4 August 2023 in IEEE Transactions on Transportation Electrification
Thermal monitoring is critical to the safety of electric vehicles. Due to the uneven surface temperature of large-format lithium-ion (Li-ion) batteries, traditional lump-based thermal monitoring methods can’t capture two-dimensional (2-D) spatiotemporal thermodynamics. This paper proposes a distributed thermal monitoring framework to solve this problem. It contains two parts. In the modeling part, the spectral method is utilized to establish the globally distributed temperature field of the pouch cell using only four sensors. In the fault detection part, there are two stages: 1) In the offline training stage, the 2-D battery thermal process is first decomposed into basis functions and time coefficients using the spectral method. The time coefficients are further decomposed by independent component analysis. Then, the dominant temporal components are formed as the monitoring statistic, which is used to derive the confidence bound through kernel density estimation. 2) In the online detection stage, the thermal fault can be detected in real-time by comparing the updated monitoring statistic with the confidence bound. Simulations and experiments on a pouch-type Li-ion battery are conducted to verify the effectiveness of the proposed method.
7 September 2023 in IEEE Transactions on Industrial Informatics
Temperature is a key factor affecting the safety of the Lithium-ion (Li-ion) battery. Therefore, real-time thermal fault diagnosis is becoming more and more prominent, as battery faults can lead to local overheating and thermal runaway in severe cases. This article proposes a Chebyshev–Galerkin-based thermal fault detection and localization framework for the pouch-type Li-ion battery under limited sensing. First, the Chebyshev function is used to construct the spatial basis functions with global and orthonormal properties. Under the time–space (T-S) separation framework, the time coefficients can be derived through the Galerkin method using six sensors. Then, by decomposing the time coefficients using the independent component analysis, the temporal and spatial reference statistics can be formed for real-time fault detection. Finally, considering the detected fault snapshots, the thermal fault location can be identified by finding the maximum contributed position through T-S synthesis. Simulations and experiments demonstrate the effectiveness of the proposed method.
15 July 2025 in Journal of Power Sources
6 March 2025 in IEEE Transactions on Instrumentation and Measurement
Sensors play a key role in monitoring distributed parameter systems (DPSs), such as thermal and chemical diffusion-reaction processes. However, sensor faults can result in data distortion, degraded system performance, and even catastrophic consequences. This paper proposes a model-based sensor fault diagnosis framework for DPSs, which can effectively detect the fault time and estimate the fault intensity. The proposed method only requires the limited measured output without any state information. Besides, noise will cause perturbations in the sampling data. Correspondingly, a linear matrix inequality considering disturbance is designed. Through theoretical analysis, the convergence of fault estimation error is ensured. The effectiveness of the proposed method is verified on a heat transfer rod and a chemical diffusion-reaction system with static and time-varying sensor faults under disturbances. The root means square errors of all sensor fault estimates are within 0.1078.
30 July 2026 in Applied Thermal Engineering