作　　者： ; ; (谢锦标）;

机构地区： 华南理工大学

出　　处： 《电网技术》 2004年第11期49-52,共4页

摘　　要： 提出了一种基于谱分析法进行建模的短期负荷预测方案,该方案利用负荷历史数据的谱分析结果进行人工神经网络(ANN)模式分类和选择输入变量。方案采用快速傅立叶变换(FFT)进行负荷数据预处理,运用滤波算法及小时负荷曲线的频谱分析来研究电网负荷的周期特性,所得结果表明四季负荷的谱特性具有明显差异,应采用不同的模型和方案进行预测。谱分析有助于各时段预测方案提取输入变量。利用该思路构造的基于人工神经网络的负荷预测方案被用于预测广东省网的负荷,与其他普遍采用的输入变量预测结果的对比表明,所提方案在短期负荷预测中的性能良好。 A short-term load forecasting method based on the modeling by spectrum analysis is put forward in which the spectrum analysis results of historical load data is used for the classification of artificial neural network (ANN) modes and the selection of input variables. In this method the load data is preprocessed by fast Fourier transform (FFT), the filtration algorithm and the spectrum analysis of hourly load curve are used to research the periodical characteristics of power load. The obtained results show that there are obvious differences among the spectrum characteristics of the loads in four seasons, therefore different models and method should be used to forecast the load in different seasons. The spectrum analysis is helpful to the extraction of the input variables of the forecasting scheme for different time intervals. According to above mentioned idea an ANN based load forecast scheme is applied in the load forecasting of Guangdong province. Comparing with the load forecasting results by other widely used input variables, it is shown that the presented scheme possesses good performance for short-term load forecasting.

关 键 词： 电力系统 短期负荷预测 人工神经网络 谱分析 建模 傅立叶变换

领　　域： [电气工程] [自动化与计算机技术] [自动化与计算机技术]