Load forecasting of microgrid based on an adaptive cuckoo search optimization improved neural network

Load forecasting is an important part of microgrid control and operation. To improve the accuracy and reliability of load forecasting in microgrid, a load forecasting method based on an adaptive cuckoo search optimization improved neural network (ICS-BP) was proposed. First, a load forecasting model in microgrid based on a neural network was designed. Then, a novel adaptive step adjustment strategy was proposed for cuckoo search optimization, and an adaptive position update law based on loss fluctuation was designed. Finally, the weights and biases of the forecasting model were optimized by the improved cuckoo search algorithm. The results showed that the BP network optimized by the improved cuckoo search optimization enhanced the global search ability, avoided the local optima, quickened the convergence speed, and presented excellent performance in load forecasting. The mean absolute percentage error (MAPE) of the ICS-BP forecasting model was 1.13%, which was very close to an ideal prediction model, and was 52.3, 32.8, and 42.3% lower than that of conventional BP, cuckoo search improved BP, and particle swarm optimization improved BP, respectively, and the root mean square error (RMSE), mean absolute error (MAE), and mean square error (MSE) of ICS-BP were reduced by 75.6, 70.6, and 94.0%, respectively, compared to conventional BP. The proposed load forecasting method significantly improved the forecasting accuracy and reliability, and can effectively realize the load forecasting of microgrid.