基于前向神经网络的分布式光伏承载力预测方法

Prediction method of distributed photovoltaic carrying capacitybased on forward neural network

光伏能源较为分散,可供容量也存在着一定的限制,并网过程中会影响电力系统的稳定性,提出基于前向神经网络的分布式光伏承载力预测方法研究。构建分布式光伏发电系统模型,深入分析分布式光伏承载力影响因素,以电力系统稳定运行为目标,构造分布式光伏承载力预测模型,并确定相应的约束条件,引入并训练前向神经网络,将分布式光伏并网相关数据输入至训练好的前向神经网络中,输出结果即为分布式光伏承载力预测结果。实验结果显示:应用提出方法获得的光伏承载力预测时间小于给定最大限值,光伏承载力预测结果与实际结果几乎保持一致,充分证实了提出方法应用性能较佳。

Photovoltaic energy is relatively scattered,and there are certain restrictions on the available capacity,which will affect the stability of the power system in the process of grid connection.A distributed photovoltaic carrying capacity based on forward neural network is proposed.Research on forecasting methods.Build a distributed photovoltaic power generation system model,in-depth analysis of the influencing factors of distributed photovoltaic carrying capacity,with the goal of stable operation of the power system,construct a distributed photovoltaic carrying capacity prediction model,and determine the corresponding constraints,introduce and train a forward neural network,The data related to distributed photovoltaic grid connection is input into the trained forward neural network,and the output result is the prediction result of distributed photovoltaic carrying capacity.The experimental results show that the photovoltaic carrying capacity prediction time obtained by the proposed method is less than the given maximum limit,and the photovoltaic carrying capacity prediction results are almost consistent with the actual results,which fully confirms the good application performance of the proposed method.