In this paper, a novel state of health (SOH) estimation method based on partial charge voltage and current data is proposed. The extraction of feature variables, which are energy signal, the Ah-throughput, and the charge duration, is discussed and analyzed. The support vector machine (SVM) with radial basis function (RBF) as kernel function is applied for the SOH estimation. The predictive performance of the SOH by the SVM are performed with full and partial charging data. Experiment results show that the addressed approach enables estimating the SOH accurately for practical application.

In this paper, an adaptive fusion algorithm is proposed to robustly estimate the state of charge of lithium-ion batteries. An improved recursive least square algorithm with a forgetting factor is employed to identify parameters of the built equivalent circuit model, and the least square support vector machine algorithm is synchronously leveraged to estimate the battery state of health. On this basis, an adaptive H-infinity filter algorithm is applied to predict the battery state of charge and to cope with uncertainty of model errors and prior noise evaluation. The proposed algorithm is comprehensively validated within a full operational temperature range of battery and with different aging status. Experimental results reveal that the maximum absolute error of the fusion estimation algorithm is less than 1.2%, manifesting its effectiveness and stability when subject to internal capacity degradation of battery and operating temperature variation.

Accurate estimation of inner status is vital for safe reliable operation of lithium-ion batteries. In this study, a temperature compensation based adaptive algorithm is proposed to simultaneously estimate the multi-state of lithium-ion batteries including state of charge, state of health and state of power. In the proposed co-estimation algorithm, the state of health is identified by the open circuit voltage-based feature point method. On the basis of accurate capacity prediction, the state of charge is estimated by the adaptive extended Kalman filter with a forgetting factor considering temperature correction. The state of power is determined according to the multi constraints subject to state of charge, operating temperature and maximum current duration. The substantial experimental validations in terms of different current profiles, aging status and time-varying temperature operating conditions highlight that the proposed algorithm furnishes preferable estimation precision with certain robustness, compared with the traditional extended Kalman filter and the adaptive extended Kalman filter. Moreover, the battery pack validation is performed to further justify the feasibility of proposed algorithm when employed in a product battery management system.

State of health is one of the most critical parameters to characterize inner status of lithium-ion batteries in electric vehicles. In this study, a uniform estimation framework is proposed to simultaneously achieve the estimation of state of health and optimize the healthy features therein, which are excavated based on the charging voltage curves within a fixed range. The fixed size least squares-support vector machine is employed to estimate the state of health with less computation intensity, and the genetic algorithm is applied to search the optimal charging voltage range and parameters of fixed size least squares-support vector machine.By this manner, the measured raw data during the charging process can be directly fed into the estimation model without any pretreatment. The estimation performance of proposed algorithm is validated in terms of different voltage ranges and sampling time, and also compared with other three traditional machine learning algorithms. The experimental results highlight that the presented estimation framework cannot only restrict the prediction error of state of health within 2%, but also feature high robustness and universality.

In this study, a machine learning method is proposed for online diagnosis of battery state of health.A prediction model for future voltage profiles is established based on the extreme learning machine algorithm with the short-term charging data. A fixed size least squares-based support vector machine with a mixed kernel function is employed to learn the dependency of state of health on feature variables generated from the charging voltage profile without preprocessing data. The simulated annealing method is employed to search and optimize the key parameters of the fixed size least squares support vector machine and the mixed kernel function. By this manner, the proposed algorithm requires only partial random and discontinuous charging data, enabling practical online diagnosis of state of health. The model training and experimental validation are conducted with different kernel functions, and the influence of voltage range and noise are also investigated.The results indicate that the proposed method can not only maintain the state of health estimation error within 2%, but also improve robustness and reliability.