Font Size: a A A

Lithium-ion Battery Parameter Identification And SOC Estimation Based On Electrochemical Model

Posted on:2019-12-01Degree:MasterType:Thesis
Country:ChinaCandidate:H DengFull Text:PDF
GTID:2492305891972609Subject:Power Engineering and Engineering Thermophysics
Abstract/Summary:PDF Full Text Request
The most developing trend of battery management system(BMS)is the model based method.The battery model mainly divided into data-driven model、equivalent circuit model and electrochemical model.The SOC estimation is the core technology of BMS.The method based on equivalent circuit model combining Kalman filter and data-driven model are the dominated SOC estimation methods.SOC on-line estimation based on electrochemical model,while online SOC estimation is hampered by the high model complexity,the mass model parameters and different parameters identifiability.Using the method of polynomial approximation,this paper simplifies the common electrochemical model-P2D(Pseudo-two-dimensional)model,and greatly reduces its computation.In this paper,Fisher information matrix was used to analyze the identification of the parameters to obtain the identifiable parameters group,and the issue of the parameters identification was solved.Then,the SOC estimation method based on the simplified electrochemical mechanism model SP2D(Simple Pseudotwo-dimensional)is proposed.In the paper,compared with COMSOL simulation software,the error of SP2 D model is verified to be less than17 m V,and the accuracy of the model is 50% higher than that of the SP(Single Particle)model,and the calculation amount is only 1.3% of the P2 D model.At the same time,the validity of the parameter identification method is verified by the model comparison,and the parameter identification error can be controlled within 4%.Finally,through the experiment confirmed the practical effect of electrochemical model,the results show that SOC estimation method based on SP2 D model is better than based on the equivalent circuit model(a first order RC model),and the average error of SOC estimation can be reduced almost 30%.In the middle and later periods of the battery discharge can be reduced up to more than60%,effectively solving the problem of SOC high precision estimation in the whole range of battery work.Finally,a BMS all-in-one system is designed based on this SOC algorithm,and the HIL test is carried out on the actual vehicle.
Keywords/Search Tags:electrochemical model, identifiability analysis, parameter identification, online SOC estimation, BMS application
PDF Full Text Request
Related items