Search results for: linear%20adsorption%3B%20chemical%20heterogeneity%3Bdispersion%3B%20fixed%20bed%3B%20porous%20media
1491 On the Algorithmic Iterative Solutions of Conjugate Gradient, Gauss-Seidel and Jacobi Methods for Solving Systems of Linear Equations
Authors: H. D. Ibrahim, H. C. Chinwenyi, H. N. Ude
Abstract:
In this paper, efforts were made to examine and compare the algorithmic iterative solutions of conjugate gradient method as against other methods such as Gauss-Seidel and Jacobi approaches for solving systems of linear equations of the form Ax = b, where A is a real n x n symmetric and positive definite matrix. We performed algorithmic iterative steps and obtained analytical solutions of a typical 3 x 3 symmetric and positive definite matrix using the three methods described in this paper (Gauss-Seidel, Jacobi and Conjugate Gradient methods) respectively. From the results obtained, we discovered that the Conjugate Gradient method converges faster to exact solutions in fewer iterative steps than the two other methods which took much iteration, much time and kept tending to the exact solutions.
Keywords: conjugate gradient, linear equations, symmetric and positive definite matrix, Gauss-Seidel, Jacobi, algorithm
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4591490 Convergence Analysis of Training Two-Hidden-Layer Partially Over-Parameterized ReLU Networks via Gradient Descent
Authors: Zhifeng Kong
Abstract:
Over-parameterized neural networks have attracted a great deal of attention in recent deep learning theory research, as they challenge the classic perspective of over-fitting when the model has excessive parameters and have gained empirical success in various settings. While a number of theoretical works have been presented to demystify properties of such models, the convergence properties of such models are still far from being thoroughly understood. In this work, we study the convergence properties of training two-hidden-layer partially over-parameterized fully connected networks with the Rectified Linear Unit activation via gradient descent. To our knowledge, this is the first theoretical work to understand convergence properties of deep over-parameterized networks without the equally-wide-hidden-layer assumption and other unrealistic assumptions. We provide a probabilistic lower bound of the widths of hidden layers and proved linear convergence rate of gradient descent. We also conducted experiments on synthetic and real-world datasets to validate our theory.Keywords: Over-parameterization, Rectified Linear Units (ReLU), convergence, gradient descent, neural networks.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 8811489 New Efficient Iterative Optimization Algorithm to Design the Two Channel QMF Bank
Authors: Ram Kumar Soni, Alok Jain, Rajiv Saxena
Abstract:
This paper proposes an efficient method for the design of two channel quadrature mirror filter (QMF) bank. To achieve minimum value of reconstruction error near to perfect reconstruction, a linear optimization process has been proposed. Prototype low pass filter has been designed using Kaiser window function. The modified algorithm has been developed to optimize the reconstruction error using linear objective function through iteration method. The result obtained, show that the performance of the proposed algorithm is better than that of the already exists methods.Keywords: Filterbank, near perfect reconstruction, Kaiserwindow, QMF.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16641488 Face Recognition using Radial Basis Function Network based on LDA
Authors: Byung-Joo Oh
Abstract:
This paper describes a method to improve the robustness of a face recognition system based on the combination of two compensating classifiers. The face images are preprocessed by the appearance-based statistical approaches such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). LDA features of the face image are taken as the input of the Radial Basis Function Network (RBFN). The proposed approach has been tested on the ORL database. The experimental results show that the LDA+RBFN algorithm has achieved a recognition rate of 93.5%
Keywords: Face recognition, linear discriminant analysis, radial basis function network.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 21091487 Adaptive Kaman Filter for Fault Diagnosis of Linear Parameter-Varying Systems
Authors: Rajamani Doraiswami, Lahouari Cheded
Abstract:
Fault diagnosis of Linear Parameter-Varying (LPV) system using an adaptive Kalman filter is proposed. The LPV model is comprised of scheduling parameters, and the emulator parameters. The scheduling parameters are chosen such that they are capable of tracking variations in the system model as a result of changes in the operating regimes. The emulator parameters, on the other hand, simulate variations in the subsystems during the identification phase and have negligible effect during the operational phase. The nominal model and the influence vectors, which are the gradient of the feature vector respect to the emulator parameters, are identified off-line from a number of emulator parameter perturbed experiments. A Kalman filter is designed using the identified nominal model. As the system varies, the Kalman filter model is adapted using the scheduling variables. The residual is employed for fault diagnosis. The proposed scheme is successfully evaluated on simulated system as well as on a physical process control system.Keywords: Keywords—Identification, linear parameter-varying systems, least-squares estimation, fault diagnosis, Kalman filter, emulators
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 12931486 Ride Control of Passenger Cars with Semi-active Suspension System Using a Linear Quadratic Regulator and Hybrid Optimization Algorithm
Authors: Ali Fellah Jahromi, Wen Fang Xie, Rama B. Bhat
Abstract:
A semi-active control strategy for suspension systems of passenger cars is presented employing Magnetorheological (MR) dampers. The vehicle is modeled with seven DOFs including the, roll pitch and bounce of car body, and the vertical motion of the four tires. In order to design an optimal controller based on the actuator constraints, a Linear-Quadratic Regulator (LQR) is designed. The design procedure of the LQR consists of selecting two weighting matrices to minimize the energy of the control system. This paper presents a hybrid optimization procedure which is a combination of gradient-based and evolutionary algorithms to choose the weighting matrices with regards to the actuator constraint. The optimization algorithm is defined based on maximum comfort and actuator constraints. It is noted that utilizing the present control algorithm may significantly reduce the vibration response of the passenger car, thus, providing a comfortable ride.Keywords: Full car model, Linear Quadratic Regulator, Sequential Quadratic Programming, Genetic Algorithm
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 29241485 Conditions on Blind Source Separability of Linear FIR-MIMO Systems with Binary Inputs
Authors: Jiashan Tang
Abstract:
In this note, we investigate the blind source separability of linear FIR-MIMO systems. The concept of semi-reversibility of a system is presented. It is shown that for a semi-reversible system, if the input signals belong to a binary alphabet, then the source data can be blindly separated. One sufficient condition for a system to be semi-reversible is obtained. It is also shown that the proposed criteria is weaker than that in the literature which requires that the channel matrix is irreducible/invertible or reversible.
Keywords: Blind source separable, FIR-MIMO system, Binary input, Bezout equality.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 12961484 Identification, Prediction and Detection of the Process Fault in a Cement Rotary Kiln by Locally Linear Neuro-Fuzzy Technique
Authors: Masoud Sadeghian, Alireza Fatehi
Abstract:
In this paper, we use nonlinear system identification method to predict and detect process fault of a cement rotary kiln. After selecting proper inputs and output, an input-output model is identified for the plant. To identify the various operation points in the kiln, Locally Linear Neuro-Fuzzy (LLNF) model is used. This model is trained by LOLIMOT algorithm which is an incremental treestructure algorithm. Then, by using this method, we obtained 3 distinct models for the normal and faulty situations in the kiln. One of the models is for normal condition of the kiln with 15 minutes prediction horizon. The other two models are for the two faulty situations in the kiln with 7 minutes prediction horizon are presented. At the end, we detect these faults in validation data. The data collected from White Saveh Cement Company is used for in this study.Keywords: Cement Rotary Kiln, Fault Detection, Delay Estimation Method, Locally Linear Neuro Fuzzy Model, LOLIMOT.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16631483 Active Linear Quadratic Gaussian Secondary Suspension Control of Flexible Bodied Railway Vehicle
Authors: Kaushalendra K. Khadanga, Lee Hee Hyol
Abstract:
Passenger comfort has been paramount in the design of suspension systems of high speed cars. To analyze the effect of vibration on vehicle ride quality, a vertical model of a six degree of freedom railway passenger vehicle, with front and rear suspension, is built. It includes car body flexible effects and vertical rigid modes. A second order linear shaping filter is constructed to model Gaussian white noise into random rail excitation. The temporal correlation between the front and rear wheels is given by a second order Pade approximation. The complete track and the vehicle model are then designed. An active secondary suspension system based on a Linear Quadratic Gaussian (LQG) optimal control method is designed. The results show that the LQG control method reduces the vertical acceleration, pitching acceleration and vertical bending vibration of the car body as compared to the passive system.
Keywords: Active suspension, bending vibration, railway vehicle, vibration control.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 7111482 Large Vibration Amplitude of Circular Functionally Graded Plates Resting on Pasternak Foundations
Authors: El Kaak Rachid, El Bikri Khalid, Benamar Rhali
Abstract:
In the present study, the problem of geometrically nonlinear free vibrations of functionally graded circular plates (FGCP) resting on Pasternak elastic foundation with immovable ends was studied. The material properties of the functionally graded composites examined were assumed to be graded in the thickness direction and estimated through the rule of mixture. The theoretical model is based on the classical Plate theory and the Von Kármán geometrical nonlinearity assumptions. Hamilton’s principle is applied and a multimode approach is derived to calculate the fundamental nonlinear frequency parameters, which are found to be in a good agreement with the published results dealing with the problem of functionally graded plates. On the other hand, the influence of the foundation parameters on the nonlinear frequency to the linear frequency ratio of the FGCP has been studied. The effect of the linear and shearing foundations is to decrease the frequency ratio, where it increases with the effect of the nonlinear foundation stiffness.
Keywords: Non-linear vibrations, Circular plates, Pasternak foundation, functionally graded materials.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 21731481 An Evaluation Method for Two-Dimensional Position Errors and Assembly Errors of a Rotational Table on a 4 Axis Machine Tool
Authors: Jooho Hwang, Chang-Kyu Song, Chun-Hong Park
Abstract:
This paper describes a method to measure and compensate a 4 axes ultra-precision machine tool that generates micro patterns on the large surfaces. The grooving machine is usually used for making a micro mold for many electrical parts such as a light guide plate for LCD and fuel cells. The ultra precision machine tool has three linear axes and one rotational table. Shaping is usually used to generate micro patterns. In the case of 50 μm pitch and 25 μm height pyramid pattern machining with a 90° wedge angle bite, one of linear axis is used for long stroke motion for high cutting speed and other linear axis are used for feeding. The triangular patterns can be generated with many times of long stroke of one axis. Then 90° rotation of work piece is needed to make pyramid patterns with superposition of machined two triangular patterns. To make a two dimensional positioning error, straightness of two axes in out of plane, squareness between the each axis are important. Positioning errors, straightness and squarness were measured by laser interferometer system. Those were compensated and confirmed by ISO230-6. One of difficult problem to measure the error motions is squareness or parallelism of axis between the rotational table and linear axis. It was investigated by simultaneous moving of rotary table and XY axes. This compensation method is introduced in this paper.Keywords: Ultra-precision machine tool, muti-axis errors, squraness, positioning errors.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15691480 Spectral Analysis of Speech: A New Technique
Authors: Neeta Awasthy, J.P.Saini, D.S.Chauhan
Abstract:
ICA which is generally used for blind source separation problem has been tested for feature extraction in Speech recognition system to replace the phoneme based approach of MFCC. Applying the Cepstral coefficients generated to ICA as preprocessing has developed a new signal processing approach. This gives much better results against MFCC and ICA separately, both for word and speaker recognition. The mixing matrix A is different before and after MFCC as expected. As Mel is a nonlinear scale. However, cepstrals generated from Linear Predictive Coefficient being independent prove to be the right candidate for ICA. Matlab is the tool used for all comparisons. The database used is samples of ISOLET.Keywords: Cepstral Coefficient, Distance measures, Independent Component Analysis, Linear Predictive Coefficients.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19431479 The Effect of Addition of Dioctyl Terephthalate and Calcite on the Tensile Properties of Organoclay/Linear Low Density Polyethylene Nanocomposites
Authors: A. Gürses, Z. Eroğlu, E. Şahin, K. Güneş, Ç. Doğar
Abstract:
In recent years, polymer/clay nanocomposites have generated great interest in the polymer industry as a new type of composite material because of their superior properties, which includes high heat deflection temperature, gas barrier performance, dimensional stability, enhanced mechanical properties, optical clarity and flame retardancy when compared with the pure polymer or conventional composites. The investigation of change of the tensile properties of organoclay/linear low density polyethylene (LLDPE) nanocomposites with the use of Dioctyl terephthalate (DOTP) (as plasticizer) and calcite (as filler) has been aimed. The composites and organoclay synthesized were characterized using the techniques such as XRD, HRTEM and FTIR techniques. The spectroscopic results indicate that platelets of organoclay were well dispersed within the polymeric matrix. The tensile properties of the composites were compared considering the stress-strain curve drawn for each composite and pure polymer. It was observed that the composites prepared by adding the plasticizer at different ratios and a certain amount of calcite exhibited different tensile behaviors compared to pure polymer.
Keywords: Linear low density polyethylene, nanocomposite, organoclay, plasticizer.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14331478 Numerical Buckling of Composite Cylindrical Shells under Axial Compression Using Asymmetric Meshing Technique (AMT)
Authors: Zia R. Tahir, P. Mandal
Abstract:
This paper presents the details of a numerical study of buckling and post buckling behaviour of laminated carbon fiber reinforced plastic (CFRP) thin-walled cylindrical shell under axial compression using asymmetric meshing technique (AMT) by ABAQUS. AMT is considered to be a new perturbation method to introduce disturbance without changing geometry, boundary conditions or loading conditions. Asymmetric meshing affects both predicted buckling load and buckling mode shapes. Cylindrical shell having lay-up orientation [0^o/+45^o/-45^o/0^o] with radius to thickness ratio (R/t) equal to 265 and length to radius ratio (L/R) equal to 1.5 is analysed numerically. A series of numerical simulations (experiments) are carried out with symmetric and asymmetric meshing to study the effect of asymmetric meshing on predicted buckling behaviour. Asymmetric meshing technique is employed in both axial direction and circumferential direction separately using two different methods, first by changing the shell element size and varying the total number elements, and second by varying the shell element size and keeping total number of elements constant. The results of linear analysis (Eigenvalue analysis) and non-linear analysis (Riks analysis) using symmetric meshing agree well with analytical results. The results of numerical analysis are presented in form of non-dimensional load factor, which is the ratio of buckling load using asymmetric meshing technique to buckling load using symmetric meshing technique. Using AMT, load factor has about 2% variation for linear eigenvalue analysis and about 2% variation for non-linear Riks analysis. The behaviour of load end-shortening curve for pre-buckling is same for both symmetric and asymmetric meshing but for asymmetric meshing curve behaviour in post-buckling becomes extraordinarily complex. The major conclusions are: different methods of AMT have small influence on predicted buckling load and significant influence on load displacement curve behaviour in post buckling; AMT in axial direction and AMT in circumferential direction have different influence on buckling load and load displacement curve in post-buckling.
Keywords: CFRP Composite Cylindrical Shell, Asymmetric Meshing Technique, Primary Buckling, Secondary Buckling, Linear Eigenvalue Analysis, Non-linear Riks Analysis.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 24751477 A Genetic Algorithm Approach for Solving Fuzzy Linear and Quadratic Equations
Authors: M. Hadi Mashinchi, M. Reza Mashinchi, Siti Mariyam H. J. Shamsuddin
Abstract:
In this paper a genetic algorithms approach for solving the linear and quadratic fuzzy equations Ãx̃=B̃ and Ãx̃2 + B̃x̃=C̃ , where Ã, B̃, C̃ and x̃ are fuzzy numbers is proposed by genetic algorithms. Our genetic based method initially starts with a set of random fuzzy solutions. Then in each generation of genetic algorithms, the solution candidates converge more to better fuzzy solution x̃b . In this proposed method the final reached x̃b is not only restricted to fuzzy triangular and it can be fuzzy number.
Keywords: Fuzzy coefficient, fuzzy equation, genetic algorithms.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 21791476 A Soft Error Rates Evaluation Method of Combinational Logic Circuit Based on Linear Energy Transfers
Authors: Man Li, Wanting Zhou, Lei Li
Abstract:
Communication stability is the primary concern of communication satellites. Communication satellites are easily affected by particle radiation to generate single event effects (SEE), which leads to soft errors (SE) of combinational logic circuit. The existing research on soft error rates (SER) of combined logic circuit is mostly based on the assumption that the logic gates being bombarded have the same pulse width. However, in the actual radiation environment, the pulse widths of the logic gates being bombarded are different due to different linear energy transfers (LET). In order to improve the accuracy of SER evaluation model, this paper proposes a soft error rates evaluation method based on LET. In this paper, we analyze the influence of LET on the pulse width of combinational logic and establish the pulse width model based on LET. Based on this model, the error rate of test circuit ISCAS’85 is calculated. Experimental results show that this model can be used for SER evaluation.
Keywords: Communication satellite, pulse width, soft error rates, linear energy transfer, LET.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3621475 On a New Numerical Analysis for the Symmetric Shortest Queue Problem
Authors: Tayeb Lardjane, Rabah Messaci
Abstract:
We consider a network of two M/M/1 parallel queues having the same poisonnian arrival stream with rate λ. Upon his arrival to the system a customer heads to the shortest queue and stays until being served. If the two queues have the same length, an arriving customer chooses one of the two queues with the same probability. Each duration of service in the two queues is an exponential random variable with rate μ and no jockeying is permitted between the two queues. A new numerical method, based on linear programming and convex optimization, is performed for the computation of the steady state solution of the system.
Keywords: Steady state solution, matrix formulation, convex set, shortest queue, linear programming.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14761474 A New Technique for Multi Resolution Characterization of Epileptic Spikes in EEG
Authors: H. N. Suresh, Dr. V. Udaya Shankara
Abstract:
A technique proposed for the automatic detection of spikes in electroencephalograms (EEG). A multi-resolution approach and a non-linear energy operator are exploited. The signal on each EEG channel is decomposed into three sub bands using a non-decimated wavelet transform (WT). The WT is a powerful tool for multi-resolution analysis of non-stationary signal as well as for signal compression, recognition and restoration. Each sub band is analyzed by using a non-linear energy operator, in order to detect spikes. A decision rule detects the presence of spikes in the EEG, relying upon the energy of the three sub-bands. The effectiveness of the proposed technique was confirmed by analyzing both test signals and EEG layouts.Keywords: EEG, Spike, SNEO, Wavelet Transform
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 13641473 Modeling and System Identification of a Variable Excited Linear Direct Drive
Authors: Heiko Weiß, Andreas Meister, Christoph Ament, Nils Dreifke
Abstract:
Linear actuators are deployed in a wide range of applications. This paper presents the modeling and system identification of a variable excited linear direct drive (LDD). The LDD is designed based on linear hybrid stepper technology exhibiting the characteristic tooth structure of mover and stator. A three-phase topology provides the thrust force caused by alternating strengthening and weakening of the flux of the legs. To achieve best possible synchronous operation, the phases are commutated sinusoidal. Despite the fact that these LDDs provide high dynamics and drive forces, noise emission limits their operation in calm workspaces. To overcome this drawback an additional excitation of the magnetic circuit is introduced to LDD using additional enabling coils instead of permanent magnets. The new degree of freedom can be used to reduce force variations and related noise by varying the excitation flux that is usually generated by permanent magnets. Hence, an identified simulation model is necessary to analyze the effects of this modification. Especially the force variations must be modeled well in order to reduce them sufficiently. The model can be divided into three parts: the current dynamics, the mechanics and the force functions. These subsystems are described with differential equations or nonlinear analytic functions, respectively. Ordinary nonlinear differential equations are derived and transformed into state space representation. Experiments have been carried out on a test rig to identify the system parameters of the complete model. Static and dynamic simulation based optimizations are utilized for identification. The results are verified in time and frequency domain. Finally, the identified model provides a basis for later design of control strategies to reduce existing force variations.Keywords: Force variations, linear direct drive, modeling and system identification, variable excitation flux.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 10261472 Variational Iteration Method for Solving Systems of Linear Delay Differential Equations
Authors: Sara Barati, Karim Ivaz
Abstract:
In this paper, using a model transformation approach a system of linear delay differential equations (DDEs) with multiple delays is converted to a non-delayed initial value problem. The variational iteration method (VIM) is then applied to obtain the approximate analytical solutions. Numerical results are given for several examples involving scalar and second order systems. Comparisons with the classical fourth-order Runge-Kutta method (RK4) verify that this method is very effective and convenient.
Keywords: Variational iteration method, delay differential equations, multiple delays, Runge-Kutta method.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 24661471 Establishment of Kinetic Zone Diagrams via Simulated Linear Sweep Voltammograms for Soluble-Insoluble Systems
Authors: Imene Atek, Abed M. Affoune, Hubert Girault, Pekka Peljo
Abstract:
Due to the need for a rigorous mathematical model that can help to estimate kinetic properties for soluble-insoluble systems, through voltammetric experiments, a Nicholson Semi Analytical Approach was used in this work for modeling and prediction of theoretical linear sweep voltammetry responses for reversible, quasi reversible or irreversible electron transfer reactions. The redox system of interest is a one-step metal electrodeposition process. A rigorous analysis of simulated linear scan voltammetric responses following variation of dimensionless factors, the rate constant and charge transfer coefficients in a broad range was studied and presented in the form of the so called kinetic zones diagrams. These kinetic diagrams were divided into three kinetics zones. Interpreting these zones leads to empirical mathematical models which can allow the experimenter to determine electrodeposition reactions kinetics whatever the degree of reversibility. The validity of the obtained results was tested and an excellent experiment–theory agreement has been showed.
Keywords: Electrodeposition, kinetics diagrams, modeling, voltammetry.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 7911470 Influence of p-y curves on Buckling Capacity of Pile Foundation
Authors: Praveen Huded M., Suresh R. Dash
Abstract:
Pile foundations are one of the most preferred deep foundation systems for high rise or heavily loaded structures. In many instances, the failure of the pile founded structures in liquefiable soils had been observed even in many recent earthquakes. Failure of pile foundation have occurred because of buckling, as the pile behaves as an unsupported slender structural element once the surrounding soil liquefies. However, the buckling capacity depends on the depth of soil liquefied and its residual strength. Hence it is essential to check the pile against the possible buckling failure. Beam on non-linear Winkler Foundation is one of the efficient methods to model the pile-soil behavior in liquefiable soil. The pile-soil interaction is modelled through p-y springs, there are different p-y curves available for modeling liquefiable soil. In the present work, the influence of two such p-y curves on the buckling capacity of pile foundation is studied considering the initial geometric and non-linear behavior of pile foundation. The proposed method is validated against experimental results. A significant difference in the buckling capacity is observed for the two p-y curves used in the analysis. A parametric study is conducted to understand the influence of pile flexural rigidity, different initial geometric imperfections, and different soil relative densities on the buckling capacity of pile foundation.
Keywords: pile foundation, liquefaction, buckling load, non-linear p-y curve
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 6491469 Face Recognition using Features Combination and a New Non-linear Kernel
Authors: Essam Al Daoud
Abstract:
To improve the classification rate of the face recognition, features combination and a novel non-linear kernel are proposed. The feature vector concatenates three different radius of local binary patterns and Gabor wavelet features. Gabor features are the mean, standard deviation and the skew of each scaling and orientation parameter. The aim of the new kernel is to incorporate the power of the kernel methods with the optimal balance between the features. To verify the effectiveness of the proposed method, numerous methods are tested by using four datasets, which are consisting of various emotions, orientations, configuration, expressions and lighting conditions. Empirical results show the superiority of the proposed technique when compared to other methods.Keywords: Face recognition, Gabor wavelet, LBP, Non-linearkerner
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15281468 Determining Optimum Time Multiplier Setting of Overcurrent Relays Using Mixed Integer Linear Programming
Authors: P. N. Korde, P. P. Bedekar
Abstract:
The time coordination of overcurrent relays (OCR) in a power distribution network is of great importance, as it reduces the power outages by avoiding the mal-operation of the backup relays. For this, the optimum value of the time multiplier setting (TMS) of OCRs should be chosen. The problem of determining the optimum value of TMS of OCRs in power distribution networks is formulated as a constrained optimization problem. The objective is to find the optimum value of TMS of OCRs to minimize the time of operation of relays under the constraint of maintaining the coordination of relays. A power distribution network can have a combination of numerical and electromechanical relays. The TMS of numerical relays can be set to any real value (which satisfies the constraints of the problem), whereas the TMS of electromechanical relays can be set in fixed step (0 to 1 in steps of 0.05). The main contribution of this paper is a formulation of the problem as a mixed-integer linear programming (MILP) problem and application of Gomory's cutting plane method to find the optimum value of TMS of OCRs. The TMS of electromechanical relays are taken as integers in the range 1 to 20 in the step of 1, and these values are mapped to 0.05 to 1 in the step of 0.05. The results obtained are compared with those obtained using a simplex method and its variants. It has been shown that the mixed-integer linear programming method outperforms the simplex method (and its variants) in the case of a system having a combination of numerical and electromechanical relays.
Keywords: Backup protection, constrained optimization, Gomory's cutting plane method, mixed-integer linear programming, overcurrent relay coordination, simplex method.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4101467 A Comparison of Adaline and MLP Neural Network based Predictors in SIR Estimation in Mobile DS/CDMA Systems
Authors: Nahid Ardalani, Ahmadreza Khoogar, H. Roohi
Abstract:
In this paper we compare the response of linear and nonlinear neural network-based prediction schemes in prediction of received Signal-to-Interference Power Ratio (SIR) in Direct Sequence Code Division Multiple Access (DS/CDMA) systems. The nonlinear predictor is Multilayer Perceptron MLP and the linear predictor is an Adaptive Linear (Adaline) predictor. We solve the problem of complexity by using the Minimum Mean Squared Error (MMSE) principle to select the optimal predictors. The optimized Adaline predictor is compared to optimized MLP by employing noisy Rayleigh fading signals with 1.8 GHZ carrier frequency in an urban environment. The results show that the Adaline predictor can estimates SIR with the same error as MLP when the user has the velocity of 5 km/h and 60 km/h but by increasing the velocity up-to 120 km/h the mean squared error of MLP is two times more than Adaline predictor. This makes the Adaline predictor (with lower complexity) more suitable than MLP for closed-loop power control where efficient and accurate identification of the time-varying inverse dynamics of the multi path fading channel is required.Keywords: Power control, neural networks, DS/CDMA mobilecommunication systems.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 25021466 Artificial Neural Network based Modeling of Evaporation Losses in Reservoirs
Authors: Surinder Deswal, Mahesh Pal
Abstract:
An Artificial Neural Network based modeling technique has been used to study the influence of different combinations of meteorological parameters on evaporation from a reservoir. The data set used is taken from an earlier reported study. Several input combination were tried so as to find out the importance of different input parameters in predicting the evaporation. The prediction accuracy of Artificial Neural Network has also been compared with the accuracy of linear regression for predicting evaporation. The comparison demonstrated superior performance of Artificial Neural Network over linear regression approach. The findings of the study also revealed the requirement of all input parameters considered together, instead of individual parameters taken one at a time as reported in earlier studies, in predicting the evaporation. The highest correlation coefficient (0.960) along with lowest root mean square error (0.865) was obtained with the input combination of air temperature, wind speed, sunshine hours and mean relative humidity. A graph between the actual and predicted values of evaporation suggests that most of the values lie within a scatter of ±15% with all input parameters. The findings of this study suggest the usefulness of ANN technique in predicting the evaporation losses from reservoirs.Keywords: Artificial neural network, evaporation losses, multiple linear regression, modeling.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19661465 Stability Criteria for Neural Networks with Two Additive Time-varying Delay Components
Authors: Qingqing Wang, Shouming Zhong
Abstract:
This paper is concerned with the stability problem with two additive time-varying delay components. By choosing one augmented Lyapunov-Krasovskii functional, using some new zero equalities, and combining linear matrix inequalities (LMI) techniques, two new sufficient criteria ensuring the global stability asymptotic stability of DNNs is obtained. These stability criteria are present in terms of linear matrix inequalities and can be easily checked. Finally, some examples are showed to demonstrate the effectiveness and less conservatism of the proposed method.
Keywords: Neural networks, Globally asymptotic stability, LMI approach, Additive time-varying delays.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15551464 Gauss-Seidel Iterative Methods for Rank Deficient Least Squares Problems
Authors: Davod Khojasteh Salkuyeh, Sayyed Hasan Azizi
Abstract:
We study the semiconvergence of Gauss-Seidel iterative methods for the least squares solution of minimal norm of rank deficient linear systems of equations. Necessary and sufficient conditions for the semiconvergence of the Gauss-Seidel iterative method are given. We also show that if the linear system of equations is consistent, then the proposed methods with a zero vector as an initial guess converge in one iteration. Some numerical results are given to illustrate the theoretical results.Keywords: rank deficient least squares problems, AOR iterativemethod, Gauss-Seidel iterative method, semiconvergence.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19181463 Studding of Number of Dataset on Precision of Estimated Saturated Hydraulic Conductivity
Authors: M. Siosemarde, M. Byzedi
Abstract:
Saturated hydraulic conductivity of Soil is an important property in processes involving water and solute flow in soils. Saturated hydraulic conductivity of soil is difficult to measure and can be highly variable, requiring a large number of replicate samples. In this study, 60 sets of soil samples were collected at Saqhez region of Kurdistan province-IRAN. The statistics such as Correlation Coefficient (R), Root Mean Square Error (RMSE), Mean Bias Error (MBE) and Mean Absolute Error (MAE) were used to evaluation the multiple linear regression models varied with number of dataset. In this study the multiple linear regression models were evaluated when only percentage of sand, silt, and clay content (SSC) were used as inputs, and when SSC and bulk density, Bd, (SSC+Bd) were used as inputs. The R, RMSE, MBE and MAE values of the 50 dataset for method (SSC), were calculated 0.925, 15.29, -1.03 and 12.51 and for method (SSC+Bd), were calculated 0.927, 15.28,-1.11 and 12.92, respectively, for relationship obtained from multiple linear regressions on data. Also the R, RMSE, MBE and MAE values of the 10 dataset for method (SSC), were calculated 0.725, 19.62, - 9.87 and 18.91 and for method (SSC+Bd), were calculated 0.618, 24.69, -17.37 and 22.16, respectively, which shows when number of dataset increase, precision of estimated saturated hydraulic conductivity, increases.Keywords: dataset, precision, saturated hydraulic conductivity, soil and statistics.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 17831462 Improved Stability Criteria for Neural Networks with Two Additive Time-Varying Delays
Authors: Miaomiao Yang, Shouming Zhong
Abstract:
This paper studies the problem of stability criteria for neural networks with two additive time-varying delays.A new Lyapunov-Krasovskii function is constructed and some new delay dependent stability criterias are derived in the terms of linear matrix inequalities(LMI), zero equalities and reciprocally convex approach.The several stability criterion proposed in this paper is simpler and effective. Finally,numerical examples are provided to demonstrate the feasibility and effectiveness of our results.
Keywords: Stability, Neural networks, Linear Matrix Inequalities (LMI) , Lyapunov function, Time-varying delays
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1440