GA/SA/TS hybrid algorithms for reactive power optimization

Reactive power optimization in a power system solved by adjusting generator voltages, transformer taps and capacitors/reactors is a mixed integer nonlinear programming problem. GA (genetic algorithm), SA (simulated annealing) and TS (tabu search) are widely used in combinatorial optimization. Combining the advantages of individual algorithms, three GA/SA/TS hybrid algorithms to solve the reactive power optimization problem are proposed in this paper. Trying to reasonably combine local and global search, they adopt the acceptance probability of SA to improve the convergence of the simple GA, and apply tabu search to find more accurate solutions. Two power systems, the IEEE 30 bus system and a 125-bus practical area power system with 64 control variables in Shandong province, China, have been tested. Comparison results of the proposed algorithms with GA, SA/GA and TS show that the proposed GA/SA/TS hybrid method has the strongest capability of finding global optimal solution within reasonable computing time