改进算法在电力系统多目标负荷分配中的应用  

Improved Hybrid Algorithm in Multi-objective Optimal Load Dispatch Application

作　　者：汪佳[1] 姚建刚[1] 孙谦[1] 吴昊[2] 李昀[3] 金鑫[1] 

机构地区：[1]湖南大学电气与信息工程学院,长沙410082 [2]湖南省电力公司长沙电业局,长沙410002 [3]湖南省电力公司超高压管理局,长沙410014

出　　处：《电力系统及其自动化学报》2013年第1期107-111,135,共6页Proceedings of the CSU-EPSA

摘　　要：针对火电厂负荷优化分配问题,将同时考虑发电成本、污染排放和负荷调整时间三个方面,建立多目标优化组合模型。在优化算法上,提出一种小生境遗传禁忌搜索混合算法。此算法以小生境遗传算法为主体,避免传统遗传算法"早熟"和解的多样性不足的问题;并结合禁忌搜索思想,使用TSR算子进行交叉操作,改进了传统遗传算法局部搜索能力较差和收敛速度有待提高的问题。以某火电厂2台200 MW和2台300 MW机组负荷优化分配为例,将所提混合算法与其他两种传统算法进行了对比验证,结果表明此改进混合算法的有效性。According to the concept of load dispatch in power plants, this paper considers the power cost, pollution e- mission and load adjusting time. Then a multi-objectives optimal load dispatch model is presented in the paper. A hy- brid optimization algorithm is used to solve the above problem. The algorithm takes niche genetic algorithm as the main part to control the pre-maturity and solve the lack of diversity; at the same time, TSR operator in crossover is used to improve local search ability and convergence speed. Take the load optimal problem of 2-machine 200 MW and 2-ma- chine 300 MW units in a thermal power as an example, the effectiveness and contrast between the hybrid algorithm and the other two traditional methods are verified. The results show the effectiveness of the hybrid algorithm.

关 键 词：厂级负荷分配 多目标优化 解集 小生境遗传算法 禁忌搜索算法 

分 类 号：TM744[电气工程—电力系统及自动化]