P. Venkata; V. Pandya; A.V. Sant
Abstract
This paper proposes a complete diagnostic analysis of faults in a typical modern power system's transmission line using the support vector machine (SVM) with time-series parameters and frequency series parameters as features. The training and testing data of the proposed method are collected by simulating ...
Read More
This paper proposes a complete diagnostic analysis of faults in a typical modern power system's transmission line using the support vector machine (SVM) with time-series parameters and frequency series parameters as features. The training and testing data of the proposed method are collected by simulating all types of faults with all possible variations on a transmission line (TL) in the IEEE-9 bus system using the PSCAD/EMTDC software. While simulating one type of fault, fault resistances and fault inception angles are also varied to account for the various behaviours of the fault. The three-phase instantaneous currents and voltages on both sides of TL are recorded at 32 samples per cycle. A thirty-two sample moving window is used to compute time-series and frequency-series parameters applied as features to the SVM. Ten-fold cross-validation is used to evaluate the performance of the proposed algorithm with evaluation metrics such as accuracy, precision, recall and F1 score. Features generation, training and testing of the proposed method, and performance comparison are done using PYTHON software. The proposed method has achieved an average accuracy of 99.996%, even in the most contaminated environment of 30 dB noise. Compared with the performance of the other popular machine learning algorithms, the proposed method has achieved more accuracy. The performance of the proposed method is also tested with different noise levels, which account for the measurement errors of 30 dB, 35 dB and 40 dB.
M.K.K. Alabdullh; M. Joorabian; S.G. Seifossadat; M. Saniei
Abstract
Transformers are one of the most important parts of the electric transmission and distribution networks, and their performance directly affects the reliability and stability of the grid. Maintenance and replacing the faulted transformers could be time-consuming and costly and accordingly, a solution ...
Read More
Transformers are one of the most important parts of the electric transmission and distribution networks, and their performance directly affects the reliability and stability of the grid. Maintenance and replacing the faulted transformers could be time-consuming and costly and accordingly, a solution should be proposed to prevent it. This led to studies in the field of transformer lifetime management. As a result, estimating the remaining lifetime of the transformer is a crucial part for the mentioned solution. Therefore, this paper aims to tackle this issue through employing a new algorithm to estimate the lifetime of a transformer by combining selection methods and Artificial Intelligence (AI)-based techniques. The main goal of this method is to reduce the estimation error and estimation time simultaneously. The proposed approach assesses transformers based on environmental conditions, power quality, oil quality, and dissolved gas analysis (DGA). Consideration of additional factors overcomes the disadvantage of traditional methods and gives a meticulous result. In this respect, the collected data from the power transformer of Iran and Iraq as well as regions with different conditions are employed in the studied algorithm. Several combinations of algorithms are investigated to choose the best one. Principal Component Analysis (PCA) is employed in the next step for weighing the various parameters to improve the accuracy and decrease execution time. Results show that the Bayesian neural network provides the best performance in the predicting remaining lifetime of the transformer with an accuracy about 98.4%.
P. Venkata; V. Pandya; A.V. Sant
Abstract
This paper reports support vector machine (SVM) based fault detection and classification in microgrid while considering distortions in voltages and currents, time and frequency series parameters, and differential parameters. For SVM-based fault classification, the data set is formed by analysing the ...
Read More
This paper reports support vector machine (SVM) based fault detection and classification in microgrid while considering distortions in voltages and currents, time and frequency series parameters, and differential parameters. For SVM-based fault classification, the data set is formed by analysing the operation of the standard IEC microgrid model, with and without grid interconnection, under different fault and non-fault scenarios. Fault scenarios also include different locations, resistances, and incident angles of fault. Whereas, for non-fault scenarios, the variation in load is considered. Voltages and currents from both ends of the distribution line (DL) are sampled at 1920 Hz. The time and frequency series parameters, total harmonic distortion (THD) in current and voltage, and differential parameters are determined. The SVM algorithm uses these parameters to detect and classify faults. The performance of this developed SVM based algorithm is compared with that of different machine learning algorithms. This comparative analysis reveals that SVM detects and classifies the faults on the microgrid with an accuracy of over 99.99%. The performance of the proposed method is also tested with 30 dB, 35 dB, and 40 dB noise in the generated data, which represent measurement errors.
Modeling and Identification of Technological Processes in the Fields of Power Engineering
D.S. Talgatkyzy; N.H. Haroon; S.A. Hussein; S.Kh. Ibrahim; K.A. Jabbar; B.A. Mohammed; S.M. Hameed
Abstract
Given the significant uncertainty surrounding future electricity prices, which is widely regarded as the most critical factor in this context, market participants must engage in forecasting to facilitate their exploitation and planning activities. The success of electricity market actors is dependent ...
Read More
Given the significant uncertainty surrounding future electricity prices, which is widely regarded as the most critical factor in this context, market participants must engage in forecasting to facilitate their exploitation and planning activities. The success of electricity market actors is dependent on the availability of more appropriate tools to address this issue. In contrast, there is a prediction of prices in the electricity market for varying periods due to the increasing use of renewable energy in global energy generation and the unsteady and disjointed configuration of renewable energy production. The fluctuating characteristics of wind energy production have increased the complexity of real-time demand management in power systems. This paper investigates the impact of renewable energy production on price forecasting using data from the Nord pool market's electricity market. The primary goal is to present a framework for forecasting market settlement prices using a hybrid wavelet-particle swarm optimization-artificial neural network (W-PSO-ANN). In two scenarios, the results showed that the proposed model accurately represents data and is more precise than the ANN and WANN models. Machine learning has demonstrated promise in predicting electricity prices, but it is not without limitations. The ANN, WANN, and W-PSO-ANN models have training phase RMSE indices of 0.09, 0.07, and 0.04 respectively. During testing, the values were 0.15, 0.11, and 0.08. This demonstrates that the proposed model outperforms previous models.