Search results for: Laplacian Smoothing

Authors: Mubarack Ahmed, Gabriele Gradoni, Stephen C. Creagh, Gregor Tanner

Abstract:

High-frequency cables commonly connect modern devices and sensors. Interestingly, the proportion of electric components is rising fast in an attempt to achieve lighter and greener devices. Modelling the propagation of signals through these cable networks in the presence of parameter uncertainty is a daunting task. In this work, we study the response of high-frequency cable networks using both Transmission Line and Quantum Graph (QG) theories. We have successfully compared the two theories in terms of reflection spectra using measurements on real, lossy cables. We have derived a generalisation of the vertex scattering matrix to include non-uniform networks – networks of cables with different characteristic impedances and propagation constants. The QG model implicitly takes into account the pseudo-chaotic behavior, at the vertices, of the propagating electric signal. We have successfully compared the asymptotic growth of eigenvalues of the Laplacian with the predictions of Weyl law. We investigate the nearest-neighbour level-spacing distribution of the resonances and compare our results with the predictions of Random Matrix Theory (RMT). To achieve this, we will compare our graphs with the generalisation of Wigner distribution for open systems. The problem of scattering from networks of cables can also provide an analogue model for wireless communication in highly reverberant environments. In this context, we provide a preliminary analysis of the statistics of communication capacity for communication across cable networks, whose eventual aim is to enable detailed laboratory testing of information transfer rates using software defined radio. We specialise this analysis in particular for the case of MIMO (Multiple-Input Multiple-Output) protocols. We have successfully validated our QG model with both TL model and laboratory measurements. The growth of Eigenvalues compares well with Weyl’s law and the level-spacing distribution agrees so well RMT predictions. The results we achieved in the MIMO application compares favourably with the prediction of a parallel on-going research (sponsored by NEMF21.)

Keywords: eigenvalues, multiple-input multiple-output, quantum graph, random matrix theory, transmission line

Procedia PDF Downloads 173

Authors: Lin Li, P. G. R. N. I. Pussella

Abstract:

When representing spatial data and their attributes on different types of maps, the scale plays a key role in the process of map generalization. The process is consisted with two main operators such as selection and omission. Once some data were selected, they would undergo of several geometrical changing processes such as elimination, simplification, smoothing, exaggeration, displacement, aggregation and size reduction. As a result of these operations at different levels of data, the geometry of the spatial features such as length, sinuosity, orientation, perimeter and area would be altered. This would be worst in the case of preparation of small scale maps, since the cartographer has not enough space to represent all the features on the map. What the GIS users do is when they wanted to analyze a set of spatial data; they retrieve a data set and does the analysis part without considering very important characteristics such as the scale, the purpose of the map and the degree of generalization. Further, the GIS users use and compare different maps with different degrees of generalization. Sometimes, GIS users are going beyond the scale of the source map using zoom in facility and violate the basic cartographic rule 'it is not suitable to create a larger scale map using a smaller scale map'. In the study, the effect of map generalization for GIS analysis would be discussed as the main objective. It was used three digital maps with different scales such as 1:10000, 1:50000 and 1:250000 which were prepared by the Survey Department of Sri Lanka, the National Mapping Agency of Sri Lanka. It was used common features which were on above three maps and an overlay analysis was done by repeating the data with different combinations. Road data, River data and Land use data sets were used for the study. A simple model, to find the best place for a wild life park, was used to identify the effects. The results show remarkable effects on different degrees of generalization processes. It can see that different locations with different geometries were received as the outputs from this analysis. The study suggests that there should be reasonable methods to overcome this effect. It can be recommended that, as a solution, it would be very reasonable to take all the data sets into a common scale and do the analysis part.

Keywords: generalization, GIS, scales, spatial analysis

Procedia PDF Downloads 328

Authors: Korpapa Srisamai, Pawee Siriruk

Abstract:

The purpose of this study is to forecast product demands and develop appropriate and adequate procurement plans to meet customer needs and reduce costs. When the product exceeds customer demands or does not move, it requires the company to support insufficient storage spaces. Moreover, some items, when stored for a long period of time, cause deterioration to dead stock. A case study of the wholesale company of electronic equipment and components, which has uncertain customer demands, is considered. The actual purchasing orders of customers are not equal to the forecast provided by the customers. In some cases, customers have higher product demands, resulting in the product being insufficient to meet the customer's needs. However, some customers have lower demands for products than estimates, causing insufficient storage spaces and dead stock. This study aims to reduce the loss of sales opportunities and the number of remaining goods in the warehouse, citing 30 product samples of the company's most popular products. The data were collected during the duration of the study from January to October 2022. The methods used to forecast are simple moving averages, weighted moving average, and exponential smoothing methods. The economic ordering quantity and reorder point are used to calculate to meet customer needs and track results. The research results are very beneficial to the company. The company can reduce the loss of sales opportunities by 20% so that the company has enough products to meet customer needs and can reduce unused products by up to 10% dead stock. This enables the company to order products more accurately, increasing profits and storage space.

Keywords: demand forecast, reorder point, lost sale, dead stock

Procedia PDF Downloads 121

Authors: Salihah Alghamdi, Surajit Ray

Abstract:

Space-time data can be observed over irregularly shaped manifolds, which might have complex boundaries or interior gaps. Most of the existing methods do not consider the shape of the data, and as a result, it is difficult to model irregularly shaped data accommodating the complex domain. We used a method that can deal with space-time data that are distributed over non-planner shaped regions. The method is based on partial differential equations and finite element analysis. The model can be estimated using a penalized least squares approach with a regularization term that controls the over-fitting. The model is regularized using two roughness penalties, which consider the spatial and temporal regularities separately. The integrated square of the second derivative of the basis function is used as temporal penalty. While the spatial penalty consists of the integrated square of Laplace operator, which is integrated exclusively over the domain of interest that is determined using finite element technique. In this paper, we applied a spatio-temporal regression model with partial differential equations regularization (ST-PDE) approach to analyze a remote sensing data measuring the greenness of vegetation, measure by an index called enhanced vegetation index (EVI). The EVI data consist of measurements that take values between -1 and 1 reflecting the level of greenness of some region over a period of time. We applied (ST-PDE) approach to irregular shaped region of the EVI data. The approach efficiently accommodates the irregular shaped regions taking into account the complex boundaries rather than smoothing across the boundaries. Furthermore, the approach succeeds in capturing the temporal variation in the data.

Keywords: irregularly shaped domain, partial differential equations, finite element analysis, complex boundray

Procedia PDF Downloads 140

Authors: V. Veeraprathap, G. S. Harish, G. Narendra Kumar

Abstract:

Uncontrolled growth of abnormal cells in the lung in the form of tumor can be either benign (non-cancerous) or malignant (cancerous). Patients with Lung Cancer (LC) have an average of five years life span expectancy provided diagnosis, detection and prediction, which reduces many treatment options to risk of invasive surgery increasing survival rate. Computed Tomography (CT), Positron Emission Tomography (PET), and Magnetic Resonance Imaging (MRI) for earlier detection of cancer are common. Gaussian filter along with median filter used for smoothing and noise removal, Histogram Equalization (HE) for image enhancement gives the best results without inviting further opinions. Lung cavities are extracted and the background portion other than two lung cavities is completely removed with right and left lungs segmented separately. Region properties measurements area, perimeter, diameter, centroid and eccentricity measured for the tumor segmented image, while texture is characterized by Gray-Level Co-occurrence Matrix (GLCM) functions, feature extraction provides Region of Interest (ROI) given as input to classifier. Two levels of classifications, K-Nearest Neighbor (KNN) is used for determining patient condition as normal or abnormal, while Artificial Neural Networks (ANN) is used for identifying the cancer stage is employed. Discrete Wavelet Transform (DWT) algorithm is used for the main feature extraction leading to best efficiency. The developed technology finds encouraging results for real time information and on line detection for future research.

Keywords: artificial neural networks, ANN, discrete wavelet transform, DWT, gray-level co-occurrence matrix, GLCM, k-nearest neighbor, KNN, region of interest, ROI

Procedia PDF Downloads 153

Authors: T. Srinivasa Krishna, K. Narendra, K. Thomas, S. S. Raju, B. Munibhadrayya

Abstract:

The densities, speeds of sound and refractive index of the binary mixture of ionic liquid (IL) 1-Butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([BMIM][Imide]) and Pyrrolidin-2-one(PY) was measured at atmospheric pressure, and over the range of temperatures T= (298.15 -323.15)K. The excess molar volume, excess isentropic compressibility, excess speed of sound, partial molar volumes, and isentropic partial molar compressibility were calculated from the values of the experimental density and speed of sound. From the experimental data excess thermal expansion coefficients and isothermal pressure coefficient of excess molar enthalpy at 298.15K were calculated. The results were analyzed and were discussed from the point of view of structural changes. Excess properties were calculated and correlated by the Redlich–Kister and the Legendre polynomial equation and binary coefficients were obtained. Values of excess partial volumes at infinite dilution for the binary system at different temperatures were calculated from the adjustable parameters obtained from Legendre polynomial and Redlich–Kister smoothing equation. Deviation in refractive indices ΔnD and deviation in molar refraction, ΔRm were calculated from the measured refractive index values. Equations of state and several mixing rules were used to predict refractive indices of the binary mixtures and compared with the experimental values by means of the standard deviation and found to be in excellent agreement. By using Prigogine–Flory–Patterson (PFP) theory, the above thermodynamic mixing functions have been calculated and the results obtained from this theory were compared with experimental results.

Keywords: density, refractive index, speeds of sound, Prigogine-Flory-Patterson theory

Procedia PDF Downloads 408

Authors: Seyi Olalekan Olawuyi, Abbyssiania Mushunje

Abstract:

The potential benefits of agricultural production to improve the welfare condition of smallholder farmers in developing countries is no more a news because it has been widely documented. Yet majority of these farming households suffer from shortfall in production output to meet both the consumption needs and market demand which adversely affects output market participation and by extension welfare condition. Therefore, this study investigated the impacts of output market participation on households’ welfare of cassava-based farmers in Oyo State, Nigeria. Multistage sampling technique was used to select 324 sample size used for this study. The findings from the data obtained and analyzed through composite score and crosstab analysis revealed that there is varying degree of output market participation among the farmers which also translate to the observed welfare profile differentials in the study area. The probit model analysis with respect to the selection equation identified gender of household head, household size, access to remittance, off-farm income and ownership of farmland as significant drivers of output market participation in the study area. Furthermore, the treatment effect model of the welfare equation and propensity score matching (PSM) technique were used as robust checks; and the findings attest to the fact that, complimentarily with other significant variables highlighted in this study, output market participation indeed has a significant impact on farming households’ welfare. As policy implication inferences, the study recommends female active inclusiveness and empowerment in farming activities, birth control strategies, secondary income smoothing activities and discouragement of land fragmentation habits, to boost productivity and output market participation, which by extension can significantly improve farming households’ welfare.

Keywords: Cassava market participation, households' welfare, propensity score matching, treatment effect model

Procedia PDF Downloads 162

Authors: Fuyuan Ma, Yuhan Wang, Junhe Zhang, Ying Wang

Abstract:

Identifying the sources of information dissemination is a pivotal task in the study of collective behaviors in networks, enabling us to discern and intercept the critical pathways through which information propagates from its origins. This allows for the control of the information’s dissemination impact in its early stages. Numerous methods for source detection rely on pre-existing, underlying propagation models as prior knowledge. Current models that eschew prior knowledge attempt to harness label propagation algorithms to model the statistical characteristics of propagation states or employ Graph Neural Networks (GNNs) for deep reverse modeling of the diffusion process. These approaches are either deficient in modeling the propagation patterns of information or are constrained by the over-smoothing problem inherent in GNNs, which limits the stacking of sufficient model depth to excavate global propagation patterns. Consequently, we introduce the ODESI model. Initially, the model employs a label propagation algorithm to delineate the distribution density of infected states within a graph structure and extends the representation of infected states from integers to state vectors, which serve as the initial states of nodes. Subsequently, the model constructs a deep architecture based on GNNs-coupled Ordinary Differential Equations (ODEs) to model the global propagation patterns of continuous propagation processes. Addressing the challenges associated with solving ODEs on graphs, we approximate the analytical solutions to reduce computational costs. Finally, we conduct simulation experiments on two real-world social network datasets, and the results affirm the efficacy of our proposed ODESI model in source identification tasks.

Keywords: source identification, ordinary differential equations, label propagation, complex networks

Procedia PDF Downloads 20

Authors: Zhang Jun, Zhai Shucheng, Bai Yaqiang, Zhang Guoping

Abstract:

The Gymnarchus Niioticus fish(GNF) cruises generally with high efficiency by undulating ribbon-fin propulsion while keeping its body for straight line. The swing amplitude of GNF fins is usually in 60° to 90°, and in normal state the amplitude is close to 90°, only in the control of hovering or swimming at very low speed, the amplitude is smaller (about 60°). It provides inspiration for underwater robot design. In the paper, the unsteady flow of undulating ribbon-fin propulsion is numerical simulated by the dynamic grid technique including spring-based smoothing model and local grid remeshing to adapt to the fin surface significantly deforming, and the swing amplitude of fin ray reaches 850. The numerical simulation method is validated by thrust experiments. The spatial vortex structure and its evolution with phase angle is analyzed. The propulsion mechanism is investigated by comprehensive analysis of the hydrodynamics, vortex structure, and pressure distribution on the fin surface. The numerical results indicates that there are mainly three kinds of vortexes, i.e. streamwise vortex, crescent vortex and toroidal vortex. The intensity of streamwise vortex is the strongest among all kinds of vortexes. Streamwise vortexes and crescent vortexes all alternately distribute on the two sides of mid-sagittal plane. Inside the crescent vortexes is high-speed flow, while outside is low-speed flow. The crescent vortexes mainly induce high-speed axial jet, which produces the primary thrust. This is hydrodynamic mechanism undulating ribbon-fin propulsion. The streamwise vortexes mainly induce the vertical jet, which generates the primary heave force. The effect on hydrodynamics of main geometry and movement parameters including wave length, amplitude and advanced coefficients is investigated. A bionic underwater robot with bilateral undulating ribbon-fins is designed, and its navigation performance and maneuverability are measured.

Keywords: bionic propulsion, mobile robot, underwater robot, undulating ribbon-fins

Procedia PDF Downloads 284

Authors: Jurriaan Gillissen

Abstract:

This work is concerned with stabilizing the numerical integration of the Navier-Stokes equation (NSE), backwards in time. Applications involve the detection of sources of, e.g., sound, heat, and pollutants. Stable reverse numerical integration of parabolic differential equations is also relevant for image de-blurring. While the literature addresses the reverse integration problem of the advection-diffusion equation, the problem of numerical reverse integration of the NSE has, to our knowledge, not yet been addressed. Owing to the presence of viscosity, the NSE is irreversible, i.e., when going backwards in time, the fluid behaves, as if it had a negative viscosity. As an effect, perturbations from the perfect solution, due to round off errors or discretization errors, grow exponentially in time, and reverse integration of the NSE is inherently unstable, regardless of using an implicit time integration scheme. Consequently, some sort of filtering is required, in order to achieve a stable, numerical, reversed integration. The challenge is to find a filter with a minimal adverse affect on the accuracy of the reversed integration. In the present work, we explore an adjoint gradient method (AGM) to achieve this goal, and we apply this technique to two-dimensional (2D), decaying turbulence. The AGM solves for the initial velocity field u0 at t = 0, that, when integrated forward in time, produces a final velocity field u1 at t = 1, that is as close as is feasibly possible to some specified target field v1. The initial field u0 defines a minimum of a cost-functional J, that measures the distance between u1 and v1. In the minimization procedure, the u0 is updated iteratively along the gradient of J w.r.t. u0, where the gradient is obtained by transporting J backwards in time from t = 1 to t = 0, using the adjoint NSE. The AGM thus effectively replaces the backward integration by multiple forward and backward adjoint integrations. Since the viscosity is negative in the adjoint NSE, each step of the AGM is numerically stable. Nevertheless, when applied to turbulence, the AGM develops instabilities, which limit the backward integration to small times. This is due to the exponential divergence of phase space trajectories in turbulent flow, which produces a multitude of local minima in J, when the integration time is large. As an effect, the AGM may select unphysical, noisy initial conditions. In order to improve this situation, we propose two remedies. First, we replace the integration by a sequence of smaller integrations, i.e., we divide the integration time into segments, where in each segment the target field v1 is taken as the initial field u0 from the previous segment. Second, we add an additional term (regularizer) to J, which is proportional to a high-order Laplacian of u0, and which dampens the gradients of u0. We show that suitable values for the segment size and for the regularizer, allow a stable reverse integration of 2D decaying turbulence, with accurate results for more then O(10) turbulent, integral time scales.

Keywords: time reversed integration, parabolic differential equations, adjoint gradient method, two dimensional turbulence

Procedia PDF Downloads 224

Authors: Kemal Efe Eseller, Göktuğ Yazici

Abstract:

Laser-induced breakdown spectroscopy (LIBS) is a rapid optical atomic emission spectroscopy which is used for material identification and analysis with the advantages of in-situ analysis, elimination of intensive sample preparation, and micro-destructive properties for the material to be tested. LIBS delivers short pulses of laser beams onto the material in order to create plasma by excitation of the material to a certain threshold. The plasma characteristics, which consist of wavelength value and intensity amplitude, depends on the material and the experiment’s environment. In the present work, medicine samples’ spectrum profiles were obtained via LIBS. Medicine samples’ datasets include two different concentrations for both paracetamol based medicines, namely Aferin and Parafon. The spectrum data of the samples were preprocessed via filling outliers based on quartiles, smoothing spectra to eliminate noise and normalizing both wavelength and intensity axis. Statistical information was obtained and principal component analysis (PCA) was incorporated to both the preprocessed and raw datasets. The machine learning models were set based on two different train-test splits, which were 70% training – 30% test and 80% training – 20% test. Cross-validation was preferred to protect the models against overfitting; thus the sample amount is small. The machine learning results of preprocessed and raw datasets were subjected to comparison for both splits. This is the first time that all supervised machine learning classification algorithms; consisting of Decision Trees, Discriminant, naïve Bayes, Support Vector Machines (SVM), k-NN(k-Nearest Neighbor) Ensemble Learning and Neural Network algorithms; were incorporated to LIBS data of paracetamol based pharmaceutical samples, and their different concentrations on preprocessed and raw dataset in order to observe the effect of preprocessing.

Keywords: machine learning, laser-induced breakdown spectroscopy, medicines, principal component analysis, preprocessing

Procedia PDF Downloads 87

Authors: Emily Injete Amondo, Alisher Mirzabaev, Emmanuel Rukundo

Abstract:

Children in rural farming households are often vulnerable to a multitude of risks, including health risks associated with climate change and variability. Cognizant of this, this study empirically traced the relationship between climate variability and nutritional health outcomes in rural children while identifying the cause-and-effect transmission mechanisms. We combined four waves of the rich Uganda National Panel Survey (UNPS), part of the World Bank Living Standards Measurement Studies (LSMS) for the period 2009-2014, with long-term and high-frequency rainfall and temperature datasets. Self-reported drought and flood shock variables were further used in separate regressions for triangulation purposes and robustness checks. Panel fixed effects regressions were applied in the empirical analysis, accounting for a variety of causal identification issues. The results showed significant negative outcomes for children’s anthropometric measurements due to the impacts of moderate and extreme droughts, extreme wet spells, and heatwaves. On the contrary, moderate wet spells were positively linked with nutritional measures. Agricultural production and child diarrhea were the main transmission channels, with heatwaves, droughts, and high rainfall variability negatively affecting crop output. The probability of diarrhea was positively related to increases in temperature and dry spells. Results further revealed that children in households who engaged in ex-ante or anticipatory risk-reducing strategies such as savings had better health outcomes as opposed to those engaged in ex-post coping such as involuntary change of diet. These results highlight the importance of adaptation in smoothing the harmful effects of climate variability on the health of rural households and children in Uganda.

Keywords: extreme weather events, undernutrition, diarrhea, agricultural production, gridded weather data

Procedia PDF Downloads 102

Authors: Paolo Sassi, Jorge Freiria, Gabriel Usera

Abstract:

In this work, a finite volume fluid flow solver is coupled with a discrete element method module for the simulation of the dynamics of free and elastic bodies in interaction with the fluid and between themselves. The open source fluid flow solver, caffa3d.MBRi, includes the capability to work with nested overlapping grids in order to easily refine the grid in the region where the bodies are moving. To do so, it is necessary to implement a recognition function able to identify the specific mesh block in which the device is moving in. The set of overlapping finer grids might be displaced along with the set of bodies being simulated. The interaction between the bodies and the fluid is computed through a two-way coupling. The velocity field of the fluid is first interpolated to determine the drag force on each object. After solving the objects displacements, subject to the elastic bonding among them, the force is applied back onto the fluid through a Gaussian smoothing considering the cells near the position of each object. The fishnet is represented as lumped masses connected by elastic lines. The internal forces are derived from the elasticity of these lines, and the external forces are due to drag, gravity, buoyancy and the load acting on each element of the system. When solving the ordinary differential equations system, that represents the motion of the elastic and flexible bodies, it was found that the Runge Kutta solver of fourth order is the best tool in terms of performance, but requires a finer grid than the fluid solver to make the system converge, which demands greater computing power. The coupled solver is demonstrated by simulating the interaction between the fluid, an elastic fishnet and a set of free bodies being captured by the net as they are dragged by the fluid. The deformation of the net, as well as the wake produced in the fluid stream are well captured by the method, without requiring the fluid solver mesh to adapt for the evolving geometry. Application of the same strategy to the simulation of elastic structures subject to the action of wind is also possible with the method presented, and one such application is currently under development.

Keywords: computational fluid dynamics, discrete element method, fishnets, nested overlapping grids

Procedia PDF Downloads 416

Authors: Mehri Sadat Alavinasab Ashgezari, Gholam Reza Nabi Bidhendi, Fatemeh Sadat Alavinasab Ashkezari

Abstract:

Atmospheric aerosol loading (AAL) from anthropogenic sources is an evidence in industrial development. The accelerated trends in material consumption at the global scale in recent years demonstrate consumption paradigms sensible to the planetary boundaries (PB). This paper is a statistical approach on recognizing the path of climate-relevant bulk materials production (CBMP) of steel, cement and plastics to AAL via an updated and validated spatiotemporal distribution. The methodology of statistical analysis used the most updated regional or global databases or instrumental technologies. This corresponded to a selection of processes and areas capable for tracking AAL within the last decade, analyzing the most validated data while leading to explore the behavior functions or models. The results also represented a correlation within socio economic metabolism idea between the materials specified as macronutrients of society and AAL as a PB with an unknown threshold. The selected country contributors of China, India, US and the sample country of Iran show comparable cumulative AAL values vs to the bulk materials domestic extraction and production rate in the study period of 2012 to 2022. Generally, there is a tendency towards gradual descend in the worldwide and regional aerosol concentration after 2015. As of our evaluation, a considerable share of human role, equivalent 20% from CBMP, is for the main anthropogenic species of aerosols, including sulfate, black carbon and organic particulate matters too. This study, in an innovative approach, also explores the potential role of AAL control mechanisms from the economy sectors where ordered and smoothing loading trends are accredited through the disordered phenomena of CBMP and aerosol precursor emissions. The equilibrium states envisioned is an approval to the well-established theory of Spin Glasses applicable in physical system like the Earth and here to AAL.

Keywords: atmospheric aeroso loading, material flows, climate bulk materials, industrial ecology

Procedia PDF Downloads 80

Authors: Seema Paul, Jesper Oppelstrup, Roger Thunvik, Vladimir Cvetkovic

Abstract:

Lake Victoria is the second largest fresh water body in the world, located in East Africa with a catchment area of 250,000 km², of which 68,800 km² is the actual lake surface. The hydrodynamic processes of the shallow (40–80 m deep) water system are unique due to its location at the equator, which makes Coriolis effects weak. The paper describes a St.Venant shallow water model of Lake Victoria developed in COMSOL Multiphysics software, a general purpose finite element tool for solving partial differential equations. Depth soundings taken in smaller parts of the lake were combined with recent more extensive data to resolve the discrepancies of the lake shore coordinates. The topography model must have continuous gradients, and Delaunay triangulation with Gaussian smoothing was used to produce the lake depth model. The model shows large-scale flow patterns, passive tracer concentration and water level variations in response to river and tracer inflow, rain and evaporation, and wind stress. Actual data of precipitation, evaporation, in- and outflows were applied in a fifty-year simulation model. It should be noted that the water balance is dominated by rain and evaporation and model simulations are validated by Matlab and COMSOL. The model conserves water volume, the celerity gradients are very small, and the volume flow is very slow and irrotational except at river mouths. Numerical experiments show that the single outflow can be modelled by a simple linear control law responding only to mean water level, except for a few instances. Experiments with tracer input in rivers show very slow dispersion of the tracer, a result of the slow mean velocities, in turn, caused by the near-balance of rain with evaporation. The numerical and hydrodynamical model can evaluate the effects of wind stress which is exerted by the wind on the lake surface that will impact on lake water level. Also, model can evaluate the effects of the expected climate change, as manifest in changes to rainfall over the catchment area of Lake Victoria in the future.

Keywords: bathymetry, lake flow and steady state analysis, water level validation and concentration, wind stress

Procedia PDF Downloads 227

Authors: José M. Liñeira Del Río, Ramón Rial, Khodor Nasser, María J.G. Guimarey

Abstract:

The need to develop new lubricants that offer better anti-friction and anti-wear performance in internal combustion vehicles is one of the great challenges of lubrication in the automotive field. The addition of nanoparticles has emerged as a possible solution and, combined with the lubricating power of ionic liquids, may become one of the alternatives to reduce friction losses and wear of the contact surfaces in the conditions to which tribo-pairs are subjected, especially in the contact of the piston rings and the cylinder liner surface. In this study, the improvement in SAE 10W-40 engine oil tribological performance after the addition of magnesium oxide (MgO) nanoadditives and two different phosphonium-based ionic liquids (ILs) was investigated. The nanoparticle characterization was performed by means of transmission electron microscopy (TEM), Fourier-transform infrared spectroscopy (FTIR), Raman spectroscopy, X-ray diffraction (XRD), and scanning electron microscopy (SEM). The tribological properties, friction coefficients and wear parameters of the formulated oil modified with 0.01 wt.% MgO and 1 wt.% ILs compared with the neat 10W-40 oil were performed and analyzed using a ball-on-three-pins tribometer and a 3D optical profilometer, respectively. Further analysis on the worn surface was carried out by Raman spectroscopy and SEM microscopy, illustrating the formation of the protective IL and MgO tribo-films as hybrid additives. In friction tests with sliding steel-steel tribo-pairs, IL3-based hybrid nanolubricant decreased the friction coefficient and wear volume by 7% and 59%, respectively, in comparison with the neat SAE 10W-40, while the one based on IL1 only achieved a reduction of these parameters by 6% and 39%, respectively. Thus, the tribological characterization also revealed that the MgO and IL3 addition has a positive synergy over the commercial lubricant, adequately meeting the requirements for their use in internal combustion engines. In summary, this study has shown that the addition of ionic liquids to MgO nanoparticles can improve the stability and lubrication behavior of MgO nanolubricant and encourages more investigations on using nanoparticle additives with green solvents such as ionic liquids to protect the environment as well as prolong the lifetime of machinery. The improvement in the lubricant properties was attributed to the following wear mechanisms: the formation of a protective tribo-film and the ability of nanoparticles to fill out valleys between asperities, thereby effectively smoothing out the shearing surfaces.

Keywords: lubricant, nanoparticles, phosphonium-based ionic liquids, tribology

Procedia PDF Downloads 82

Authors: Adel S. Alanazi

Abstract:

The transition of a student with a disability from school to work is the most crucial phase while moving from the stage of adolescence into early adulthood. In this process, young individuals face various difficulties and challenges in order to accomplish the next venture of life successfully. In this respect, this paper aims to examine the challenges encountered by the individuals with intellectual disabilities in transition to work in Saudi Arabia. For this purpose, this study has undertaken a qualitative research-based methodology; wherein interpretivist philosophy has been followed along with inductive approach and exploratory research design. The data for the research has been gathered with the help of semi-structured interviews, whose findings are analysed with the help of thematic analysis. Semi-structured interviews were conducted with parents of persons with intellectual disabilities, officials, supervisors and specialists of two vocational rehabilitation centres providing training to intellectually disabled students, in addition to that, directors of companies and websites in hiring those individuals. The total number of respondents for the interview was 15. The purposive sampling method was used to select the respondents for the interview. This sampling method is a non-probability sampling method which draws respondents from a known population and allows flexibility and suitability in selecting the participants for the study. The findings gathered from the interview revealed that the lack of awareness among their parents regarding the rights of their children who are intellectually disabled; the lack of adequate communication and coordination between various entities; concerns regarding their training and subsequent employment are the key difficulties experienced by the individuals with intellectual disabilities. Training in programmes such as bookbinding, carpentry, computing, agriculture, electricity and telephone exchange operations were involved as key training programmes. The findings of this study also revealed that information technology and media were playing a significant role in smoothing the transition to employment of individuals with intellectual disabilities. Furthermore, religious and cultural attitudes have been identified to be restricted for people with such disabilities in seeking advantages from job opportunities. On the basis of these findings, it can be implied that the information gathered through this study will serve to be highly beneficial for Saudi Arabian schools/ rehabilitation centres for individuals with intellectual disability to facilitate them in overcoming the problems they encounter during the transition to work.

Keywords: intellectual disability, transition services, rehabilitation centre, employment

Procedia PDF Downloads 159

Authors: Antoine Lauvray, Fabien Poulhaon, Pierre Michaud, Pierre Joyot, Emmanuel Duc

Abstract:

Additive Friction Stir Manufacturing (AFSM) is a new industrial process that follows the emergence of friction-based processes. The AFSM process is a solid-state additive process using the energy produced by the friction at the interface between a rotating non-consumable tool and a substrate. Friction depends on various parameters like axial force, rotation speed or friction coefficient. The feeder material is a metallic rod that flows through a hole in the tool. Unlike in Friction Stir Welding (FSW) where abundant literature exists and addresses many aspects going from process implementation to characterization and modeling, there are still few research works focusing on AFSM. Therefore, there is still a lack of understanding of the physical phenomena taking place during the process. This research work aims at a better AFSM process understanding and implementation, thanks to numerical simulation and experimental validation performed on a prototype effector. Such an approach is considered a promising way for studying the influence of the process parameters and to finally identify a process window that seems relevant. The deposition of material through the AFSM process takes place in several phases. In chronological order these phases are the docking phase, the dwell time phase, the deposition phase, and the removal phase. The present work focuses on the dwell time phase that enables the temperature rise of the system composed of the tool, the filler material, and the substrate and due to pure friction. Analytic modeling of heat generation based on friction considers as main parameters the rotational speed and the contact pressure. Another parameter considered influential is the friction coefficient assumed to be variable due to the self-lubrication of the system with the rise in temperature or the materials in contact roughness smoothing over time. This study proposes, through numerical modeling followed by experimental validation, to question the influence of the various input parameters on the dwell time phase. Rotation speed, temperature, spindle torque, and axial force are the main monitored parameters during experimentations and serve as reference data for the calibration of the numerical model. This research shows that the geometry of the tool as well as fluctuations of the input parameters like axial force and rotational speed are very influential on the temperature reached and/or the time required to reach the targeted temperature. The main outcome is the prediction of a process window which is a key result for a more efficient process implementation.

Keywords: numerical model, additive manufacturing, friction, process

Procedia PDF Downloads 147

Authors: Itissam Abuiziah

Abstract:

This work presents a case study of water-hammer analysis and control for the Oum Azza pumping station project in the coastal area of Rabat to Casablanca from the dam Sidi Mohamed Ben Abdellah (SMBA). This is a typical pumping system with a long penstock and is currently at design and executions stages. Since there is no ideal location for construction of protection devices, the protection devices were provisionally designed to protect the whole conveying pipeline. The simulation results for the transient conditions caused by a sudden pumping stopping without including any protection devices, show that there is a negative beyond 1300m to the station 5725m near the arrival of the reservoir, therefore; there is a need for the protection devices to protect the conveying pipeline. To achieve the goal behind the transient flow analysis which is to protect the conveying pipeline system, four scenarios had been investigated in this case study with two types of protecting devices (pressure relief valve and desurging tank with automatic air control). The four scenarios are conceders as with pressure relief valve, with pressure relief valve and a desurging tank with automatic air control, with pressure relief valve and tow desurging tanks with automatic air control and with pressure relief valve and three desurging tanks with automatic air control. The simulation result for the first scenario shows that overpressure corresponding to an instant pumping stopping is reduced from 263m to 240m, and the minimum hydraulic grad line for the length approximately from station 1300m to station 5725m is still below the pipeline profile which means that the pipe must be equipped with another a protective devices for smoothing depressions. The simulation results for the second scenario show that the minimum and maximum pressures envelopes are decreases especially in the depression phase but not effectively protects the conduct in this case study. The minimum pressure increased from -77.7m for the previous scenario to -65.9m for the current scenario. Therefore the pipeline is still requiring additional protective devices; another desurging tank with automatic air control is installed at station2575.84m. The simulation results for the third scenario show that the minimum and maximum pressures envelopes are decreases but not effectively protects the conduct in this case study since the depression is still exist and varies from -0.6m to– 12m. Therefore the pipeline is still requiring additional protective devices; another desurging tank with automatic air control is installed at station 5670.32 m. Examination of the envelope curves of the minimum pressuresresults for the fourth scenario, we noticed that the piezometric pressure along the pipe remains positive over the entire length of the pipe. We can, therefore, conclude that such scenario can provide effective protection for the pipeline.

Keywords: analysis methods, protection devices, transient flow, water hammer

Procedia PDF Downloads 187

Authors: Anastasios Georgoulas, Manolia Andredaki, Marco Marengo

Abstract:

The present paper investigates the hydrodynamics and heat transfer characteristics of slug-plug flows under saturated flow boiling conditions within circular micro-channels. Numerical simulations are carried out, using an enhanced version of the open-source CFD-based solver ‘interFoam’ of OpenFOAM CFD Toolbox. The proposed user-defined solver is based in the Volume Of Fluid (VOF) method for interface advection, and the mentioned enhancements include the implementation of a smoothing process for spurious current reduction, the coupling with heat transfer and phase change as well as the incorporation of conjugate heat transfer to account for transient solid conduction. In all of the considered cases in the present paper, a single phase simulation is initially conducted until a quasi-steady state is reached with respect to the hydrodynamic and thermal boundary layer development. Then, a predefined and constant frequency of successive vapour bubbles is patched upstream at a certain distance from the channel inlet. The proposed numerical simulation set-up can capture the main hydrodynamic and heat transfer characteristics of slug-plug flow regimes within circular micro-channels. In more detail, the present investigation is focused on exploring the interaction between subsequent vapour slugs with respect to their generation frequency, the hydrodynamic characteristics of the liquid film between the generated vapour slugs and the channel wall as well as of the liquid plug between two subsequent vapour slugs. The proposed investigation is carried out for the 3 different working fluids and three different values of applied heat flux in the heated part of the considered microchannel. The post-processing and analysis of the results indicate that the dynamics of the evolving bubbles in each case are influenced by both the upstream and downstream bubbles in the generated sequence. In each case a slip velocity between the vapour bubbles and the liquid slugs is evident. In most cases interfacial waves appear close to the bubble tail that significantly reduce the liquid film thickness. Finally, in accordance with previous investigations vortices that are identified in the liquid slugs between two subsequent vapour bubbles can significantly enhance the convection heat transfer between the liquid regions and the heated channel walls. The overall results of the present investigation can be used to enhance the present understanding by providing better insight of the complex, underpinned heat transfer mechanisms in saturated boiling within micro-channels in the slug-plug flow regime.

Keywords: slug-plug flow regime, micro-channels, VOF method, OpenFOAM

Procedia PDF Downloads 267

Authors: Michael McCann

Abstract:

Managerial theories propose that, in joint stock companies, executives may be tempted to waste excess free cash on unprofitable projects to keep control of resources. In order to conceal their projects' poor performance, they may seek to engage in earnings management. On the one hand, managers may manipulate earnings upwards in order to post ‘good’ performances and safeguard their position. On the other, since managers pursuit of unrewarding investments are likely to lead to low long-term profitability, managers will use negative accruals to reduce current year’s earnings, smoothing earnings over time in order to conceal the negative effects. Agency models argue that boards of directors are delegated by shareholders to ensure that companies are governed properly. Part of that responsibility is ensuring the reliability of financial information. Analyses of the impact of board characteristics, particularly board independence on the misuse of free cash flow and earnings management finds conflicting evidence. However, existing characterizations of board independence do not account for such directors gaining firm-specific knowledge over time, influencing their monitoring ability. Further, there is little analysis of the influence of the relative experience of independent directors and executives on decisions surrounding the use of free cash. This paper contributes to this literature regarding the heterogeneous characteristics of boards by investigating the influence of independent director tenure on earnings management and the relative tenures of independent directors and Chief Executives. A balanced panel dataset comprising 51 companies across 11 annual periods from 2005 to 2015 is used for the analysis. In each annual period, firms were classified as conducting earnings management if they had discretionary accruals in the bottom quartile (downwards) and top quartile (upwards) of the distributed values for the sample. Logistical regressions were conducted to determine the marginal impact of independent board tenure and a number of control variables on the probability of conducting earnings management. The findings indicate that both absolute and relative measures of board independence and experience do not have a significant impact on the likelihood of earnings management. It is the level of free cash flow which is the major influence on the probability of earnings management. Higher free cash flow increases the probability of earnings management significantly. The research also investigates whether board monitoring of earnings management is contingent on the level of free cash flow. However, the results suggest that board monitoring is not amplified when free cash flow is higher. This suggests that the extent of earnings management in companies is determined by a range of company, industry and situation-specific factors.

Keywords: corporate governance, boards of directors, agency theory, earnings management

Procedia PDF Downloads 233

Authors: M. Zasada, A. Markiewicz, A. Erkiert-Polguj, E. Budzisz

Abstract:

The epidermis is a keratinized stratified squamous epithelium covered by the hydro-lipid barrier. Therefore, active substances should be able to penetrate through this hydro-lipid coating. L-ascorbic acid is one of the vitamins which plays an important role in stimulation fibroblast to produce collagen type I and in hyperpigmentation lightening. Vitamin C is a water-soluble antioxidant, which protects skin from oxidation damage and rejuvenates photoaged skin. No-needle mesotherapy is a non-invasive rejuvenation technique depending on electric pulses, electroporation, and ultrasounds. These physicals factors result in deeper penetration of cosmetics. It is important to increase the penetration of L-ascorbic acid, thereby increasing the spectrum of its activity. The aim of the work was to assess the effectiveness of pure L-ascorbic acid activity in anti-aging therapy using a needle-free and micro-needling mesotherapy. The study was performed on a group of 35 healthy volunteers in accordance with the Declaration of Helsinki of 1964 and agreement of the Ethics Commissions no RNN/281/16/KE 2017. Women were randomized to mesotherapy or control group. Control group applied topically 2,5 ml serum containing 20% L-ascorbic acid with hydrate from strawberries, every 10 days for a period of 9 weeks. No-needle mesotherapy, on the left half of the face and micro-needling on the right with the same serum, was done in mesotherapy group. The pH of serum was 3.5-4, and the serum was prepared directly prior to the facial treatment. The skin parameters were measured at the beginning and before each treatment. The measurement of the forehead skin was done using Cutometer® (measurement of skin elasticity and firmness), Corneometer® (skin hydration measurement), Mexameter® (skin tone measurement). Also, the photographs were taken by Fotomedicus system. Additionally, the volunteers fulfilled the questionnaire. Serum was tested for microbiological purity and stability after the opening of the cosmetic. During the study, all of the volunteers were taken care of a dermatologist. The regular application of the serum has caused improvement of the skin parameters. Respectively, after 4 and 8 weeks improvement in hydration and elasticity has been seen (Corneometer®, Cutometer® results). Moreover, the number of hyper-pigmentated spots has decreased (Mexameter®). After 8 weeks the volunteers has claimed that the tested product has smoothing and moisturizing features. Subjective opinions indicted significant improvement of skin color and elasticity. The product containing the L-ascorbic acid used with intercellular penetration promoters demonstrates higher anti-aging efficiency than control. In vivo studies confirmed the effectiveness of serum and the impact of the active substance on skin firmness and elasticity, the degree of hydration and skin tone. Mesotherapy with pure L-ascorbic acid provides better diffusion of active substances through the skin.

Keywords: anti-aging, l-ascorbic acid, mesotherapy, promoters

Procedia PDF Downloads 265

Authors: Kamyar Tolouei, Ehsan Moosavi

Abstract:

In open-pit mines, the long-term production scheduling optimization problem (LTPSOP) is a complicated problem that contains constraints, large datasets, and uncertainties. Uncertainty in the output is caused by several geological, economic, or technical factors. Due to its dimensions and NP-hard nature, it is usually difficult to find an ideal solution to the LTPSOP. The optimal schedule generally restricts the ore, metal, and waste tonnages, average grades, and cash flows of each period. Past decades have witnessed important measurements of long-term production scheduling and optimal algorithms since researchers have become highly cognizant of the issue. In fact, it is not possible to consider LTPSOP as a well-solved problem. Traditional production scheduling methods in open-pit mines apply an estimated orebody model to produce optimal schedules. The smoothing result of some geostatistical estimation procedures causes most of the mine schedules and production predictions to be unrealistic and imperfect. With the expansion of simulation procedures, the risks from grade uncertainty in ore reserves can be evaluated and organized through a set of equally probable orebody realizations. In this paper, to synthesize grade uncertainty into the strategic mine schedule, a stochastic integer programming framework is presented to LTPSOP. The objective function of the model is to maximize the net present value and minimize the risk of deviation from the production targets considering grade uncertainty simultaneously while satisfying all technical constraints and operational requirements. Instead of applying one estimated orebody model as input to optimize the production schedule, a set of equally probable orebody realizations are applied to synthesize grade uncertainty in the strategic mine schedule and to produce a more profitable and risk-based production schedule. A mixture of metaheuristic procedures and mathematical methods paves the way to achieve an appropriate solution. This paper introduced a hybrid model between the augmented Lagrangian relaxation (ALR) method and the metaheuristic algorithm, the Harris Hawks optimization (HHO), to solve the LTPSOP under grade uncertainty conditions. In this study, the HHO is experienced to update Lagrange coefficients. Besides, a machine learning method called Random Forest is applied to estimate gold grade in a mineral deposit. The Monte Carlo method is used as the simulation method with 20 realizations. The results specify that the progressive versions have been considerably developed in comparison with the traditional methods. The outcomes were also compared with the ALR-genetic algorithm and ALR-sub-gradient. To indicate the applicability of the model, a case study on an open-pit gold mining operation is implemented. The framework displays the capability to minimize risk and improvement in the expected net present value and financial profitability for LTPSOP. The framework could control geological risk more effectively than the traditional procedure considering grade uncertainty in the hybrid model framework.

Keywords: grade uncertainty, metaheuristic algorithms, open-pit mine, production scheduling optimization

Procedia PDF Downloads 105

Authors: V. Saccà, A. Sarica, F. Novellino, S. Barone, T. Tallarico, E. Filippelli, A. Granata, P. Valentino, A. Quattrone

Abstract:

The work aim was to evaluate how well Random Forest (RF) and Support Vector Machine (SVM) algorithms could support the early diagnosis of Multiple Sclerosis (MS) from resting-state functional connectivity data. In particular, we wanted to explore the ability in distinguishing between controls and patients of mean signals extracted from ICA components corresponding to 15 well-known networks. Eighteen patients with early-MS (mean-age 37.42±8.11, 9 females) were recruited according to McDonald and Polman, and matched for demographic variables with 19 healthy controls (mean-age 37.55±14.76, 10 females). MRI was acquired by a 3T scanner with 8-channel head coil: (a)whole-brain T1-weighted; (b)conventional T2-weighted; (c)resting-state functional MRI (rsFMRI), 200 volumes. Estimated total lesion load (ml) and number of lesions were calculated using LST-toolbox from the corrected T1 and FLAIR. All rsFMRIs were pre-processed using tools from the FMRIB's Software Library as follows: (1) discarding of the first 5 volumes to remove T1 equilibrium effects, (2) skull-stripping of images, (3) motion and slice-time correction, (4) denoising with high-pass temporal filter (128s), (5) spatial smoothing with a Gaussian kernel of FWHM 8mm. No statistical significant differences (t-test, p < 0.05) were found between the two groups in the mean Euclidian distance and the mean Euler angle. WM and CSF signal together with 6 motion parameters were regressed out from the time series. We applied an independent component analysis (ICA) with the GIFT-toolbox using the Infomax approach with number of components=21. Fifteen mean components were visually identified by two experts. The resulting z-score maps were thresholded and binarized to extract the mean signal of the 15 networks for each subject. Statistical and machine learning analysis were then conducted on this dataset composed of 37 rows (subjects) and 15 features (mean signal in the network) with R language. The dataset was randomly splitted into training (75%) and test sets and two different classifiers were trained: RF and RBF-SVM. We used the intrinsic feature selection of RF, based on the Gini index, and recursive feature elimination (rfe) for the SVM, to obtain a rank of the most predictive variables. Thus, we built two new classifiers only on the most important features and we evaluated the accuracies (with and without feature selection) on test-set. The classifiers, trained on all the features, showed very poor accuracies on training (RF:58.62%, SVM:65.52%) and test sets (RF:62.5%, SVM:50%). Interestingly, when feature selection by RF and rfe-SVM were performed, the most important variable was the sensori-motor network I in both cases. Indeed, with only this network, RF and SVM classifiers reached an accuracy of 87.5% on test-set. More interestingly, the only misclassified patient resulted to have the lowest value of lesion volume. We showed that, with two different classification algorithms and feature selection approaches, the best discriminant network between controls and early MS, was the sensori-motor I. Similar importance values were obtained for the sensori-motor II, cerebellum and working memory networks. These findings, in according to the early manifestation of motor/sensorial deficits in MS, could represent an encouraging step toward the translation to the clinical diagnosis and prognosis.

Keywords: feature selection, machine learning, multiple sclerosis, random forest, support vector machine

Procedia PDF Downloads 240

Authors: E. K. Tripsanas, D. Spanos, I. Oikonomopoulos, K. Stathopoulou, A. S. Abdelsamad, A. Pagoulatos

Abstract:

Patraikos Gulf is located offshore western Greece, and it is limited to the west by the Zante, Cephalonia, and Lefkas islands. The Plio/Pleistocene sequence is characterized by two depocenters, the east and west Patraikos basins separated from each other by a prominent sill. This study is based on the Plio/Pleistocene seismic stratigraphy analysis of a newly acquired 3D PSDM (Pre-Stack depth migration) seismic survey in the west Patraikos Basin and few 2D seismic profiles throughout the entire Patraikos Gulf. The eastern Patraikos Basin, although completely buried today with water depths less than 100 m, it was a deep basin during Pliocene ( > 2 km of Pliocene-Pleistocene sediments) and appears to have gathered most of Achelous River discharges. The west Patraikos Gulf was shallower ( < 1300 m of Pliocene-Pleistocene sediments) and characterized by a hummocky relief due to thrust-belt tectonics and Miocene to Pleistocene halokinetic processes. The transition from Pliocene to Miocene is expressed by a widespread erosional unconformity with evidence of fluvial drainage patterns. This indicates that west Patraikos Basin was aerially exposed during the Messinian Salinity Crisis. Continuous to semi-continuous, parallel reflections in the lower, early- to mid-Pliocene seismic packet provides evidence that the re-connection of the Mediterranean Sea with the Atlantic Ocean during Zanclean resulted in the flooding of the west Patraikos basin and the domination of hemipelagic sedimentation interrupted by occasional gravity flows. This is evident in amplitude and semblance horizon slices, which clearly show the presence of long-running, meandering submarine channels sourced from the southeast (northwest Peloponnese) and north. The long-running nature of the submarine channels suggests mobile efficient turbidity currents, probably due to the participation of a sufficient amount of clay minerals in their suspended load. The upper seismic section in the study area mainly consists of several successions of clinoforms, interpreted as progradational delta complexes of Achelous River. This sudden change from marine to shallow marine sedimentary processes is attributed to climatic changes and eustatic perturbations since late Pliocene onwards (~ 2.6 Ma) and/or a switch of Achelous River from the east Patraikos Basin to the west Patraikos Basin. The deltaic seismic unit consists of four delta complexes. The first two complexes result in the infill of topographic depressions and smoothing of an initial hummocky bathymetry. The distribution of the upper two delta complexes is controlled by compensational stacking. Amplitude and semblance horizon slices depict the development of several almost straight and short (a few km long) distributary submarine channels at the delta slopes and proximal prodeltaic plains with lobate sand-sheet deposits at their mouths. Such channels are interpreted to result from low-efficiency turbidity currents with low content in clay minerals. Such a differentiation in the nature of the gravity flows is attributed to the switch of the sediment supply from clay-rich sediments derived from the draining of flysch formations of the Ionian and Gavrovo zones, to the draining of poor in clay minerals carbonate formations of Gavrovo zone through the Achelous River.

Keywords: sequence stratigraphy, basin analysis, river deltas, submarine channels

Procedia PDF Downloads 322

Authors: Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Abstract:

Time series data consist of successive data points collected over a period of time. Accurate prediction of future values is essential for informed decision-making in several real-world applications, including electricity load demand forecasting, lifetime estimation of industrial machinery, traffic planning, weather prediction, and the stock market. Due to their critical relevance and wide application, there has been considerable interest in time series forecasting in recent years. However, the proliferation of sensors and IoT devices, real-time monitoring systems, and high-frequency trading data introduce significant intricate temporal variations, rapid changes, noise, and non-linearities, making time series forecasting more challenging. Classical methods such as Autoregressive integrated moving average (ARIMA) and Exponential Smoothing aim to extract pre-defined temporal variations, such as trends and seasonality. While these methods are effective for capturing well-defined seasonal patterns and trends, they often struggle with more complex, non-linear patterns present in real-world time series data. In recent years, deep learning has made significant contributions to time series forecasting. Recurrent Neural Networks (RNNs) and their variants, such as Long short-term memory (LSTMs) and Gated Recurrent Units (GRUs), have been widely adopted for modeling sequential data. However, they often suffer from the locality, making it difficult to capture local trends and rapid fluctuations. Convolutional Neural Networks (CNNs), particularly Temporal Convolutional Networks (TCNs), leverage convolutional layers to capture temporal dependencies by applying convolutional filters along the temporal dimension. Despite their advantages, TCNs struggle with capturing relationships between distant time points due to the locality of one-dimensional convolution kernels. Transformers have revolutionized time series forecasting with their powerful attention mechanisms, effectively capturing long-term dependencies and relationships between distant time points. However, the attention mechanism may struggle to discern dependencies directly from scattered time points due to intricate temporal patterns. Lastly, Multi-Layer Perceptrons (MLPs) have also been employed, with models like N-BEATS and LightTS demonstrating success. Despite this, MLPs often face high volatility and computational complexity challenges in long-horizon forecasting. To address intricate temporal variations in time series data, this study introduces Times2D, a novel framework that parallelly integrates 2D spectrogram and derivative heatmap techniques. The spectrogram focuses on the frequency domain, capturing periodicity, while the derivative patterns emphasize the time domain, highlighting sharp fluctuations and turning points. This 2D transformation enables the utilization of powerful computer vision techniques to capture various intricate temporal variations. To evaluate the performance of Times2D, extensive experiments were conducted on standard time series datasets and compared with various state-of-the-art algorithms, including DLinear (2023), TimesNet (2023), Non-stationary Transformer (2022), PatchTST (2023), N-HiTS (2023), Crossformer (2023), MICN (2023), LightTS (2022), FEDformer (2022), FiLM (2022), SCINet (2022a), Autoformer (2021), and Informer (2021) under the same modeling conditions. The initial results demonstrated that Times2D achieves consistent state-of-the-art performance in both short-term and long-term forecasting tasks. Furthermore, the generality of the Times2D framework allows it to be applied to various tasks such as time series imputation, clustering, classification, and anomaly detection, offering potential benefits in any domain that involves sequential data analysis.

Keywords: derivative patterns, spectrogram, time series forecasting, times2D, 2D representation

Procedia PDF Downloads 42