Electric vehicle charging station load prediction method based on deep learning
By combining the model structures and hyperparameter optimizations of MS-CNN and LSTM-AM, the problem of difficulty in mining local time-scale variation features in electric vehicle charging station load forecasting was solved, achieving high-precision and stable load forecasting results.
CN122153396APending Publication Date: 2026-06-05STATE GRID ANHUI ELECTRIC POWER CO LTD ELECTRIC POWER SCI RES INST
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- STATE GRID ANHUI ELECTRIC POWER CO LTD ELECTRIC POWER SCI RES INST
- Filing Date
- 2026-03-06
- Publication Date
- 2026-06-05
Smart Images
Figure CN122153396A_ABST
Abstract
The application relates to a kind of electric vehicle charging station load prediction methods based on deep learning, including electric vehicle charging station load data preprocessing and prediction sample construction, MS-CNN-based time series load feature extraction, based on LSTMAm's charging load prediction model training and based on the prediction model parameter optimization and prediction result generation of hyperparameter optimization.This application effectively improves the expression ability of load time series feature, thereby improving the accuracy of prediction result, enhancing the adaptability of model to complex load form, avoiding the information loss problem caused by single scale feature extraction, and maintaining high prediction accuracy and stability under different charging stations and different operating conditions.
Need to check novelty before this filing date? Find Prior Art