广东电力2017，Vol.30Issue(8)：32-37,6.DOI:10.3969/j.issn.1007-290X.2017.08.006
基于ARIMA和SVR的光伏电站超短期功率预测
Ultra-short Term Power Forecast Based on ARIMA and SVR for Photovoltaic Power Station
摘要
Abstract
According to historical actual monitoring data of the meteorological station in photovoltaic power station, an ultra-short term power forecast model for the photovoltaic power station based on autoregressive integrated moving average (ARIMA) and support vector regression (SVR) was established so as to realize real-time tracking and forecast on ultra-short term power of the photovoltaic power station.This model firstly made use of measured data of the forecasting day for single-step prediction on ARIMA time sequences of irradiation intensity and temperature.Secondly, it obtained forecast value of output power of the next forecasting point by inputting forecasting results of irradiation intensity and temperature into the SVR model.Finally, data of measured irradiation intensity, temperature and power of the forecasting point was adopted to update original ARIMA time sequences in real time.Based on actual examples of forecasting ultra-short term output power under four different weather conditions, effectiveness of ARIMA-SVR model was verified.关键词
光伏电站/超短期/功率预测/ARIMA/SVRKey words
photovoltaic power station/ultra-short term/power forecast/ARIMA/SVR分类
信息技术与安全科学引用本文复制引用
赫卫国,郝向军,郭雅娟,曹潇,陈锦铭,梅飞,刘皓明..基于ARIMA和SVR的光伏电站超短期功率预测[J].广东电力,2017,30(8):32-37,6.基金项目
国家电网公司科技项目(NY71-16-034) （NY71-16-034）
