Physics-informed Online Engine Abnormal-Noise Detection
Acoustic-physical fusion for online detection and isolation of engine/bearing abnormal noise; patents granted in the US, Germany, and China
Project Overview
A barely audible noise can be the earliest signal of bearing failure. This project detects, isolates, and prognoses abnormal noise of vehicle bearings and engines acoustically.
Research Objectives
Acoustic-based wheel-bearing fault isolation and prognosis
Physics-data fused abnormal-noise judgment
On-vehicle online deployment
Methodology
Acoustic feature extraction fused with physical mechanism models for online detection, source isolation, and degradation prognosis.
Key Results
US (US20200149993A1), German (DE102019115237B4), and Chinese (CN111175044A) patents granted
Applied in General Motors on-vehicle monitoring scenarios
Online Equipment Fault Diagnosis