超短期风电功率预测及风电并网暂态切机控制研究

发布时间：2018-04-21 03:18

本文选题：风电预测 + 风速修正　；参考：《浙江大学》2014年硕士论文

【摘要】：随着风电装机容量的迅猛增长,由于风电的高度波动性和随机性,大规模风电并网已成为制约中国风电发展的最大瓶颈。对风电场输出功率进行准确的超短期预测可以为实时调整电网调度计划、安排备用容量提供依据,有利于提高系统的安全稳定和经济运行。同时,大规模风电基地接入比例增加,也对电力系统的暂态安全稳定性造成了影响,使电力系统在严重故障下的紧急控制必须考虑风电机组的影响,制定相应的紧急控制方案。本文通过研究风电功率预测和风电并网暂态切机控制的相关文献,主要进行了以下几个方面的研究： 1、介绍了风电功率预测的主要方法及国内外研究现状,结合实际风场历史风能数据和数值天气预报信息,对SCADA数据进行补充修正,分析影响风申.机组输出功率的各项因素,并利用模糊C-均值聚类算法进行了相似日聚类。 2、对大规模风电并网的暂态稳定性研究进行了综述,介绍了不同风电并网输送通道的特点和PSASP风机仿真模型,最后给出了东北某区域算例电网的参数和风电并网消纳存在的问题。 3、提出了一种基于在线序贯极限学习机(OS-ELM)的超短期风电功率预测方法。利用OS-ELM学习速度快、泛化能力强的优点,将批处理和逐次迭代相结合,不断更新训练数据和网络结构,实现了对数值天气预报风速的快速实时修正和风机输出功率的快速预测。 4、构造了基于计算机自助法(Bootstrap)的多ELM误差评估网络,产生Bootstrap样本,通过修正偏差后的百分位数Bootstrap(BCPB)法给出了风电预测功率的置信区间评估。 5、分析了双馈风机的运行特性,采用PSASP对算例系统进行了暂态稳定校验以及严重故障下的风火电联合切机控制方案的仿真。结果表明系统存在最优的风火比例切机方案
[Abstract]:With the rapid growth of wind power installed capacity due to the high volatility and randomness of wind power large-scale wind power grid connection has become the biggest bottleneck restricting the development of wind power in China. Accurate ultra-short-term prediction of wind farm output power can provide the basis for real-time adjustment of power grid dispatching plan and arrangement of reserve capacity, which is conducive to improving the system's safety, stability and economic operation. At the same time, the proportion of large-scale wind power base access is increased, which also has an impact on the transient security and stability of power system. Therefore, the emergency control of power system under serious faults must consider the impact of wind turbine, and formulate the corresponding emergency control scheme. Based on the research of wind power prediction and the related literature of wind power grid-connected transient cutting machine control, the following aspects are studied in this paper:--- 1. The main methods of wind power prediction and the present research situation at home and abroad are introduced. Combined with the historical wind energy data of actual wind field and numerical weather forecast information, the SCADA data are supplemented and corrected to analyze the wind impact. The factors of the output power of the unit are analyzed, and the similar day clustering is carried out by using the fuzzy C-means clustering algorithm. 2. The research on transient stability of large-scale wind power grid is summarized. The characteristics of different wind power transmission channels and the simulation model of PSASP fan are introduced. Finally, the parameters of a regional grid and the existing problems of wind power grid absorption are given. 3. A prediction method of ultra-short term wind power based on online sequential learning machine (OS-ELM) is proposed. Based on the advantages of fast learning speed and strong generalization ability of OS-ELM, the training data and network structure are continuously updated by combining batch processing with successive iterations. The fast real-time correction of wind speed and the fast prediction of fan output power are realized. 4. A multiple ELM error evaluation network based on the self-help method of computer is constructed to generate Bootstrap samples. The confidence interval evaluation of the predicted power of wind power is given by using the modified percentile Bootstrapper BCPB method. 5. The operation characteristics of the doubly-fed fan are analyzed, and the transient stability check of the system with PSASP and the simulation of the control scheme of the wind power combined cutting machine under the serious fault are carried out. The results show that the system has an optimal scheme of wind and fire proportional cutting machine.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM614

【参考文献】