A Day-Ahead Power Prediction Algorithm for PV Power Plants Based on Power Market Benefit and BP

With the continuous depletion of fossil energy, new energy represented by PV energy has received widespread global attention. However, new energy is easily affected by the environment, which makes the participation of new energy in the electricity Market depend on energy forecasting technology. At present, the research on the power prediction of Photovoltaic (PV) power plants is mainly biased towards the accuracy of the prediction, and in the general trading Market, the impact of economic factors such as price and cost is more important for Market participants. In this paper, a day-ahead forecasting model for PV power plants considering the power Market benefits is presented. By constructing the objective function with benefit as the target, the model can better sense the transformation of electricity Market price. By comparing with the BP model trained with precision, it is proved that the cost of energy prediction can be greatly reduced under the condition of sacrificing certain precision.