【作者】 黄鹏洲；

【导师】 吴俊勇；

【作者基本信息】 北京交通大学 ， 电气工程， 2010， 硕士

【摘要】 摘要：在风光互补独立供电系统的设计中,如何配置风力发电机、太阳能电池板和蓄电池,在满足负荷需求的前提下,使风能和太阳能这样的清洁能源得到充分利用,负荷的供电可靠性较高,而系统成本较低,这是一个多目标优化设计的问题。本文就是针对这样的风光互补独立供电系统,进行多目标的优化配置。首先建立了影响光伏发电和风力发电的天气模型,并使用了尖峰负荷小时(PSH)的方法和威布尔分布的方法分别模拟太阳能辐射和风速状况,采用蒙特卡罗仿真对天气情况进行仿真计算；其次建立了风力发电、光伏发电和蓄电池储能的模型,然后针对风光互补供电系统的研究重点,本文提出了衡量供电可靠性的失负荷概率LOLP、衡量清洁能源浪费的清洁能源浪费概率LOEP和系统成本三个指标,把它们作为多目标问题的优化目标。为了解决风光互补独立供电系统中提出的多目标优化问题,本文提出了一种混沌自适应进化算法(CSEA),新算法的混沌初始种群算子提高了初代种群的多样性,分组选择策略保证了各代中一定数量的劣势个体可以参与进化,自适应遗传算子增加了劣势个体的交叉和变异概率,从而避免算法早熟,增强了算法的全局搜索能力。算例表明,CSEA算法比传统单目标遗传算法的结果更加接近实际运行的Pareto优化前端,综合效果更优。另外,本文运用CSEA算法对风光互补独立供电方式与纯光伏、纯风力独立供电方式进行比较,证实了风光互补独立供电方式在经济性和可靠性等方面更为合理。可见,采用混沌自适应进化算法进行风光互补独立供电系统的优化设计,对于提高供电系统可靠性,降低成本,减少能源浪费具有非常重要的意义。更多还原

【Abstract】 ABSTRACT:In optimal sizing of stand-alone hybrid wind/PV system, how to make more full use of wind and solar energy source, and how to configure PV modules, wind generators and batteries based on the load demand with higher reliability, lower system cost and lower lost of clean energy is a multi-objective optimization problem in nature.In this paper, weather modeling which is build at fist is one of the most important factors. Peak-Sun-Hours method is used to calculate solar radiation data, while the weibull distribution is used to simulate the wind speed. Monte Carlo method is used to simulate weather condition after considering the random condition. Then wind generators and PV Systems are modeling, LOLP and LOEP indices which reflect the reliability level and the lost of clean energy are proposed and calculated through Monte Carlo simulation.To solve the multi-objective optimization problem, a so called CSEA Multi-Objective Evolutionary Algorithm has been proposed in detail. Simulation results show that, The Chaotic Initial Generation helps to improve the initial diversity, the Grouping Selection Strategy and Self-adaptive Genetic Operator help to avoid pre-maturity and enhance the global searching ability of the algorithm. Comparisons with the single-objective optimization show that CSEA outperforms GA in terms of diversity preservation and in converging closer to the pareto-optimal frontier in one-run-time.Photovoltaic system and wind power system are compared with the hybrid wind/PV system in double-objective evolutionary algorithm. The result shows that stand-alone hybrid wind/PV system is more reasonable comparing with PV system or wind farm in power system.Application of CSEA on the Optimal Sizing of Stand-alone hybrid wind/PV system can improve system reliability, reduce the cost and energy waste, which has great significance.更多还原