Lightweight lithium battery health state estimation method based on L-TCN and GL-Former

The invention relates to a lightweight lithium battery health state estimation method based on L-TCN and GL-Former, and belongs to the technical field of new energy, and the method comprises the following steps: S1, data collection: collecting real lithium battery charge and discharge data through a sensor, including discharge voltage, discharge current, battery body temperature and battery capacity; s2, establishing a hybrid neural network model of a lightweight improved time convolutional neural network (L-TCN) and a global-local Transform architecture (GL-Former), and extracting time features and spatial features of the lithium battery data; s3, selecting an optimal model hyper-parameter by adopting a genetic algorithm; and S4, inputting the data in the step S1 and the hyper-parameters in the step S3 into a hybrid neural network model for training to obtain a trained lithium battery health state estimation model, and then performing battery health state prediction based on the lithium battery health state