Research On The Application Of Photovoltaic Power Forecasting In Wind-solar-battery Hybrid Generation Systems

Photovoltaic (PV) output power forecasting is a effective way to reduce the impact of power system caused by the intermittent solar energy. So it is of great significance to forecast the output power of PV array. First of all, the paper has overviewed the research background and some common methods of PV power forecasting. Due to the accuracy of PV power forecasting is still need to be improved, the paper introduced acombination forecasting method based on entropy weight and similar day forecasting method based on combination weights to predict the output power of PV array. And then a control strategy using the results of PV power forecasting is proposed to smooth the output of wind-solar-battery hybrid generation systems.Firstly, A combination method for PV power forecasting based on entropy weight method is proposed. The method uses entropy weight combination method to combine three single forecasting methods. These three single forecasting methods are a past-predicts-futuremodel, a support vector machine(SVM) model and a similar data prediction model.And Matlab simulation resultsshow that the combination model proposed has higher accuracy than others obviously.Moreover, according to the facts that days which have similar weather condition also have high correlation of PV output power. Combination weights method for selecting similar days and photovoltaic output Power Forecasting based on it is proposed. In this method, solarirradiance is considered as similarity variable. The combination weight method is used to choose three days which have smallest similar error as the similar days of forecasting day. The combination weight is obtained by combining subjective weight and entropy objective using theory of the smallest discrimination information. The Matlab simulations indicate that photovoltaic output power forecasting based on similar day selected by combination weight improve the forecasting accuracy.Finally, the paper introduces a control method for wind-solar-battery hybrid generation systems to smooth output power. The method is based on the photovoltaic forecasting results. The control method is able to realize the control of smoothing the output power of wind-solar-battery hybrid generation systems. And it can control the value SOC of the battery to stay in the normal range. Matlab is used to simulate and test the control strategy. And the result indicates the method is a effective control strategy.