OPTIMAL ALLOCATION OF SLOW AND FAST VAR RESOURCES CONSIDERING MULTI LOAD LEVEL STATES

摘要

This paper deals with optimal allocation of fast and slow VAR devices under different load levels. These devices are supposed to be utilized to maintain system security in normal and contingency states, where corrective and preventive controls are implemented for the contingency cases. Load shedding and fast VAR devices are used in the corrective state in order to quickly restore system stability even they are expensive, while cheap slow VAR devices can be used in the preventive state to obtain the desired security level. The main objective of this paper is to make a trade-off between economy and security by determining the optimal combination of fast and slow controls (load shedding, new slow and fast VAR devices). To meet desired security limits, huge numbers of constraints during the investigated transitions states, have to be considered. Therefore, overall problem is formulated as a large scale mixed integer nonlinear programming problem, which cannot be solved by conventional optimization methods. Due to the proved effectiveness of Particle Swarm Optimization method (PSO) in performing difficult optimization tasks, this paper discovers the efficiency of PSO approach in solving our problem. The results and convergence of GA and PSO methods are also compared. The proposed approach has been successfully tested on IEEE 14 bus system.