Polarization characteristic-based LSTM battery temperature estimation method

The invention discloses an LSTM battery temperature estimation method based on polarization characteristics, and the method comprises the steps: carrying out the multi-rate discharge test of a sample battery, and obtaining the first working condition data of the sample battery; wherein the first working condition data comprises voltage values, current values and temperature values of the sample battery under different discharge rates in a preset temperature environment; constructing a first LSTM temperature estimation model based on a genetic algorithm, and training the first model through the first working condition data; performing precision test on the trained first model, and determining heat production characteristics under different discharge rates according to a heat production mechanism of the sample battery based on a precision test result; wherein the heat production characteristics comprise an electric pressure heat production characteristic and a polarization heat production characteristic; based