Intelligent Force-Measurement System Use in Shock Tunnel

Area of Science:

• Aerospace Engineering
• Measurement Science
• Artificial Intelligence

Background:

• Inertial vibrations in force measurement systems (FMS) significantly impact aircraft force test results, particularly in high-enthalpy impulse facilities like shock tunnels.
• Low-frequency vibrations and FMS motion in shock tunnels are challenging to address with digital filtering due to inertial forces from initial flow impact.

Purpose of the Study:

• To investigate the dynamic characteristics of force measurement systems (FMS) in shock tunnel environments.
• To propose and validate a novel deep-learning-based single-vector dynamic self-calibration (DL-based SV-DSC) method for impulse FMS.

Main Methods:

• A deep-learning technique, specifically convolutional neural networks, was employed to train the dynamic model of the FMS.
• The DL-based SV-DSC method was developed to eliminate low-frequency vibration signals from shock tunnel test results.
• The trained model was validated using force test data obtained from a shock tunnel.

Main Results:

• The developed deep-learning model successfully performs intelligent processing of FMS balance signals.
• The DL-based SV-DSC method effectively eliminates low-frequency vibration signals, a common issue in shock tunnel testing.
• Validation confirmed the model's capability to improve the accuracy of aerodynamic force measurements.

Conclusions:

• The proposed DL-based SV-DSC method offers a reliable and accurate approach for dynamic calibration of impulse FMS.
• This technique enhances the precision of aerodynamic force measurements in challenging shock tunnel environments.
• The study demonstrates the potential of deep learning for intelligent signal processing in advanced experimental facilities.