Prognostic Bearing Assessment Testbed for High-Load Non-Steady State Systems

Maintaining reliable and maintainable high-load bearings is critical to the operational availability of U.S. Navy aircraft carrier arresting systems. Bearing failures in these systems can arise from a combination of mechanical and environmental factors such as lubricant degradation, fatigue, extreme temperature fluctuations, and high dynamic loads during operations. Maintenance and operational costs are significantly impacted by frequent greasing and teardown inspections. The implementation of condition-based maintenance (CBM) through sensorization could provide real-time health assessments of bearings, reducing the frequency of shutdowns and increasing overall operational efficiency. Existing approaches lack the capability to precisely simulate the complex failure modes observed in Navy aircraft carrier arresting system bearings, making it difficult to validate CBM methodologies. To address this gap, a dedicated test rig is necessary to replicate real-world failure mechanisms, enabling the systematic study of degradation patterns, wear progression, and sensor response under varying operational conditions. Construction of a test rig will provide reliable data for estimating maintenance intervals and detecting fault signatures in arresting gear systems utilizing ML techniques. During Phase I, Luna Labs and VTNSI will develop a conceptual design for a sensorized high-load non-steady-state testbed to produce data to train and test ML models, which will form the basis of CBM nodes. This testbed will serve as a critical platform for the future development of CBM fault detection models. Construction and validation of the operational of the testbed for data collection will be performed during Phase I Option and Phase II periods, utilizing the conceptional design developed in the Phase I base. Luna Labs and VTNSI will leverage open-source datasets, simulations, and previously collected data from analogous testbeds to inform the design process. This iterative and data-driven approach will ensure the testbed is robust to provide high-quality fault data in environments relevant to on-asset sheave bearings and effectively support the development of advanced CBM solutions in subsequent phases.