A Novel Data Fusion and Operational State Awareness Method for Distribution Network Feeders Based on Deep Learning Algorithm

Le résumé fourni par la source

The effective integration of multi-source heterogeneous data and accurate perception of operational status in distribution networks are essential for advancing grid intelligence. This paper presents a novel data fusion and operational state awareness method for distribution network feeders using a deep learning algorithm. A virtual data generator simulates electrical and weather time-series data, capturing seasonal, daily, and weekly patterns alongside occasional anomalies. A dedicated data processor fuses and standardises these multi-source inputs. A Long Short-Term Memory (LSTM) network is then constructed to predict key operational parameters such as voltage and current. Experimental results demonstrate that the proposed model achieves a Mean Squared Error (MSE) below 0.005 on the test set, with Mean Absolute Error (MAE) for all target variables remaining under 5% of their mean values. The model exhibits high prediction accuracy and robust generalisation, offering effective support for real-time state awareness and fault warning in distribution networks.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.