基于矩阵实数编码遗传算法求解大规模机组组合问题

该文提出了一种采用矩阵实数编码遗传算法(MRCGA)进行机组组合优化的新方法：采用矩阵实数编码方式对整体发电计划进行编码后，可直接运用遗传操作求解机组组合问题，避免将其分解成机组启停安排和经济负荷分配的两层优化问题进行求解；采用多窗口变异技术，增强了算法的搜索能力。此方法提出了一种新的个体调整方法，可以处理各项约束条件，保证了结果的可行性。文中通过2个算例及与其它算法的对比分析，验证了所提出的方法在大规模机组组合问题求解时具有很强的适应性和全局搜索能力。

A solution to the Unit Commitment Problem Based on Matrix Real-coded Genetic Algorithm

An approach for solving large scale unit commitment (UC) problems based on matrix real-coded genetic algorithm (MRCGA) with multi-window mutations and a new repairing mechanism is presented. The GA chromosome consists of a two-dimensional real number matrix representing the generation schedule. Using proposed coding manner, the MRCGA can directly solve UC through genetic operations and avoid coping with economic dispatch problem in each hour. Multi-window mutations improve the search performance of MRCGA and the new repairing mechanism is applied to the infeasible solutions. The algorithm is tested and validated in two cases. The results show that the MRCGA for large UC is versatile and efficient.