机组优化组合问题的随机tabu搜索算法

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

针对机组组合问题的特点,作者对tabu搜索策略和模拟退火算法(SA)的组合算法随机tabu搜索策略的编码方式、移动规则和tabu表的构造等作了适应性的改进,并对实际机组组合算例进行了求解。仿真结果表明,tabu-SA组合算法有效地结合了SA算法的大规模寻优特性和tabu搜索的局部搜索能力强的特点,能快速搜索到高质量的系统优化解。而且随机tabu搜索算法具有框架性算法不易陷入局部最优的特点,易于与其他算法结合来构造新的高效组合算法。

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.