【作者】 杭银丽；

【导师】 张俊芳；

【作者基本信息】 南京理工大学 ， 电力系统及其自动化， 2010， 硕士

【摘要】 随着分布式电源(DG)越来越多的引入配电网,正确评估分布式电源对系统谐波分布的影响以及合理配置DG,对保证电网和设备可靠优化运行,具有十分重要的意义。论文主要研究分布式电源对电网谐波分布的影响,并在此基础上对DG进行了优化配置。论文从谐波分析的角度,把分布式电源分为线性模型DG和非线性模型DG两类,建立它们的谐波计算模型,理论推导出在理想配电网中不同位置接入不同容量和模型的DG时,系统谐波畸变水平的变化规律。利用PCFLO软件对一实际的链式配电网算例进行谐波分析,仿真结果与理论推导一致。在上述分析的基础上,考虑DG对系统谐波分布的影响,提出两种DG优化配置模型：考虑引入DG使系统谐波畸变水平最小的单目标优化模型和综合考虑有功网损和谐波畸变指标的多目标优化模型。针对DG配置的多目标性,对粒子群算法进行了改进,建立混合PSO计算模型、自适应调整惯性权重系数,提出了混合最优解评估选取和孤立点搜索的双向搜索多目标粒子群(MOPSO)算法；该算法收敛速度快,可有效地逼近Pareto最优前端,且优化结果分布均匀,能保持多目标优化群体的多样性。最后,对含DG的配电网基波和谐波潮流算法进行研究与改进,结合改进的MOPSO算法和潮流计算解决了多约束、多目标和整数寻优的DG优化配置问题。对IEEE33节点配电系统进行仿真计算,验证改进的前推回代潮流算法、解耦谐波潮流算法、DG优化配置模型和改进MOPSO算法的正确性和适用性；表明合理地对非线性模型和线性模型DG进行选址和定容,不仅可以有效降低配电网谐波畸变水平,减少系统的有功网损,还能提高系统电压水平。更多还原

【Abstract】 With more and more DG used in distribution network, it’s important for guaranteeing the reliable and optimal operation of grid and equipment by planning DG rationally and correctly assessing the influence of harmonic distortion after adding DG This paper studies the harmonic impact of DG, based on which, it optimizes the site and size of DG through improved multi-objective particle swarm optimization (MOPSO) algorithm.This paper classifies DG as the linear model and nonlinear model from the harmonic analysis perspective, and establishes their harmonic calculation model. In theory, the rule of THD (Total Harmonic Distortion) variety is deduced when two types of DG access to an ideal distribution network with the different site and size. The simulation of 21-bus chain distribution network shows the same result as the ideal system’s using PCFLO software.Based on the above analysis, two optimal models of DG planning are proposed with considering the harmonic effct of DG:one is the single-objective optimization model for minimizing THD; the other is multi-objective optimization model considering active power loss and THD. The particle swarm algorithm is improved by linear adjustment of inertia weight and introducing an item in formula of updating particle’s velocity to form a hybrid PSO model. Then MOPSO algorithm with bidirectional searching strategy is proposed mixed optimal solution and isolated point search. Combining the improved MOPSO algorithm with fundamental and harmonic power flow can solve DG optimization configuration of multi-objectives constraints and integer optimization. The simulation result of IEEE 33-bus distribution network proves that the improved back/forward sweep flow algorithm, decoupled harmonic power flow calculation, DG optimization configuration model and improved MOPSO algorithm are correct and suitable. Easonable planning of DG not only can effectively decrease THD and power loss of distribution network but raise the voltage level, which has high practical value.更多还原