A cnn-biGRU-based electric vehicle charging load prediction method and system
By extracting the spatial features of electric vehicle charging load using CNN and combining them with BiGRU to capture bidirectional temporal dependencies, a CNN-BiGRU hybrid model was constructed. This solved the problem of high-precision prediction of electric vehicle load under extreme weather conditions and improved the resilience of the power grid.
CN122288005APending Publication Date: 2026-06-26JIBEI ELECTRIC POWER TRADING CENT CO LTD +2
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JIBEI ELECTRIC POWER TRADING CENT CO LTD
- Filing Date
- 2026-03-25
- Publication Date
- 2026-06-26
Smart Images
Figure CN122288005A_ABST
Abstract
This invention belongs to the field of power system electricity market technology, and relates to a method and system for predicting electric vehicle charging load based on CNN-BiGRU, including: CNN spatial feature extraction, BiGRU temporal feature extraction, collaborative mining of spatial patterns and bidirectional temporal dependencies in load data, and CNN-BiGRU hybrid model training and optimization. This invention utilizes convolutional neural networks to extract the spatial distribution features of charging load, and combines bidirectional gated recurrent units to capture the bidirectional temporal dependencies in the disaster evolution process, thereby achieving high-precision prediction of electric vehicle charging load under extreme scenarios.
Need to check novelty before this filing date? Find Prior Art