COMPUTATIONAL INTELLIGENCE IN ELECTRICAL ENGINEERING (INTELLIGENT SYSTEMS IN ELECTRICAL ENGINEERING)
Year:2014 | Volume:5 | Issue:2|Papers Count:8

In recent years with notice increase reliability in power system and Intelligent Systems and also notice that transformers are one of the main part of the transmission and distribution systems, online monitoring of these equipment in power system are require. In this paper, a new method for online interturn fault detection base on leakage flux in power transformer are propose. When an interturn fault occur the symmetry of flux destruction and leakage flux increase or decrease and for various location and severity of fault leakage flux is different and it can be used for fault detection. In this paper for measure these flux we using search coils that mounted on HV winding. To fault detection and classify we using probabilistic neural network. and for decrease the information volume PCA is used. The simulation results are compare and verified with experimental result and show that this propose method is very good.

Active and reactive power components of a Distributed Generation (DG) is normally controlled by a conventional dq-current control strategy however, after islanding the dq-current which is not able to successfully complete the control task is disabled and a lead-lag control strategy based optimized by simulated annealing is proposed for control of DG unit in islanding mode. Integral of Time multiply by Absolute Error (ITEA) criterion is used as cost function of simulated annealing in order to achieve smooth response and robust behavior. The proposed controller improved robust stability margins of the system. Simulations with different load and input operating conditions verify advantages of the proposed controller in comparison with a previously developed classic controller in terms of robustness and response time.

Microarray data have an important role in identification and classification of the cancer tissues. In cancer researches always a few samples of microarrays are led to some problems in designing the classifiers, so non-informative genes have been removed from microarray data before classification using the preprocessing gene selection techniques. Basically, appropriate gene selection method can significantly improve the performance of cancer classification. In this paper, a new method is proposed based on hybrid model Binary Particle Swarm Optimization algorithm and Bayesian Linear Discriminant Analysis in order to classification of large scale microarray data. First, the position of each particle is represented in the form of binary vector and random, as each bit illustrates a gene. The zero and one bits represent that the corresponding feature (gene) is not/is selected, respectively. So the position of each particle clarifies a gene subset and fitness of each particle is calculated using Bayesian Linear Discriminant Analysis algorithm to quality evaluation of selected gene subset by that particle. The proposed algorithm is applied on four cancer datasets and its results are compared with other existed methods. Simulation results illustrate that proposed algorithm has high accuracy and validity compared to other existed methods and enables to select the small subset of informative genes in order to increase the classification accuracy.

Reliability of Wind Energy Conversion Systems (WECSs) is greatly important regarding to extract the maximum amount of available wind energy. In order to accurately study WECSs during occurrence of faults and to explore the impact of faults on each component of WECSs, a detailed model is required in which mechanical and electrical parts of WECSs are properly involved. In addition, a Fault Detection and Isolation System (FDIS) is required by which occurred faults can be diagnosed at the appropriate time in order to ensure safe system operation and avoid heavy economic losses. This can be performed by subsequent actions through fast and accurate detection and isolation of faults. In this paper, by utilizing a comprehensive dynamic model of the WECS, an FDIS is presented using dynamic recurrent neural networks. In industrial processes, dynamic neural networks are known as a good mathematical tool for fault detection. Simulation results show that the proposed FDIS detects faults of the generator’s angular velocity sensor, pitch angle sensors and pitch actuators appropriately. The suggested FDIS is capable to detect and isolate the faults shortly while owing very low false alarms rate. The presented FDIS scheme can be used to identify faults in other parts of the WECS.

Digital image has special cryptography algorithms for its specific properties. A mathematics sequence in most image cryptography has been used for image scrambling. The used mathematics sequence has a recursive equation which it has some coefficients that changes of these coefficients can generate different sequences. Performance of this sequence in image cryptography is evaluated with different standard criteria. Due to complexity of system and no direct relation between the coefficient and evaluation criteria, selection of the suitable coefficient is not easily possible. In this article, by considering a general form of recursive equation and define a fitness function, the proper coefficients are calculated by genetic algorithm that satisfies the evaluation criteria. The experimental results show that recursive equation that is computed by the genetic algorithm has satisfactory performance from some schemes.

This paper presents a hybrid optimization technique in order to solve large scale mixed integer optimization problems. The aforementioned technique is modeled as a two-level optimization problem consists of a master problem and some slave sub-problems. In the proposed method, a virtual database has been considered in line with the master problem to store evaluated cases during simulation. The virtual database accelerates the simulation process and also could be incorporated in multi-processor simulators. Composite generation and transmission expansion planning problem is modeled as a dynamic mixed integer optimization problem has been considered here to evaluate the proposed technique. The simulation results show that the presented method is satisfactory and consistent with the expectation.

In recent years, utilization of phasor measurement units (PMUs) has increased in monitoring, control and protection of power systems. In reality, power systems are large scale, accordingly, financial limitations (due to PMU cost) and technical problems are avoiding to install all necessary PMUs in one stage. Therefore, the PMUs usually are installed in several stages. This paper proposes a new dynamic multi-stage PMU placement approach by introducing a new index related to network observability in planning stages. Despite of conventional methods, the proposed multi-stage PMU placement is investigated dependently, simultaneously, and dynamically. Moreover, the phasing of PMUs of all stages is achieved in a single optimization process. Furthermore, in order to consider the practical aspect, the channel and communication limitations are covered in this study. According to the complexity of the proposed model, Bat Algorithm is used as an optimization tool to solve the proposed dynamic multi-stage PMU placement model. The proposed approach is applied on standard IEEE 14-, 57- and 118- bus test systems as well as Iranian 230- and 400-kV transmission network. Finally, the obtained results are compared with the results of conventional methods and ability of the proposed approach is investigated.

The Seeker Optimization Algorithm (SOA), as it is called now, is currently gaining popularity in the community of researchers, for its effectiveness in solving certain difficult real world optimization problems. This paper proposes SOA based Load Frequency Control (LFC) for the suppression of oscillations in power system. A two area system and four area system deregulated is considered to be equipped with proportional plus integral (PI) controllers. SOA is employed to search for optimal controller parameters by minimizing the time domain objective function. The performance of the proposed controller has been evaluated with the performance of the conventional PI controller and PI controller tuned by Particle Swarm Optimization (PSO) and Bacterial Foraging Algorithm (BFA) in order to demonstrate the superior efficiency of the proposed SOA in tuning PI controller. Simulation results emphasis on the better performance of the optimized PI controller based on SOA in compare to optimized PI controller based on PSO and BFA and conventional one over wide range of operating conditions, and system parameters variations.