Optimization and application of the electricity charge trial calculation technology within the intelligent electricity billing management system

Abstract

In response to the poor performance of the existing electricity billing management system in optimizing enterprise electricity costs, a method for predicting electricity demand based on the CNN-BiLSTM-Attention model has been proposed. This method has been implemented in the intelligent electricity optimization cloud platform to enhance the prediction accuracy of enterprise power load. The objective of this research is to forecast the continuous variation curve of 24-h demand in a day, in order to calculate the peak demand value and provide users with rational and effective energy efficiency optimization recommendations. Leveraging historical data, a prediction model was constructed using MATLAB in this study, and case simulations were conducted. By analyzing the performance of this model in predicting the demand for different energy usage characteristics, it was observed that the model performs better in enterprises with stable or fluctuating energy usage profiles compared to those with random energy usage profiles. Compared with other models, the MAE value and RMSE value of this model are reduced. Furthermore, the prediction results were compared with those of other models to validate the accuracy of the prediction model. Finally, a strategy for electricity cost optimization control was proposed, which involves controlling the charging or discharging of energy storage systems based on the predicted demand curve.