Real-Time Optimal Design of Experiment for Parameter Identification of Li-Ion Cell Electrochemical Model

TL;DR

This paper tackles the problem of efficient parameter identification for Li-ion battery electrochemical models in dynamic environments. It proposes a real-time reinforcement-learning-based optimal experimental design (RL-OED) implemented in a hardware-in-the-loop (HIL) testbed and guided by a reduced-order electrochemical model (E-ECM) to identify a rate constant $k_n$. A data-driven state estimator based on SINDYc feeds the RL agent, while the policy is learned with Twin Delayed DDPG (TD3) to maximize Fisher Information $FI$ subject to voltage constraints. Experiments show that RL-HIL can achieve high identifiability and accurate drive-cycle prediction with significantly shorter experiment durations compared with RCID or fixed-current tests, highlighting its practical impact for BMS verification and SOC estimation.

Abstract

Accurately identifying the parameters of electrochemical models of li-ion battery (LiB) cells is a critical task for enhancing the fidelity and predictive ability. Traditional parameter identification methods often require extensive data collection experiments and lack adaptability in dynamic environments. This paper describes a Reinforcement Learning (RL) based approach that dynamically tailors the current profile applied to a LiB cell to optimize the parameters identifiability of the electrochemical model. The proposed framework is implemented in real-time using a Hardware-in-the-Loop (HIL) setup, which serves as a reliable testbed for evaluating the RL-based design strategy. The HIL validation confirms that the RL-based experimental design outperforms conventional test protocols used for parameter identification in terms of both reducing the modeling errors on a verification test and minimizing the duration of the experiment used for parameter identification.

Figures (6)

• Figure 1: Schematic of HIL experimental setup.
• Figure 2: Current profile for training
• Figure 3: Battery model state $\overline{c}_{se,p}$ prediction for training set
• Figure 4: Battery model state $\overline{c}_{se,p}$ prediction under a section of training and validation sets
• Figure 5: RL-HIL experimental data for $k_n$ identification