Manufacturing, Ultrasonic Testing

Nondestructive Evaluation of Lithium-Ion Battery Health Using Low-Frequency Resonant Vibration

Abstract

The structural integrity of lithium-ion batteries is crucial for safety and reliability, particularly in electric vehicles and energy storage systems. Electrical diagnostics alone often cannot reveal internal defects such as delamination, lithium plating, or layer separation, particularly after mechanical impacts. This study introduces a nondestructive evaluation (NDE) method based on low-frequency resonant vibration (LFRV) using a shaker and a laser Doppler vibrometer. Four groups of commercial LiFePO4 battery packs, including healthy, aged, and mechanically impacted samples at different states of charge, were tested under controlled excitation. The measured vibration spectra were used to extract features, including peak frequency, spectral centroid, and energy-related metrics. These features were reduced and visualized using principal component analysis (PCA) to identify differences between battery conditions. PCA separated damaged and healthy packs with strong clustering, demonstrating the sensitivity of the vibration-based features to internal structural changes. The proposed LFRV method, combined with statistical pattern recognition, offers a fast and noninvasive approach for assessing battery health and post-impact screening.

Keywords: lithium-ion battery (LiB), nondestructive evaluation (NDE), low-frequency resonant vibration (LFRV), laser Doppler vibrometer (LDV), structural health monitoring (SHM), modal analysis

1. Introduction

Lithium-ion batteries (LiBs) play a central role in modern energy storage, particularly in electric vehicles (EVs), portable electronics, and renewable energy systems. Global demand is projected to reach 2.6 TWh by 2030 [1], which increases the importance of ensuring their safety and reliability. Recent high-profile failures, including smartphone recalls and EV fires, demonstrate the risks associated with hidden internal defects [2]. Mechanical abuse scenarios, such as crashes and drops, are of particular concern, as they can introduce latent internal damage that may not appear immediately but can trigger thermal runaway hours or even days after the event [2–4].

Chat Window Trigger