Article directory
Effect list
Article overview
Pytorch implements Transformer-based lithium battery life prediction, the environment is pytorch 1.8.0, pandas 0.24.2
As the number of charges and discharges increases, the performance of lithium batteries gradually decreases. The performance of the battery can be expressed in terms of capacity, so the life prediction (RUL) can be defined as follows:
SOH(t)=CtC0×100%,
where C0 represents the rated capacity and Ct represents the capacity at time t. When the SOH drops to 70-80%, the battery can be scrapped. What we want to do is use the battery's historical data, such as current, voltage and capacity, to model the battery's downward trend. Then, use the trained model to predict the RUL of the battery.