【作者】 刘杰；

【导师】 张怡；

【作者基本信息】 西南交通大学 ， 电气工程， 2012， 硕士

【摘要】 能源是人类社会发展的必需条件,电能作为现代社会最主要的二次能源之一,在国民经济和和人民日常生活中具有极广泛的应用。随着电力技术的发展,现代电力工业进入了大系统、超高压、远距离、大容量的发展时代,电力系统运行中的各种问题日益突出,对系统的经济、安全、稳定运行具有重要的影响。电力系统无功优化和经济负荷分配是其中比较典型的两个优化问题,本文采用粒子群算法及其改进算法对这两个问题进行了相关研究。针对电力系统无功优化问题,本文建立了一个以有功损耗最小和电压质量最好为目标函数的多目标优化模型,并采用固定权重法将多目标函数转变为单目标函数；采用了三种优化算法：粒子群算法(PSO)、自适应权重粒子群算法(AWPSO)和继承学习粒子群算法(ILPSO),对IEEE30、IEEE118节点系统进行了仿真计算,结果表明ILPSO算法可以有效地求解无功优化问题。同时,在三种粒子群算法中选取了不同的种群数及迭代次数对无功优化问题进行了计算和分析,结果表明不同的算法参数对优化结果具有较大的影响。在电力系统经济负荷分配问题中,本文以总发电成本最小为目标函数的优化模型,同时还考虑了系统传输网损,采用罚函数的形式处理功率平衡约束：采用了PSO算法、AWPSO算法和ILPSO算法三种优化算法,对3机组、6机组、15机组算例进行仿真计算,与文献结果进行了比较,表明ILPSO算法对于电力系统经济负荷分配问题的求解是有效、可行的。分析了不同的算法参数对优化结果的影响,结果表明选取正确的参数可以得到合适的优化结果。对于40机组,还考虑了阀点效应,通过ILPSO算法进行了仿真计算,获得的结果优于文献结果。本文利用三种粒子群算法在不同参数的条件下对电力系统无功优化和经济负荷分配问题进行了研究,可以得出选取不同的算法参数设置对优化结果具有较大的影响以及ILPSO算法在求解这两个优化问题中有效性和优越性的结论。更多还原

【Abstract】 Energy is a necessary condition for the development of human society. In the modern society, electricity is one of the most important secondary energy, which has a very wide range of applications in the national economy and people’s daily life. With the development of electricity technology, modern power industry has entered the developing period of large-scale system, extra high voltage, long-distance, high-capacity, but various problems in power system operation have also become increasingly prominent, which has an great impact on economy, safe and stable operation of power system. Reactive power optimization and economic load dispatch is typical of two optimization problems. In this paper, which two problems are research by particle swarm optimization algorithm and its improved.In reactive power optimization problem, the mathematical model of multi-objective optimization has been established in this paper. The objectives consist of real power loss and voltage quality. This paper uses fixed-weight method to transform the multi-objective functions to the single objective function. Three optimization algorithms are used:they are particle swarm optimization (PSO), adaptive weight particle swarm optimization (AWPSO) and inheritance learning particle swarm optimization (ILPSO), which has been used to solve the IEEE30, IEEE118bus system.The results show that ILPSO algorithm can effectively solve the reactive power optimization problem. Meanwhile, the different species and iterations of algorithm which are selected to calculate and analyze reactive power optimization problem. The results show that different parameters of the algorithm have a great impact on the results of optimization.In power system economic load dispatch problem, the optimization model of total generation minimum cost has been established in this paper. Also the system transmission loss has been considered and the penalty function has been considered to handle the power balance constraints. The three optimization algorithms are used:they are particle swarm optimization (PSO), adaptive weight particle swarm optimization (AWPSO), inheritance learning particle swarm optimization (ILPSO). Three simulated examples including3-generators,6-generators and15-generators are calculated by the three optimization algorithms. The calculated results are compared with literature results; ILPSO algorithm is an effective and feasible method to solve the power system economic load dispatch problem. The impact of different algorithm parameters to optimize on the results is analyzed. The influence of different algorithm parameters on the optimization results is analyzed.The results show that selecting the correct parameters can get suitable optimization results. For40-generators, the valve point effect is considered, and the calculation results with ILPSO are better than the literature results.In this paper, reactive power optimization and economic load dispatch are studied by three kinds of particle swarm optimization of the different parameters.It can be concluded to select the correct algorithm parameter, which has a great influence on optimization results. And ILPSO algorithm is an effective and superior algorithm for solving these two problems.更多还原