Methods And Research On Forecasting Output Power Of Photovoltaic Power Generation

Solar energy as the most ideal of human clean energy,compared with other energy advantages of its natural self-evident,and photovoltaic power generation has become an important way to use this energy.The most obvious factor affecting the effect of photovoltaic power generation is environmental factors,due to the randomness and uncertainty of the environment,also led to the instability of the photovoltaic output power,which will undoubtedly affect the power grid scheduling and the impact of the regulation.So if you can pre-estimate the output power of the next period of photovoltaic power generation,the corresponding output power of photovoltaic power generation forecast,will reduce the risk of grid operation and also more convenient for people to rational distribution and use of electricity,so that the efficiency of energy utilization has been Improve,thus indirectly reducing the use of energy in the process of environmental pollution.In this paper,MATLAB software is used as a modeling platform.First,a large amount of data is analyzed and processed.The relationship between the three environmental factors,such as ambient temperature,relative humidity and radiation intensity,and the output power of photovoltaic power generation is confirmed by data.The regression equation is obtained by the analysis of large numbers of data and the seven outliers,and then the regression equation is used to test the significance of the regression equation,and then the regression equation is used to test the regression equation.,Will eventually be tested by the regression model applied to the sunny weather conditions under the photovoltaic power forecast,the effect is stable to meet the requirements.However,in the non-sunny weather conditions,the predictive model based on the multiple linear regression model is not ideal.Therefore,the BP neural network is used to reconstruct the model in the non-sunny weather conditions,and the input layer is improved and optimized.Three environmental factors based on the addition of the previous three days of historical power together as a model of input for training,the results meet the expected requirements,to achieve in non-sunny weather conditions,photovoltaic output power prediction.