law, can be directly implemented in online and real-time feedback control for periodic (or semi-periodic) disturbances with invariant or time-varying parameters, as well as for non-periodic, transient or random disturbances, what is a limitation for some other known solutions. No offline calculation, excitations/disturbances assumption or vibration frequency determination is necessary, moreover, all of the nonlinear actuator (MR damper) force constraints, i.e., no active forces, lower and upper saturation limits, hysteresis-type dynamics, etc., are embedded in the control technique, thus the solution is optimal or suboptimal for the assumed actuator, respecting its limitations. Depending on the selected method variant, a moderate or decisive reduction in the computational load is possible compared to other methods of nonlinear optimal control, while assuring the quality and robustness of the vibration reduction system, as well as considering multi-pronged operational aspects, such as possible minimization of the amplitude of the deflection and acceleration of the vibrating structure, its potential and/or kinetic energy, required actuator force, control input (e.g. electric current in the MR damper coil) and/or stroke amplitude. The developed solutions are characterized by high vibration reduction efficiency – the obtained maximum values of the dynamic amplification factor are close to 2.0, while for the best of the passive systems, these values exceed 3.5.

Keywords: magnetorheological damper, nonlinear tuned vibration absorber, optimal control, real-time structural vibration attenuation, wind turbines

Procedia PDF Downloads 124

Authors: Ja-Keoung Koo, Kensuke Nakamura, Hyohun Kim, Dongwha Shin, Yeonseok Kim, Ji-Su Ahn, Byung-Woo Hong

Abstract:

The machine learning techniques based on a convolutional neural network (CNN) have been actively developed and successfully applied to a variety of image analysis tasks including reconstruction, noise reduction, resolution enhancement, segmentation, motion estimation, object recognition. The classical visual information processing that ranges from low level tasks to high level ones has been widely developed in the deep learning framework. It is generally considered as a challenging problem to derive visual interpretation from high dimensional imagery data. A CNN is a class of feed-forward artificial neural network that usually consists of deep layers the connections of which are established by a series of non-linear operations. The CNN architecture is known to be shift invariant due to its shared weights and translation invariance characteristics. However, it is often computationally intractable to optimize the network in particular with a large number of convolution layers due to a large number of unknowns to be optimized with respect to the training set that is generally required to be large enough to effectively generalize the model under consideration. It is also necessary to limit the size of convolution kernels due to the computational expense despite of the recent development of effective parallel processing machinery, which leads to the use of the constantly small size of the convolution kernels throughout the deep CNN architecture. However, it is often desired to consider different scales in the analysis of visual features at different layers in the network. Thus, we propose a CNN model where different sizes of the convolution kernels are applied at each layer based on the random projection. We apply random filters with varying sizes and associate the filter responses with scalar weights that correspond to the standard deviation of the random filters. We are allowed to use large number of random filters with the cost of one scalar unknown for each filter. The computational cost in the back-propagation procedure does not increase with the larger size of the filters even though the additional computational cost is required in the computation of convolution in the feed-forward procedure. The use of random kernels with varying sizes allows to effectively analyze image features at multiple scales leading to a better generalization. The robustness and effectiveness of the proposed CNN based on random kernels are demonstrated by numerical experiments where the quantitative comparison of the well-known CNN architectures and our models that simply replace the convolution kernels with the random filters is performed. The experimental results indicate that our model achieves better performance with less number of unknown weights. The proposed algorithm has a high potential in the application of a variety of visual tasks based on the CNN framework. Acknowledgement—This work was supported by the MISP (Ministry of Science and ICT), Korea, under the National Program for Excellence in SW (20170001000011001) supervised by IITP, and NRF-2014R1A2A1A11051941, NRF2017R1A2B4006023.

Keywords: deep learning, convolutional neural network, random kernel, random projection, dimensionality reduction, object recognition

Procedia PDF Downloads 290

Authors: Aamir Shahzad, Mao-Gang He

Abstract:

Currently, dusty plasmas motivated the researchers' widespread interest. Since the last two decades, substantial efforts have been made by the scientific and technological community to investigate the transport properties and their nonlinear behavior of three-dimensional and two-dimensional nonideal complex (dusty plasma) liquids (NICDPLs). Different calculations have been made to sustain and utilize strongly coupled NICDPLs because of their remarkable scientific and industrial applications. Understanding of the thermophysical properties of complex liquids under various conditions is of practical interest in the field of science and technology. The determination of thermal conductivity is also a demanding question for thermophysical researchers, due to some reasons; very few results are offered for this significant property. Lack of information of the thermal conductivity of dense and complex liquids at different parameters related to the industrial developments is a major barrier to quantitative knowledge of the heat flux flow from one medium to another medium or surface. The exact numerical investigation of transport properties of complex liquids is a fundamental research task in the field of thermophysics, as various transport data are closely related with the setup and confirmation of equations of state. A reliable knowledge of transport data is also important for an optimized design of processes and apparatus in various engineering and science fields (thermoelectric devices), and, in particular, the provision of precise data for the parameters of heat, mass, and momentum transport is required. One of the promising computational techniques, the homogenous nonequilibrium molecular dynamics (HNEMD) simulation, is over viewed with a special importance on the application to transport problems of complex liquids. This proposed work is particularly motivated by the FIRST TIME to modify the problem of heat conduction equations leads to polynomial velocity and temperature profiles algorithm for the investigation of transport properties with their nonlinear behaviors in the NICDPLs. The aim of proposed work is to implement a NEMDS algorithm (Poiseuille flow) and to delve the understanding of thermal conductivity behaviors in Yukawa liquids. The Yukawa system is equilibrated through the Gaussian thermostat in order to maintain the constant system temperature (canonical ensemble ≡ NVT)). The output steps will be developed between 3.0×105/ωp and 1.5×105/ωp simulation time steps for the computation of λ data. The HNEMD algorithm shows that the thermal conductivity is dependent on plasma parameters and the minimum value of lmin shifts toward higher G with an increase in k, as expected. New investigations give more reliable simulated data for the plasma conductivity than earlier known simulation data and generally the plasma λ0 by 2%-20%, depending on Γ and κ. It has been shown that the obtained results at normalized force field are in satisfactory agreement with various earlier simulation results. This algorithm shows that the new technique provides more accurate results with fast convergence and small size effects over a wide range of plasma states.

Keywords: molecular dynamics simulation, thermal conductivity, nonideal complex plasma, Poiseuille flow

Procedia PDF Downloads 274

Authors: Mayank Gupta, Siba Prasad Samal, Vasu Kakkirala

Abstract:

The internet world and its priorities have changed considerably in the last decade. Browsing on smart phones has increased manifold and is set to explode much more. Users spent considerable time browsing different websites, that gives a great deal of insight into user’s preferences. Instead of plain information classifying different aspects of browsing like Bookmarks, History, and Download Manager into useful categories would improve and enhance the user’s experience. Most of the classification solutions are server side that involves maintaining server and other heavy resources. It has security constraints and maybe misses on contextual data during classification. On device, classification solves many such problems, but the challenge is to achieve accuracy on classification with resource constraints. This on device classification can be much more useful in personalization, reducing dependency on cloud connectivity and better privacy/security. This approach provides more relevant results as compared to current standalone solutions because it uses content rendered by browser which is customized by the content provider based on user’s profile. This paper proposes a Naive Bayes based lightweight classification engine targeted for a resource constraint devices. Our solution integrates with Web Browser that in turn triggers classification algorithm. Whenever a user browses a webpage, this solution extracts DOM Tree data from the browser’s rendering engine. This DOM data is a dynamic, contextual and secure data that can’t be replicated. This proposal extracts different features of the webpage that runs on an algorithm to classify into multiple categories. Naive Bayes based engine is chosen in this solution for its inherent advantages in using limited resources compared to other classification algorithms like Support Vector Machine, Neural Networks, etc. Naive Bayes classification requires small memory footprint and less computation suitable for smartphone environment. This solution has a feature to partition the model into multiple chunks that in turn will facilitate less usage of memory instead of loading a complete model. Classification of the webpages done through integrated engine is faster, more relevant and energy efficient than other standalone on device solution. This classification engine has been tested on Samsung Z3 Tizen hardware. The Engine is integrated into Tizen Browser that uses Chromium Rendering Engine. For this solution, extensive dataset is sourced from dmoztools.net and cleaned. This cleaned dataset has 227.5K webpages which are divided into 8 generic categories ('education', 'games', 'health', 'entertainment', 'news', 'shopping', 'sports', 'travel'). Our browser integrated solution has resulted in 15% less memory usage (due to partition method) and 24% less power consumption in comparison with standalone solution. This solution considered 70% of the dataset for training the data model and the rest 30% dataset for testing. An average accuracy of ~96.3% is achieved across the above mentioned 8 categories. This engine can be further extended for suggesting Dynamic tags and using the classification for differential uses cases to enhance browsing experience.

Keywords: chromium, lightweight engine, mobile computing, Naive Bayes, Tizen, web browser, webpage classification

Procedia PDF Downloads 163

Authors: Jara Kaye S. Villanueva, M. Rosario Concepcion O. Ang

Abstract:

For the past few years, Philippines has depended most of its energy source on oil, coal, and fossil fuel. According to the Department of Energy (DOE), the dominance of coal in the energy mix will continue until the year 2020. The expanding energy needs in the country have led to increasing efforts to promote and develop renewable energy. This research is a part of the government initiative in preparation for renewable energy development and expansion in the country. The Philippine Renewable Energy Resource Mapping from Light Detection and Ranging (LiDAR) Surveys is a three-year government project which aims to assess and quantify the renewable energy potential of the country and to put them into usable maps. This study focuses on the site suitability analysis of the four renewable energy sources – biomass (coconut, corn, rice, and sugarcane), hydro, solar, and wind energy. The site assessment is a key component in determining and assessing the most suitable locations for the construction of renewable energy power plants. This method maximizes the use of both the technical methods in resource assessment, as well as taking into account the environmental, social, and accessibility aspect in identifying potential sites by utilizing and integrating two different methods: the Multi-Criteria Decision Analysis (MCDA) method and Geographic Information System (GIS) tools. For the MCDA, Analytical Hierarchy Processing (AHP) is employed to determine the parameters needed for the suitability analysis. To structure these site suitability parameters, various experts from different fields were consulted – scientists, policy makers, environmentalists, and industrialists. The need to have a well-represented group of people to consult with is relevant to avoid bias in the output parameter of hierarchy levels and weight matrices. AHP pairwise matrix computation is utilized to derive weights per level out of the expert’s gathered feedback. Whereas from the threshold values derived from related literature, international studies, and government laws, the output values were then consulted with energy specialists from the DOE. Geospatial analysis using GIS tools translate this decision support outputs into visual maps. Particularly, this study uses Euclidean distance to compute for the distance values of each parameter, Fuzzy Membership algorithm which normalizes the output from the Euclidean Distance, and the Weighted Overlay tool for the aggregation of the layers. Using the Natural Breaks algorithm, the suitability ratings of each of the map are classified into 5 discrete categories of suitability index: (1) not suitable (2) least suitable, (3) suitable, (4) moderately suitable, and (5) highly suitable. In this method, the classes are grouped based on the best groups similar values wherein each subdivision are set from the rest based on the big difference in boundary values. Results show that in the entire Philippine area of responsibility, biomass has the highest suitability rating with rice as the most suitable at 75.76% suitability percentage, whereas wind has the least suitability percentage with score 10.28%. Solar and Hydro fall in the middle of the two, with suitability values 28.77% and 21.27%.

Keywords: site suitability, biomass energy, hydro energy, solar energy, wind energy, GIS

Procedia PDF Downloads 149

Authors: Óscar Muñoz Garrigós, José Manuel Bernabeu Aubán

Abstract:

Time-sensitive services are the base of the cloud services industry. Keeping low service saturation is essential for controlling response time. All auto-scalable services make use of reactive auto-scaling. However, reactive auto-scaling has few in-depth studies. This presentation shows a model for reactive auto-scaling methodologies with a MAPE-k architecture. Queuing theory can compute different properties of static services but lacks some parameters related to the transition between models. Our model uses queuing theory parameters to relate the transition between models. It associates MAPE-k related times, the sampling frequency, the cooldown period, the number of requests that an instance can handle per unit of time, the number of incoming requests at a time instant, and a function that describes the acceleration in the service's ability to handle more requests. This model is later used as a solution to horizontally auto-scale time-sensitive services composed of microservices, reevaluating the model’s parameters periodically to allocate resources. The solution requires limiting the acceleration of the growth in the number of incoming requests to keep a constrained response time. Business benefits determine such limits. The solution can add a dynamic number of instances and remains valid under different system sizes. The study includes performance recommendations to improve results according to the incoming load shape and business benefits. The exposed methodology is tested in a simulation. The simulator contains a load generator and a service composed of two microservices, where the frontend microservice depends on a backend microservice with a 1:1 request relation ratio. A common request takes 2.3 seconds to be computed by the service and is discarded if it takes more than 7 seconds. Both microservices contain a load balancer that assigns requests to the less loaded instance and preemptively discards requests if they are not finished in time to prevent resource saturation. When load decreases, instances with lower load are kept in the backlog where no more requests are assigned. If the load grows and an instance in the backlog is required, it returns to the running state, but if it finishes the computation of all requests and is no longer required, it is permanently deallocated. A few load patterns are required to represent the worst-case scenario for reactive systems: the following scenarios test response times, resource consumption and business costs. The first scenario is a burst-load scenario. All methodologies will discard requests if the rapidness of the burst is high enough. This scenario focuses on the number of discarded requests and the variance of the response time. The second scenario contains sudden load drops followed by bursts to observe how the methodology behaves when releasing resources that are lately required. The third scenario contains diverse growth accelerations in the number of incoming requests to observe how approaches that add a different number of instances can handle the load with less business cost. The exposed methodology is compared against a multiple threshold CPU methodology allocating/deallocating 10 or 20 instances, outperforming the competitor in all studied metrics.

Keywords: reactive auto-scaling, auto-scaling, microservices, cloud computing

Procedia PDF Downloads 93

Authors: Dario Milani, Guido Morgenthal

Abstract:

Fluid dynamic computation of wind caused forces on bluff bodies e.g light flexible civil structures or high incidence of ground approaching airplane wings, is one of the major criteria governing their design. For such structures a significant dynamic response may result, requiring the usage of small scale devices as guide-vanes in bridge design to control these effects. The focus of this paper is on the numerical simulation of the bluff body problem involving multiscale phenomena induced by small scale devices. One of the solution methods for the CFD simulation that is relatively successful in this class of applications is the Vortex Particle Method (VPM). The method is based on a grid free Lagrangian formulation of the Navier-Stokes equations, where the velocity field is modeled by particles representing local vorticity. These vortices are being convected due to the free stream velocity as well as diffused. This representation yields the main advantages of low numerical diffusion, compact discretization as the vorticity is strongly localized, implicitly accounting for the free-space boundary conditions typical for this class of FSI problems, and a natural representation of the vortex creation process inherent in bluff body flows. When the particle resolution reaches the Kolmogorov dissipation length, the method becomes a Direct Numerical Simulation (DNS). However, it is crucial to note that any solution method aims at balancing the computational cost against the accuracy achievable. In the classical VPM method, if the fluid domain is discretized by Np particles, the computational cost is O(Np2). For the coupled FSI problem of interest, for example large structures such as long-span bridges, the aerodynamic behavior may be influenced or even dominated by small structural details such as barriers, handrails or fairings. For such geometrically complex and dimensionally large structures, resolving the complete domain with the conventional VPM particle discretization might become prohibitively expensive to compute even for moderate numbers of particles. It is possible to reduce this cost either by reducing the number of particles or by controlling its local distribution. It is also possible to increase the accuracy of the solution without increasing substantially the global computational cost by computing a correction of the particle-particle interaction in some regions of interest. In this paper different strategies are presented in order to extend the conventional VPM method to reduce the computational cost whilst resolving the required details of the flow. The methods include temporal sub stepping to increase the accuracy of the particles convection in certain regions as well as dynamically re-discretizing the particle map to locally control the global and the local amount of particles. Finally, these methods will be applied on a test case and the improvements in the efficiency as well as the accuracy of the proposed extension to the method are presented. The important benefits in terms of accuracy and computational cost of the combination of these methods will be thus presented as long as their relevant applications.

Keywords: adaptation, fluid dynamic, remeshing, substepping, vortex particle method

Procedia PDF Downloads 262

Authors: Hiroshi Kinoshita, Yasuto Nakanishi, Ryuhei Okuno, Toshio Higashi

Abstract:

Various kinds of pegboards have been developed and used widely in research and clinics of rehabilitation for evaluation and training of patient’s hand function. A common measure in these peg boards is a total time of performance execution assessed by a tester’s stopwatch. Introduction of electrical and automatic measurement technology to the apparatus, on the other hand, has been delayed. The present work introduces the development of a pegboard with an electric sensor to detect moments of individual peg’s insertion and removal. The work also gives fundamental data obtained from a group of healthy young individuals who performed peg transfer tasks using the pegboard developed. Through trails and errors in pilot tests, two 10-hole peg-board boxes installed with a small photo-reflector and a DC amplifier at the bottom of each hole were designed and built by the present authors. The amplified electric analogue signals from the 20 reflectors were automatically digitized at 500 Hz per channel, and stored in a PC. The boxes were set on a test table at different distances (25, 50, 75, and 125 mm) in parallel to examine the effect of hole-to-hole distance. Fifty healthy young volunteers (25 in each gender) as subjects of the study performed successive fast 80 time peg transfers at each distance using their dominant and non-dominant hands. The data gathered showed a clear-cut light interruption/continuation moment by the pegs, allowing accurately (no tester’s error involved) and precisely (an order of milliseconds) to determine the pull out and insertion times of each peg. This further permitted computation of individual peg movement duration (PMD: from peg-lift-off to insertion) apart from hand reaching duration (HRD: from peg insertion to lift-off). An accidental drop of a peg led to an exceptionally long ( < mean + 3 SD) PMD, which was readily detected from an examination of data distribution. The PMD data were commonly right-skewed, suggesting that the median can be a better estimate of individual PMD than the mean. Repeated measures ANOVA using the median values revealed significant hole-to-hole distance, and hand dominance effects, suggesting that these need to be fixed in the accurate evaluation of PMD. The gender effect was non-significant. Performance consistency was also evaluated by the use of quartile variation coefficient values, which revealed no gender, hole-to-hole, and hand dominance effects. The measurement reliability was further examined using interclass correlation obtained from 14 subjects who performed the 25 and 125 mm hole distance tasks at two 7-10 days separate test sessions. Inter-class correlation values between the two tests showed fair reliability for PMD (0.65-0.75), and for HRD (0.77-0.94). We concluded that a sensor peg board developed in the present study could provide accurate (excluding tester’s errors), and precise (at a millisecond rate) time information of peg movement separated from that used for hand movement. It could also easily detect and automatically exclude erroneous execution data from his/her standard data. These would lead to a better evaluation of hand dexterity function compared to the widely used conventional used peg boards.

Keywords: hand, dexterity test, peg movement time, performance consistency

Procedia PDF Downloads 133

Authors: Mingxin Li, Liya Ni

Abstract:

Autonomous parking is a valuable feature applicable to many robotics applications such as tour guide robots, UV sanitizing robots, food delivery robots, and warehouse robots. With auto-parking, the robot will be able to park at the charging zone and charge itself without human intervention. As compared to self-driving vehicles, auto-parking is more challenging for a small-scale mobile robot only equipped with a front camera due to the camera view limited by the robot’s height and the narrow Field of View (FOV) of the inexpensive camera. In this research, auto-parking of a small-scale mobile robot with a front camera only was achieved in a four-step process: Firstly, transfer learning was performed on the AlexNet, a popular pre-trained convolutional neural network (CNN). It was trained with 150 pictures of empty parking slots and 150 pictures of occupied parking slots from the view angle of a small-scale robot. The dataset of images was divided into a group of 70% images for training and the remaining 30% images for validation. An average success rate of 95% was achieved. Secondly, the image of detected empty parking space was processed with edge detection followed by the computation of parametric representations of the boundary lines using the Hough Transform algorithm. Thirdly, the positions of the entrance point and center of available parking space were predicted based on the robot kinematic model as the robot was driving closer to the parking space because the boundary lines disappeared partially or completely from its camera view due to the height and FOV limitations. The robot used its wheel speeds to compute the positions of the parking space with respect to its changing local frame as it moved along, based on its kinematic model. Lastly, the predicted entrance point of the parking space was used as the reference for the motion control of the robot until it was replaced by the actual center when it became visible again by the robot. The linear and angular velocities of the robot chassis center were computed based on the error between the current chassis center and the reference point. Then the left and right wheel speeds were obtained using inverse kinematics and sent to the motor driver. The above-mentioned four subtasks were all successfully accomplished, with the transformed learning, image processing, and target prediction performed in MATLAB, while the motion control and image capture conducted on a self-built small scale differential drive mobile robot. The small-scale robot employs a Raspberry Pi board, a Pi camera, an L298N dual H-bridge motor driver, a USB power module, a power bank, four wheels, and a chassis. Future research includes three areas: the integration of all four subsystems into one hardware/software platform with the upgrade to an Nvidia Jetson Nano board that provides superior performance for deep learning and image processing; more testing and validation on the identification of available parking space and its boundary lines; improvement of performance after the hardware/software integration is completed.

Keywords: autonomous parking, convolutional neural network, image processing, kinematics-based prediction, transfer learning

Procedia PDF Downloads 132

Authors: Adriana Haulica

Abstract:

Powered by Machine Learning, Precise medicine is tailored by now to use genetic and molecular profiling, with the aim of optimizing the therapeutic benefits for cohorts of patients. As the majority of Machine Language algorithms come from heuristics, the outputs have contextual validity. This is not very restrictive in the sense that medicine itself is not an exact science. Meanwhile, the progress made in Molecular Biology, Bioinformatics, Computational Biology, and Precise Medicine, correlated with the huge amount of human biology data and the increase in computational power, opens new healthcare challenges. A more accurate diagnosis is needed along with real-time treatments by processing as much as possible from the available information. The purpose of this paper is to present a deeper vision for the future of Artificial Intelligence in Precise medicine. In fact, actual Machine Learning algorithms use standard mathematical knowledge, mostly Euclidian metrics and standard computation rules. The loss of information arising from the classical methods prevents obtaining 100% evidence on the diagnosis process. To overcome these problems, we introduce MEDICOMPILLS, a new architectural concept tool of information processing in Precise medicine that delivers diagnosis and therapy advice. This tool processes poly-field digital resources: global knowledge related to biomedicine in a direct or indirect manner but also technical databases, Natural Language Processing algorithms, and strong class optimization functions. As the name suggests, the heart of this tool is a compiler. The approach is completely new, tailored for omics and clinical data. Firstly, the intrinsic biological intuition is different from the well-known “a needle in a haystack” approach usually used when Machine Learning algorithms have to process differential genomic or molecular data to find biomarkers. Also, even if the input is seized from various types of data, the working engine inside the MEDICOMPILLS does not search for patterns as an integrative tool. This approach deciphers the biological meaning of input data up to the metabolic and physiologic mechanisms, based on a compiler with grammars issued from bio-algebra-inspired mathematics. It translates input data into bio-semantic units with the help of contextual information iteratively until Bio-Logical operations can be performed on the base of the “common denominator “rule. The rigorousness of MEDICOMPILLS comes from the structure of the contextual information on functions, built to be analogous to mathematical “proofs”. The major impact of this architecture is expressed by the high accuracy of the diagnosis. Detected as a multiple conditions diagnostic, constituted by some main diseases along with unhealthy biological states, this format is highly suitable for therapy proposal and disease prevention. The use of MEDICOMPILLS architecture is highly beneficial for the healthcare industry. The expectation is to generate a strategic trend in Precise medicine, making medicine more like an exact science and reducing the considerable risk of errors in diagnostics and therapies. The tool can be used by pharmaceutical laboratories for the discovery of new cures. It will also contribute to better design of clinical trials and speed them up.

Keywords: bio-semantic units, multiple conditions diagnosis, NLP, omics

Procedia PDF Downloads 70

Authors: Zachary Blanks, Solomon Sonya

Abstract:

Poaching presents a serious threat to endangered animal species, environment conservations, and human life. Additionally, some poaching activity has even been linked to supplying funds to support terrorist networks elsewhere around the world. Consequently, agencies dedicated to protecting wildlife habitats have a near intractable task of adequately patrolling an entire area (spanning several thousand kilometers) given limited resources, funds, and personnel at their disposal. Thus, agencies need predictive tools that are both high-performing and easily implementable by the user to help in learning how the significant features (e.g. animal population densities, topography, behavior patterns of the criminals within the area, etc) interact with each other in hopes of abating poaching. This research develops a classification model using machine learning algorithms to aid in forecasting future attacks that is both easy to train and performs well when compared to other models. In this research, we demonstrate how data imputation methods (specifically predictive mean matching, gradient boosting, and random forest multiple imputation) can be applied to analyze data and create significant predictions across a varied data set. Specifically, we apply these methods to improve the accuracy of adopted prediction models (Logistic Regression, Support Vector Machine, etc). Finally, we assess the performance of the model and the accuracy of our data imputation methods by learning on a real-world data set constituting four years of imputed data and testing on one year of non-imputed data. This paper provides three main contributions. First, we extend work done by the Teamcore and CREATE (Center for Risk and Economic Analysis of Terrorism Events) research group at the University of Southern California (USC) working in conjunction with the Department of Homeland Security to apply game theory and machine learning algorithms to develop more efficient ways of reducing poaching. This research introduces ensemble methods (Random Forests and Stochastic Gradient Boosting) and applies it to real-world poaching data gathered from the Ugandan rain forest park rangers. Next, we consider the effect of data imputation on both the performance of various algorithms and the general accuracy of the method itself when applied to a dependent variable where a large number of observations are missing. Third, we provide an alternate approach to predict the probability of observing poaching both by season and by month. The results from this research are very promising. We conclude that by using Stochastic Gradient Boosting to predict observations for non-commercial poaching by season, we are able to produce statistically equivalent results while being orders of magnitude faster in computation time and complexity. Additionally, when predicting potential poaching incidents by individual month vice entire seasons, boosting techniques produce a mean area under the curve increase of approximately 3% relative to previous prediction schedules by entire seasons.

Keywords: ensemble methods, imputation, machine learning, random forests, statistical analysis, stochastic gradient boosting, wildlife protection

Procedia PDF Downloads 292

Authors: Despoina Chochtoula, Aristidis Ilias, Yannis Stamatiou

Abstract:

Modbus is a protocol that enables the communication among devices which are connected to the same network. This protocol is, often, deployed in connecting sensor and monitoring units to central supervisory servers in Supervisory Control and Data Acquisition, or SCADA, systems. These systems monitor critical infrastructures, such as factories, power generation stations, nuclear power reactors etc. in order to detect malfunctions and ignite alerts and corrective actions. However, due to their criticality, SCADA systems are vulnerable to attacks that range from simple eavesdropping on operation parameters, exchanged messages, and valuable infrastructure information to malicious modification of vital infrastructure data towards infliction of damage. Thus, the SCADA research community has been active over strengthening SCADA systems with suitable data protection mechanisms based, to a large extend, on cryptographic methods for data encryption, device authentication, and message integrity protection. However, due to the limited computation power of many SCADA sensor and embedded devices, the usual public key cryptographic methods are not appropriate due to their high computational requirements. As an alternative, Elliptic Curve Cryptography has been proposed, which requires smaller key sizes and, thus, less demanding cryptographic operations. Until now, however, no such implementation has been proposed in the SCADA literature, to the best of our knowledge. In order to fill this gap, our methodology was focused on integrating Modbus, a frequently used SCADA communication protocol, with Elliptic Curve based cryptography and develop a server/client application to demonstrate the proof of concept. For the implementation we deployed two C language libraries, which were suitably modify in order to be successfully integrated: libmodbus (https://github.com/stephane/libmodbus) and ecc-lib https://www.ceid.upatras.gr/webpages/faculty/zaro/software/ecc-lib/). The first library provides a C implementation of the Modbus/TCP protocol while the second one offers the functionality to develop cryptographic protocols based on Elliptic Curve Cryptography. These two libraries were combined, after suitable modifications and enhancements, in order to give a modified version of the Modbus/TCP protocol focusing on the security of the data exchanged among the devices and the supervisory servers. The mechanisms we implemented include key generation, key exchange/sharing, message authentication, data integrity check, and encryption/decryption of data. The key generation and key exchange protocols were implemented with the use of Elliptic Curve Cryptography primitives. The keys established by each device are saved in their local memory and are retained during the whole communication session and are used in encrypting and decrypting exchanged messages as well as certifying entities and the integrity of the messages. Finally, the modified library was compiled for the Android environment in order to run the server application as an Android app. The client program runs on a regular computer. The communication between these two entities is an example of the successful establishment of an Elliptic Curve Cryptography based, secure Modbus wireless communication session between a portable device acting as a supervisor station and a monitoring computer. Our first performance measurements are, also, very promising and demonstrate the feasibility of embedding Elliptic Curve Cryptography into SCADA systems, filling in a gap in the relevant scientific literature.

Keywords: elliptic curve cryptography, ICT security, modbus protocol, SCADA, TCP/IP protocol

Procedia PDF Downloads 271

Authors: Adrian W. Egger, Savvas P. Triantafyllou, Eleni N. Chatzi

Abstract:

The scaled boundary finite element method (SBFEM) is a semi-analytical numerical method, which introduces a scaling center in each element’s domain, thus transitioning from a Cartesian reference frame to one resembling polar coordinates. Consequently, an analytical solution is achieved in radial direction, implying that only the boundary need be discretized. The only limitation imposed on the resulting polygonal elements is that they remain star-convex. Further arbitrary p- or h-refinement may be applied locally in a mesh. The polygonal nature of SBFEM elements has been exploited in quadtree meshes to alleviate all issues conventionally associated with hanging nodes. Furthermore, since in 2D this results in only 16 possible cell configurations, these are precomputed in order to accelerate the forward analysis significantly. Any cells, which are clipped to accommodate the domain geometry, must be computed conventionally. However, since SBFEM permits polygonal elements, significantly coarser meshes at comparable accuracy levels are obtained when compared with conventional quadtree analysis, further increasing the computational efficiency of this scheme. The generalized stress intensity factors (gSIFs) are computed by exploiting the semi-analytical solution in radial direction. This is initiated by placing the scaling center of the element containing the crack at the crack tip. Taking an analytical limit of this element’s stress field as it approaches the crack tip, delivers an expression for the singular stress field. By applying the problem specific boundary conditions, the geometry correction factor is obtained, and the gSIFs are then evaluated based on their formal definition. Since the SBFEM solution is constructed as a power series, not unlike mode superposition in FEM, the two modes contributing to the singular response of the element can be easily identified in post-processing. Compared to the extended finite element method (XFEM) this approach is highly convenient, since neither enrichment terms nor a priori knowledge of the singularity is required. Computation of the gSIFs by SBFEM permits exceptional accuracy, however, when combined with hybrid quadtrees employing linear elements, this does not always hold. Nevertheless, it has been shown that crack propagation schemes are highly effective even given very coarse discretization since they only rely on the ratio of mode one to mode two gSIFs. The absolute values of the gSIFs may still be subject to large errors. Hence, we propose a post-processing scheme, which minimizes the error resulting from the approximation space of the cracked element, thus limiting the error in the gSIFs to the discretization error of the quadtree mesh. This is achieved by h- and/or p-refinement of the cracked element, which elevates the amount of modes present in the solution. The resulting numerical description of the element is highly accurate, with the main error source now stemming from its boundary displacement solution. Numerical examples show that this post-processing procedure can significantly improve the accuracy of the computed gSIFs with negligible computational cost even on coarse meshes resulting from hybrid quadtrees.

Keywords: linear elastic fracture mechanics, generalized stress intensity factors, scaled finite element method, hybrid quadtrees

Procedia PDF Downloads 146

Authors: Rita Sassine, Yassine Hassani, Mohamad Al Omari, Stéphanie Guibert

Abstract:

Building Information Modelling (BIM) is the process of creating, managing, and centralizing information during the building lifecycle. BIM can be used all over a construction project, from the initiation phase to the planning and execution phases to the maintenance and lifecycle management phase. For existing buildings, BIM can be used for specific applications such as lifecycle management. However, most of the existing buildings don’t have a BIM model. Creating a compatible BIM for existing buildings is very challenging. It requires special equipment for data capturing and efforts to convert these data into a BIM model. The main difficulties for such projects are to define the data needed, the level of development (LOD), and the methodology to be adopted. In addition to managing information for an existing building, studying the impact of the built environment is a challenging topic. So, integrating the existing terrain that surrounds buildings into the digital model is essential to be able to make several simulations as flood simulation, energy simulation, etc. Making a replication of the physical model and updating its information in real-time to make its Digital Twin (DT) is very important. The Digital Terrain Model (DTM) represents the ground surface of the terrain by a set of discrete points with unique height values over 2D points based on reference surface (e.g., mean sea level, geoid, and ellipsoid). In addition, information related to the type of pavement materials, types of vegetation and heights and damaged surfaces can be integrated. Our aim in this study is to define the methodology to be used in order to provide a 3D BIM model for the site and the existing building based on the case study of “Ecole Spéciale des Travaux Publiques (ESTP Paris)” school of engineering campus. The property is located on a hilly site of 5 hectares and is composed of more than 20 buildings with a total area of 32 000 square meters and a height between 50 and 68 meters. In this work, the campus precise levelling grid according to the NGF-IGN69 altimetric system and the grid control points are computed according to (Réseau Gédésique Français) RGF93 – Lambert 93 french system with different methods: (i) Land topographic surveying methods using robotic total station, (ii) GNSS (Global Network Satellite sytem) levelling grid with NRTK (Network Real Time Kinematic) mode, (iii) Point clouds generated by laser scanning. These technologies allow the computation of multiple building parameters such as boundary limits, the number of floors, the floors georeferencing, the georeferencing of the 4 base corners of each building, etc. Once the entry data are identified, the digital model of each building is done. The DTM is also modeled. The process of altimetric determination is complex and requires efforts in order to collect and analyze multiple data formats. Since many technologies can be used to produce digital models, different file formats such as DraWinG (DWG), LASer (LAS), Comma-separated values (CSV), Industry Foundation Classes (IFC) and ReViT (RVT) will be generated. Checking the interoperability between BIM models is very important. In this work, all models are linked together and shared on 3DEXPERIENCE collaborative platform.

Keywords: building information modeling, digital terrain model, existing buildings, interoperability

Procedia PDF Downloads 112

Authors: Norma Binti Alias, Che Rahim Che The, Norfarizan Mohd Said, Sakinah Abdul Hanan, Akhtar Ali

Abstract:

This paper discusses on the transportation of magnetic drug targeting through blood within vessels, tissues and cells. There are three integrated mathematical models to be discussed and analyze the concentration of drug and blood flow through magnetic nanoparticles. The cell therapy brought advancement in the field of nanotechnology to fight against the tumors. The systematic therapeutic effect of Single Cells can reduce the growth of cancer tissue. The process of this nanoscale phenomena system is able to measure and to model, by identifying some parameters and applying fundamental principles of mathematical modeling and simulation. The mathematical modeling of single cell growth depends on three types of cell densities such as proliferative, quiescent and necrotic cells. The aim of this paper is to enhance the simulation of three types of models. The first model represents the transport of drugs by coupled partial differential equations (PDEs) with 3D parabolic type in a cylindrical coordinate system. This model is integrated by Non-Newtonian flow equations, leading to blood liquid flow as the medium for transportation system and the magnetic force on the magnetic nanoparticles. The interaction between the magnetic force on drug with magnetic properties produces induced currents and the applied magnetic field yields forces with tend to move slowly the movement of blood and bring the drug to the cancer cells. The devices of nanoscale allow the drug to discharge the blood vessels and even spread out through the tissue and access to the cancer cells. The second model is the transport of drug nanoparticles from the vascular system to a single cell. The treatment of the vascular system encounters some parameter identification such as magnetic nanoparticle targeted delivery, blood flow, momentum transport, density and viscosity for drug and blood medium, intensity of magnetic fields and the radius of the capillary. Based on two discretization techniques, finite difference method (FDM) and finite element method (FEM), the set of integrated models are transformed into a series of grid points to get a large system of equations. The third model is a single cell density model involving the three sets of first order PDEs equations for proliferating, quiescent and necrotic cells change over time and space in Cartesian coordinate which regulates under different rates of nutrients consumptions. The model presents the proliferative and quiescent cell growth depends on some parameter changes and the necrotic cells emerged as the tumor core. Some numerical schemes for solving the system of equations are compared and analyzed. Simulation and computation of the discretized model are supported by Matlab and C programming languages on a single processing unit. Some numerical results and analysis of the algorithms are presented in terms of informative presentation of tables, multiple graph and multidimensional visualization. As a conclusion, the integrated of three types mathematical modeling and the comparison of numerical performance indicates that the superior tool and analysis for solving the complete set of magnetic drug delivery system which give significant effects on the growth of the targeted cancer cell.

Keywords: mathematical modeling, visualization, PDE models, magnetic nanoparticle drug delivery model, drug release model, single cell effects, avascular tumor growth, numerical analysis

Procedia PDF Downloads 428

Authors: Stefan K. Behfar

Abstract:

The rapid growth of the Ethereum network has brought forth the urgent need for scalable analysis methods to handle the increasing volume of blockchain data. In this research, we propose efficient methodologies for making Ethereum network analysis scalable. Our approach leverages a combination of graph-based data representation, probabilistic sampling, and parallel processing techniques to achieve unprecedented scalability while preserving critical network insights. Data Representation: We develop a graph-based data representation that captures the underlying structure of the Ethereum network. Each block transaction is represented as a node in the graph, while the edges signify temporal relationships. This representation ensures efficient querying and traversal of the blockchain data. Probabilistic Sampling: To cope with the vastness of the Ethereum blockchain, we introduce a probabilistic sampling technique. This method strategically selects a representative subset of transactions and blocks, allowing for concise yet statistically significant analysis. The sampling approach maintains the integrity of the network properties while significantly reducing the computational burden. Graph Convolutional Networks (GCNs): We incorporate GCNs to process the graph-based data representation efficiently. The GCN architecture enables the extraction of complex spatial and temporal patterns from the sampled data. This combination of graph representation and GCNs facilitates parallel processing and scalable analysis. Distributed Computing: To further enhance scalability, we adopt distributed computing frameworks such as Apache Hadoop and Apache Spark. By distributing computation across multiple nodes, we achieve a significant reduction in processing time and enhanced memory utilization. Our methodology harnesses the power of parallelism, making it well-suited for large-scale Ethereum network analysis. Evaluation and Results: We extensively evaluate our methodology on real-world Ethereum datasets covering diverse time periods and transaction volumes. The results demonstrate its superior scalability, outperforming traditional analysis methods. Our approach successfully handles the ever-growing Ethereum data, empowering researchers and developers with actionable insights from the blockchain. Case Studies: We apply our methodology to real-world Ethereum use cases, including detecting transaction patterns, analyzing smart contract interactions, and predicting network congestion. The results showcase the accuracy and efficiency of our approach, emphasizing its practical applicability in real-world scenarios. Security and Robustness: To ensure the reliability of our methodology, we conduct thorough security and robustness evaluations. Our approach demonstrates high resilience against adversarial attacks and perturbations, reaffirming its suitability for security-critical blockchain applications. Conclusion: By integrating graph-based data representation, GCNs, probabilistic sampling, and distributed computing, we achieve network scalability without compromising analytical precision. This approach addresses the pressing challenges posed by the expanding Ethereum network, opening new avenues for research and enabling real-time insights into decentralized ecosystems. Our work contributes to the development of scalable blockchain analytics, laying the foundation for sustainable growth and advancement in the domain of blockchain research and application.

Keywords: Ethereum, scalable network, GCN, probabilistic sampling, distributed computing

Procedia PDF Downloads 76

Authors: Manuel I. Capel, Antonio Tomeu, Alberto Salguero

Abstract:

Tumor growth from a transformed cancer-cell up to a clinically apparent mass spans through a range of spatial and temporal magnitudes. Through computer simulations, Cellular Automata (CA) can accurately describe the complexity of the development of tumors. Tumor development prognosis can now be made -without making patients undergo through annoying medical examinations or painful invasive procedures- if we develop appropriate CA-based software tools. In silico testing mainly refers to Computational Biology research studies of application to clinical actions in Medicine. To establish sound computer-based models of cellular behavior, certainly reduces costs and saves precious time with respect to carrying out experiments in vitro at labs or in vivo with living cells and organisms. These aim to produce scientifically relevant results compared to traditional in vitro testing, which is slow, expensive, and does not generally have acceptable reproducibility under the same conditions. For speeding up computer simulations of cellular models, specific literature shows recent proposals based on the CA approach that include advanced techniques, such the clever use of supporting efficient data structures when modeling with deterministic stochastic cellular automata. Multiparadigm and multiscale simulation of tumor dynamics is just beginning to be developed by the concerned research community. The use of stochastic cellular automata (SCA), whose parallel programming implementations are open to yield a high computational performance, are of much interest to be explored up to their computational limits. There have been some approaches based on optimizations to advance in multiparadigm models of tumor growth, which mainly pursuit to improve performance of these models through efficient memory accesses guarantee, or considering the dynamic evolution of the memory space (grids, trees,…) that holds crucial data in simulations. In our opinion, the different optimizations mentioned above are not decisive enough to achieve the high performance computing power that cell-behavior simulation programs actually need. The possibility of using multicore and GPU parallelism as a promising multiplatform and framework to develop new programming techniques to speed-up the computation time of simulations is just starting to be explored in the few last years. This paper presents a model that incorporates parallel processing, identifying the synchronization necessary for speeding up tumor growth simulations implemented in Java and C++ programming environments. The speed up improvement that specific parallel syntactic constructs, such as executors (thread pools) in Java, are studied. The new tumor growth parallel model is proved using implementations with Java and C++ languages on two different platforms: chipset Intel core i-X and a HPC cluster of processors at our university. The parallelization of Polesczuk and Enderling model (normally used by researchers in mathematical oncology) proposed here is analyzed with respect to performance gain. We intend to apply the model and overall parallelization technique presented here to solid tumors of specific affiliation such as prostate, breast, or colon. Our final objective is to set up a multiparadigm model capable of modelling angiogenesis, or the growth inhibition induced by chemotaxis, as well as the effect of therapies based on the presence of cytotoxic/cytostatic drugs.

Keywords: cellular automaton, tumor growth model, simulation, multicore and manycore programming, parallel programming, high performance computing, speed up

Procedia PDF Downloads 244

Authors: Asma Omri, Iheb Sifaoui, Sofiane Sayahi, Hichem Besbes

Abstract:

Future vehicle systems demand advanced capabilities, notably in-cabin life detection and driver monitoring systems, with a particular emphasis on drowsiness detection. To meet these requirements, several techniques employ artificial intelligence methods based on real-time vital sign measurements. In parallel, Frequency-Modulated Continuous-Wave (FMCW) radar technology has garnered considerable attention in the domains of healthcare and biomedical engineering for non-invasive vital sign monitoring. FMCW radar offers a multitude of advantages, including its non-intrusive nature, continuous monitoring capacity, and its ability to penetrate through clothing. In this paper, we propose a system utilizing the AWR6843AOP radar from Texas Instruments (TI) to extract precise vital sign information. The radar allows us to estimate Ballistocardiogram (BCG) signals, which capture the mechanical movements of the body, particularly the ballistic forces generated by heartbeats and respiration. These signals are rich sources of information about the cardiac cycle, rendering them suitable for heart rate estimation. The process begins with real-time subject positioning, followed by clutter removal, computation of Doppler phase differences, and the use of various filtering methods to accurately capture subtle physiological movements. To address the challenges associated with FMCW radar-based vital sign monitoring, including motion artifacts due to subjects' movement or radar micro-vibrations, Long Short-Term Memory (LSTM) networks are implemented. LSTM's adaptability to different heart rate patterns and ability to handle real-time data make it suitable for continuous monitoring applications. Several crucial steps were taken, including feature extraction (involving amplitude, time intervals, and signal morphology), sequence modeling, heart rate estimation through the analysis of detected cardiac cycles and their temporal relationships, and performance evaluation using metrics such as Root Mean Square Error (RMSE) and correlation with reference heart rate measurements. For dataset construction and LSTM training, a comprehensive data collection system was established, integrating the AWR6843AOP radar, a Heart Rate Belt, and a smart watch for ground truth measurements. Rigorous synchronization of these devices ensured data accuracy. Twenty participants engaged in various scenarios, encompassing indoor and real-world conditions within a moving vehicle equipped with the radar system. Static and dynamic subject’s conditions were considered. The heart rate estimation through LSTM outperforms traditional signal processing techniques that rely on filtering, Fast Fourier Transform (FFT), and thresholding. It delivers an average accuracy of approximately 91% with an RMSE of 1.01 beat per minute (bpm). In conclusion, this paper underscores the promising potential of FMCW radar technology integrated with artificial intelligence algorithms in the context of automotive applications. This innovation not only enhances road safety but also paves the way for its integration into the automotive ecosystem to improve driver well-being and overall vehicular safety.

Keywords: ballistocardiogram, FMCW Radar, vital sign monitoring, LSTM

Procedia PDF Downloads 72