SOH (state of health) estimation method, device and equipment of energy storage system

The invention provides an energy storage system SOH estimation method, device and equipment, and relates to the technical field of energy storage systems, and the method comprises the steps: obtaining the actual charging and discharging data of an energy storage system in actual working condition operation; the actual data are input into a pre-constructed actual working condition feature extraction model, actual time sequence features are obtained, and the actual working condition feature extraction model is obtained through combined training of transfer learning and adversarial training; the actual time sequence features are input into a preset SOH prediction model, an SOH estimated value of the energy storage system is obtained, and the SOH prediction model is correspondingly obtained when the actual working condition feature extraction model is trained. Thus, through transfer learning and adversarial training, features learned from a small amount of data with SOH calibration in a laboratory can be realized