The Output Forecasting Of Lower Solar Power Plant

With the rapid development of world economy,the consumption of energy has risen sharply.Under the dual pressures of a gradual shortage of fossil fuels and environmental protection,seeking a clean and renewable energy has become an urgent issue for the development of human society.As a clean and renewable energy,solar energy is getting more and more attention.In addition,concentrated solar power,as a way of solar energy utilization,will have broad application prospects in the future.Due to the volatility of solar energy,accurate power forecasting is one of the most important conditions for solar power plants to be successfully integrated into the power grid.Some works about the output forecasting of tower solar power plant have been done in this paper,which are as follows:1.The direct normal irradiance(DNI)forecasting.In this section,firstly,some forecasting methods have been introduced.Then,we selected three different artificial neural network(ANN)as research methods,including the back propagation neural networks,wavelet neural network and long-short term memory neural network.Last,we analyzed and compared the results of three artificial neural networks.2.The establishment of the mathematical model of heliostat field.In this section,at start,we had learned about the optical efficiency calculation of heliostat field and established the mathematical model of heliostat field.At last,we validated the mathematical model.3.The output forecasting of tower solar power plant.In this section,we had established a simple model of thermal system of tower solar power plant and used the model and the value that had been calculated by the section 1 and 2 to get the prediction of the output of receiver and the heat storage of storage system.Last,we also gave the period of stable operation of the power station.