Battery SOC state estimation method based on nonlinear Kalman filtering

The invention provides a battery SOC state estimation method based on nonlinear Kalman filtering, and the method comprises the steps: building a state space expression of a battery through building a battery model, taking the SOC and polarization voltage of the battery as state variables, and taking the output voltage of the battery as an output variable; a series of point sets which have the same statistical characteristics as the original state quantity SOC at each k moment in a state equation are sampled through proportional correction to represent the distribution characteristics of state variables, and through transmission of a nonlinear function, a posteriori mean value and a variance are approximated by using a weighted statistical linear regression technology; the positive semi-definite variance can be ensured, and the non-local sampling problem can be solved; and finally, through a neural network algorithm, the output result of the SOC is corrected, and the SOC estimation precision is further improve