【作者】 张恒；

【导师】 袁晓辉；

【作者基本信息】 华中科技大学 ， 水利工程， 2018， 硕士

【摘要】 目前全球气候条件的变化和传统能源的价格上涨,给太阳能、风能、水能等分布式可再生清洁能源提供了良好的推广和发展时机。微电网能将这些分布式电源与负荷进行消纳整合,并最大程度发挥各分布式电源的特点,可解决分布式发电接入大电网时功率波动大、成本高、难以控制等问题,是作为大电网系统的强有力补充。其中,冷热电联供型微电网因其系统负荷形式更加多样复杂及能源利用率更高而在各领域受到关注,逐渐成为未来第二代能源技术重点之一。本文针对冷热电联供型微电网的功率预测及日前经济优化运行展开研究,主要研究内容如下:(1)在微电网及冷热电多联供系统相关概念基础上,构建冷热电多联供型微电网的总体结构和供能框架;建立了包含多种分布式电源如光伏、风机、抽水蓄能电站和微型燃气轮机等供电储能设备数学模型,同时对冷热电联供系统的辅助供能设备进行数学模型描述,为冷热电联供型微电网的研究提供基本单元设备模型支撑。(2)考虑到冷热电联供型微电网日前经济优化运行是在系统内多种预测数据的基础上进行的,本文选取微电网系统光伏发电和电负荷功率作为能源供给侧和消耗侧代表进行预测对象研究;采用改进的蜻蜓算法(IDA)对最小二乘支持向量机(LSSVM)核参数优化从而建立IDA-LSSVM预测模型,并对光伏发电和电负荷进行短期功率预测仿真分析,将结果与DA-LSSVM和LSSVM模型预测结果比较,证明IDA-LSSVM预测模型具有较高预测精度和较好的泛化推广能力。(3)建立以冷热电联供型微电网经济运行成本最小化为目标的优化调度模型,同时结合考虑功率预测数据、市场分时电价、多设备耦合出力约束和冷热电三种负荷需求平衡等方面复杂因素,提出采用改进的混合蛙跳算法进行并网运行方式下的微电网经济运行调度计算,算例分析给出不同季节典型日下微电源和供能设备的出力情况,验证了本文提出的冷热电联供型微电网经济运行策略及其优化算法的可行性。更多还原

【Abstract】 At present,changes in global climatic conditions and rising prices of traditional energy sources provide good opportunities for the promotion and development of distributed renewable clean energy,such as solar energy,wind power,and hydropower.The microgrid can integrate these distributed power sources and loads,and maximize the characteristics of each distributed power source.As a powerful supplement to the large power grid system,microgrid can solve the problems of large power fluctuation,high cost and difficult control when distributed generation is connected to the large power grid.Among them,the combined cooling,heating and power(CCHP)microgrid has attracted attention in various fields due to its more complex system load and higher energy utilization rate,which has gradually become one of the second generation energy technology priorities in the future.This article focuses on the power forecast and day-ahead economic optimization operation of CCHP microgrid.The main work is shown as followed:(1)Based on the relevant concepts of the microgrid and the CCHP system,the overall structure and energy supply framework of the CCHP microgrid is built.The mathematical models for a variety of distributed power sources,such as photovoltaics,fans,pumpedstorage power plants,and gas turbines,are established.At the same time,the part of energy supply equipments for CCHP system are described with mathematical formulas,which provide unit equipment model support for the research of CCHP microgrid.(2)Considering the optimized scheduling of CCHP microgrid is based on forecast data,in this paper,photovoltaic power generation and electrical load power are selected as the representative of energy supply side and consumption side to predict.In order to obtain higher prediction accuracy,the improved dragonfly algorithm(IDA)is proposed to optimize the kernel parameters of least squares support vector machine(LSSVM),and the IDALSSVM prediction model is established which carries out the short-term power forecasting for photovoltaic generation and electrical load.Comparing the results with the prediction results of DA-LSSVM model and LSSVM model,it shows that the IDA-LSSVM model has higher prediction accuracy and better generalization ability.(3)Establishing an optimal dispatching model with the goal of minimizing the economic operating costs of the CCHP microgrid.At the same time,considering the complex factors,such as power forecast data,market time-sharing tariff,multi device coupling output constraint and the balance between cold,heat and electric load demand,etc.The improved shuffled frog leaping algorithm(MSFLA)is proposed to calculate the economic operation for grid-connected CCHP microgrid,and the example analysis gives the output of micro-sources under typical daily conditions in different season,which verified the hypothesis presented about micro-net economic operation strategy and the feasibility of the optimization algorithm.更多还原