【作者】 张德玲；

【导师】 陈根永；

【作者基本信息】 郑州大学 ， 电力系统及其自动化， 2007， 硕士

【摘要】 短期电力负荷预测是电力系统运行调度中一项非常重要的内容，可以经济合理地安排电网内部发电机组的启停，减少不必要的旋转备用容量；可以合理安排机组检修计划，保证社会的正常生产和生活，提高发电企业的经济效益和社会效益；在目前能源日益紧张的情况下，国家能源发展战略要求逐步降低单位GDP的能源损耗，电力生产和消费日益市场化，短期负荷预测结果成为制定电力市场交易计划的重要依据，这就对短期负荷预测提出了更高的要求。本文在分析了目前短期电力负荷预测的现状及各种预测方法、预测模型的基础上，针对近年来，商业用电、居民生活用电在社会总用电中所占的比重越来越大，而且这些负荷易受到预测日类型、气候等条件的影响；以及地区负荷水平一般都不是太高，且负荷构成相对简单，因此更易受到气象因素的影响的特点，本文根据某地区电网运行实际，提出把气象因素和日类型作为影响地区负荷预测的主要影响因素，由于考虑到某些影响因素具有不确定性的特点，采用了模糊化的方法对数据进行了处理。常规算法不能较好地反映气象条件等外界因素对负荷的影响，而近年来人工神经网络法等智能算法具有高度的非线性映射能力，可以较好地考虑气象条件等因素对电网负荷的影响，本文采用了RBF(Radial Basis Function径向基函数)神经网络；但另一方面因许多智能方法在学习收敛的速度方面、收敛的稳定性方面、收敛至全局极小点方面，尚缺乏指导模型自动选择的一般规则，只是在一定程度上加快了收敛速度，本文作者结合近年来有关学者对负荷预测长期研究和不断实践，提出了把模糊控制规则引入到RBF神经网络中，较好的解决了上述问题，并具有较好的应用前景。根据负荷的数学模型，结合多种负荷预测方法，开发了的短期电力负荷预测应用软件。更多还原

【Abstract】 Short-Term Load Forecasting(STLF) is one of the most important contents of running and dispatching of power system. It can be economic and reasonable to arrange start and stop of the Generator in wire net, reduce otiose revolve the storage capacity. It can be reasonable to arrange Generator the maintain plan, assurance normal produce and live of society, raise the economic efficiency and social efficiency of Electric power enterprise; Under the situation that the energy is increasingly lacking currently, the national energy development strategy requests gradually to reduce energy consumption of the unit GDP, the production and consumption of electric power increasingly go to market , short-term load forecasting result become importance basis of drawing up the electric power market bargain plan. So these put short-term load forecasting forward a higher request.This text analyze the present condition and Various methods and mathematics model of the short-term load forecasting. Considering the fact, the specific Comparison of the commercial electricity and life electricity of residents of the recent years is more and more big in the all society electricity, these loads are subjected to influence by load day type and weather and etc; Moreover, the region load level isn’t all too high generally, and the load constitute is simple generally, so these loads are also effected easily by load day type and weather factors . According to some area fact ,this text takes day type and weather as the main influence factors of the region short-term load forecasting. Because of indetermination characteristic of the some influence factors, this text takes fuzzy method to process data.The normal calculate way can’t reflect goodly weather condition and other outside factors to the influence for load forecasting, in recent years,. the artificial neural network method etc have height nonlinear to reflect the ability of shoot, can reflect goodly the weather factor etc. But lack of the general rule that instruct model to chooses automatically in the speed, stability and the overall situation to order smallest aspects of studying to refraining from rash action, just speed to refrain from rash action at some extent. This text author combine relevant the scholar studies and practices continuously towards short-term load forecasting over a long period of time in recent years, put fuzzy control rule to the RBF (Radial Basis Function) neural network, resolve the above-mentioned problem more goodly, and have better applied foreground.According to mathematics model and Various methods of load forecasting, a design of short-term load forecasting software system is presented in this paper.更多还原