基于自适应神经模糊推理系统的城市电力需求及预测模型

Model of City Electricity Demand and Forecasting Based on Adaptive Neuro-Fuzzy Inference System

城市年度电能消耗数据具有较强的非线性特点。科学预测城市电力需求,是一项重要的基础性工作,需要有一种较为简易又有足够精度的预测方法。运用神经网络较强的非线性映射能力和模糊推理系统计算量小的特点构造了基于自适应神经模糊推理系统的城市电力需求及预测模型。经实证分析及与ARIMA方法的对比,结果表明,该方法适用于长中短各期预测并有很好的预测效果。

There are strong nonlinear characteristics in the city＇s year data of electricity consume. Forecasting city＇s electricity demand scientifically is an important basic tasks, and need an forecasting method with easier to use and having sufficient precision. For this purpose we apply the strong nonlinear characteristics of neural network and there is a little of calculation in fuzzy inference systems to build a Model of City Electricity Demand and Forecasting based on Adaptive Neuro - Fuzzy inference systems. To evaluate the prediction accuracy of the model,we compare its performance with ARIMA. The experiment results show the model can be applied to long,medium and short term＇s Forecasting of City Electricity Demand and has a very good prediction accuracy.