【作者】 秦梁栋；

【导师】 曹一家；

【作者基本信息】 华中科技大学 ， 电力系统及其自动化， 2004， 硕士

【摘要】 本文着重研究了改进型遗传算法在电力系统中的应用,其主要包括经济调度和无功优化两个方面。论文首先对遗传算法在电力系统中的应用进行了综述,然后基于Queen-bee进化(Queen-bee进化是模拟蜂王繁殖这一生态行为发展而来的自适应寻优算法),提出了一种改进型遗传算法以提高遗传算法的寻优能力。并以电力系统经济调度问题为背景,通过数字仿真实例验证了该改进算法的有效性。针对电力系统经济调度问题的多目标特性,提出了一种混合多目标遗传算法。该算法除了改进传统目标函数加权法使之更适于多目标优化问题之外,还利用模糊逻辑技术来自适应调整交叉概率、交叉点位置以及变异概率,使算法具有优良的收敛性能。数字仿真实例验证了该改进算法的有效性,最后,提出了一种应用于电力系统无功优化问题的改进型遗传算法。该算法在一般遗传算法的基础上,利用模糊逻辑技术来自适应调整交叉概率、交叉点位置及变异概率。通过对IEEE30节点系统的计算分析表明,该算法优于一般遗传算法。更多还原

【Abstract】 This paper aims at providing us with improved genetic algorithms and their applications in power systems, especially on the aspects of economic dispatch and reactive power optimization. Firstly, this paper surveys applications of genetic algorithms to reactive power optimization and economic dispatch of power system, secondly, it proposes a queen-bee evolution based genetic algorithm for solving the economic dispatch problem of power system. This queen-bee evolution is similar to nature in that the queen-bee plays a major role in reproduction process and this proposed algorithm can enhance the optimization capability of genetic algorithms. Numerical results on two actual systems of 6 generators and 13 generators respectively show that the proposed algorithm is faster and more robust than the conventional genetic algorithm, thirdly, it proposes a hybrid multi-objective genetic algorithm for optimizing the multi-objective problems in the economic dispatch of power system. This proposed algorithm differs from other multi-objective genetic algorithms in its selection procedure, crossover procedure and mutation procedure. The selection procedure selects individuals for a crossover operation based on a weighted sum of multiple objective functions. The characteristic feature of the selection procedure is that the weights attached to the multiple functions are not constant but randomly specified for each selection. Furthermore, the crossover procedure and mutation procedure adaptively adjust the crossover probability, crossover position and mutation probability based on fuzzy logic technology. Numerical results on an actual system of 13 generators show that this proposed algorithm is effective, finally, it presents an improved genetic algorithm to reactive power optimization. This approach adaptively adjusts the crossover probability、crossover position and mutation probability based on fuzzy logic technology. The proposed method has been applied to the IEEE30 buses power systems. The computation results show that this approach can find optimal solution more efficiently.更多还原