Secure Estimation of Battery Voltage Under Sensor Attacks: A Self-Learning Koopman Approach

TL;DR

This work addresses secure terminal voltage estimation for Li-ion batteries in cloud-based BMS under sensor attacks during EV charging. It introduces a self-learning Koopman estimator with two-stage error compensation: Stage I corrects Koopman model error, and Stage II handles higher-order dynamics via an OCV-SOC based empirical mapping or a GPR data-driven correction. The approach requires limited offline data and minimal model information, and it adapts to aging and varying operating conditions. Simulation on a nonlinear ECM for NMC cells shows accurate voltage tracking under DoS and FDI attacks, validating robustness and practical potential for safe, efficient charging.

Abstract

Cloud-based battery management system (BMS) requires accurate terminal voltage measurement data to ensure optimal and safe charging of Lithium-ion batteries. Unfortunately, an adversary can corrupt the battery terminal voltage data as it passes from the local-BMS to the cloud-BMS through the communication network, with the objective of under- or over-charging the battery. To ensure accurate terminal voltage data under such malicious sensor attacks, this paper investigates a Koopman-based secure terminal voltage estimation scheme using a two-stage error-compensated self-learning feedback. During the first stage of error correction, the potential Koopman prediction error is estimated to compensate for the error accumulation due to the linear approximation of Koopman operator. The second stage of error compensation aims to recover the error amassing from the higher-order dynamics of the Lithium-ion batteries missed by the self-learning strategy. Specifically, we have proposed two different methods for this second stage error compensation. First, an interpretable empirical correction strategy has been obtained using the open circuit voltage to state-of-charge mapping for the battery. Second, a Gaussian process regression-based data-driven method has been explored. Finally, we demonstrate the efficacy of the proposed secure estimator using both empirical and data-driven corrections.

Figures (5)

• Figure 1: Block diagram shows the secure terminal voltage estimation generation under sensor attack.
• Figure 2: Block diagram showing the components of the secure estimator along with the sensor attack triggered self-learning feedback mechanism.
• Figure 3: For the prismatic cell NMC Lithium-ion battery, the figure exhibits OCV-SOC (top), $\Delta OCV$ (middle), and $\Delta^2 OCV$ graphs.
• Figure 4: DoS sensor attack: The plot shows the true and corrupt terminal voltage for the battery, the voltage estimation from open-loop and closed model-based observers, self-learning Koopman-based estimation with stage I correction only, with stage II empirical correction, and stage II GPR correction.
• Figure 5: FDI sensor attack on aging battery: The plot shows the true and corrupt terminal voltage for the battery, the voltage estimation from open-loop and closed model-based observers, self-learning Koopman-based estimation with stage I correction only, with stage II empirical correction, and stage II GPR correction.