机组负荷优化的遗传-禁忌混合算法

针对遗传算法“爬山”能力差的缺点，提出在遗传算法中引入禁忌搜索操作来提高局部寻优能力，并应用遗传-禁忌混合算法进行了火电机组负荷优化研究.在遗传算法每计算一定代数后，将每个个体作为禁忌搜索的初始解进行搜索，可以改善群体质量，加快搜索速度.混合算法有效地综合了遗传算法的全局搜索能力和禁忌搜索算法的局部搜索能力，从而避免出现“早熟”现象，改善了算法的收敛性.应用某实例进行负荷优化计算，结果表明混合算法大大改善了局部搜索能力和搜索速度，可以有效解决机组在线负荷优化难题.

Genetic-tabu hybrid algorithm for optimization of unit commitment

Due to the poor hill-climbing ability of genetic algorithm(GA),tabu search(TS) operation was integrated with GA to improve the local search performance.A genetic-tabu hybrid algorithm was employed to solve the thermal unit commitment problem.Each individual in GA was used as the initial solution of TS algorithm after certain generations of GA iteration,accordingly the population quality and the search speed were improved.The hybrid algorithm has the global search ability of GA and the local search ability of TS to avoid the prematurity and improve the convergence performance.A case study shows that the hybrid algorithm can improve the local search performance and efficiently solve the unit commitment problem.