Short-term Power Load Prediction Method for High Voltage Power Cables Based on IGOA-VMD-LSTM-MHSAM

Short-term power load prediction of high-voltage power cables (HVPC) is an important basis for the development of its maintenance plan. Aiming at the current problem of low accuracy of HVPC short-term power load prediction, this paper proposes a HVPC power load prediction method based on IGOA-VMD-LSTM-MHSAM. Firstly, for the typical nonlinear characteristics of HVPC power load, the parameter optimization of Vibrational Mode Decomposition (VMD) is achieved by using Gannet Optimization Algorithm (GOA) improved by Sine chaotic mapping and nonlinear convergence factor, and the parameter optimization of VMD is adopted. IGOA to decompose the HVPC power load into several Intrinsic Mode Functions (IMFs). Then Multi-head Self-Attention Mechanism (MHSAM) is introduced into Long Short-Term Memory (LSTM) to improve its feature extraction capability, thus obtaining the LSTM-MHSAM HVPC power load prediction model. Finally, the effectiveness of the LSTM-MHSAM model is verified using the actual power load data of a 220kV HVPC in a city in China. The experimental results show that the proposed prediction model for HVPC short-term power load prediction results are closer to the actual load. Compared with the traditional forecasting methods, the proposed model has higher accuracy and stability, and has certain potential for practical application.