MSIE: Multi-Scale Informer for Precise Electricity Consumption Forecasting of Distributed User

The increasing complexity of electricity consumption patterns necessitates advanced forecasting models that can account for both large-scale trends and granular consumption behaviors. This paper introduces multi-scale informer for precise electricity consumption forecasting model for distributed users, which leverages multi-scale feature extraction and optimized encoder-decoder configurations, incorporating distributed user consumption prediction to better capture variations across different user types. Our method was evaluated against ARIMA and LSTM models using RMSE and MAE metrics across daily, weekly, and monthly time scales. The results show that the our method achieved the lowest RMSE and MAE values, demonstrating superior accuracy in capturing complex consumption patterns. This highlights its potential as a robust tool for precise electricity consumption forecasting and improved energy management, particularly in systems with distributed energy resources.