Researchers have proposed a model-based diagnostic framework for electric vertical take-off and landing aircraft battery systems, aiming to isolate concurrent electro-thermal faults under demanding aviation conditions. The approach uses structural analysis, analytically decoupled residuals, Kalman filtering, and adaptive thresholds to improve fault detection and isolation in battery modules.
Battery fault diagnosis is a prerequisite for the safety certification of electric vertical take-off and landing, or eVTOL, aircraft. Unlike many ground vehicle applications, eVTOL systems operate in a zero-failure tolerance environment, where even small battery anomalies may create unacceptable flight-safety risks. This makes battery diagnostics not only a battery management problem, but also a certification and system-safety problem.
Existing diagnostic methods face important limitations in this context. Many approaches are structurally built around single-fault assumptions, even though real battery systems may experience concurrent faults. Others may have limited adaptability under high-rate electro-thermal coupling, a condition that is especially relevant to aviation duty cycles. In eVTOL applications, batteries can experience rapid power demands, tight thermal margins, and strong interactions between electrical and thermal behavior.
The new study addresses these challenges by proposing a structurally decoupled, model-based diagnostic framework for battery modules. According to the article record, the researchers establish a cell-level electro-thermal coupling model to capture dynamic electrical and thermal interactions under high discharge rates. This modeling step is important because faults in battery systems may appear through both voltage and temperature behavior, and the two domains can influence one another.
Based on structural analysis, the framework constructs four minimal structurally over-constrained subsystems. These subsystems are then used to generate analytically decoupled residuals, which are diagnostic signals designed to respond differently to different fault types. In practical terms, this residual design allows the system to separate the signatures of faults that might otherwise be confused, especially when more than one fault occurs at the same time.
The framework is designed to isolate seven fault categories, including short circuits, interconnection faults, and multiple sensor failures. This scope is important for eVTOL systems because sensor faults and electrical faults can both undermine battery-state assessment. If a diagnostic system cannot distinguish a genuine cell or interconnection fault from a sensor failure, it may either miss a hazardous condition or trigger unnecessary responses.
The study also incorporates Kalman filtering to improve state-estimation robustness and an adaptive threshold strategy to accommodate different operational regimes. These additions matter because aircraft battery systems may not operate under steady or uniform conditions. A threshold that works in one regime may be too sensitive or too permissive in another, especially under high-rate discharge or changing thermal conditions.
The reported results suggest that the proposed framework performs strongly under both single-fault and concurrent-fault scenarios. Under single-fault scenarios, the method achieved a detection rate of 93.88%, enabling complete isolation and parameter estimation of all seven fault types. More critically, under concurrent-fault scenarios in a 10S3P module with 4,323 possible fault-pair combinations, 85.5% were uniquely identified and 98.3% were reduced to fewer than three candidates.
These results indicate that structural residual design may be especially important for battery systems that need certifiable multi-fault diagnosability. Data-driven pattern recognition can be valuable in many battery applications, but the study suggests that explicit structural modeling and decoupled residuals are essential when concurrent faults must be diagnosed systematically. Further work will still be needed to validate the framework across broader eVTOL architectures, operating profiles, and certification workflows. Even so, the study offers a strong indication that model-based electro-thermal fault isolation could help make next-generation eVTOL battery systems safer and more diagnosable.
Reference
Author:
Chenxu Wang a , Yanbo Jia a , Cong Hou b , Hailong Li c , Rui Xiong a
Title of original paper:
Model-based structural isolation of concurrent electro-thermal faults for eVTOL battery systems
Article link:
https://www.sciencedirect.com/science/article/pii/S2773153726000113
Journal:
Green Energy and Intelligent Transportation
DOI:
10.1016/j.geits.2026.100399
Affiliations:
a National Engineering Research Center of Electric Vehicles, School of Mechanical Engineering, Beijing Institute of Technology, Beijing, 100081, China
b Guangdong Huitian Aerospace Co. Limited, Guangzhou, 510653, China
c School of Business, Society and Engineering, Mälardalen University, Västerås, 72123, Sweden
Green Energy and Intelligent Transportation
Experimental study
Not applicable
Model-based structural isolation of concurrent electro-thermal faults for eVTOL battery systems
11-Apr-2026