Delta-Based Target Reformulation for Short-Term Electricity Load Forecasting Using LSTM and Transformer Models

2026-06-16

AI Summary

This study addresses the non-stationarity in short-term electricity load forecasting caused by weather fluctuations, calendar effects, and evolving consumption patterns. To tackle this challenge, the authors propose a delta-based target reconstruction approach that incorporates the differencing principle from ARIMA as an inductive bias within deep learning frameworks. Specifically, LSTM, Transformer, and LightGBM models are trained to predict the incremental change in load between consecutive time steps, and the final forecasts are reconstructed by combining these deltas with the most recent observed load values. Evaluated on a comprehensive dataset integrating multi-year hourly load data from India, NASA POWER meteorological variables, and calendar features, the delta strategy reduces MAPE by over 50% for all models in hourly forecasting tasks and substantially enhances the day-ahead performance of LSTM and Transformer, while LightGBM remains competitive when trained on absolute targets.

Abstract

Accurate short-term electricity load forecasting is critical for the reliable and economic operation of modern power systems, under non-stationarity arising from weather variability, calendar effects, and evolving consumption patterns. While deep learning models such as LSTMs and Transformers show promising performance, most existing studies focus on direct absolute load prediction without explicitly addressing target non-stationarity. Motivated by classical time-series differencing techniques in ARIMA models, this paper investigates a delta-based target reformulation for short-term electricity load forecasting using deep learning. Instead of directly predicting absolute load values, the proposed formulation trains models to predict the change in load between consecutive time steps, with final forecasts reconstructed using the last observed load. This aims to stabilize the learning target and reduce forecasting difficulty. Using multi-year, hourly real-world electricity load data from India, augmented with meteorological variables from the NASA POWER project and calendar features, this study evaluates LSTM and Transformer models under both formulations, benchmarking them against LightGBM. Experiments are conducted for hour-ahead and day-ahead horizons, assessing performance via Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Results show that delta-based reformulation consistently improves forecasting accuracy for hour-ahead prediction across all evaluated models, yielding MAPE reductions of over 50% compared to absolute formulations. For day-ahead forecasting, delta targets specifically benefit deep sequence models (LSTM and Transformer), while LightGBM remains competitive under the absolute formulation. These findings indicate that while delta reformulation is a powerful inductive bias for neural networks, its efficacy is model- and horizon-dependent.

Problem

• short-term electricity load forecasting
• non-stationarity
• delta-based target reformulation
• deep learning
• time-series forecasting

Innovation

• delta-based reformulation
• non-stationarity
• load forecasting
• LSTM
• Transformer