COMPUTATIONAL INTELLIGENCE IN ELECTRICAL ENGINEERING (INTELLIGENT SYSTEMS IN ELECTRICAL ENGINEERING)
Year:2021 | Volume:12 | Issue:1|Papers Count:9

Heart sound signal processing consists of different phases. After applying necessary preprocessing and segmenting heart sound cycles, some distinctive features of heart sound are extracted. Since the appropriate operation of the classifier has a high impact on the performance of the system, in this study we propose a proper classification algorithm. One of the commonly used methods to build accurate classifiers is to use a group of classifiers and make decision based on the outputs of these classifiers. By far, the performance of the ensemble methods has been investigated in different fields of classification problems by researchers. However, in the field of heart valve diagnosis there are almost no studies investigating these methods. In this study, we train several linear classifiers and the final decision is made according to the outputs of them based on the majority voting algorithm. The training samples of each classifier are chosen randomly with replacement from the whole training set. The proposed method is implemented for 5 datasets and also compared with 3 other methods using different criteria including sensitivity, specificity, diagnostic odds ratio, precision and error. Results show that the proposed method has higher accuracy and faster prediction time. The noise label problem and the robustness of the proposed method against this noise are also investigated. Statistical tests show that the proposed method significantly outperforms other methods. ]

Recently, due to environmental issues and lack of fossil resources, the widespread use of renewable resources has been considered. The use of renewable resources in the power systems leads to decentralized generation and due to the changing nature of such resources, many fluctuations have been added to the network. DC microgrids, as a small power grid, make it easy to use and control such resources in the distribution network. Excessive penetration of power electronic based renewable sources has reduced the inertia and damping making the system more sensitive to disturbances and reducing the system stability margin. This paper uses a DC microgrid structure connected to an AC network equipped with a virtual synchronous machine to manage automatically the power of the DC microgrid and AC networks and analyzing the role of virtual inertia in small signal stability. In order to verify the proposed method, the network has been studied for different operating conditions and inertias. In this method, with the use of the voltage source converter response and proper design dual droop control of the voltage, frequency and active output controllers, the necessary support for adjusting the system frequency are provided to withstand disturbances leading to enhancing stability for different conditions.

In this paper, a hybrid model for increasing the precision of the support vector machine classifier is proposed to detect stator windings short circuit fault detection in induction motors. The proposed method consists of three different phases, wherein the first phase the statistical features of a healthy and defective data set are extracted. The principal component analysis is used to reduce the dimensions of the obtained features. Then, different SVMs are constructed based on training data sets. To achieve a better result, the parameters of the SVM are determined by the fractionalorder chaotic particle swarm optimization algorithm. Finally, a hybrid model for combining SVMs with type-2 Fuzzy logic is implemented. The proposed approach is then applied on measured stator current data for stator winding short circuit fault detection in a three-phase induction motor with 2. 2kW, 50Hz, 6 Pole. The average accuracy of 98. 4% of the detection of stator winding error on laboratory data under different load conditions indicates the performance and validity of the proposed algorithm.

In recent years, the use of renewable energy systems has grown significantly, among which photovoltaic systems have received much attention. Solar cells are known as the building blocks of a photovoltaic system. Because of the nonlinear nature of solar cells and the continuous changes in atmospheric conditions, maximum power point tracking (MPPT) is essential to extract the maximum power of a photovoltaic system. In this study, in order to achieve the maximum power, it was proposed to apply flower pollination algorithm (FPA) combined with a comprehensive selection algorithm, named as improved FPA. In addition, to evaluate the proposed algorithm, its performance was compared with genetic algorithm (GA) and standard FPA under rapid changes in atmospheric conditions. The calculated results showed that the improved FPA has a better accuracy than GA; moreover it has a higher convergence rate as compared with other applied algorithms.

Estimating the status of battery charge (SOC) in lithium-ion batteries is important not only for optimum energy management but also for ensuring safe operation and preventing charge and discharge and thus reducing battery life. However, this parameter cannot be directly measured from the battery terminals. Therefore, SOC needs to be estimated. In this paper, the recursive least squares method (RLS) is used to estimate the battery parameters and the modified particle filter is used to estimate the SOC of lithium-ion batteries. The standard particle filter has the problem of particle degeneracy phenomenon, which reduces estimation accuracy. Therefore, in modified particle filter, the difference evolutionary algorithm and the Markov chain Monte Carlo) MCMC (method are applied to the standard PF, that makes the estimation of SOC more accurate and consistent. In order to evaluate the performance of the proposed method, this method is compared with the classical methods. The results show the effective performance of the proposed method compared to other methods.

Electric field intensity is one of the factors affecting the corona discharge and insulation of high voltage composite insulators. Accordingly, it is necessary to adjust the potential distribution and electric field along the insulator. Using the corona ring on the high voltage side has a great potential to minimize the intensity of the electric field. Since the design and installation conditions of the corona ring can change the electric field, in this paper, Derivative-Free Solvers (DFS) based numerical solution methods are used to obtain optimal parameters. Three-dimensional finite element method (FEM) in COMSOL software is employed to simulate and compute the electric field. Comparison of the results has shown that Derivative-Free Solvers have acceptable speed and good convergence. The parameters obtained from the optimization methods can reduce the electric field by up to %66. According to the results, FEM-DFS hybridization technique could be very helpful in optimization of corona ring design.

In this paper, a method for evaluating the impact of load response on microgrids is presented. Several different modes have been stimulated to better understand the problem. A hybrid optimization algorithm including gray wolf optimization algorithm and shark olfactory optimization algorithm to optimize the multi-purpose objective function under limited conditions and constraints. In addition, to determine the uncertainty in the production of renewable energy sources, the Monte Carlo method has been used to produce the scenario. The parameters considered in this method include technical parameters such as network losses, generation cost, and reliability index and voltage deviation. The proposed method is implemented using a hybrid optimization algorithm using MATLAB software on a modified 69-bus system, including wind turbines, solar power plants and energy storage systems. The results show that the proposed method will increase network productivity.

In this study, a hybrid fuzzy intelligent method for diagnosing and managing uncertainty in input features to identify breast tumors in mammography images has been proposed. Moreover, a hybrid fuzzy evolutionary model has been applied to increase the system efficiency and optimize results. Soft computing models were used to detect the type of breast mass based on the analysis of features in mammography images. The combined models proposed in this study include FuzzyTBO and Fuzzy-PSO-TLBO models. The performance evaluation was conducted using the Receiver Operator Characterization (ROC) analysis in terms of accuracy and area under the ROC curve. A 10-fold cross-validation technique was conducted to divide the data into training and testing sections. The obtained results reveal an accuracy of 96. 27% for determining different types of mass based on tumors’ features according to the images. The presented model competes and outperforms other proposed models in previous studies. The outcome of this study may be promising for apropos diagnosis and effective treatments.