Photovoltaic power prediction based on VMD and optimized BiLSTM model

Accurate PV power prediction helps power stations to develop scheduling plans and maintain safety activity of the power system. Aiming at the problems of noise interference in the PV power prediction model and the unstable performance of a single prediction model, a PV power prediction method with variational modal decomposition, fox optimisation algorithm and CNN-BiLSTM is proposed. Experiments are carried out utilizing actual photovoltaic energy data from a specific location, the efficacy of the suggested model is measured numerically through metrics including the square root of the root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE). The findings indicate that the recommended model demonstrates enhanced performance when juxtaposed with alternative photovoltaic power forecasting models, offering theoretical insights for photovoltaic power forecasting challenges across various scenarios.