基于ARMA-GARCH模型的超短期风功率预测研究  

The Ultra short-term prediction of wind power based on ARMA-GARCH model

作　　者：田波[1] 朴在林[1] 郭丹[1] 王慧[1] Tian Bo Piao Zailin Guo Dan Wang Hui(College of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110161, China)

机构地区：[1]沈阳农业大学信息与电气工程学院,沈阳110161

出　　处：《电测与仪表》2016年第17期12-17,共6页Electrical Measurement & Instrumentation

基　　金：十二五国家科技支撑项目(2012BAJ26B00)

摘　　要：风功率预测对提高电能质量和电力系统的安全运行具有重要意义。基于时间序列的方法,对内蒙古赤峰地区某风场的风功率数据进行了超短期预测,通过对数据平稳性检验的结果,建立了时间序列的ARMA模型,利用拉格朗日乘子检验的方法,检验ARMA模型具有ARCH效应,并建立适合的ARMA-GARCH模型。结论通过对比ARMA模型,ARMA-ARCH模型和ARMA-GARCH模型的风功率预测精度可知,在解决数据的残差序列异方差函数具有长期相关性时,ARMA-GARCH模型能够有效的提高预测精度。Wind power prediction is very important to improve the power quality and the safe operation of power sys- tem. The ultra short-term prediction of wind power data is carried out in a wind farm in Chifeng of Inner Mongola based on time series analysis, and the ARMA （ Autoregressive Moving Average） model of time series is built through the stationary test of data. Through ARCH effect of the residual of ARMA model by Lagrange Multiplier （ LM ） , the corresponding ARMA-GARCH model is set up. Through the comparison of the wind power prediction by using ARMA model, ARMA-ARCH model and ARMA-GARCH model respectively, ARMA-GARCH model possesses higher accura- cy on the residual sequence of the data with the long term correlation.

关 键 词：时间序列 风电功率 预测 ARMA模型 ARCH模型 GARCH模型 

分 类 号：TM711[电气工程—电力系统及自动化]