风光柴储独立微网分布式电源多目标优化配置研究

发布时间：2018-01-03 12:38

本文关键词：风光柴储独立微网分布式电源多目标优化配置研究　出处：《广西大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着经济社会的发展,电力需求不断增长,集中式大电网的弊端日益凸显。微电网由分布式电源、储能装置、负荷、监控保护装置以及电力电子能量转换装置组成,在整合分布式电源优势、提高供电可靠性、改善系统性能等方面具有重要作用,是传统电网的有力补充。合理对微电网分布式电源的类型、容量大小进行优化配置,减少污染排放,提高能源利用率,对微电网经济、环保、可靠地运行具有重要意义。本文针对风光柴储独立运行微电网分布式电源优化配置进行了研究,主要研究内容分为以下方面：首先,研究了典型风光柴储混合微电网的系统结构和运行方式,考虑风速、光照的随机性和波动性特点,研究风机、光伏、柴油发电机及储能装置的输出功率时序模型。接着,针对独立运行的风光柴储混合微电网,应用HOMER软件进行仿真优化,并对不同空气污染惩罚、柴油价格、年均日照强度、年均风速、设备投资成本以及储能技术等条件下的优化配置结果,围绕系统的污染排放、经济性和供电可靠性进行灵敏度分析,给出独立运行微电网电源优化配置的一些实际建议。然后,以分布式电源出力、储能装置容量、系统供电可靠性为约束条件,构建微电网经济成本最小、污染排放最少和能源利用率最高的多目标优化配置模型,并引入熵理论以确定多目标函数各指标的权重系数；制定了改进的循环充放控制(ICC)策略,在运行控制策略中增加了对储能装置的充放电管理；采用改进种群移动、自适应迭代步长、自适应变异策略和搜索空间缩减等措施来改善仿电磁学(ELM)算法的收敛性；应用改进的ELM并结合ICC控制策略求解配置模型,通过仿真实验验证优化模型和求解算法的正确性和有效性。最后,针对优化过程中需遍历全年数据进行时序分析所导致的计算量大、运行效率低的问题,提出一种基于Markov的ELM优化方法：采用模糊C-均值(FCM)算法从风速、光照和负荷8760h的数据中提炼出典型的场景,用随机过程方法建立其对应的Markov模型；将前述的Markov模型嵌入改进的ELM中进行优化求解,以广西涠洲岛为背景进行算例分析,验证模型和算法的有效性。
[Abstract]:With the development of economy and society, the demand for electricity is increasing, and the disadvantages of centralized large power grid are becoming more and more obvious. Microgrid is made up of distributed generation, energy storage device and load. The components of monitoring and protecting device and power electronic energy conversion device play an important role in integrating the advantages of distributed power supply, improving the reliability of power supply and improving the performance of the system. It is a powerful supplement to the traditional power grid. It optimizes the type and capacity of distributed power generation in microgrid, reduces pollution emissions, improves energy utilization, and is economical and environmentally friendly to microgrid. Reliable operation is of great significance. In this paper, the optimal configuration of distributed power generation for the independent operation of wind-firewood storage microgrid is studied. The main research contents are as follows: first of all. The system structure and operation mode of typical wind-firewood storage hybrid microgrid are studied. Considering the stochastic and fluctuating characteristics of wind speed and illumination, the wind turbine and photovoltaic system are studied. The output power timing model of diesel generator and energy storage device. Then, aiming at the independently running wind-firewood mixed microgrid, the simulation optimization is carried out by using HOMER software, and the different air pollution is punished. Under the conditions of diesel price, annual sunshine intensity, average annual wind speed, equipment investment cost and energy storage technology, the optimal allocation results are analyzed around the system pollution emissions, economy and reliability of power supply. Some practical suggestions on the optimal configuration of power supply for independent running microgrid are given. Then, the economic cost of building microgrid is minimized by taking distributed power generation, capacity of energy storage device and system power supply reliability as constraints. The multi-objective optimal allocation model with the least pollution emission and the highest energy efficiency is introduced, and the entropy theory is introduced to determine the weight coefficient of each index of the multi-objective function. An improved cyclic charging and discharge control (ICCC) strategy is developed, and the charge and discharge management of the energy storage unit is added to the operation control strategy. Improved population movement, adaptive iterative step size, adaptive mutation strategy and search space reduction are adopted to improve the convergence of ELM algorithm. Using the improved ELM and ICC control strategy to solve the configuration model through simulation experiments to verify the correctness and effectiveness of the optimization model and the algorithm. Finally. In order to solve the problem of large computation and low running efficiency caused by the time series analysis of traversing the whole year data in the optimization process. A ELM optimization method based on Markov is proposed. The fuzzy C-means algorithm is used to extract the typical scene from the data of wind speed, illumination and load 8760h. The corresponding Markov model is established by the stochastic process method. The Markov model mentioned above is embedded in the improved ELM to solve the optimization problem. A case study on Weizhou Island in Guangxi is carried out to verify the validity of the model and the algorithm.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TM727

【相似文献】