COMPUTATIONAL INTELLIGENCE IN ELECTRICAL ENGINEERING (INTELLIGENT SYSTEMS IN ELECTRICAL ENGINEERING)
Year:2011 | Volume:2 | Issue:1|Papers Count:6

In recent years the use of intelligent systems in science and engineering, especially in the diagnosis of disease, is increasingly growing. In this paper a smart way to diagnose heart disease (cardiac arrhythmias) is presented. This method is based on a combination of structures using neural networks for classification of normal operation and four abnormal heart functions. In the combination of these structures, some neural networks as a mediator, and some of them have been used as a specialist. In the proposed method firstly for removing noise from ECG signal, preprocessing was performed. The various time features (including fifteen properties) and wavelet features (includes fifteen feature) are extracted from the noise free signal and given the large number of selected features, principal components analysis is used for feature reduction to eight features. The proposed structures of MLP neural networks and RBF neural networks are appropriately trained for classification of arrhythmias and their performance has been evaluated. The results of the implementation of the proposed method on MIT / BIH database show the better performance in the diagnosis of cardiac arrhythmias compared to previous approaches.

In path planning of mobile robots dealing with concave obstacles is a major challenge. More specifically in real-time planning where there is no complete representation of the environment, this challenge would be much more problematic. In such cases local minimums and high computations cost are the most important problems. In this paper, in order to reduce computational cost, cellular automata as a distributed computational method with parallel processing properties is employed as tool for path planning purposes. The environment of the robot is modeled as a two dimensional cellular automata with four states. Evolutionary rules of the automata are proposed to perform the planning task. The proposed method is appropriate for single robot systems as well as multi robot systems. The proposed method is afterwards extended to be employed for concave obstacles using an ant colony inspired technique. The most superior advantage of the proposed method is its capability of real-time path planning of mobile robots with no need to prior representation of the environment.

In this paper, Generation Expansion Planning (GEP), is modeled as an optimization problem in which the objective function is to minimize the total investment, operation, and outage (energy not served) costs of power system as well as salvage value of investment costs. Generation system reliability is assessed and provided by means of EENS and LOLP indices. To solve the GEP problem, a new Modified Shuffled Frog Leaping namely MSFL algorithm is proposed. A new frog leaping rule and a new strategy for frog distribution into memeplexes is introduced to improve the local exploration and performance of the original SFL algorithm. To show the effectiveness of the MSFL algorithm, it is applied to a test system with 15 existing power plants and 5 types of new candidates, for a 12-years and a 24-years planning horizon. The original SFL algorithm and the Genetic Algorithm (GA) are also applied to solve the GEP problem. Simulation results show the advantages of the proposed MSFL algorithm over the original SFL and GA.

The niche methods for search algorithms are important techniques in optimization. Most niche techniques need some extra tunable parameters to get a better performance. Achieving a good method for multimodal optimization by heuristic algorithms will be possible if and only if the population diversity is preserved. Shuffled Frog leaping (SFL) algorithm is a new heuristic algorithm that its ability is not proved for solving the multimodal problems. This paper proposes a niche method for SFL. Several benchmark problems are considered for testing the robustness and effectiveness of the proposed method over the results available in the literature. The results show that the proposed method performs well.

Recent developments of Facts devices increase the importance of their coordination with the power system controllers. With regard to nonlinearities of power system, changes in the operating points, reaction between power system and STATCOM, linear methods cannot be used to design parameters of stabilizers. Therefore, in this paper, a nonlinear model of power system is considered for the coordination design of PSS and STATCOM. A hybrid method which combines bacterial foraging (BF) algorithm with Nelder-Mead (NM) method (BF-NM) is employed to coordinately design the PSS and STATCOM controllers. By combining these two methods, the search power of the intelligent methods and the precision of conventional methods are simultaneously employed. To evaluate the performance of the proposed method, it is applied on a four machine power system. Simulation results confirm the efficiency of the proposed method for stabilizing power system oscillations.

Brushless permanent magnet motors play an important role in many applications. Power density and efficiency are two important factors in designing such motors. This paper proposes a novel approach to design a brushless permanent magnet motor based on optimization of a combination of power density and efficiency. First of all, this paper presents the equation related to the design and dimensions of brushless permanent magnet motor. Then, an optimum design based on bee's algorithm (BA) with the purpose of increasing power density is presented. Finally, Simulation results of a 2-D finite element analysis have well validated the efficiency of the applied method.