Search results for: Rectified Linear Units (ReLU)

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 814

Authors: Apoorva Vinod, Archana Mathur, Snehanshu Saha

Abstract:

Though there exists a plethora of Activation Functions (AFs) used in single and multiple hidden layer Neural Networks (NN), their behavior always raised curiosity, whether used in combination or singly. The popular AFs – Sigmoid, ReLU, and Tanh – have performed prominently well for shallow and deep architectures. Most of the time, AFs are used singly in multi-layered NN, and, to the best of our knowledge, their performance is never studied and analyzed deeply when used in combination. In this manuscript, we experiment on multi-layered NN architecture (both on shallow and deep architectures; Convolutional NN and VGG16) and investigate how well the network responds to using two different AFs (Sigmoid-Tanh, Tanh-ReLU, ReLU-Sigmoid) used alternately against a traditional, single (Sigmoid-Sigmoid, Tanh-Tanh, ReLU-ReLU) combination. Our results show that on using two different AFs, the network achieves better accuracy, substantially lower loss, and faster convergence on 4 computer vision (CV) and 15 Non-CV (NCV) datasets. When using different AFs, not only was the accuracy greater by 6-7%, but we also accomplished convergence twice as fast. We present a case study to investigate the probability of networks suffering vanishing and exploding gradients when using two different AFs. Additionally, we theoretically showed that a composition of two or more AFs satisfies Universal Approximation Theorem (UAT).

Keywords: Activation Function, Universal Approximation function, Neural Networks, convergence.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 96

Authors: Zukisa Nante, Wang Zenghui

Abstract:

Face recognition is the problem of identifying or recognizing individuals in an image. This paper investigates a possible method to bring a solution to this problem. The method proposes an amalgamation of Principal Component Analysis (PCA), K-Means clustering, and Convolutional Neural Network (CNN) for a face recognition system. It is trained and evaluated using the ORL dataset. This dataset consists of 400 different faces with 40 classes of 10 face images per class. Firstly, PCA enabled the usage of a smaller network. This reduces the training time of the CNN. Thus, we get rid of the redundancy and preserve the variance with a smaller number of coefficients. Secondly, the K-Means clustering model is trained using the compressed PCA obtained data which select the K-Means clustering centers with better characteristics. Lastly, the K-Means characteristics or features are an initial value of the CNN and act as input data. The accuracy and the performance of the proposed method were tested in comparison to other Face Recognition (FR) techniques namely PCA, Support Vector Machine (SVM), as well as K-Nearest Neighbour (kNN). During experimentation, the accuracy and the performance of our suggested method after 90 epochs achieved the highest performance: 99% accuracy F1-Score, 99% precision, and 99% recall in 463.934 seconds. It outperformed the PCA that obtained 97% and KNN with 84% during the conducted experiments. Therefore, this method proved to be efficient in identifying faces in the images.

Keywords: Face recognition, Principal Component Analysis, PCA, Convolutional Neural Network, CNN, Rectified Linear Unit, ReLU, feature extraction.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 435

Authors: Dursun Aydın

Abstract:

In this paper, estimation of the linear regression model is made by ordinary least squares method and the partially linear regression model is estimated by penalized least squares method using smoothing spline. Then, it is investigated that differences and similarity in the sum of squares related for linear regression and partial linear regression models (semi-parametric regression models). It is denoted that the sum of squares in linear regression is reduced to sum of squares in partial linear regression models. Furthermore, we indicated that various sums of squares in the linear regression are similar to different deviance statements in partial linear regression. In addition to, coefficient of the determination derived in linear regression model is easily generalized to coefficient of the determination of the partial linear regression model. For this aim, it is made two different applications. A simulated and a real data set are considered to prove the claim mentioned here. In this way, this study is supported with a simulation and a real data example.

Keywords: Partial Linear Regression Model, Linear RegressionModel, Residuals, Deviance, Smoothing Spline.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1834

Authors: Navid Mehdizadeh Afroozi, Khodakhast Isapour, Mojtaba Hakimzadeh, Abdolmohammad Davodi

Abstract:

The objective of the Economic Dispatch(ED) Problems of electric power generation is to schedule the committed generating units outputs so as to meet the required load demand at minimum operating cost while satisfying all units and system equality and inequality constraints. This paper presents a new method of ED problems utilizing the Max-Min Ant System Optimization. Historically, traditional optimizations techniques have been used, such as linear and non-linear programming, but within the past decade the focus has shifted on the utilization of Evolutionary Algorithms, as an example Genetic Algorithms, Simulated Annealing and recently Ant Colony Optimization (ACO). In this paper we introduce the Max-Min Ant System based version of the Ant System. This algorithm encourages local searching around the best solution found in each iteration. To show its efficiency and effectiveness, the proposed Max-Min Ant System is applied to sample ED problems composed of 4 generators. Comparison to conventional genetic algorithms is presented.

Keywords: Economic Dispatch (ED), Ant Colony Optimization, Fuel Cost, Algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2541

Authors: M. Esmaeeli Shahrakht, A. Kazemi

Abstract:

Due to uncertainty of wind velocity, wind power generators don’t have deterministic output power. Utilizing wind power generation and thermal power plants together create new concerns for operation engineers of power systems. In this paper, a model is presented to implement the uncertainty of load and generated wind power which can be utilized in power system operation planning. Stochastic behavior of parameters is simulated by generating scenarios that can be solved by deterministic method. A mixed-integer linear programming method is used for solving deterministic generation scheduling problem. The proposed approach is applied to a 12-unit test system including 10 thermal units and 2 wind farms. The results show affectivity of piecewise linear model in unit commitment problems. Also using linear programming causes a considerable reduction in calculation times and guarantees convergence to the global optimum. Neglecting the uncertainty of wind velocity causes higher cost assessment of generation scheduling.

Keywords: Load uncertainty, linear programming, scenario generation, unit commitment, wind farm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2892

Authors: Smajo Bisanovic, Mensur Hajro, Muris Dlakic

Abstract:

This paper presents one comprehensive modelling approach for maintenance scheduling problem of thermal power units in competitive market. This problem is formulated as a 0/1 mixedinteger linear programming model. Model incorporates long-term bilateral contracts with defined profiles of power and price, and weekly forecasted market prices for market auction. The effectiveness of the proposed model is demonstrated through case study with detailed discussion.

Keywords: Maintenance scheduling, bilateral contracts, market prices, profit.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1559

Authors: D. Charvatova

Abstract:

This contribution deals with the relationship between communication effectiveness and the extent of communication among organizational units. To facilitate communication between employees and to increase the level of understanding, the knowledge of communication tools is necessary. Recent experience has shown that personal communication is critical for smooth running of companies and cannot be fully replaced by any form of technical communication devices. Below are presented the outcomes of the research on the relationship between the extent of communication among organisational units and its efficiency.

Keywords: Company meetings, corporate culture, efficiency, effective communication, form of communication, information, organizational units, quality of communication, strategy, subordinates, superiors.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2463

Authors: S.H. Nasseri, E. Ardil

Abstract:

Fuzzy linear programming is an application of fuzzy set theory in linear decision making problems and most of these problems are related to linear programming with fuzzy variables. A convenient method for solving these problems is based on using of auxiliary problem. In this paper a new method for solving fuzzy variable linear programming problems directly using linear ranking functions is proposed. This method uses simplex tableau which is used for solving linear programming problems in crisp environment before.

Keywords: Fuzzy variable linear programming, fuzzy number, ranking function, simplex method.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3300

Authors: Elias K. Maragos, Petros E. Maravelakis

Abstract:

In Dynamic Data Envelopment Analysis (DDEA), which is a subfield of Data Envelopment Analysis (DEA), the productivity of Decision Making Units (DMUs) is considered in relation to time. In this case, as it is accepted by the most of the researchers, there are outputs, which are produced by a DMU to be used as inputs in a future time. Those outputs are known as intermediates. The common models, in DDEA, do not take into account the shape of the distribution of those inputs, outputs or intermediates data, assuming that the distribution of the virtual value of them does not deviate from linearity. This weakness causes the limitation of the accuracy of the analytical power of the traditional DDEA models. In this paper, the authors, using the concept of piecewise linear inputs and outputs, propose an extended DDEA model. The proposed model increases the flexibility of the traditional DDEA models and improves the measurement of the dynamic performance of DMUs.

Keywords: Data envelopment analysis, Dynamic DEA, Piecewise linear inputs, Piecewise linear outputs.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 622

Authors: Mayank Shakya, Soundra Pandian. K. K

Abstract:

A linear feedback shift register (LFSR) is proposed which targets to reduce the power consumption from within. It reduces the power consumption during testing of a Circuit Under Test (CUT) at two stages. At first stage, Control Logic (CL) makes the clocks of the switching units of the register inactive for a time period when output from them is going to be same as previous one and thus reducing unnecessary switching of the flip-flops. And at second stage, the LFSR reorders the test vectors by interchanging the bit with its next and closest neighbor bit. It keeps fault coverage capacity of the vectors unchanged but reduces the Total Hamming Distance (THD) so that there is reduction in power while shifting operation.

Keywords: Linear Feedback Shift Register, Total Hamming Distance, Fault Coverage, Control Logic

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1988

Authors: Nooshin Salari, Viliam Makis

Abstract:

In this paper, we propose a condition-based maintenance policy for multi-unit systems considering the existence of economic dependency among units. We consider a system composed of N identical units, where each unit deteriorates independently. Deterioration process of each unit is modeled as a three-state continuous time homogeneous Markov chain with two working states and a failure state. The average production rate of units varies in different working states and demand rate of the system is constant. Units are inspected at equidistant time epochs, and decision regarding performing maintenance is determined by the number of units in the failure state. If the total number of units in the failure state exceeds a critical level, maintenance is initiated, where units in failed state are replaced correctively and deteriorated state units are maintained preventively. Our objective is to determine the optimal number of failed units to initiate maintenance minimizing the long run expected average cost per unit time. The problem is formulated and solved in the semi-Markov decision process (SMDP) framework. A numerical example is developed to demonstrate the proposed policy and the comparison with the corrective maintenance policy is presented.

Keywords: Reliability, production, maintenance optimization, Semi-Markov Decision Process.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 749

Authors: Chao Wang, Ting-Zhu Huang Chen Jia

Abstract:

In this paper, we propose a new linear complementarity problem named as bi-linear complementarity problem (BLCP) and the method for solving BLCP. In addition, the algorithm for error estimation of BLCP is also given. Numerical experiments show that the algorithm is efficient.

Keywords: Bi-linear complementarity problem, Linear complementarity problem, Extended linear complementarity problem, Error estimation, P-matrix, M-matrix.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1686

Authors: H. Taghvafard, G. H. Erjaee

Abstract:

A method for solving linear and non-linear Goursat problem is given by using the two-dimensional differential transform method. The approximate solution of this problem is calculated in the form of a series with easily computable terms and also the exact solutions can be achieved by the known forms of the series solutions. The method can easily be applied to many linear and non-linear problems and is capable of reducing the size of computational work. Several examples are given to demonstrate the reliability and the performance of the presented method.

Keywords: Quadrature, Spline interpolation, Trapezoidal rule, Numericalintegration, Error analysis.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2338

Authors: Reza Nadimi, Fariborz Jolai

Abstract:

This article combines two techniques: data envelopment analysis (DEA) and Factor analysis (FA) to data reduction in decision making units (DMU). Data envelopment analysis (DEA), a popular linear programming technique is useful to rate comparatively operational efficiency of decision making units (DMU) based on their deterministic (not necessarily stochastic) input–output data and factor analysis techniques, have been proposed as data reduction and classification technique, which can be applied in data envelopment analysis (DEA) technique for reduction input – output data. Numerical results reveal that the new approach shows a good consistency in ranking with DEA.

Keywords: Effectiveness, Decision Making, Data EnvelopmentAnalysis, Factor Analysis

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2382

Authors: I Made Darmayuda, Chai Tshun Chuan Kevin, Je Minkyu

Abstract:

Plenty researches have reported techniques to harvest energy from piezoelectric transducer. In the earlier years, the researches mainly report linear energy harvesting techniques whereby interface circuitry is designed to have input impedance that match with the impedance of the piezoelectric transducer. In recent years non-linear techniques become more popular. The non-linear technique employs voltage waveform manipulation to boost the available-for-extraction energy at the time of energy transfer.  The fact that non-linear energy extraction provides larger available-for-extraction energy doesn’t mean the linear energy extraction is completely obsolete. In some scenarios, such as where initial power is not available, linear energy extraction is still preferred. A modified Buck Boost circuit which is capable of harvesting piezoelectric energy using both linear and non-linear techniques is reported in this paper. Efficiency of at least 64% can be achieved using this circuit. For linear extraction, the modified Buck Boost circuit is controlled using a fix frequency and duty cycle clock. A voltage sensor and a pulse generator are added as the controller for the non-linear extraction technique. 

Keywords: Buck boost, energy harvester, linear energy harvester, non-linear energy harvester, piezoelectric, synchronized charge extraction.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2397

Authors: Baoguang Tian, Nan Chen

Abstract:

A new relative efficiency in linear model in reference is instructed into the linear weighted regression, and its upper and lower bound are proposed. In the linear weighted regression model, for the best linear unbiased estimation of mean matrix respect to the least-squares estimation, two new relative efficiencies are given, and their upper and lower bounds are also studied.

Keywords: Linear weighted regression, Relative efficiency, Mean matrix, Trace.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2420

Authors: Sergey Kopysov, Nikita Nedozhogin, Leonid Tonkov

Abstract:

The article presents a parallel iterative solver for large sparse linear systems which can be used on a heterogeneous platform. Traditionally, the problem of solving linear systems do not scale well on cluster containing multiple Central Processing Units (multi-CPUs cluster) or cluster containing multiple Graphics Processing Units (multi-GPUs cluster). For example, most of the attempts to implement the classical conjugate gradient method were at best counted in the same amount of time as the problem was enlarged. The paper proposes the pipelined variant of the conjugate gradient method (PCG), a formulation that is potentially better suited for hybrid CPU/GPU computing since it requires only one synchronization point per one iteration, instead of two for standard CG (Conjugate Gradient). The standard and pipelined CG methods need the vector entries generated by current GPU and other GPUs for matrix-vector product. So the communication between GPUs becomes a major performance bottleneck on miltiGPU cluster. The article presents an approach to minimize the communications between parallel parts of algorithms. Additionally, computation and communication can be overlapped to reduce the impact of data exchange. Using pipelined version of the CG method with one synchronization point, the possibility of asynchronous calculations and communications, load balancing between the CPU and GPU for solving the large linear systems allows for scalability. The algorithm is implemented with the combined use of technologies: MPI, OpenMP and CUDA. We show that almost optimum speed up on 8-CPU/2GPU may be reached (relatively to a one GPU execution). The parallelized solver achieves a speedup of up to 5.49 times on 16 NVIDIA Tesla GPUs, as compared to one GPU.

Keywords: Conjugate Gradient, GPU, parallel programming, pipelined algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 327

Authors: Jung Ho Yang, Yong Soo Kim

Abstract:

Electrical units are operated by electrical and electronic components. An environmental test sequence is useful for testing electrical units to reduce reliability issues. This study introduces test sequence guidelines based on relevant principles and considerations for electronic testing according to International Standard IEC-60068-1 and the United States Military Standard MIL-STD-810G. Test sequences were then proposed based on the descriptions for each test. Finally, General Motors (GM) specification GMW3172 was interpreted and compared to IEC-60068-1 and MIL-STD-810G.

Keywords: Reliability, Environmental test sequence, Electrical units, IEC 60068-1, MIL-STD-810G.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4268

Authors: Boukhzer Abdenbi, El Bikri Khalid, Benamar Rhali

Abstract:

In the present study, the problem of geometrically non-linear free vibrations of functionally graded rectangular plates (FGRP) is studied. The theoretical model, previously developed and based on Hamilton’s principle, is adapted here to determine the fundamental non-linear mode shape of these plates. Frequency parameters, displacements and stress are given for various power-law distributions of the volume fractions of the constituents and various aspect ratios. Good agreement with previous published results is obtained in the case of linear and non-linear analyses.

Keywords: Non-linear vibration, functionally graded materials, rectangular plates.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2201

Authors: S. H. Nasseri, E. Ardil, A. Yazdani, R. Zaefarian

Abstract:

The fuzzy set theory has been applied in many fields, such as operations research, control theory, and management sciences, etc. In particular, an application of this theory in decision making problems is linear programming problems with fuzzy numbers. In this study, we present a new method for solving fuzzy number linear programming problems, by use of linear ranking function. In fact, our method is similar to simplex method that was used for solving linear programming problems in crisp environment before.

Keywords: Fuzzy number linear programming, rankingfunction, simplex method.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3471

Authors: Jian Du, Shui Hong Hong, Lu Meng, Qing Wei Meng

Abstract:

This work addresses the problem of optimizing completely batch water-using network with multiple contaminants where the flow change caused by mass transfer is taken into consideration for the first time. A mathematical technique for optimizing water-using network is proposed based on source-tank-sink superstructure. The task is to obtain the freshwater usage, recycle assignments among water-using units, wastewater discharge and a steady water-using network configuration by following steps. Firstly, operating sequences of water-using units are determined by time constraints. Next, superstructure is simplified by eliminating the reuse and recycle from water-using units with maximum concentration of key contaminants. Then, the non-linear programming model is solved by GAMS (General Algebra Model System) for minimum freshwater usage, maximum water recycle and minimum wastewater discharge. Finally, numbers of operating periods are calculated to acquire the steady network configuration. A case study is solved to illustrate the applicability of the proposed approach.

Keywords: Completely batch process, flow change, multiple contaminants, water-using network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1397

Authors: R. B. Ogunrinde, C. C. Jibunoh

Abstract:

In this paper, a spectral decomposition method is developed for the direct integration of stiff and nonstiff homogeneous linear (ODE) systems with linear, constant, or zero right hand sides (RHSs). The method does not require iteration but obtains solutions at any random points of t, by direct evaluation, in the interval of integration. All the numerical solutions obtained for the class of systems coincide with the exact theoretical solutions. In particular, solutions of homogeneous linear systems, i.e. with zero RHS, conform to the exact analytical solutions of the systems in terms of t.

Keywords: Spectral decomposition, eigenvalues of the Jacobian, linear RHS, homogeneous linear systems.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1107

Authors: Hadj Bourdoucen, Adel Gastli

Abstract:

In this paper, an approach for finding optimized layouts for connecting PV units delivering maximum array output power is suggested. The approach is based on considering the different varying parameters of PV units that might be extracted from a general two-diode model. These are mainly, solar irradiation, reverse saturation currents, ideality factors, series and shunt resistances in addition to operating temperature. The approach has been tested on 19 possible 2×3 configurations and allowed to determine the optimized configurations as well as examine the effects of the different units- parameters on the maximum output power. Thus, using this approach, standard arrays with n×m units can be configured for maximum generated power and allows designing PV based systems having reduced surfaces to fit specific required power, as it is the case for solar cars and other mobile systems.

Keywords: Photovoltaic, PV unit, optimum configuration, maximum power, Orcad.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1707

Authors: Jianqiao. Yu, Jianbo. Wang, Xinzhen. He

Abstract:

This paper presents a longitudinal quasi-linear model for the ADMIRE model. The ADMIRE model is a nonlinear model of aircraft flying in the condition of high angle of attack. So it can-t be considered to be a linear system approximately. In this paper, for getting the longitudinal quasi-linear model of the ADMIRE, a state transformation based on differentiable functions of the nonscheduling states and control inputs is performed, with the goal of removing any nonlinear terms not dependent on the scheduling parameter. Since it needn-t linear approximation and can obtain the exact transformations of the nonlinear states, the above-mentioned approach is thought to be appropriate to establish the mathematical model of ADMIRE. To verify this conclusion, simulation experiments are done. And the result shows that this quasi-linear model is accurate enough.

Keywords: quasi-linear model, simulation, state transformation approach, the ADMIRE model.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1465

Authors: Yun Jong Choi, Nam Woong, PooGyeon Park

Abstract:

Variable Structure Control (VSC) is one of the most useful tools handling the practical system with uncertainties and disturbances. Up to now, unfortunately, not enough studies on the input-saturated system with linear-growth-bound disturbances via VSC have been presented. Therefore, this paper proposes an asymp¬totic stability condition for the system via VSC. The designed VSC controller consists of two control parts. The linear control part plays a role in stabilizing the system, and simultaneously, the nonlinear control part in rejecting the linear-growth-bound disturbances perfectly. All conditions derived in this paper are expressed with Linear Matrices Inequalities (LMIs), which can be easily solved with an LMI toolbox in MATLAB.

Keywords: Input saturation, linear-growth bounded disturbances, linear matrix inequality (LMI), variable structure control

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1593

Authors: Hazem Al-Mofleh, John Daniels, Joseph McKean

Abstract:

Within geostatistics research, effective estimation of the variogram points has been examined, particularly in developing robust alternatives. The parametric fit of these variogram points which eventually defines the kriging weights, however, has not received the same attention from a robust perspective. This paper proposes the use of the non-linear Wilcoxon norm over weighted non-linear least squares as a robust variogram fitting alternative. First, we introduce the concept of variogram estimation and fitting. Then, as an alternative to non-linear weighted least squares, we discuss the non-linear Wilcoxon estimator. Next, the robustness properties of the non-linear Wilcoxon are demonstrated using a contaminated spatial data set. Finally, under simulated conditions, increasing levels of contaminated spatial processes have their variograms points estimated and fit. In the fitting of these variogram points, both non-linear Weighted Least Squares and non-linear Wilcoxon fits are examined for efficiency. At all levels of contamination (including 0%), using a robust estimation and robust fitting procedure, the non-weighted Wilcoxon outperforms weighted Least Squares.

Keywords: Non-Linear Wilcoxon, robust estimation, Variogram estimation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 932

Authors: Dursun Aydın, Bilgin Senel

Abstract:

In this paper, the sum of squares in linear regression is reduced to sum of squares in semi-parametric regression. We indicated that different sums of squares in the linear regression are similar to various deviance statements in semi-parametric regression. In addition to, coefficient of the determination derived in linear regression model is easily generalized to coefficient of the determination of the semi-parametric regression model. Then, it is made an application in order to support the theory of the linear regression and semi-parametric regression. In this way, study is supported with a simulated data example.

Keywords: Semi-parametric regression, Penalized LeastSquares, Residuals, Deviance, Smoothing Spline.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1815

Authors: Fituri H Belgassem, Abdulbasit Nigrat, Seddeeq Ghrari

Abstract:

In this paper, the construction of fast algorithms for the computation of Periodic Walsh Piecewise-Linear PWL transform and the Periodic Haar Piecewise-Linear PHL transform will be presented. Algorithms for the computation of the inverse transforms are also proposed. The matrix equation of the PWL and PHL transforms are introduced. Comparison of the computational requirements for the periodic piecewise-linear transforms and other orthogonal transforms shows that the periodic piecewise-linear transforms require less number of operations than some orthogonal transforms such as the Fourier, Walsh and the Discrete Cosine transforms.

Keywords: Piece wise linear transforms, Fast transforms, Fast algorithms.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1624

Authors: Richardson M. Abraham-A., Colombo Celso Gaeta Tassinari

Abstract:

The correlation between Total Organic Carbon (TOC) and flow units have been carried out to predict and compare the carbon dioxide (CO₂) storage potential of the shale and carbonate rocks in Irati Formation of the Parana Basin. The equations for permeability (K), reservoir quality index (RQI) and flow zone indicator (FZI) are redefined and engaged to evaluate the flow units in both potential reservoir rocks. Shales show higher values of TOC compared to carbonates, as such,  porosity (Ф) is most likely to be higher in shales compared to carbonates. The increase in Ф corresponds to the increase in K (in both rocks). Nonetheless, at lower values of Ф, K is higher in carbonates compared to shales. This shows that at lower values of TOC in carbonates, Ф is low, yet, K is likely to be high compared to shale. In the same vein, at higher values of TOC in shales, Ф is high, yet, K is expected to be low compared to carbonates.  Overall, the flow unit factors (RQI and FZI) are better in the carbonates compared to the shales. Moreso, within the study location,  there are some portions where the thicknesses of the carbonate units are higher compared to the shale units. Most parts of the carbonate strata in the study location are fractured in situ, hence,  this could provide easy access for the storage of CO₂. Therefore, based on these points and the disparities between the flow units in the evaluated rock types, the carbonate units are expected to show better potentials for the storage of CO₂. The shale units may be considered as potential cap rocks or seals.

Keywords: Total organic carbon, flow units, carbon dioxide storage.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 793