Signal Processing Using a Circular Sensor Array to Measure the Torsional Angle of a Bolted Joint

Area of Science:

• Mechanical Engineering
• Sensor Technology
• Signal Processing

Background:

• Accurate preload force determination in bolted joints is critical for structural integrity and safety.
• Traditional methods for measuring torsional angle often involve contact, limiting application scope and potentially introducing errors.
• The small torsional angles (<4 degrees) in bolted joints necessitate high-resolution measurement techniques.

Purpose of the Study:

• To develop and evaluate a novel, non-contact approach for measuring the torsional angle in bolted joints.
• To integrate a circular magnetic sensor array into a torque wrench for high-resolution angle measurement.
• To assess the efficacy of two distinct signal processing methods for calculating torsional angle.

Main Methods:

• Implementation of a circular magnetic sensor array within a torque wrench for non-contact torsional angle measurement.
• Application of the discrete Fourier transformation (DFT) for direct calculation of rotation angle from signal phase.
• Utilization of Gaussian process regression for an angle error minimization approach involving a learning phase.

Main Results:

• Both the direct DFT method and the learning-based Gaussian process regression demonstrated promising results in experimental validation.
• The direct method achieved excellent angular resolution without the need for prior training or calibration.
• The learning-based method showed potential for significant resolution enhancement in complex systems with initial reference data.

Conclusions:

• The developed non-contact method using a magnetic sensor array offers a viable solution for determining bolted joint preload force.
• The direct method is suitable for mobile and less-complex applications due to its robustness and independence from calibration.
• The learning-based method provides superior resolution in complex setups after an initial training phase, enhancing measurement accuracy.