Battery charge state estimation method and device based on noise adaptive particle filtering

Abstract

The invention discloses a battery charge state estimation method and device based on noise adaptive particle filtering, and the method comprises the steps: obtaining single data sampled at a fixed period, and selecting a characteristic single body for parameter identification and state tracking; performing parameter initialization on a standard particle filtering algorithm, and obtaining an SOC initial state particle set according to a target state estimated value and an initial noise covariance; carrying out one-step prediction on an SOC value, guiding sequential importance sampling by combining terminal voltage estimation deviation of particles, and updating a normalized weight; weighting calculation is carried out to obtain a predicted value of the SOC, and random resampling is carried out according to normalized weight to complete particle set updating and new particle set measurement deviation recording; determining an adaptive process noise standard deviation range for next cycle prediction; and circularly executing to complete the battery SOC estimation based on the noise self-adaptive particle filter. The defect that when a traditional particle filtering method is applied to SOC estimation, the size of the process noise variance needs to be determined through multiple times of trial and error is overcome.

Description

technical field [0001] The invention relates to the technical field of electric vehicle battery management, in particular to a method and device for estimating the state of charge of a battery based on noise adaptive particle filtering. Background technique [0002] With the increasing pressure of energy shortage and environmental pollution, the popularity of pure electric vehicles and hybrid vehicles in life has been greatly accelerated. Battery state of charge SOC (State of Charge) is one of the key state parameters of electric vehicles, and its accurate estimation is an important means to ensure the normal operation of the vehicle. When the car is running continuously for a long time, it often brings a large cumulative deviation of Ah, and more complex prediction algorithms need to be added to improve the accuracy of real-time estimation of the state of charge SOC. However, considering the cost pressure and technical difficulty, the BMS of the vehicle-end battery managem...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the defect that the traditional particle filter method with fixed process noise is applied to the SOC estimation of the state of charge, multiple trials and errors are required to determine the variance of the process noise. The large deviation leads to unstable state tracking, which in turn leads to poor state tracking effect and reduced prediction robustness. A method and device for battery state of charge estimation based on noise adaptive particle filter is proposed

Images