摘要
随着光伏并网容量的增加,光伏发电功率的波动对电网调度运行的影响不容忽视,电网对光伏发电功率预测精度提出了更高要求。在分析了光伏发电功率波动影响因素的基础上,基于BP神经网络建立光伏发电功率预测模型。通过大唐吐鲁番光伏发电实测数据验证该方法,预测结果 RMSE为3.544,表明该方法可以准确预测光伏发电功率。
With the increase of the capacity of photovoltaic connected to grid,influence of photovoltaic power fluctuation on power grid operation can not be ignored. The grid puts forward higher requirements on the accuracy of photovoltaic power prediction. Based on the analysis of the factors affecting the PV power fluctuation,the prediction model of photovoltaic power based on BP neural network is established. Institute of technology in Datang Turpan photovoltaic power generation data validate this method. The prediction results of RMSE is 3. 544,showing that this method can accurately predict the PV power.
作者
陈德会
杨海艳
曲宏伟
CHEN Dehui;YANG Haiyan;QU Hongwei(Datang Xinjiang Clean Energy Co., Ltd., Urumchi, Xinjiang 830001, China;Shool of Energy and Power Engineering, Northeast Electric Power University, Jilin, Jilin 132012, China)
出处
《东北电力技术》
2018年第4期42-44,共3页
Northeast Electric Power Technology
关键词
光伏发电功率
预测
神经网络
均方根误差
photovoltaic power generation
prediction
neural network
root mean square error