Research On Power Prediction And Access Impact Of Photovoltaic Power Generation Systems

With the appearance of environmental problems caused by fossil fuels and energy shortage,photovoltaic power generation systems has become an important part of energy structure in the world.However,the distributed photovoltaic power generation system mostly depends on the external environment,so its output power changes with the environmental factors.When it accesses to the main power grid will cause uncertainty on the power grid.With the application of the distributed photovoltaic power generation system in the power grid,the complexity and higher risk of the power system are greater.Therefore,the power forecasting of the distributed photovoltaic power generation system and the distribution trend of the distribution network after the grid connection help the power dispatching department to make the scheduling plan and improve the safety of the power system.The specific work of this paper is as follows:(1)The solar radiation intensity,ambient temperature,weather category as the PV system output instability of the main factors,and based on momentum adaptive learning rate BP neural network,RBF neural network and GRNN neural network,the short-term prediction model of the output power of photovoltaic power generation system is established.(2)A combination forecasting model based on decision theoretic minimax criterion is proposed,which is composed of BP neural network,RBF neural network and GRNN neural network.In order to predict the output power of the photovoltaic power generation system and improve the prediction accuracy of the output power of photovoltaic power generation system.(3)An improved pre-push back method is proposed in this paper,and different access schemes are designed,the use of photovoltaic system output power of the predicted value of the simulation experiment,and the distribution trend of distribution network is calculated by combining the pre-push back method.In order to study the influence of distributed photovoltaic power generation system connected to power distribution network.Research indicates that the solar radiation intensity,environmental temperature,weather category as the main factors affecting the output power of photovoltaic power generation system can be used in artificial neural network prediction model is established;the GRNN neural network has higher prediction accuracy in the three single prediction models;the prediction accuracy of the combined forecasting model based on the decision theoretic minimax criterion is the highest;impact on the distribution network of the photovoltaic power generation system connected to the grid distributed as follows:the access of the distributed power supply improves the distribution of the feeder voltage and can reduce the network loss of the system to a certain extent.