Remaining useful life prediction of lithium‐ion battery based on an improved particle filter algorithm

Abstract

Because lithium-ion batteries are the main power source of industrial electronic equipment, their degradation process modelling and remaining useful life (RUL) prediction problems have attracted wide attention. The particle filter (PF) method has been successfully applied to suppress the model uncertainty and predict the RUL of the lithium-ion battery. In order to further enhance the stability of the PF method and realize a more satisfactory prediction result, a RUL prediction method based on the hybrid algorithm, which combines the PF algorithm and extended unbiased finite impulse response (EFIR) filter, is proposed. Firstly, the state space model of capacity degradation for the lithium-ion battery is established, and the model parameters are estimated by the extended Kalman filter (EKF) algorithm. Secondly, a preliminary battery capacity is predicted by using a regularized particle filter. The preliminary predictions with large deviations are diagnosed and repaired by combining the EFIR filter and diagnostic strategy. Finally, the optimized RUL prediction results of the lithium-ion battery are extrapolated based on the failure threshold. The experiment results demonstrate that the proposed method has good stability and accuracy in predicting the RUL of a lithium-ion battery. K E Y W O R D Sdegradation model, lithium-ion battery, particle filter, remaining useful life

Research on the remaining useful life prediction method of lithium‐ion batteries based on aging feature extraction and multi‐kernel relevance vector machine optimization model

“…Among them, f(x) is the original objective function, δ(t)H(x) is the penalty term, δ(t)is the penalty intensity, and H(x) is the penalty factor. Also, the penalty term needs to vary with the number of iterations t. Drawing on the non-fixed multi-segment mapping penalty function method in References [39], there are the following update Equations (8)(9)(10)(11)(12):…”

“…There are currently three types of RUL prediction methods 8,9 for lithium-ion batteries, namely mechanistic model, data-driven model, and hybrid methods. The mechanistic model method builds the RUL prediction model by analyzing the internal mechanism of lithiumion batteries, 10,11 but the physical and chemical properties of different batteries are different. And it needs to be re-modeled for different lithium-ion batteries, 12,13 which is a large workload and the model generalization rate is low, so the model-based method is not suitable for RUL prediction.…”

A novel combined estimation method of online full‐parameter identification and adaptive unscented particle filter for Li ‐ion batteries SOC based on fractional‐order modeling

“…19,20 Particle filtering (PF) has unique advantages in processing nonlinear systems that are not constrained by assumptions such as noise and models and are gradually becoming a research hotspot. [21][22][23][24][25] Nevertheless, there are few PF algorithms based on FOM. [26][27][28] Li et al 29 introduced the FOM combined with the PF method for SOC estimation, without considering the large SOC estimation error caused by insufficient particle diversity and scarcity in the PF algorithm.…”

Motion trajectory prediction based on a CNN-LSTM sequential model

“…It can be seen from historical traffic accidents that ensuring operation safety of traffic transportation is extremely important. Currently, there are many transportation safety performance evaluation methods, such as fault diagnosis [2][3][4][5], remaining useful life prediction [6,7], and parameters identification [8].…”