This conference paper was submitted for presentation at the NAFEMS World Congress 2025, held in Salzburg, Austria from May 19–22, 2025.

Abstract

In traditional internal combustion engine vehicles, a small, lightweight battery is typically attached to the body in white. In contrast, modern electric vehicles (EVs) integrate the battery as a crucial part of the chassis. The primary challenges for an EV battery include its lifespan, range, and structural durability. The EV battery, comparable in weight to the chassis, significantly influences the vehicle'™s dynamic response and structural integrity. It must withstand road loads, provide crash protection, and manage inertial loads from heavy cells. This complex system comprises a load-bearing chassis and numerous joints, making the EV battery not only an electrochemical system but also a sophisticated mechanical structure. The structural durability of EV batteries encompasses fatigue design, fatigue simulation, and qualification testing under structural, thermal, and inertial loads. To assess these factors, EV batteries undergo vibration profiles (including proving ground road load data test '“ measure vibration levels for each event) on a shaker table during testing. This setup characterizes fatigue damage and shock spectra for various duty cycles, with the option to accelerate tests if needed. Beyond structural durability assessments, quantifying warranty exposure is crucial for companies. This involves balancing excessive customer loading against the quality and strength of components. The presentation addresses the challenges and issues in the qualification process, including fatigue simulation, verification, and validation of reliability tests to account for uncertainties. The design of the battery pack is a significant uncertainty factor, with uncertainties categorized as either epistemic (reducible through better knowledge, measurement errors, and model simplifications) or aleatoric (irreducible, stemming from natural phenomena like material variability). The presentation will explore probabilistic design simulation methods and the statistical correlation of simulation and tests with small sample sizes to estimate product reliability and reduce uncertainties. It will conclude with a summary of system-level reliability simulation, highlighting quantifiable risk exposure, reliability growth analysis, and optimal maintenance planning.