Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Pattern recognition

Plasma Chemistry: References

Gravitational Search Algorithm (GSA) is an optimization method inspired by the theory of Newtonian gravity in physics. Till now, many variants of GSA have been introduced, most of them are motivated by gravity-related theories such as relativity and astronomy. On the one hand, to solve different kinds of optimization problems, modified versions of GSA have been presented such as continuous (real), binary, discrete, multimodal, constraint, single-objective, and multi-objective GSA. On the other hand, to tackle the difficulties in real-world problems, the efficiency of GSA has been improved using specialized operators, hybridization, local search, and designing the self-adaptive algorithms. Researchers have utilized GSA to solve various engineering optimization problems in diverse fields of applications ranging from electrical engineering to bioinformatics. Here, we discussed a comprehensive investigation of GSA and a brief review of GSA developments in solving different engineering problems to build up a global picture and to open the mind to explore possible applications. We also made a number of suggestions that can be undertaken to help move the area forward.

A comprehensive survey on gravitational search algorithm

http://ieeexplore.ieee.org/iel5/57/17476/00806380.pdf

Outdoor insulators

This paper deals with the use of corona ring at the HV end fitting for improving the electric field and potential distributions and then for minimizing the corona discharges on 230 kV AC transmission line composite insulator. A single corona ring is installed at the energized end side. Three-dimensional finite element method (FEM) software is employed to compute the electric field. As the performance of high voltage insulator strings closely depends on designs and locations of corona ring, the effects of the corona ring radius, the ring tube radius and the ring vertical position are examined. The minimization of the electric field necessitates the optimization of corona ring. For this purpose, new nonlinear mathematical objective function linking the electric field strength to the corona ring structure parameters is established. The optimization problem is achieved by minimizing the objective function using a modified particles swarm optimization (PSO) algorithm with a dynamic population size. The algorithm adjusts the size of population for each iteration. Based on the average value and the best solution of the objective function, we propose a new mathematical model to update the population size. This algorithm enables the population size reduction leading to computing time decrease. According to the results, FEM-PSO hybridization technique could be very helpful in optimization of corona ring design.

Optimal design of corona ring on HV composite insulator using PSO approach with dynamic population size

The carried out investigations deal with the application of machine learning algorithms to Duval Pentagon 1 graphical method for the diagnosis of transformer oil. In fact, combined to graphical methods, pattern recognition aims to may complement. For this purpose, we have used the Support Vector Machine (SVM) and the K-Nearest Neighbor (KNN) algorithms combined to the Duval method. The SVM parameters have been optimized with the Particle Swarm Optimization (PSO). Inspired from IEC and IEEE, five classes namely PD, D1, D2, T1&T2, and T3 have been adopted. The combined algorithms were verified using 155 samples from IEC TC 10 and related databases. We found that KNN, SVM may complement the Duval Pentagon 1 diagnosis method.

Application of SVM and KNN to Duval Pentagon 1 for transformer oil diagnosis

At lightning discharge, the grounding systems’ transient behavior is influenced by time-dependent nonlinear behavior and frequency-dependent effects. These effects are related to complex phenomena arising from the soil ionization surrounding the conductors at high current and to the fast risetime of the impulse current, respectively. Most of the performed investigations were either focused on soil ionization effects or frequency-dependent effects. This paper proposes a model that incorporates both aspects including the mutual coupling and air–soil interface effects, giving a more realistic representation of the grounding systems behavior. Based on the transmission line approach, the model can compute the grounding systems transient response buried in uniform and nonuniform soil. The model is applied to horizontal, vertical electrodes, and grounding grid subjected to impulse currents of variable magnitudes and shapes. The model results are validated against published measurement results obtained with low-current and high-current impulses. The time-domain transient potential and transient impedances show good agreement with experimental results. Other impulse impedance parameters used for the grounding systems analysis are reproduced quite accurately. As an application, the impulse impedance of a horizontal electrode and a grounding grid buried in two-layer soil is discussed when the upper layer depth and the resistivity of soil layers are varied.

Time-Domain Modeling of Grounding Systems’ Impulse Response Incorporating Nonlinear and Frequency-Dependent Aspects

This study aims at elucidating the correlations between structural changes and the variation of dielectric properties of cross-linked polyethylene (XLPE) during thermal aging. For this purpose, thermal aging experiments were carried out on XLPE samples at 80, 100, 120 and 140 °C for different aging times for a maximum total duration of 5040 hours. Chemical changes were assessed by Fourier transform infrared spectrum (FTIR). Besides, the crystallinity ratio and the melting temperature were evaluated using differential scanning calorimetry (DSC). Frequency domain dielectric spectroscopy was used to investigate the evolution of the dielectric constant and losses of XLPE during aging. The results show that aging at 80 and 100 °C could help to improve the crystalline state of XLPE which leads to the decrease of the dielectric constant. However, for higher thermal stress, the main aging mechanism of XLPE is thermal oxidation which leads to deteriorate the morphology and increase both the polarization and losses, mainly at low frequencies. It has been found also that in some cases, the non-homogeneity of the color of XLPE seems to indicate the increase of interfaces in the material which could lead to increase the dielectric losses due to the interfacial polarization.

/pdf/correlations-between-structural-changes-and-dielectric-2h6seyow6e.pdf

Correlations between structural changes and dielectric behavior of thermally aged XLPE

Dissolved gas analysis (DGA) is the standard technique to diagnose the fault types of oil-immersed power transformers. Various traditional DGA methods have been employed to detect the transformer faults, but their accuracies were mostly poor. In this light, the current work aims to improve the diagnostic accuracy of power transformer faults using artificial intelligence. A KNN algorithm is combined with the decision tree principle as an improved DGA diagnostic tool. A total of 501 dataset samples are used to train and test the proposed model. Based on the number of correct detections, the neighbor’s number and distance type of the KNN algorithm are optimized in order to improve the classifier’s accuracy rate. For each fault, indeed, several input vectors are assessed to select the most appropriate one for the classifier’s corresponding layer, increasing the overall diagnostic accuracy. On the basis of the accuracy rate obtained by knots and type of defect, two models are proposed where their results are compared and discussed. It is found that the global accuracy rate exceeds 93% for the power transformer diagnosis, demonstrating the effectiveness of the proposed technique. An independent database is employed as a complimentary validation phase of the proposed research.

/pdf/accuracy-improvement-of-power-transformer-faults-diagnostic-50h2jv9em6.pdf

Madjid Teguar

Papers

Optimal design of corona ring on HV composite insulator using PSO approach with dynamic population size

Application of SVM and KNN to Duval Pentagon 1 for transformer oil diagnosis

Time-Domain Modeling of Grounding Systems’ Impulse Response Incorporating Nonlinear and Frequency-Dependent Aspects

Correlations between structural changes and dielectric behavior of thermally aged XLPE

Accuracy Improvement of Power Transformer Faults Diagnostic Using KNN Classifier With Decision Tree Principle