Electric load forecasting method based on DQPSO-LSTM with joint tax and electricity data

In the context of the current era where energy demands are increasingly growing, accurate electric load forecasting is crucial for ensuring the stability and economic efficiency of power systems. This paper presents a novel electric load forecasting method based on tax and electricity joint data, known as the DQPSO-LSTM model. This model integrates Quantum Particle Swarm Optimization (QPSO) with Long Short-Term Memory networks (LSTM), aiming to enhance forecasting accuracy. To overcome the limitation of QPSO's tendency to converge prematurely, we introduce a dynamic nonlinear varying inertia weight strategy, which enhances the algorithm's global search efficiency and local search capability. By balancing extensive exploration in the early stages and rapid convergence in the later stages, the model parameters are optimized. Validation using electricity consumption data and tax data from Heilongjiang Province from 2018 to 2021 demonstrates that the DQPSO-LSTM model significantly outperforms benchmark models such as LSTM, ARIMA, and Prophet in terms of prediction error (RMSE), mean absolute percentage error (MAPE), and the coefficient of determination (R²), highlighting the significant advantages of the proposed method in electric load forecasting.