摘要
随着并网光伏发电容量的持续增加及多能源发电协同利用的需要,光伏发电功率的高精度预测对于提高规模化光伏发电的优化调度和安全运行日益重要。为了解决单一预测模型精度低的问题,提出了一种基于季节气象特征划分的光伏发电多模型预测方法。通过不同季节下光伏发电系统的电气特性和出力特性分析,说明了按照季节来划分功率预测多模型的必要性。以某光伏电站为例,利用BP神经网络建立不同季节的光伏发电预测模型,通过遗传算法优化了季节模型参数。利用实测数据对2种功率预测方法进行了比较,结果表明,该方法能有效提高光伏电站的功率预测精度。
With the continuous increase of grid-connected photovoltaic capacity and the need for collaborative utilization of multi-energy generation,the high-precision prediction of photovoltaic power generation is increasingly important for the optimal scheduling and safe operation of the grid.In order to solve the problem of low accuracy of the single prediction model,a multi-model prediction method for photovoltaic power generation is proposed based on the seasonal meteorological features.First,through the analysis of the output characteristics of typical seasons,the necessity of dividing multi-model of power prediction according to seasons is explained.Second,taking a photovoltaic power plant as an example,the forecasting model of photovoltaic power generation is established by using BP neural network in different seasons,and the parameters of the seasonal models are optimized by genetic algorithm.Finally,the measured data is used to compare the two power forecasting methods.The results show that the proposed method can effectively improve the power forecasting accuracy of photovoltaic plants.
作者
时珉
周海
韩雨彤
尹瑞
SHI Min;ZHOU Hai;HAN Yutong;YIN Rui(State Grid Hebei Electric Power Company,Shijiazhuang 050022,Hebei,China;State Key Laboratory of Operation and Control of Renewable Energy&Storage Systems,Beijing 100192,China;China Electric Power Research Institute,Nanjing 210003,Jiangsu,China;School of Renewable Energy,North China Electric Power University,Beijing 102206,China)
出处
《电网与清洁能源》
2019年第7期75-82,共8页
Power System and Clean Energy
基金
国网河北省电力有限公司科技项目(5204BB170007)~~
关键词
光伏功率预测
太阳辐照度
人工神经网络
多季节模型
photovoltaic power forecasting
solar irradiance
artificial neural network
multi-season model
