University of IsfahanComputational Intelligence in Electrical Engineering2821-06899220180904Wind Power Forecasting by a New Hybrid Forecast Engine Composed of GA/EPSO-Based Mutual Information and Group Method of Data Handling (GMDH)Wind Power Forecasting by a New Hybrid Forecast Engine Composed of GA/EPSO-Based Mutual Information and Group Method of Data Handling (GMDH)1142289010.22108/isee.2018.89972.0FAVahidVahidinasabElectric Energy Systems Planning and Operation Group, Department of Electrical Engineering, Abbaspour School of Engineering, Shahid Beheshti University, Tehran, IranHosseinSohrabi VafaAbbaspour School of Engineering, Shahid Beheshti University, Tehran, IranJournal Article20140726a really tough and complicated issue. In this paper, a successful method is proposed for predicting the wind power productions which is based on the self-organized neural networks called Group Method of Data Handling (GMDH). By analyzing and discovering the hidden relationships between the inputs, the GMDH-based neural network intelligently presents the optimal model and predicts the output variable. Patterns used in this study are based on the two methods of artificial intelligence and information theory. At first, the effective variables are selected based on Mutual Information (MI) technique and the mixed particle swarm and genetic algorithm and after that the proposed forecast engine is used. In contrast to the mutual correlation method, in the proposed cross-entropy-based approach of this paper, non-linear relations between the variables are considered and the selection of effective variables in the forecasting of wind power in which nonlinear fluctuations and trends are observed are chosen more precisely and more accurately. In order to evaluate the ability, speed and accuracy of the proposed framework, real-world data of Sotavento wind farm in the Spain were used. The results of the study indicate that the proposed technique has a higher speed and accuracy than other methods.a really tough and complicated issue. In this paper, a successful method is proposed for predicting the wind power productions which is based on the self-organized neural networks called Group Method of Data Handling (GMDH). By analyzing and discovering the hidden relationships between the inputs, the GMDH-based neural network intelligently presents the optimal model and predicts the output variable. Patterns used in this study are based on the two methods of artificial intelligence and information theory. At first, the effective variables are selected based on Mutual Information (MI) technique and the mixed particle swarm and genetic algorithm and after that the proposed forecast engine is used. In contrast to the mutual correlation method, in the proposed cross-entropy-based approach of this paper, non-linear relations between the variables are considered and the selection of effective variables in the forecasting of wind power in which nonlinear fluctuations and trends are observed are chosen more precisely and more accurately. In order to evaluate the ability, speed and accuracy of the proposed framework, real-world data of Sotavento wind farm in the Spain were used. The results of the study indicate that the proposed technique has a higher speed and accuracy than other methods.https://isee.ui.ac.ir/article_22890_3677613289ee25dde9be88960c5d1a95.pdfUniversity of IsfahanComputational Intelligence in Electrical Engineering2821-06899220181003Localization of Epileptic Focus Using Simultaneously Acquired EEG-FMRI DataLocalization of Epileptic Focus Using Simultaneously Acquired EEG-FMRI Data15282297510.22108/isee.2018.111024.1123FAEliasEbrahimzadehSchool of Electrical and Computer Engineering, College of Engineering, University of Tehran, N Kargar St., Tehran, IranHamidSoltanian-ZadehSchool of Electrical and Computer Engineering, College of Engineering, University of Tehran, N Kargar St., Tehran, Iran -
Image Analysis Laboratory, Department of Radiology, Henry Ford Hospital, Detroit, MI, USABabakNadjar AraabiSchool of Electrical and Computer Engineering, College of Engineering, University of Tehran, N Kargar St., Tehran, IranJournal Article20180517Affecting daily lives of millions of people, Epilepsy is a common central nervous system (neurological) disorder where cell activity in the brain is disturbed, causing recurrent seizures. Epilepsy can be treated commonly by medications. Be that as it may, medications do not always work as one may hope, and thus, some patients tend to resort to surgeries. The primary challenge in such surgeries, and by extension any other surgery where some part of brain may need to be disabled, disconnected or removed, is managing to pose no threat to the critical healthy textures adjacent or close to the part being operated on. Therefore, the precise localization of epileptic focus is a matter of vital importance in treating this condition. Various algorithms have been proposed to localize the brain sources and thus to determine the epileptic focus; however, none has yet been able to offer a solution to effectively address this issue. With EEG signal containing temporal information and fMRI carrying spatial information, it is hoped that the combination of the two can yield optimal results. In this research, we first remove the artifacts caused by EEG gradients, and proceed to study the signal in and outside the scanner by localizing the brain sources. The simultaneous processing of EEG-fMRI enables us to make use of the temporal information in EEG to analyze fMRI. Epileptic foci are finally localized based on GLM method. This study has been conducted on 10 medication-resistant patients with epilepsy whose data was recorded in Iran National Brain Mapping Centre. The results suggest a significant improvement in localization accuracy compared to existing methods in the literature. Affecting daily lives of millions of people, Epilepsy is a common central nervous system (neurological) disorder where cell activity in the brain is disturbed, causing recurrent seizures. Epilepsy can be treated commonly by medications. Be that as it may, medications do not always work as one may hope, and thus, some patients tend to resort to surgeries. The primary challenge in such surgeries, and by extension any other surgery where some part of brain may need to be disabled, disconnected or removed, is managing to pose no threat to the critical healthy textures adjacent or close to the part being operated on. Therefore, the precise localization of epileptic focus is a matter of vital importance in treating this condition. Various algorithms have been proposed to localize the brain sources and thus to determine the epileptic focus; however, none has yet been able to offer a solution to effectively address this issue. With EEG signal containing temporal information and fMRI carrying spatial information, it is hoped that the combination of the two can yield optimal results. In this research, we first remove the artifacts caused by EEG gradients, and proceed to study the signal in and outside the scanner by localizing the brain sources. The simultaneous processing of EEG-fMRI enables us to make use of the temporal information in EEG to analyze fMRI. Epileptic foci are finally localized based on GLM method. This study has been conducted on 10 medication-resistant patients with epilepsy whose data was recorded in Iran National Brain Mapping Centre. The results suggest a significant improvement in localization accuracy compared to existing methods in the literature. https://isee.ui.ac.ir/article_22975_590bf38fa6d63305facbb157a0a64b20.pdfUniversity of IsfahanComputational Intelligence in Electrical Engineering2821-06899220181010Immunity Improvement of a Low-Impedance Restricted Earth Fault Relay by Making its Mechanism Intelligent based on Support Vector MachineImmunity Improvement of a Low-Impedance Restricted Earth Fault Relay by Making its Mechanism Intelligent based on Support Vector Machine29402299110.22108/isee.2018.111464.1132FAAliEbadiDept. of Electrical and Computer Engineering, Babol Noshirvani University of Technology, Babol, IranSeyyed MehdiHosseiniDept. of Electrical and Computer Engineering, Babol Noshirvani University of Technology, Babol, IranAli AkbarAbdoosDept. of Electrical and Computer Engineering, Babol Noshirvani University of Technology, Babol, IranJournal Article20180607Low-impedance Restricted Earth Fault (REF) relay is a unit of digital protection system of power transformers that its duty is detecting single phase to ground faults on the transformer terminal and turning it to core faults on its windings. It is a fast and sensitive type of protection schemes. However, during external fault with high magnitude fault currents and flowing magnetizing inrush current of power transformer, due to Current Transformer (CT) saturation, it may be exposed to maloperation. In this paper, immunity improvement of a REF relay is shown by making its algorithm intelligent. To do so, in the first step, major parameters used in the operating characteristic of a conventional REF relay and neutral current are chosen to be employed as the training input of Support Vector Machine (SVM) classifier. Then, a sample power system including a power transformer and related CTs is simulated by using PSCAD/EMTDC software under many different operating conditions involving internal fault, external fault, and inrush current. Finally, support vector machine is trained using obtained simulation results and after validating its accuracy, is employed as the intelligent core of the new REF protection scheme.Low-impedance Restricted Earth Fault (REF) relay is a unit of digital protection system of power transformers that its duty is detecting single phase to ground faults on the transformer terminal and turning it to core faults on its windings. It is a fast and sensitive type of protection schemes. However, during external fault with high magnitude fault currents and flowing magnetizing inrush current of power transformer, due to Current Transformer (CT) saturation, it may be exposed to maloperation. In this paper, immunity improvement of a REF relay is shown by making its algorithm intelligent. To do so, in the first step, major parameters used in the operating characteristic of a conventional REF relay and neutral current are chosen to be employed as the training input of Support Vector Machine (SVM) classifier. Then, a sample power system including a power transformer and related CTs is simulated by using PSCAD/EMTDC software under many different operating conditions involving internal fault, external fault, and inrush current. Finally, support vector machine is trained using obtained simulation results and after validating its accuracy, is employed as the intelligent core of the new REF protection scheme.https://isee.ui.ac.ir/article_22991_e48cd6bb7da8837975feda1fd82df1f7.pdfUniversity of IsfahanComputational Intelligence in Electrical Engineering2821-06899220181021A Hybrid Approach for Mid-Term Electricity Price Forecasting based on Support Vector Machine and Neural NetworksA Hybrid Approach for Mid-Term Electricity Price Forecasting based on Support Vector Machine and Neural Networks41542301610.22108/isee.2018.110266.1113FANavidNazeriSmart Microgrid Research Center, Najafabad Branch, Islamic Azad University, Najafabad, Iran.
Department of Electrical Engineering, Najafabad Branch, Islamic Azad University, Najafabad, IranMajidMoazzamiSmart Microgrid Research Center, Najafabad Branch, Islamic Azad University, Najafabad, Iran.
Department of Electrical Engineering, Najafabad Branch, Islamic Azad University, Najafabad, IranGhazanfarShahgholianSmart Microgrid Research Center, Najafabad Branch, Islamic Azad University, Najafabad, Iran.
Department of Electrical Engineering, Najafabad Branch, Islamic Azad University, Najafabad, IranJournal Article20180331In future smart grids, it's imperative to know the price of electricity market to guide the behavior of consumers and suppliers. This paper presents a hybrid approach for mid-term electricity price forecasting based on support vector machine and neural networks. In this method, at first, the price upper bound is considered. Then, the training set is divided into two parts including normal price and price spikes. Feature extraction applies on input data sets using stacked auto-encoders and a prediction model trained using each training set. Support Vector Machine (SVM) models with different kernel functions and a two layered feed-forward neural network were trained and tested with the proposed method. Simulation results using the proposed method show that this method has a significant effect on the speed of model training and improves forecasting accuracy.In future smart grids, it's imperative to know the price of electricity market to guide the behavior of consumers and suppliers. This paper presents a hybrid approach for mid-term electricity price forecasting based on support vector machine and neural networks. In this method, at first, the price upper bound is considered. Then, the training set is divided into two parts including normal price and price spikes. Feature extraction applies on input data sets using stacked auto-encoders and a prediction model trained using each training set. Support Vector Machine (SVM) models with different kernel functions and a two layered feed-forward neural network were trained and tested with the proposed method. Simulation results using the proposed method show that this method has a significant effect on the speed of model training and improves forecasting accuracy.https://isee.ui.ac.ir/article_23016_1cd164e456c34584a6962b7247ad4967.pdfUniversity of IsfahanComputational Intelligence in Electrical Engineering2821-06899220181021Optimal Multi-objective Development Scheduling of Electric Vehicles in Distribution Network using Particle Swarm OptimizationOptimal Multi-objective Development Scheduling of Electric Vehicles in Distribution Network using Particle Swarm Optimization55642301710.22108/isee.2018.106972.1068FAImanGoroohi SardouDept. of Electrical and Computer Engineering, University of Jiroft, Jiroft, IranJournal Article20170929The use of the electric energy stored in batteries of electric vehicles connected to the grid (V2G) will play a significant role in the development of distribution systems in the future. Electric vehicles (EVs) are able to charge during base load hours and inject energy into the grid during peak hours. Besides, the grid-connected EVs increase the system reliability under the outages. In this paper, Particle Swarm Optimization (PSO) algorithm is employed to solve the multi-objective problem of development scheduling of electric vehicles in the distribution network. ε-constraint method is employed to solve the proposed multi-objective problem. Besides, a fuzzy decision making approach is employed to determine the most compromise solution among the Pareto solutions obtained by the solving the sub-problems generated by the ε-constraint method. Decision variables include the location, and charge and discharge capacity of the smart parking lots of the EVs. IEEE 54-bus distribution test system is employed as the studied test network. Operation costs of the distribution network are compared in both states of with or without EVs. The results demonstrate the effectiveness of the proposed method for EVs’ development scheduling in the distribution network.The use of the electric energy stored in batteries of electric vehicles connected to the grid (V2G) will play a significant role in the development of distribution systems in the future. Electric vehicles (EVs) are able to charge during base load hours and inject energy into the grid during peak hours. Besides, the grid-connected EVs increase the system reliability under the outages. In this paper, Particle Swarm Optimization (PSO) algorithm is employed to solve the multi-objective problem of development scheduling of electric vehicles in the distribution network. ε-constraint method is employed to solve the proposed multi-objective problem. Besides, a fuzzy decision making approach is employed to determine the most compromise solution among the Pareto solutions obtained by the solving the sub-problems generated by the ε-constraint method. Decision variables include the location, and charge and discharge capacity of the smart parking lots of the EVs. IEEE 54-bus distribution test system is employed as the studied test network. Operation costs of the distribution network are compared in both states of with or without EVs. The results demonstrate the effectiveness of the proposed method for EVs’ development scheduling in the distribution network.https://isee.ui.ac.ir/article_23017_bfde48e458f0d8a96994e5be734c456e.pdfUniversity of IsfahanComputational Intelligence in Electrical Engineering2821-06899220181127Optimized Fuzzy PID Controller Design to Reduce Low Frequency Oscillations using the Improved Cryogenic Search Virus AlgorithmOptimized Fuzzy PID Controller Design to Reduce Low Frequency Oscillations using the Improved Cryogenic Search Virus Algorithm65822303910.22108/isee.2018.110543.1119FAMohammadEslamiDepartment of Electrical Engineering, Yazd Branch, Islamic Azad University, Yazd, IranMohammadRezaShayestehDepartment of Electrical Engineering, Yazd Branch, Islamic Azad University, Yazd, IranMajidPourahmadiDepartment of Electrical Engineering, Yazd Branch, Islamic Azad University, Yazd, IranVahidAyatollahi TaftiDepartment of Computer Engineering, Taft Branch, Islamic Azad University, Taft, IranJournal Article20180501In this paper, the modeling of the system under study was first proposed and then a fuzzy-PID controller was designed considering the downstream filter for optimal system performance to simulate the stability of a nonlinear system and to simulate low frequency oscillations. In designing the proposed controller, the parameters of controller and fuzzy members were considered as the variable, which ultimately turns into an optimization issue. The controlling structure of the proposed PID is such that, regardless of type and structure of the system under study, it can guarantee the stability of the system and minimize the volatility and frequency fluctuations. The most important feature of the proposed method is its independence on the structure of the system and operating conditions. On the other hand, this paper attempts to reduce the frequency deviation of the dynamic mode by improving the frequency controller and implementing a new method. The new method is based on minimizing total sum of summing time, peak time, peak value, and permanent error state over load changes using the virus search optimization algorithm. Investigations done by different criteria in time and frequency domains, using the proposed algorithm shows high efficiency in comparison with other methods in the articles. Also, the proposed fuzzy controller has a better performance for damping system disturbances in bad situations.In this paper, the modeling of the system under study was first proposed and then a fuzzy-PID controller was designed considering the downstream filter for optimal system performance to simulate the stability of a nonlinear system and to simulate low frequency oscillations. In designing the proposed controller, the parameters of controller and fuzzy members were considered as the variable, which ultimately turns into an optimization issue. The controlling structure of the proposed PID is such that, regardless of type and structure of the system under study, it can guarantee the stability of the system and minimize the volatility and frequency fluctuations. The most important feature of the proposed method is its independence on the structure of the system and operating conditions. On the other hand, this paper attempts to reduce the frequency deviation of the dynamic mode by improving the frequency controller and implementing a new method. The new method is based on minimizing total sum of summing time, peak time, peak value, and permanent error state over load changes using the virus search optimization algorithm. Investigations done by different criteria in time and frequency domains, using the proposed algorithm shows high efficiency in comparison with other methods in the articles. Also, the proposed fuzzy controller has a better performance for damping system disturbances in bad situations.https://isee.ui.ac.ir/article_23039_0d30add3b9f0fdbd432d46e2e79d4c39.pdf