Search results for: optimization algorithms
3475 A Comparative Study of Sampling-Based Uncertainty Propagation with First Order Error Analysis and Percentile-Based Optimization
Authors: M. Gulam Kibria, Shourav Ahmed, Kais Zaman
Abstract:
In system analysis, the information on the uncertain input variables cause uncertainty in the system responses. Different probabilistic approaches for uncertainty representation and propagation in such cases exist in the literature. Different uncertainty representation approaches result in different outputs. Some of the approaches might result in a better estimation of system response than the other approaches. The NASA Langley Multidisciplinary Uncertainty Quantification Challenge (MUQC) has posed challenges about uncertainty quantification. Subproblem A, the uncertainty characterization subproblem, of the challenge posed is addressed in this study. In this subproblem, the challenge is to gather knowledge about unknown model inputs which have inherent aleatory and epistemic uncertainties in them with responses (output) of the given computational model. We use two different methodologies to approach the problem. In the first methodology we use sampling-based uncertainty propagation with first order error analysis. In the other approach we place emphasis on the use of Percentile-Based Optimization (PBO). The NASA Langley MUQC’s subproblem A is developed in such a way that both aleatory and epistemic uncertainties need to be managed. The challenge problem classifies each uncertain parameter as belonging to one the following three types: (i) An aleatory uncertainty modeled as a random variable. It has a fixed functional form and known coefficients. This uncertainty cannot be reduced. (ii) An epistemic uncertainty modeled as a fixed but poorly known physical quantity that lies within a given interval. This uncertainty is reducible. (iii) A parameter might be aleatory but sufficient data might not be available to adequately model it as a single random variable. For example, the parameters of a normal variable, e.g., the mean and standard deviation, might not be precisely known but could be assumed to lie within some intervals. It results in a distributional p-box having the physical parameter with an aleatory uncertainty, but the parameters prescribing its mathematical model are subjected to epistemic uncertainties. Each of the parameters of the random variable is an unknown element of a known interval. This uncertainty is reducible. From the study, it is observed that due to practical limitations or computational expense, the sampling is not exhaustive in sampling-based methodology. That is why the sampling-based methodology has high probability of underestimating the output bounds. Therefore, an optimization-based strategy to convert uncertainty described by interval data into a probabilistic framework is necessary. This is achieved in this study by using PBO.Keywords: aleatory uncertainty, epistemic uncertainty, first order error analysis, uncertainty quantification, percentile-based optimization
Procedia PDF Downloads 2463474 Utilization of Mustard Leaves (Brassica juncea) Powder for the Development of Cereal Based Extruded Snacks
Authors: Maya S. Rathod, Bahadur Singh Hathan
Abstract:
Mustard leaves are rich in folates, vitamin A, K and B-complex. Mustard greens are low in calories and fats and rich in dietary fiber. They are rich in potassium, manganese, iron, copper, calcium, magnesium and low in sodium. It is very rich in antioxidants and Phytonutrients. For the optimization of process variables (moisture content and mustard leave powder), the experiments were conducted according to central composite Face Centered Composite design of RSM. The mustard leaves powder was replaced with composite flour (a combination of rice, chickpea and corn in the ratio of 70:15:15). The extrudate was extruded in a twin screw extruder at a barrel temperature of 120°C. The independent variables were mustard leaves powder (2-10 %) and moisture content (12-20 %). Responses analyzed were bulk density, water solubility index, water absorption index, lateral expansion, hardness, antioxidant activity, total phenolic content and overall acceptability. The optimum conditions obtained were 7.19 g mustard leaves powder in 100 g premix having 16.8 % moisture content (w.b).Keywords: extrusion, mustard leaves powder, optimization, response surface methodology
Procedia PDF Downloads 5493473 Modeling and Optimization of Performance of Four Stroke Spark Ignition Injector Engine
Authors: A. A. Okafor, C. H. Achebe, J. L. Chukwuneke, C. G. Ozoegwu
Abstract:
The performance of an engine whose basic design parameters are known can be predicted with the assistance of simulation programs into the less time, cost and near value of actual. This paper presents a comprehensive mathematical model of the performance parameters of four stroke spark ignition engine. The essence of this research work is to develop a mathematical model for the analysis of engine performance parameters of four stroke spark ignition engine before embarking on full scale construction, this will ensure that only optimal parameters are in the design and development of an engine and also allow to check and develop the design of the engine and it’s operation alternatives in an inexpensive way and less time, instead of using experimental method which requires costly research test beds. To achieve this, equations were derived which describe the performance parameters (sfc, thermal efficiency, mep and A/F). The equations were used to simulate and optimize the engine performance of the model for various engine speeds. The optimal values obtained for the developed bivariate mathematical models are: sfc is 0.2833kg/kwh, efficiency is 28.77% and a/f is 20.75.Keywords: bivariate models, engine performance, injector engine, optimization, performance parameters, simulation, spark ignition
Procedia PDF Downloads 3343472 Initial Dip: An Early Indicator of Neural Activity in Functional Near Infrared Spectroscopy Waveform
Authors: Mannan Malik Muhammad Naeem, Jeong Myung Yung
Abstract:
Functional near infrared spectroscopy (fNIRS) has a favorable position in non-invasive brain imaging techniques. The concentration change of oxygenated hemoglobin and de-oxygenated hemoglobin during particular cognitive activity is the basis for this neuro-imaging modality. Two wavelengths of near-infrared light can be used with modified Beer-Lambert law to explain the indirect status of neuronal activity inside brain. The temporal resolution of fNIRS is very good for real-time brain computer-interface applications. The portability, low cost and an acceptable temporal resolution of fNIRS put it on a better position in neuro-imaging modalities. In this study, an optimization model for impulse response function has been used to estimate/predict initial dip using fNIRS data. In addition, the activity strength parameter related to motor based cognitive task has been analyzed. We found an initial dip that remains around 200-300 millisecond and better localize neural activity.Keywords: fNIRS, brain-computer interface, optimization algorithm, adaptive signal processing
Procedia PDF Downloads 2283471 Machine Learning Techniques in Seismic Risk Assessment of Structures
Authors: Farid Khosravikia, Patricia Clayton
Abstract:
The main objective of this work is to evaluate the advantages and disadvantages of various machine learning techniques in two key steps of seismic hazard and risk assessment of different types of structures. The first step is the development of ground-motion models, which are used for forecasting ground-motion intensity measures (IM) given source characteristics, source-to-site distance, and local site condition for future events. IMs such as peak ground acceleration and velocity (PGA and PGV, respectively) as well as 5% damped elastic pseudospectral accelerations at different periods (PSA), are indicators of the strength of shaking at the ground surface. Typically, linear regression-based models, with pre-defined equations and coefficients, are used in ground motion prediction. However, due to the restrictions of the linear regression methods, such models may not capture more complex nonlinear behaviors that exist in the data. Thus, this study comparatively investigates potential benefits from employing other machine learning techniques as statistical method in ground motion prediction such as Artificial Neural Network, Random Forest, and Support Vector Machine. The results indicate the algorithms satisfy some physically sound characteristics such as magnitude scaling distance dependency without requiring pre-defined equations or coefficients. Moreover, it is shown that, when sufficient data is available, all the alternative algorithms tend to provide more accurate estimates compared to the conventional linear regression-based method, and particularly, Random Forest outperforms the other algorithms. However, the conventional method is a better tool when limited data is available. Second, it is investigated how machine learning techniques could be beneficial for developing probabilistic seismic demand models (PSDMs), which provide the relationship between the structural demand responses (e.g., component deformations, accelerations, internal forces, etc.) and the ground motion IMs. In the risk framework, such models are used to develop fragility curves estimating exceeding probability of damage for pre-defined limit states, and therefore, control the reliability of the predictions in the risk assessment. In this study, machine learning algorithms like artificial neural network, random forest, and support vector machine are adopted and trained on the demand parameters to derive PSDMs for them. It is observed that such models can provide more accurate estimates of prediction in relatively shorter about of time compared to conventional methods. Moreover, they can be used for sensitivity analysis of fragility curves with respect to many modeling parameters without necessarily requiring more intense numerical response-history analysis.Keywords: artificial neural network, machine learning, random forest, seismic risk analysis, seismic hazard analysis, support vector machine
Procedia PDF Downloads 1103470 Methodology of Construction Equipment Optimization for Earthwork
Authors: Jaehyun Choi, Hyunjung Kim, Namho Kim
Abstract:
Earthwork is one of the critical civil construction operations that require large-quantities of resources due to its intensive dependency upon construction equipment. Therefore, efficient construction equipment management can highly contribute to productivity improvements and cost savings. Earthwork operation utilizes various combinations of construction equipment in order to meet project requirements such as time and cost. Identification of site condition and construction methods should be performed in advance in order to develop a proper execution plan. The factors to be considered include capacity of equipment assigned, the method of construction, the size of the site, and the surrounding condition. In addition, optimal combination of various construction equipment should be selected. However, in real world practice, equipment utilization plan is performed based on experience and intuition of management. The researchers evaluated the efficiency of various alternatives of construction equipment combinations by utilizing the process simulation model, validated the model from a case study project, and presented a methodology to find optimized plan among alternatives.Keywords: earthwork operation, construction equipment, process simulation, optimization
Procedia PDF Downloads 4343469 Remote Sensing Approach to Predict the Impacts of Land Use/Land Cover Change on Urban Thermal Comfort Using Machine Learning Algorithms
Authors: Ahmad E. Aldousaria, Abdulla Al Kafy
Abstract:
Urbanization is an incessant process that involves the transformation of land use/land cover (LULC), resulting in a reduction of cool land covers and thermal comfort zones (TCZs). This study explores the directional shrinkage of TCZs in Kuwait using Landsat satellite data from 1991 – 2021 to predict the future LULC and TCZ distribution for 2026 and 2031 using cellular automata (CA) and artificial neural network (ANN) algorithms. Analysis revealed a rapid urban expansion (40 %) in SE, NE, and NW directions and TCZ shrinkage in N – NW and SW directions with 25 % of the very uncomfortable area. The predicted result showed an urban area increase from 44 % in 2021 to 47 % and 52 % in 2026 and 2031, respectively, where uncomfortable zones were found to be concentrated around urban areas and bare lands in N – NE and N – NW directions. This study proposes an effective and sustainable framework to control TCZ shrinkage, including zero soil policies, planned landscape design, manmade water bodies, and rooftop gardens. This study will help urban planners and policymakers to make Kuwait an eco–friendly, functional, and sustainable country.Keywords: land cover change, thermal environment, green cover loss, machine learning, remote sensing
Procedia PDF Downloads 2303468 Adsorption of Cerium as One of the Rare Earth Elements Using Multiwall Carbon Nanotubes from Aqueous Solution: Modeling, Equilibrium and Kinetics
Authors: Saeb Ahmadi, Mohsen Vafaie Sefti, Mohammad Mahdi Shadman, Ebrahim Tangestani
Abstract:
Carbon nanotube has shown great potential for the removal of various inorganic and organic components due to properties such as large surface area and high adsorption capacity. Central composite design is widely used method for determining optimal conditions. Also due to the economic reasons and wide application, the rare earth elements are important components. The analyses of cerium (Ce(III)) adsorption as one of the Rare Earth Elements (REEs) adsorption on Multiwall Carbon Nanotubes (MWCNTs) have been studied. The optimization process was performed using Response Surface Methodology (RSM). The optimum amount conditions were pH of 4.5, initial Ce (III) concentration of 90 mg/l and MWCNTs dosage of 80 mg. Under this condition, the optimum adsorption percentage of Ce (III) was obtained about 96%. Next, at the obtained optimum conditions the kinetic and isotherm studied and result showed the pseudo-second order and Langmuir isotherm are more fitted with experimental data than other models.Keywords: cerium, rare earth element, MWCNTs, adsorption, optimization
Procedia PDF Downloads 1693467 Optimization of Temperature for Crystal Violet Dye Adsorption Using Castor Leaf Powder by Response Surface Methodology
Authors: Vipan Kumar Sohpal
Abstract:
Temperature effect on the adsorption of crystal violet dye (CVD) was investigated using a castor leaf powder (CLP) that was prepared from the mature leaves of castor trees, through chemical reaction. The optimum values of pH (8), adsorbent dose (10g/L), initial dye concentration (10g/L), time (2hrs), and stirrer speed (120 rpm) were fixed to investigate the influence of temperature on adsorption capacity, percentage of removal of dye and free energy. A central composite design (CCD) was successfully employed for experimental design and analysis of the results. The combined effect of temperature, absorbance, and concentration on the dye adsorption was studied and optimized using response surface methodology. The optimum values of adsorption capacity, percentage of removal of dye and free energy were found to be 0.965(mg/g), 93.38 %, -8202.7(J/mol) at temperature 55.97 °C having desirability > 90% for removal of crystal violet dye respectively. The experimental values were in good agreement with predicted values.Keywords: crystal violet dye, CVD, castor leaf powder, CLP, response surface methodology, temperature, optimization
Procedia PDF Downloads 1403466 The Asymmetric Proximal Support Vector Machine Based on Multitask Learning for Classification
Authors: Qing Wu, Fei-Yan Li, Heng-Chang Zhang
Abstract:
Multitask learning support vector machines (SVMs) have recently attracted increasing research attention. Given several related tasks, the single-task learning methods trains each task separately and ignore the inner cross-relationship among tasks. However, multitask learning can capture the correlation information among tasks and achieve better performance by training all tasks simultaneously. In addition, the asymmetric squared loss function can better improve the generalization ability of the models on the most asymmetric distributed data. In this paper, we first make two assumptions on the relatedness among tasks and propose two multitask learning proximal support vector machine algorithms, named MTL-a-PSVM and EMTL-a-PSVM, respectively. MTL-a-PSVM seeks a trade-off between the maximum expectile distance for each task model and the closeness of each task model to the general model. As an extension of the MTL-a-PSVM, EMTL-a-PSVM can select appropriate kernel functions for shared information and private information. Besides, two corresponding special cases named MTL-PSVM and EMTLPSVM are proposed by analyzing the asymmetric squared loss function, which can be easily implemented by solving linear systems. Experimental analysis of three classification datasets demonstrates the effectiveness and superiority of our proposed multitask learning algorithms.Keywords: multitask learning, asymmetric squared loss, EMTL-a-PSVM, classification
Procedia PDF Downloads 1443465 Optimal Scheduling of Trains in Complex National Scale Railway Networks
Authors: Sanat Ramesh, Tarun Dutt, Abhilasha Aswal, Anushka Chandrababu, G. N. Srinivasa Prasanna
Abstract:
Optimal Schedule Generation for a large national railway network operating thousands of passenger trains with tens of thousands of kilometers of track is a grand computational challenge in itself. We present heuristics based on a Mixed Integer Program (MIP) formulation for local optimization. These methods provide flexibility in scheduling new trains with varying speed and delays and improve utilization of infrastructure. We propose methods that provide a robust solution with hundreds of trains being scheduled over a portion of the railway network without significant increases in delay. We also provide techniques to validate the nominal schedules thus generated over global correlated variations in travel times thereby enabling us to detect conflicts arising due to delays. Our validation results which assume only the support of the arrival and departure time distributions takes an order of few minutes for a portion of the network and is computationally efficient to handle the entire network.Keywords: mixed integer programming, optimization, railway network, train scheduling
Procedia PDF Downloads 1623464 Design and Implementation of a Hardened Cryptographic Coprocessor with 128-bit RISC-V Core
Authors: Yashas Bedre Raghavendra, Pim Vullers
Abstract:
This study presents the design and implementation of an abstract cryptographic coprocessor, leveraging AMBA(Advanced Microcontroller Bus Architecture) protocols - APB (Advanced Peripheral Bus) and AHB (Advanced High-performance Bus), to enable seamless integration with the main CPU(Central processing unit) and enhance the coprocessor’s algorithm flexibility. The primary objective is to create a versatile coprocessor that can execute various cryptographic algorithms, including ECC(Elliptic-curve cryptography), RSA(Rivest–Shamir–Adleman), and AES (Advanced Encryption Standard) while providing a robust and secure solution for modern secure embedded systems. To achieve this goal, the coprocessor is equipped with a tightly coupled memory (TCM) for rapid data access during cryptographic operations. The TCM is placed within the coprocessor, ensuring quick retrieval of critical data and optimizing overall performance. Additionally, the program memory is positioned outside the coprocessor, allowing for easy updates and reconfiguration, which enhances adaptability to future algorithm implementations. Direct links are employed instead of DMA(Direct memory access) for data transfer, ensuring faster communication and reducing complexity. The AMBA-based communication architecture facilitates seamless interaction between the coprocessor and the main CPU, streamlining data flow and ensuring efficient utilization of system resources. The abstract nature of the coprocessor allows for easy integration of new cryptographic algorithms in the future. As the security landscape continues to evolve, the coprocessor can adapt and incorporate emerging algorithms, making it a future-proof solution for cryptographic processing. Furthermore, this study explores the addition of custom instructions into RISC-V ISE (Instruction Set Extension) to enhance cryptographic operations. By incorporating custom instructions specifically tailored for cryptographic algorithms, the coprocessor achieves higher efficiency and reduced cycles per instruction (CPI) compared to traditional instruction sets. The adoption of RISC-V 128-bit architecture significantly reduces the total number of instructions required for complex cryptographic tasks, leading to faster execution times and improved overall performance. Comparisons are made with 32-bit and 64-bit architectures, highlighting the advantages of the 128-bit architecture in terms of reduced instruction count and CPI. In conclusion, the abstract cryptographic coprocessor presented in this study offers significant advantages in terms of algorithm flexibility, security, and integration with the main CPU. By leveraging AMBA protocols and employing direct links for data transfer, the coprocessor achieves high-performance cryptographic operations without compromising system efficiency. With its TCM and external program memory, the coprocessor is capable of securely executing a wide range of cryptographic algorithms. This versatility and adaptability, coupled with the benefits of custom instructions and the 128-bit architecture, make it an invaluable asset for secure embedded systems, meeting the demands of modern cryptographic applications.Keywords: abstract cryptographic coprocessor, AMBA protocols, ECC, RSA, AES, tightly coupled memory, secure embedded systems, RISC-V ISE, custom instructions, instruction count, cycles per instruction
Procedia PDF Downloads 743463 Resilient Machine Learning in the Nuclear Industry: Crack Detection as a Case Study
Authors: Anita Khadka, Gregory Epiphaniou, Carsten Maple
Abstract:
There is a dramatic surge in the adoption of machine learning (ML) techniques in many areas, including the nuclear industry (such as fault diagnosis and fuel management in nuclear power plants), autonomous systems (including self-driving vehicles), space systems (space debris recovery, for example), medical surgery, network intrusion detection, malware detection, to name a few. With the application of learning methods in such diverse domains, artificial intelligence (AI) has become a part of everyday modern human life. To date, the predominant focus has been on developing underpinning ML algorithms that can improve accuracy, while factors such as resiliency and robustness of algorithms have been largely overlooked. If an adversarial attack is able to compromise the learning method or data, the consequences can be fatal, especially but not exclusively in safety-critical applications. In this paper, we present an in-depth analysis of five adversarial attacks and three defence methods on a crack detection ML model. Our analysis shows that it can be dangerous to adopt machine learning techniques in security-critical areas such as the nuclear industry without rigorous testing since they may be vulnerable to adversarial attacks. While common defence methods can effectively defend against different attacks, none of the three considered can provide protection against all five adversarial attacks analysed.Keywords: adversarial machine learning, attacks, defences, nuclear industry, crack detection
Procedia PDF Downloads 1623462 Quantitative Evaluation of Endogenous Reference Genes for ddPCR under Salt Stress Using a Moderate Halophile
Authors: Qinghua Xing, Noha M. Mesbah, Haisheng Wang, Jun Li, Baisuo Zhao
Abstract:
Droplet digital PCR (ddPCR) is being increasingly adopted for gene detection and quantification because of its higher sensitivity and specificity. According to previous observations and our lab data, it is essential to use endogenous reference genes (RGs) when investigating gene expression at the mRNA level under salt stress. This study aimed to select and validate suitable RGs for gene expression under salt stress using ddPCR. Six candidate RGs were selected based on the tandem mass tag (TMT)-labeled quantitative proteomics of Alkalicoccus halolimnae at four salinities. The expression stability of these candidate genes was evaluated using statistical algorithms (geNorm, NormFinder, BestKeeper and RefFinder). There was a small fluctuation in cycle threshold (Ct) value and copy number of the pdp gene. Its expression stability was ranked in the vanguard of all algorithms, and was the most suitable RG for quantification of expression by both qPCR and ddPCR of A. halolimnae under salt stress. Single RG pdp and RG combinations were used to normalize the expression of ectA, ectB, ectC, and ectD under four salinities. The present study constitutes the first systematic analysis of endogenous RG selection for halophiles responding to salt stress. This work provides a valuable theory and an approach reference of internal control identification for ddPCR-based stress response models.Keywords: endogenous reference gene, salt stress, ddPCR, RT-qPCR, Alkalicoccus halolimnae
Procedia PDF Downloads 1113461 Multi-Objective Optimization (Pareto Sets) and Multi-Response Optimization (Desirability Function) of Microencapsulation of Emamectin
Authors: Victoria Molina, Wendy Franco, Sergio Benavides, José M. Troncoso, Ricardo Luna, Jose R. PéRez-Correa
Abstract:
Emamectin Benzoate (EB) is a crystal antiparasitic that belongs to the avermectin family. It is one of the most common treatments used in Chile to control Caligus rogercresseyi in Atlantic salmon. However, the sea lice acquired resistance to EB when it is exposed at sublethal EB doses. The low solubility rate of EB and its degradation at the acidic pH in the fish digestive tract are the causes of the slow absorption of EB in the intestine. To protect EB from degradation and enhance its absorption, specific microencapsulation technologies must be developed. Amorphous Solid Dispersion techniques such as Spray Drying (SD) and Ionic Gelation (IG) seem adequate for this purpose. Recently, Soluplus® (SOL) has been used to increase the solubility rate of several drugs with similar characteristics than EB. In addition, alginate (ALG) is a widely used polymer in IG for biomedical applications. Regardless of the encapsulation technique, the quality of the obtained microparticles is evaluated with the following responses, yield (Y%), encapsulation efficiency (EE%) and loading capacity (LC%). In addition, it is important to know the percentage of EB released from the microparticles in gastric (GD%) and intestinal (ID%) digestions. In this work, we microencapsulated EB with SOL (EB-SD) and with ALG (EB-IG) using SD and IG, respectively. Quality microencapsulation responses and in vitro gastric and intestinal digestions at pH 3.35 and 7.8, respectively, were obtained. A central composite design was used to find the optimum microencapsulation variables (amount of EB, amount of polymer and feed flow). In each formulation, the behavior of these variables was predicted with statistical models. Then, the response surface methodology was used to find the best combination of the factors that allowed a lower EB release in gastric conditions, while permitting a major release at intestinal digestion. Two approaches were used to determine this. The desirability approach (DA) and multi-objective optimization (MOO) with multi-criteria decision making (MCDM). Both microencapsulation techniques allowed to maintain the integrity of EB in acid pH, given the small amount of EB released in gastric medium, while EB-IG microparticles showed greater EB release at intestinal digestion. For EB-SD, optimal conditions obtained with MOO plus MCDM yielded a good compromise among the microencapsulation responses. In addition, using these conditions, it is possible to reduce microparticles costs due to the reduction of 60% of BE regard the optimal BE proposed by (DA). For EB-GI, the optimization techniques used (DA and MOO) yielded solutions with different advantages and limitations. Applying DA costs can be reduced 21%, while Y, GD and ID showed 9.5%, 84.8% and 2.6% lower values than the best condition. In turn, MOO yielded better microencapsulation responses, but at a higher cost. Overall, EB-SD with operating conditions selected by MOO seems the best option, since a good compromise between costs and encapsulation responses was obtained.Keywords: microencapsulation, multiple decision-making criteria, multi-objective optimization, Soluplus®
Procedia PDF Downloads 1343460 Resistivity Tomography Optimization Based on Parallel Electrode Linear Back Projection Algorithm
Authors: Yiwei Huang, Chunyu Zhao, Jingjing Ding
Abstract:
Electrical Resistivity Tomography has been widely used in the medicine and the geology, such as the imaging of the lung impedance and the analysis of the soil impedance, etc. Linear Back Projection is the core algorithm of Electrical Resistivity Tomography, but the traditional Linear Back Projection can not make full use of the information of the electric field. In this paper, an imaging method of Parallel Electrode Linear Back Projection for Electrical Resistivity Tomography is proposed, which generates the electric field distribution that is not linearly related to the traditional Linear Back Projection, captures the new information and improves the imaging accuracy without increasing the number of electrodes by changing the connection mode of the electrodes. The simulation results show that the accuracy of the image obtained by the inverse operation obtained by the Parallel Electrode Linear Back Projection can be improved by about 20%.Keywords: electrical resistivity tomography, finite element simulation, image optimization, parallel electrode linear back projection
Procedia PDF Downloads 1583459 Hyper Parameter Optimization of Deep Convolutional Neural Networks for Pavement Distress Classification
Authors: Oumaima Khlifati, Khadija Baba
Abstract:
Pavement distress is the main factor responsible for the deterioration of road structure durability, damage vehicles, and driver comfort. Transportation agencies spend a high proportion of their funds on pavement monitoring and maintenance. The auscultation of pavement distress was based on the manual survey, which was extremely time consuming, labor intensive, and required domain expertise. Therefore, the automatic distress detection is needed to reduce the cost of manual inspection and avoid more serious damage by implementing the appropriate remediation actions at the right time. Inspired by recent deep learning applications, this paper proposes an algorithm for automatic road distress detection and classification using on the Deep Convolutional Neural Network (DCNN). In this study, the types of pavement distress are classified as transverse or longitudinal cracking, alligator, pothole, and intact pavement. The dataset used in this work is composed of public asphalt pavement images. In order to learn the structure of the different type of distress, the DCNN models are trained and tested as a multi-label classification task. In addition, to get the highest accuracy for our model, we adjust the structural optimization hyper parameters such as the number of convolutions and max pooling, filers, size of filters, loss functions, activation functions, and optimizer and fine-tuning hyper parameters that conclude batch size and learning rate. The optimization of the model is executed by checking all feasible combinations and selecting the best performing one. The model, after being optimized, performance metrics is calculated, which describe the training and validation accuracies, precision, recall, and F1 score.Keywords: distress pavement, hyperparameters, automatic classification, deep learning
Procedia PDF Downloads 993458 Load Management Using Multiple Sequential Load Shaping Techniques
Authors: Amira M. Attia, Karim H. Youssef, Nabil H. Abbasi
Abstract:
Demand Side Management (DSM) is an essential characteristic of current and future smart grid systems. As one of DSM functions, load management aims to control customers’ total electric consumption and utility’s load factor by using various load shaping techniques. However, applying load shaping techniques such as load shifting, peak clipping, or strategic conservation individually does not provide the desired level of improvement for load factor increment and/or customer’s bill reduction. In this paper, two load shaping techniques will be simulated as constrained optimization problems. The purpose is to reflect the application of combined load shifting and strategic conservation model together at the same time, and the application of combined load shifting and peak clipping model as well. The problem will be formulated and solved by using disciplined convex programming (CVX) based MATLAB® R2013b. Simulation results will be evaluated and compared for studying the most impactful multi-techniques model in improving load curve.Keywords: convex programing, demand side management, load shaping, multiple, building energy optimization
Procedia PDF Downloads 3183457 Optimal Allocation of PHEV Parking Lots to Minimize Dstribution System Losses
Authors: Mohsen Mazidi, Ali Abbaspour, Mahmud Fotuhi-Firuzabad, Mohamamd Rastegar
Abstract:
To tackle the air pollution issues, Plug-in Hybrid Electric Vehicles (PHEVs) are proposed as an appropriate solution. Charging a large amount of PHEV batteries, if not controlled, would have negative impacts on the distribution system. The control process of charging of these vehicles can be centralized in parking lots that may provide a chance for better coordination than the individual charging in houses. In this paper, an optimization-based approach is proposed to determine the optimum PHEV parking capacities in candidate nodes of the distribution system. In so doing, a profile for charging and discharging of PHEVs is developed in order to flatten the network load profile. Then, this profile is used in solving an optimization problem to minimize the distribution system losses. The outputs of the proposed method are the proper place for PHEV parking lots and optimum capacity for each parking. The application of the proposed method on the IEEE-34 node test feeder verifies the effectiveness of the method.Keywords: loss, plug-in hybrid electric vehicle (PHEV), PHEV parking lot, V2G
Procedia PDF Downloads 5463456 Optimization of Ultrasound Assisted Extraction of Polysaccharides from Plant Waste Materials: Selected Model Material is Hazelnut Skin
Abstract:
In this study, optimization of ultrasound assisted extraction (UAE) of hemicellulose based polysaccharides from plant waste material has been studied. Selected material is hazelnut skin. Extraction variables for the operation are extraction time, amplitude and application temperature. Optimum conditions have been evaluated depending on responses such as amount of wet crude polysaccharide, total carbohydrate content and dried sample. Pretreated hazelnut skin powders were used for the experiments. 10 grams of samples were suspended in 100 ml water in a jacketed vessel with additional magnetic stirring. Mixture was sonicated by immersing ultrasonic probe processor. After the extraction procedures, ethanol soluble and insoluble sides were separated for further examinations. The obtained experimental data were analyzed by analysis of variance (ANOVA). Second order polynomial models were developed using multiple regression analysis. The individual and interactive effects of applied variables were evaluated by Box Behnken Design. The models developed from the experimental design were predictive and good fit with the experimental data with high correlation coefficient value (R2 more than 0.95). Extracted polysaccharides from hazelnut skin are assumed to be pectic polysaccharides according to the literature survey of Fourier Transform Spectrometry (FTIR) analysis results. No more change can be observed between spectrums of different sonication times. Application of UAE at optimized condition has an important effect on extraction of hemicellulose from plant material by satisfying partial hydrolysis to break the bounds with other components in plant cell wall material. This effect can be summarized by varied intensity of microjets and microstreaming at varied sonication conditions.Keywords: hazelnut skin, optimization, polysaccharide, ultrasound assisted extraction
Procedia PDF Downloads 3343455 A Comparison of Alternative Traffic Controls for Interchange Ramp Areas Using Synchro Software
Authors: Mohamed Mesbah, Bruce Janson
Abstract:
An interchange is the most important component of freeway and highway facilities. It is working as a connector between the highway’s elements. The main goal of designing interchanges is to provide an acceptable level of service and delay to make vehicles move smoothly when they are entering and exiting the interchange. There are many factors that can have a significant impact on the level of service; the main factors are traffic volumes, and type of interchange. This paper will discuss interchange with roundabouts under various values of traffic volumes to determine the level of service of the interchanges that will be studied in this paper and replace the system of interchange from roundabout to traffic signal to make a significant compression between these systems. A secondary goal is to propose improvements for scenarios where the level of service is deemed unacceptable. This will be achieved using Synchro traffic simulation software, which facilitates the simulation and optimization of interchanges to enhance operational efficiency and safety.Keywords: interchange, roundabout, traffic signal, Synchro, delay, level of service, traffic volumes, vehicles, simulation, optimization, adjustment
Procedia PDF Downloads 453454 A Deterministic Approach for Solving the Hull and White Interest Rate Model with Jump Process
Authors: Hong-Ming Chen
Abstract:
This work considers the resolution of the Hull and White interest rate model with the jump process. A deterministic process is adopted to model the random behavior of interest rate variation as deterministic perturbations, which is depending on the time t. The Brownian motion and jumps uncertainty are denoted as the integral functions piecewise constant function w(t) and point function θ(t). It shows that the interest rate function and the yield function of the Hull and White interest rate model with jump process can be obtained by solving a nonlinear semi-infinite programming problem. A relaxed cutting plane algorithm is then proposed for solving the resulting optimization problem. The method is calibrated for the U.S. treasury securities at 3-month data and is used to analyze several effects on interest rate prices, including interest rate variability, and the negative correlation between stock returns and interest rates. The numerical results illustrate that our approach essentially generates the yield functions with minimal fitting errors and small oscillation.Keywords: optimization, interest rate model, jump process, deterministic
Procedia PDF Downloads 1623453 Estimation of Fourier Coefficients of Flux Density for Surface Mounted Permanent Magnet (SMPM) Generators by Direct Search Optimization
Authors: Ramakrishna Rao Mamidi
Abstract:
It is essential for Surface Mounted Permanent Magnet (SMPM) generators to determine the performance prediction and analyze the magnet’s air gap flux density wave shape. The flux density wave shape is neither a pure sine wave or square wave nor a combination. This is due to the variation of air gap reluctance between the stator and permanent magnets. The stator slot openings and the number of slots make the wave shape highly complicated. To reduce the complexity of analysis, approximations are made to the wave shape using Fourier analysis. In contrast to the traditional integration method, the Fourier coefficients, an and bn, are obtained by direct search method optimization. The wave shape with optimized coefficients gives a wave shape close to the desired wave shape. Harmonics amplitudes are worked out and compared with initial values. It can be concluded that the direct search method can be used for estimating Fourier coefficients for irregular wave shapes.Keywords: direct search, flux plot, fourier analysis, permanent magnets
Procedia PDF Downloads 2203452 Optimization of Ultrasonic Assisted Extraction of Antioxidants and Phenolic Compounds from Coleus Using Response Surface Methodology
Authors: Reihaneh Ahmadzadeh Ghavidel
Abstract:
Free radicals such as reactive oxygen species (ROS) have detrimental effects on human health through several mechanisms. On the other hand, antioxidant molecules reduce free radical generation in biologic systems. Synthetic antioxidants, which are used in food industry, have also negative impact on human health. Therefore recognition of natural antioxidants such as anthocyanins can solve these problems simultaneously. Coleus (Solenostemon scutellarioides) with red leaves is a rich source of anthocyanins compounds. In this study we evaluated the effect of time (10, 20 and 30 min) and temperature (40, 50 and 60° C) on optimization of anthocyanin extraction using surface response method. In addition, the study was aimed to determine maximum extraction for anthocyanin from coleus plant using ultrasound method. The results indicated that the optimum conditions for extraction were 39.84 min at 69.25° C. At this point, total compounds were achieved 3.7451 mg 100 ml⁻¹. Furthermore, under optimum conditions, anthocyanin concentration, extraction efficiency, ferric reducing ability, total phenolic compounds and EC50 were registered 3.221931, 6.692765, 223.062, 3355.605 and 2.614045, respectively.Keywords: anthocyanin, antioxidant, coleus, extraction, sonication
Procedia PDF Downloads 3233451 A Study of Using Multiple Subproblems in Dantzig-Wolfe Decomposition of Linear Programming
Authors: William Chung
Abstract:
This paper is to study the use of multiple subproblems in Dantzig-Wolfe decomposition of linear programming (DW-LP). Traditionally, the decomposed LP consists of one LP master problem and one LP subproblem. The master problem and the subproblem is solved alternatively by exchanging the dual prices of the master problem and the proposals of the subproblem until the LP is solved. It is well known that convergence is slow with a long tail of near-optimal solutions (asymptotic convergence). Hence, the performance of DW-LP highly depends upon the number of decomposition steps. If the decomposition steps can be greatly reduced, the performance of DW-LP can be improved significantly. To reduce the number of decomposition steps, one of the methods is to increase the number of proposals from the subproblem to the master problem. To do so, we propose to add a quadratic approximation function to the LP subproblem in order to develop a set of approximate-LP subproblems (multiple subproblems). Consequently, in each decomposition step, multiple subproblems are solved for providing multiple proposals to the master problem. The number of decomposition steps can be reduced greatly. Note that each approximate-LP subproblem is nonlinear programming, and solving the LP subproblem must faster than solving the nonlinear multiple subproblems. Hence, using multiple subproblems in DW-LP is the tradeoff between the number of approximate-LP subproblems being formed and the decomposition steps. In this paper, we derive the corresponding algorithms and provide some simple computational results. Some properties of the resulting algorithms are also given.Keywords: approximate subproblem, Dantzig-Wolfe decomposition, large-scale models, multiple subproblems
Procedia PDF Downloads 1713450 Aggregation Scheduling Algorithms in Wireless Sensor Networks
Authors: Min Kyung An
Abstract:
In Wireless Sensor Networks which consist of tiny wireless sensor nodes with limited battery power, one of the most fundamental applications is data aggregation which collects nearby environmental conditions and aggregates the data to a designated destination, called a sink node. Important issues concerning the data aggregation are time efficiency and energy consumption due to its limited energy, and therefore, the related problem, named Minimum Latency Aggregation Scheduling (MLAS), has been the focus of many researchers. Its objective is to compute the minimum latency schedule, that is, to compute a schedule with the minimum number of timeslots, such that the sink node can receive the aggregated data from all the other nodes without any collision or interference. For the problem, the two interference models, the graph model and the more realistic physical interference model known as Signal-to-Interference-Noise-Ratio (SINR), have been adopted with different power models, uniform-power and non-uniform power (with power control or without power control), and different antenna models, omni-directional antenna and directional antenna models. In this survey article, as the problem has proven to be NP-hard, we present and compare several state-of-the-art approximation algorithms in various models on the basis of latency as its performance measure.Keywords: data aggregation, convergecast, gathering, approximation, interference, omni-directional, directional
Procedia PDF Downloads 2343449 Evaluation of Beam Structure Using Non-Destructive Vibration-Based Damage Detection Method
Authors: Bashir Ahmad Aasim, Abdul Khaliq Karimi, Jun Tomiyama
Abstract:
Material aging is one of the vital issues among all the civil, mechanical, and aerospace engineering societies. Sustenance and reliability of concrete, which is the widely used material in the world, is the focal point in civil engineering societies. For few decades, researchers have been able to present some form algorithms that could lead to evaluate a structure globally rather than locally without harming its serviceability and traffic interference. The algorithms could help presenting different methods for evaluating structures non-destructively. In this paper, a non-destructive vibration-based damage detection method is adopted to evaluate two concrete beams, one being in a healthy state while the second one contains a crack on its bottom vicinity. The study discusses that damage in a structure affects modal parameters (natural frequency, mode shape, and damping ratio), which are the function of physical properties (mass, stiffness, and damping). The assessment is carried out to acquire the natural frequency of the sound beam. Next, the vibration response is recorded from the cracked beam. Eventually, both results are compared to know the variation in the natural frequencies of both beams. The study concludes that damage can be detected using vibration characteristics of a structural member considering the decline occurred in the natural frequency of the cracked beam.Keywords: concrete beam, natural frequency, non-destructive testing, vibration characteristics
Procedia PDF Downloads 1163448 Optimal Protection Coordination in Distribution Systems with Distributed Generations
Authors: Abdorreza Rabiee, Shahla Mohammad Hoseini Mirzaei
Abstract:
The advantages of distributed generations (DGs) based on renewable energy sources (RESs) leads to high penetration level of DGs in distribution network. With incorporation of DGs in distribution systems, the system reliability and security, as well as voltage profile, is improved. However, the protection of such systems is still challenging. In this paper, at first, the related papers are reviewed and then a practical scheme is proposed for coordination of OCRs in distribution system with DGs. The coordination problem is formulated as a nonlinear programming (NLP) optimization problem with the object function of minimizing total operating time of OCRs. The proposed method is studied based on a simple test system. The optimization problem is solved by General Algebraic Modeling System (GAMS) to calculate the optimal time dial setting (TDS) and also pickup current setting of OCRs. The results show the effectiveness of the proposed method and its applicability.Keywords: distributed generation, DG, distribution network, over current relay, OCR, protection coordination, pickup current, time dial setting, TDS
Procedia PDF Downloads 1413447 An IM-COH Algorithm Neural Network Optimization with Cuckoo Search Algorithm for Time Series Samples
Authors: Wullapa Wongsinlatam
Abstract:
Back propagation algorithm (BP) is a widely used technique in artificial neural network and has been used as a tool for solving the time series problems, such as decreasing training time, maximizing the ability to fall into local minima, and optimizing sensitivity of the initial weights and bias. This paper proposes an improvement of a BP technique which is called IM-COH algorithm (IM-COH). By combining IM-COH algorithm with cuckoo search algorithm (CS), the result is cuckoo search improved control output hidden layer algorithm (CS-IM-COH). This new algorithm has a better ability in optimizing sensitivity of the initial weights and bias than the original BP algorithm. In this research, the algorithm of CS-IM-COH is compared with the original BP, the IM-COH, and the original BP with CS (CS-BP). Furthermore, the selected benchmarks, four time series samples, are shown in this research for illustration. The research shows that the CS-IM-COH algorithm give the best forecasting results compared with the selected samples.Keywords: artificial neural networks, back propagation algorithm, time series, local minima problem, metaheuristic optimization
Procedia PDF Downloads 1573446 Estimating Algae Concentration Based on Deep Learning from Satellite Observation in Korea
Authors: Heewon Jeong, Seongpyo Kim, Joon Ha Kim
Abstract:
Over the last few tens of years, the coastal regions of Korea have experienced red tide algal blooms, which are harmful and toxic to both humans and marine organisms due to their potential threat. It was accelerated owing to eutrophication by human activities, certain oceanic processes, and climate change. Previous studies have tried to monitoring and predicting the algae concentration of the ocean with the bio-optical algorithms applied to color images of the satellite. However, the accurate estimation of algal blooms remains problems to challenges because of the complexity of coastal waters. Therefore, this study suggests a new method to identify the concentration of red tide algal bloom from images of geostationary ocean color imager (GOCI) which are representing the water environment of the sea in Korea. The method employed GOCI images, which took the water leaving radiances centered at 443nm, 490nm and 660nm respectively, as well as observed weather data (i.e., humidity, temperature and atmospheric pressure) for the database to apply optical characteristics of algae and train deep learning algorithm. Convolution neural network (CNN) was used to extract the significant features from the images. And then artificial neural network (ANN) was used to estimate the concentration of algae from the extracted features. For training of the deep learning model, backpropagation learning strategy is developed. The established methods were tested and compared with the performances of GOCI data processing system (GDPS), which is based on standard image processing algorithms and optical algorithms. The model had better performance to estimate algae concentration than the GDPS which is impossible to estimate greater than 5mg/m³. Thus, deep learning model trained successfully to assess algae concentration in spite of the complexity of water environment. Furthermore, the results of this system and methodology can be used to improve the performances of remote sensing. Acknowledgement: This work was supported by the 'Climate Technology Development and Application' research project (#K07731) through a grant provided by GIST in 2017.Keywords: deep learning, algae concentration, remote sensing, satellite
Procedia PDF Downloads 188