DISCRETE VAR OPTIMIZATION NEAR TO GLOBAL OPTIMUMBY GENETIC ALGORITHM
Fan Mingtian; Xiang Niande; Zhang Zuping ;Fu Chengjun(Tsinghua University)(Eectric Power Research institute, Ministry of Electric Power)Beijng 100084, China Beijing 100085, China
The allocations of reactive sources andthe control of discrete reactive facilities create a largescale non-linear discrete VAR optimization problem,which is a non-convex problem with multiple integerfeasible solutions mathematically. This research attemptsto solve the discrete VAR optimization problem based onsuch a technique that a Genetic Algorithm (GA) isemployed to find out an approximation solution near tothe global optimum. And to speed up the solutionprocedure, an expert system is utilized to provide powersystem information. A real distribution system with 18changeable taps and 7 capacitors is used to demonstratethe validity ofthe approach.
【CateGory Index】： TM769