改进单亲遗传算法在电源规划中的应用  

APPLICATION OF IMPROVED PARTHENO-GENETIC ALGORITHM IN GENERATION EXPANSION PLANNING OF POWER SYSTEM

作　　者：冯凯 李远德 侯云鹤[2] 吴耀武[2] 熊信艮[2] 

机构地区：[1]四川省电力公司,四川省成都市610021 [2]华中科技大学电力工程系,湖北省武汉市430074

出　　处：《电网技术》2004年第3期11-15,共5页Power System Technology

摘　　要：电源规划问题本质上是一多阶段组合优化问题,使用常规遗传操作会产生大量无效解,从而降低了求解效率文章提出了一种基于单亲遗传算法的电力系统电源规划模型,并采用分段编码法解决了单亲遗传算法用于电源规划的编码问题。该模型容易计及电源规划中需要考虑的各种约束条件。文中还研究了精英保留算子和扰动算子对收敛的影响。算例仿真表明:该模型可靠有效,既能获得最优解,也能获得次优解;加入特殊算子改进后,可进一步提高精度,加快收敛速度。The generation expansion planning (GEP) is essentially a complicated multi-stage combinatorial optimization problem. Traditional genetic operation may produce many ineffective chromosomes which would decrease the efficiency of search. A model for the generation expansion planning of power system based on partheno-genetic algorithm is presented and a subsection coding method is successfully used to solve the chromosome-coding problem. A variety of constraints which should be considered in GEP can be easily taken into account. The influence of several special operators, such as elitist reserved operators and disturbance operators, on the convergence is researched. Simulation results show that the presented model is reliable and effective, and using this model the global optimal solution with some sub-optimal solutions can be obtained. Acceding special operator into the presented model, both the calculation accuracy and the convergence speed can be improved.

关 键 词：电源规划 电力系统 人工智能 单亲遗传算法 目标函数 电网 

分 类 号：TM715[电气工程—电力系统及自动化] O242.23[理学—计算数学]