A Mobile Energy Storage Power Supply Service Cabin Status Monitoring Method Based on Multi-source Perception and Improved CNN-BiLSTM

A mobile energy storage power supply service cabin status monitoring method based on multi-source perception and improved CNN-BiLSTM is proposed to address the issues of difficult analysis of multi-source data and low accuracy of state prediction using traditional methods. Firstly, a mobile energy storage power supply service cabin status monitoring system architecture is designed, which achieves status monitoring and early warning through multi-source data perception and analysis. Then, a CNN-BiLSTM model is constructed by combining Convolutional Neural Network (CNN) and Bidirectional Long Short Term Memory Network (BiLSTM), and attention mechanism is used to improve the accuracy of model analysis. Finally, the improved CNN-BiLSTM model will be deployed at the platform layer of the system for state prediction, making it easier for the application layer to call and achieve real-time monitoring of the system’s state. Based on the collected operational data of the mobile energy storage power supply service cabin, the proposed method was experimentally demonstrated. The results shows that the predicted energy storage charge state values are highly consistent with the actual values, and the RMSE and MAE of the health state prediction results are 1.14 × 10^-3 and 1.05 × 10-³, respectively.