基于函数系数自回归模型的风速时间序列预测

准确预测风电场风速是解决风能对电力系统所造成的安全、稳定运行和电能质量等问题的有效途径之一.风速的难以预测是由于它的高度随机和非线性.基于一种非参数的非线性自回归随机模型来预测风速,模型的自回归系数随模型依赖变量的变化而变化,因而它有灵活的非线性结构.数值实验和比较结果表明了这种函数系数自回归模型在风电场风速预测中的有效性.

Wind Speed Prediction Based on Functional Coefficient Autoregressive Models

Accurate predciton of wind speed is one of the most valuable ways to solve the problems of electricity security,stability and quality which are caused by the wind energy production for power system.The major difficulties for accurate prediction of wind speed are its high stochastictiy and noninearity.This paper predicts the wind speed using a nonlinear autoregressive model.The autoregressive coefficients of the model vary with the state-dependent variables.Such kind of model offers a very flexible structure for nonlinear time series modeling.A simulation example and the comparsion results show that the effectiveness of the proposed approach.