摘要
建立了模糊决策模型分析所给历史负荷数据的关联度,以提取对预测有贡献的有用数据,除去"坏数据",即对原始数据进行预处理。建立了神经网络模型,并结合小波分析和神经网络的优势建立改进小波神经网络的结构模型,预测出待测日96个时间点的电力负荷值。通过准确度分析,验证了模型的合理性。进而将所建模型应用于某地区进行负荷预测,并且对该地区的负荷特点及规律性进行了讨论。
This paper, firstly, establishes a fuzzy decision model for analyzing the historical load data correlation to extract the useful data for forecasting and remove ' bad data', namely preprocessing of the original data. Secondly, a neural network model is established, and combined with the advantages of wavelet analysis and neural network, the structural model of improved wavelet neural network is established, and the power load on the proposed day of measurement is predicted at 96 time points. The rationality of the model is verified through the accuracy analysis. Furthermore, the model is applied in the power load prediction of a region, and the load characteristics and regularity of the region's power load are discussed.
出处
《电网与清洁能源》
北大核心
2015年第2期16-20,27,共6页
Power System and Clean Energy
基金
国家自然科学基金(50977029)~~
关键词
负荷数据
小波分析
神经网络
电力负荷
负荷预测
load data
wavelet analysis
neural network
electrical load
load forecasting
