《差分进化算法最优潮流》.pdf

第 34 卷 第 8 期 电 网 技 术 Vol. 34 No. 8 2010 年 8 月 Power System Technology Aug. 2010 文章编号：1000-3673 （2010 ）08-0123-06 中图分类号：TM 72 文献标志码：A 学科代码：470·4051 基于改进差分进化算法的电力系统最优潮流计算 赵树本，张伏生 （西安交通大学电气工程学院，陕西省 西安市 710049 ） Solution of Optimal Power Flow Based on Differential Evolution and Its Modified Algorithm ZHAO Shu-ben, ZHANG Fu-sheng (School of Electrical Engineering, Xi’an Jiaotong University, Xi’an 710049, Shaanxi Province, China) ABSTRACT: The calculation of optimal power flow (OPF) is 算法包括简化梯度法[1] [2] [3] 、牛顿法 、解耦法 和内点 the process to solve a multi-dimensional nonlinear optimization 法[4-5]等。这类算法的特点是利用目标函数对控制变 problem. By means of adjusting active output and terminal 量一阶或二阶梯度求解，由于这类算法对起始点非 voltage of generation units as well as transformer voltage ratio 常敏感，当搜索起始点在局部最优点的收敛域以内 to make the fuel cost function minimized and ensure all 时，常常收敛到局部最优解，所以这些传统的数学 constraint conditions satisfied. For this purpose, it is proposed to solve OPF by improved differential evolution algorithm. 优化方法难以解决大规模非线性优化问题。现代优 Calculation results of IEEE 30-bus system show that the 化算法主要是进化类算法。自 1960 年首次提出遗 algorithm utilized in this paper possesses good search ability 传算法(genetic algorithm，GA)后，产生了很多其他 and convergence performance, and the fuel cost for generation 进化类算法，如进化规划[6] 、差分进化(differential units can be effectively reduced by this algorithm than by other evolution ，DE)[7]算法、粒子群优化(particle swarm evolutional algorithms. optimization，PSO)[8-10]、蚁群优化[11-12] 。这类算法 K