COMPARE THE OPTIMAL ALGORITHMS FOR WAVENET APPLICATIONS IN LOAD FORECASTING

Abstract

Power load forecasting is an important issue in microgrid energy management. Accurate load forecasting is urgently required for effective power management for microgrid. This paper considers the evaluation of the effectiveness of applying different optimization algorithms to the proposed Deep Learning Neural Network, which is Wavenet. Models combining optimization algorithms with Wavenet are applied for short-term load forecasting. In order to evaluate the accuracy of the predictive models, this study used optimization algorithms (HHO, Adam, RMSprop, SGD, Adagrad) to calculate the Wavenet network. To perform calculations for the model, we work with the load data set of a microgrid model belonging to the Ho Chi Minh City power grid. The results show that our HHO model outperforms the model based on other optimization algorithms in terms of Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE).