一种适用于机组组合优化的改进整数编码粒子群算法  

An modified integer-coded particle swarm optimization algorithm for unit commitment optimization

作　　者：吴和海 熊高峰[1] 袁晋蓉 秦跃杰 

机构地区：[1]湖南大学电气与信息工程学院,湖南长沙410082

出　　处：《现代电子技术》2017年第11期167-171,共5页Modern Electronics Technique

摘　　要：针对机组组合这一高维、非线性混合整数规划问题,提出一种结合修补策略的整数编码粒子群(ICPSO)算法。用正负整数分别表示机组开停机的时间长度,有效减少待优化变量个数。基于机组组合问题的特点,采用修补策略处理不满足约束条件的个体,使算法只在可行解区域内搜索,有效提高收敛速度,通过切除冗余机组,提高解的质量。仿真算例表明,相比整型编码遗传(r-ICGA)算法、改进粒子群(IPSO)算法、社会演化(SEP)算法,提出的ICPSO算法能够更有效地处理大规模机组组合优化问题,执行时间较短、求解精度更高。Aiming at the high dimension and nonlinear mixed integer programming problems of unit commitment,an integercoded particle swarm optimization(ICPSO)algorithm combined with repairing scheme is proposed. The positive and negative integers are used to represent the time length of unit startup and shutdown to reduce the quantity of variables under optimization effectively. On the basis of the characteristic of unit commitment,the repairing scheme is adopted to handle the individuals which can′t conform to the constraint conditions,so as to make it only search in the feasible solution region,and improve the convergence rate. The redundancy unit is cut out to improve the quality of solution. The simulation example results show that,in comparison with r-ICGA,IPSO algorithm and SEP algorithm,the ICPSO algorithm can handle the combinatorial optimization of largescale units effectively,has shorter execution time and higher solving accuracy.

关 键 词：机组组合 粒子群算法 整数编码 修补策略 

分 类 号：TN919-34[电子电信—通信与信息系统] TP18[电子电信—信息与通信工程]