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

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.