.png){kind=icon}
.png){kind=icon}
Return
Battery pack temperature field compression sensing based on deep learning algorithm
.png){kind=icon}
PRE
AI
DOI：10.1109/icemi46757.2019.9101493.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
We propose a method that relates to a real-time monitoring method for internal temperature of a battery pack based on compression sensing and deep learning theory, belonging to the technical field of battery pack thermal management. According to the experimental temperature of all positions when charging, discharging and under different load conditions Jew the same type of battery pack, we apply the neural network algorithm in deep learning to train the simulated temperature field model which is suitable fir the specific kind of battery packs. Then we call the model by software to predict the temperature of all the positions in the battery packs, thereby completing global real-time monitoring of the internal temperature of the battery packs. In this paper, we use LSTM, and machine learning algorithms to achieve the compressive sensing of the battery packs. The results show that the LSTM algorithm requires 32-channel detection, with its average absolute error between the predicted result and the actual detection result is 0.21 degrees C and the maximum error is 1 degrees C The 'DNN algorithm only needs 8 channels for detection and the average absolute error between the prediction result and the actual detection result is 0.25 degrees. This algorithm has an excellent application prospect in the field of battery pack thermal management and real-time condition detection of power batteries packs.
Keywords:
Power battery packs
Real-thne condition detection
compressive sensing theory
Neural network algorithm
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
P
IF:
0
Papers:
1.8K
Citations:
0