Monitoring Bolt Tightness Using Percussion and Machine Learning
Many civil structural failures and equipment accidents occurred due to the looseness of the bolted joints. This invention proposes a new percussion-based non-destructive approach to determine the health condition of bolted joints. Due to the different interfacial properties among the bolts, nuts and the host structure, bolted joints can generate unique sounds when it is excited by impacts, such as from tapping. Characteristics of the sounds can be analyzed by data processing algorithms to obtain unique features and classify recorded tapping data. A simple machine learning model using the decision tree method can be employed to identify the bolt looseness level. The proposed approach can be used to produce a cyber physical system that can automatically inspect and determine the health of a structure.
  • System used by infrastructure inspectors to perform examination of bolted connections
  • Inspection of other critical structural components such as welds, hinges, valves, etc.
Problems Addressed
  • The existing technologies require the sensors to be in constant contact with the bolted joint or its host structure
  • Existing technologies are sensitive to particular bolt type and geometry of host structure, which limits those technologies for in-situ application
Competitive Advantages
  • This approach of bolt looseness monitoring does not require specialized training for technicians
  • There is no installation or sustained contact with the bolt necessary
  • This method of monitoring reaches 96-98% accuracy; much more precise than ultrasonic methods
  • US Patent No.62/664,366
Case ID
Dr. Gangbing Song
Professor, Department of Mechanical Engineering