Bolt Looseness Monitoring via Percussion and Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bolt looseness monitoring technologies face challenges in harsh environments due to the need for fragile piezoelectric elements, costly equipment, and the impracticality of direct measurement methods, especially for large bolts and complex geometries, which often require constant contact and sophisticated signal processing.
Innovation Solution
A percussion-based method utilizing impact-induced sound analysis combined with machine learning, where a decision tree model is trained to predict bolt looseness by extracting power spectrum density features from acoustic signals recorded using a smartphone, allowing for non-destructive, in-situ monitoring without the need for constant contact or complex equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If piezoelectric transducers are used to monitor bolt preload, then measurement precision is improved, but device complexity and reliability worsen due to fragile elements requiring constant contact
Solution Approach 1:
The patent extracts the monitoring function from fragile contact-based piezoelectric transducers and implements it through contactless acoustic sensing. The acoustic sensor captures vibration signals from the bolt without requiring physical contact, eliminating the reliability issues associated with fragile elements while maintaining measurement capability through signal processing and machine learning analysis
Solution Approach 2:
The patent replaces the mechanical contact-based piezoelectric sensing system with an acoustic vibration analysis system. Instead of using piezoelectric elements that require constant mechanical contact with the bolt, the system uses acoustic sensors to capture vibration signals generated by percussive excitation, substituting a more reliable contactless mechanical-acoustic system for the fragile electro-mechanical system
2Measurement precision
If direct measurement methods are used for bolt load, then measurement precision is improved, but ease of operation worsens due to impracticality for large bolts and complex geometries
Solution Approach 1:
The patent introduces acoustic vibration signals as an intermediary to indirectly measure bolt preload. Instead of directly measuring the load on large or complex bolts where direct measurement is impractical, the system uses percussive excitation to generate vibrations that propagate through the bolt, and analyzes these vibrations to infer preload information, making measurement accessible in difficult-to-reach locations
Solution Approach 2:
The patent replaces direct mechanical measurement methods with acoustic vibration analysis. By using acoustic sensors to capture vibration signals and applying machine learning algorithms to interpret these signals, the system achieves precise preload measurement without requiring physical access to the bolt or installation of measurement devices on large or complex geometries
3Measurement precision
If acoustic-elastic methods with piezoelectric actuators are used, then measurement precision is improved, but device complexity and cost worsen due to sophisticated equipment requirements
Solution Approach 1:
The patent extracts the essential measurement function from complex piezoelectric actuator-sensor systems and implements it using simple percussive excitation and acoustic sensing. Instead of using sophisticated piezoelectric actuators to generate controlled vibrations, the system uses simple impact excitation that naturally generates vibration signals, which are then captured by acoustic sensors and analyzed through machine learning
Solution Approach 2:
The patent replaces expensive, sophisticated piezoelectric transducer systems with inexpensive acoustic sensors and processing equipment. The system uses off-the-shelf acoustic sensors, basic signal processing, and machine learning algorithms to achieve measurement precision comparable to or exceeding traditional methods, while dramatically reducing equipment cost and complexity
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables efficient and accurate monitoring of bolt looseness across various conditions, including difficult-to-reach locations, with an average accuracy of 96.94% in identifying bolt torque levels, facilitating automated inspection and reducing reliance on human expertise and costly instrumentation.
Implementation Method 1
an acoustic sensor to capture vibration signals from the bolted structure
Implementation Method 2
an impact device to generate vibration signals by impacting the bolted structure
Data Source
AI summary
The systems and methods described herein are for monitoring the tightness of bolts. The systems and methods may be used with a mechanism to apply a percussive tap and with a recording or monitoring device for detecting and recording the acoustic signals that are generated by the percussive tap. The acoustic signals generated by percussive taps applied to bolts in various looseness states are analyzed and a machine learning model is developed that allows for determining bolt looseness based on the acoustic signals.