Search

Power system security poses an important issue in planning and operation stages of a power system. Conventional method of security evaluation involves solving full AC load flow equations and transient stability analysis by time domain simulation. This technique is highly time consuming and infeasible for real time applications. The Pattern Recognition (PR) approach is recognized as an alternative tool for online security evaluation. This paper proposes a Support Vector Machine (SVM) based classification for transient security assessment. The proposed SVM-PR approach is tested on 9 Bus WSCC, 30 Bus and 57 Bus IEEE standard systems and the transient security status of the system is accessed. The performance of the SVM classifier is gauged in terms of measures like classification accuracy, execution time and misclassification rate and the results are compared with the Multilayer Perceptron (MLP) classifier. The proposed method helps to identify the most dominant system dependent features responsible for transient security classification and yields fairly high classification accuracy with less computation time. Further, the proposed approach can realistically be applied to large scale systems, as the number of input attributes selected for classification is independent of the system size. Copyright � JES 2009.

A new method for the solution of the problems of unit commitment (UC) and economic dispatch (ED) using Hopfield neural network (HNN) is proposed. The difficulty associated with combining these two problems is overcome with a new neural network architecture which uses the discrete type HNN for solving UC and the continuous type HNN for solving ED. The associated advantage of using HNN for the combined solution is the decoupling of their interdependence ie, both the UC and ED are iteratively solved using respective HNNfor the particular period. The implementation of the proposed method causes a considerable reduction in the HNN size, complexity and computation requirements. The proposed method was successfully tested on different systems (3, 5, 6,10, 26 generator units), with varying load pattern of different durations (24 h and 168 h).

Constrained evolutionary programming algorithm approach for two distinct studies of static and dynamic economic dispatch is presented. Economic dispatch (ED) is one of the main functions of power generation operation and control. The objective is to operate an electric power system most economically while the system is operating within its security limits. Multiple objective functions subjected to power flow constraints; Ramp rate limits for units, minimum and maximum allowable generations for each units are considered. Comparison of the computational efficiency and accuracy of constrained evolutionary programming (CEP) is provided. This paper proposes a novel methodology for solving dynamic economic dispatch. Comparison with other evolutionary methods like genetic algorithm, particle swarm optimization has been provided for 10-generator system with nonsmooth fuel cost functions to illustrate the suitability and effectiveness of the proposed method.

Artificial neural network approach to the problem of static security assessment of power system is presented. This paper utilizes the artificial neural net of Kohonen's self-organizing feature map (SOFM) technique that transforms input patterns into neurons on the two-dimensional grid to classify the secure/insecure status of the power system. SOFM uses the line flows under different component cases as inputs and self-organizes to obtain the cluster of the components based on their loading limits. The output of SOFM provides information about the violation of the constraints from which the operating state of the power system can be identified, which can be classified as secure or insecure. The proposed method of security assessment was initially demonstrated for a model 3 generator 6-bus system and later extended to IEEE-14, -30 and -57 bus systems. � Indian Institute of Science.