基于隐马尔科夫修正的光伏中长期电量预测及调度计划应用  

Mid-long Term Available Quantity of Electricity Forecasting with Error Calibration by Hidden Markov Model in Photovoltaic and Application of Dispatching Plan

作　　者：刘大贵 王维庆 张慧娥 李国庆 肖桂莲 张斌 LIU Dagui;WANG Weiqing;ZHANG Huie;LI Guoqing;XIAO Guilian;ZHANG Bin(Engineering Research Center of Education Ministry for Renewable Energy Power Generation and Grid Control,Xinjiang University,Urumqi 830047,China;Power Dispatching Control Center,State Grid Xinjiang Electric Power Co.,Ltd.,Urumqi 830063,China;College of Energy Engineering,Xinjiang Institute of Engineering,Urumqi 830023,China)

机构地区：[1]新疆大学可再生能源发电与并网技术教育部工程研究中心,乌鲁木齐830047 [2]国网新疆电力有限公司电力调度控制中心,乌鲁木齐830063 [3]新疆工程学院能源工程学院,乌鲁木齐830023

出　　处：《高电压技术》2023年第2期840-848,共9页High Voltage Engineering

基　　金：自治区重点实验室开放课题(2018D04005);国家自然科学基金(51667020);自治区高校科研计划自然科学重点项目(XJEDU2019I009);教育部创新团队(IRT_16R63)。

摘　　要：构建高精度的光伏中长期可用电量预测模型,对电力市场调度模式下的月度计划制定具有重要意义。为此,首先建立了基于差分自回归移动平均模型的光伏发电可用电量预测统计模型,实现了横向逐月移动的未来年际预测;然后,考虑光资源月度的差异性和同季节的平稳性,提出了基于隐马尔科夫模型的光伏发电可用电量预测修正方法,实现了纵向同月递推的差异月度预测修正。基于新疆电网某地区光伏运行数据,对方法的有效性进行了验证,结果表明所提出方法预测精度较高。最后,通过在新疆新能源月度计划控制系统中进行应用,实现了月度计划和日前计划动态滚动跟踪相结合的调度计划模式,满足了调度生产运行的需求。The development of a high-precision model for forecasting the medium-and long-term power availability of photovoltaic(PV) power is critical for the development of monthly plans in the electricity market dispatch model. Therefore, a statistical model for forecasting the PV power generation availability based on a differential autoregressive moving average model is developed to realize a horizontal monthly-moving forecast on the future inter-annual basis. Then, the monthly variability of light resources and the smoothness of the same season are taken into account, and a correction method for forecasting the PV power generation availability based on the hidden Markov model is proposed to achieve a vertical monthly forecast correction with the same monthly recurrence of variability. The validity and accuracy of the method are confirmed by using PV operation data from a Xinjiang power grid region.

关 键 词：ARIMA模型 HMM模型 中长期预测 光伏发电 调度计划 电力市场 

分 类 号：TM615[电气工程—电力系统及自动化]