Nonlinear Kalman Filtering for acoustic emission source localization in anisotropic panels

Area of Science:

• Applied Probability
• Control Systems
• Materials Science
• Nonlinear Dynamics

Background:

• Nonlinear Kalman Filtering is crucial for applications like target tracking and prediction.
• Its use in acoustic emission (AE) source localization remains underexplored.
• Accurate AE source localization is vital for structural health monitoring in composite materials.

Purpose of the Study:

• To apply and evaluate Extended Kalman Filter (EKF) and Unscented Kalman Filter (UKF) for AE source localization in anisotropic panels.
• To compare the performance of EKF and UKF against traditional nonlinear least squares methods.
• To assess the algorithms' efficacy in both known and unknown velocity profile scenarios.

Main Methods:

• Implementation of EKF and UKF algorithms for AE source localization.
• Experimental validation using a carbon-fiber reinforced polymer (CFRP) composite panel.
• Simulation of AE sources with a miniature impulse hammer and piezoelectric transducers.
• Evaluation based on localization accuracy and computational cost.

Main Results:

• Both EKF and UKF demonstrated capability in estimating AE source locations in CFRP panels.
• The algorithms provided confidence intervals for the estimated source locations.
• Performance was assessed for cases with known and unknown material velocity profiles.
• Comparison with nonlinear least squares highlighted the advantages of Kalman filtering approaches.

Conclusions:

• EKF and UKF are effective nonlinear filtering techniques for AE source localization in anisotropic composite materials.
• These methods offer robust performance and provide uncertainty quantification, crucial for practical applications.
• The study validates the application of advanced filtering techniques in a challenging materials science context.