基于非平稳太阳辐照强度计算和LMS滤波器光伏电站功率预测方法

摘要

光伏电站功率预测对电力系统调度及安全稳定运行具有重要意义,对于光伏电站输出功率时间序列表征出来的周期性非平稳特性,采用计算理论太阳辐照强度的方法完成去趋势化的数据预处理,在此基础上提出了一种理论太阳辐照强度计算和LMS自适应滤波器相结合的光伏电站功率预测方法.将光伏电站过去时刻运行值作为滤波器的输入,当前时刻运行值作为滤波器的期望值,白适应滤波器即可建立光伏电站功率预测模型的有限脉冲响应逼近模型.该算法简单,能够实时显式结果.使用某光伏电站运行数据进行验证结果表明该方法对超短期功率预测精度较高.%With increase of installed capacity of photovoltaic power plant,the power prediction techniques of photovoltaic power station play an important role in dispatch and stable operation of power system.Aiming at periodic non-stationary characteristics of photovoltaic power station output,a theoretical calculation of solar radiation intensity is used to eliminate the periodic non-stationary trend for data pre-processing.And a power prediction method of photovoltaic power plant based on the non-stationary solar irradiation intensity calculation and least-mean-square adaptive filter is proposed.The least-mean-square filter takes history running value of PV station as input value of the filter and current running value as the expected value.The advantage of this method is to obtain a real-time and explicit identification result at the same time the algorithm is simple.A test based on running data of one photovoltaic power station shows that the prediction method in the paper has good accuracy in terms of ultra-short term power prediction.