Parameter identification method for lithium-ion batteries based on recursive least square with sliding window difference forgetting factor

In order to ensure battery management system (BMS) operating safely and reliably, it is of critical importance to accurately identify lithium-ion battery model parameters. A recursive least square parameter identification approach is proposed by variable forgetting factor with the difference between open circuit and terminal voltages in sliding window mode (SDFF-RLS). The contributions of the approach include: Firstly, a novel discrete form of second-order resistance-capacitance (RC) model for lithium-ion battery is derived by the difference between open circuit and terminal voltages. Secondly, self-adaptive selection of sliding window size is applied according to working conditions, which effectively response to data perturbations. On this basis, forgetting factor is obtained by utilizing mean square value of the difference between open circuit and terminal voltages within sliding window, which strengthens the amount of information available from new data. Besides, the model parameters are counter-solved using recursive least square (RLS) approach. Thirdly, the effectiveness of the proposed approach is validated by comprehensive experiments on three different working conditions. The experimental results demonstrate that the approach has the following strengths: (1) self-adaptive selection of sliding window size is able to improve efficiency of data utilization; (2) taking sliding window mode and the difference between open circuit and terminal voltages as the basis for forgetting factor contributes to promoting convergence speed of parameters; (3) the proposed SDFF-RLS algorithm makes it possible to identify the parameters accurately while keeping the terminal voltage error within a relatively narrow range.

Automated summary

This paper introduces a recursive least square parameter identification method using variable forgetting factor and the difference between open circuit and terminal voltages in sliding window mode, aiming to accurately identify lithium-ion battery model parameters. The approach adapts sliding window size according to working conditions, enhances information utilization by considering mean square value of voltage difference, and achieves accurate parameter identification with relatively narrow terminal voltage error range.