Lithium iron phosphate battery SOC estimation method based on dynamic optimization forgetting factor recursive least square online identification

Abstract

The invention provides a lithium iron phosphate battery state of charge (SOC) estimation method based on dynamic optimization forgetting factor recursive least square online identification. The method comprises the steps: carrying out segmented equal-interval discharge standing test on a lithium iron phosphate battery to obtain the curve relation between the open-circuit voltage and the state of charge of the lithium iron phosphate battery, carrying out particle swarm optimization on a forgetting factor, recurring a forgetting factor of a least square method, and estimating the SOC of the battery by combining optimized least square online identification and extended Kalman filtering. According to the method, the forgetting factor selected in real time can enable the terminal voltage following error to be extremely small so as to ensure the accuracy of parameter online identification, and the optimized forgetting factor recursive least square and extended Kalman filtering are combined to estimate the SOC of the lithium iron phosphate battery; and the combined estimation precision is high, and the dynamic requirements of combined estimation on forgetting factor values under different batteries and different battery use environments can be adapted.

Description

technical field [0001] The invention relates to a method for estimating the state of charge (SOC) of a power battery. It provides a method for estimating the SOC of a lithium iron phosphate battery based on a dynamic optimal forgetting factor recursive least squares online identification. SOC estimation accuracy. Background technique [0002] In recent years, the development of electric vehicles has become increasingly rapid, and the role of power batteries as the power source of electric vehicles is self-evident. SOC is an important parameter of power batteries, and researchers are eager to estimate SOC more accurately to serve vehicle energy management and safety. [0003] The equivalent circuit model observation method is currently the mainstream battery SOC estimation method. Based on the circuit model, the battery SOC is estimated by using the charging and discharging data to identify the parameters and combining with the filtering algorithm. Model parameters affect ...

AI Technical Summary

Problems solved by technology

Offline identification has a relatively simple least squares fitting method, but it generally requires appropriate initial parameters, but as the battery environment and cycle times change, its internal parameters will inevitably change. The parameters of the environment

Images