RANDOM TABU SEARCH WITH SIMULATED ANNEALING APPROACH FOR THE UNIT COMMITMENT PROBLEM

According to the features of unit commitment, some improvements are made to the hybrid algorithm which combines the tabu search strategy with simulated annealing algorithm, namely the random tabu search strategy, the implements include following items: the numerical coding for random tabu search strategy, tabu moving rule and the structure of tabu list. Besides, the examples of practical unit commitment are calculated. The simulation results show that in the combinational algorithm, which combines the tabu search algorithm with SA algorithm, the feature of large scale search of SA algorithm and the powerful local search ability of tabu search algorithm is effectively combined and the high quality optimized solution of the system can be quickly searched. Otherwise, because the random tabu search strategy possesses the character of not easy to fall into local optimization, therefore, in the further study and application this search strategy can be easily combined with other algorithm to form new hybrid algorithm with high efficiency.