A comparison of deconvolution techniques for stress relaxation

Area of Science:

• Materials Science
• Rheology
• Mechanical Engineering

Background:

• Stress relaxation experiments are crucial for understanding material behavior across a wide spectrum.
• Higher frequency data (around Hertz) is often lost due to specimen strain time limitations.
• This lost information is vital for comprehensive material characterization.

Purpose of the Study:

• To investigate the effectiveness of deconvolution techniques for recovering high-frequency data from stress relaxation experiments.
• To compare the performance of numerical (Fourier), semi-analytical, and linear approximation deconvolution methods.
• To assess the applicability of these methods to both theoretical and experimental data.

Main Methods:

• Numerical (Fourier) deconvolution.
• Semi-analytical deconvolution using theoretical strain functions.
• Deconvolution via a linear approximation method.
• Validation using both synthetic and real-world experimental data.

Main Results:

• All three deconvolution techniques successfully improved high-frequency data recovery up to Hertz frequencies.
• The linear approximation method demonstrated superior resolution for high-frequency analysis of theoretical data.
• Consistent improvements were observed across all methods when applied to experimental data.

Conclusions:

• Deconvolution of stress and strain during loading is a practical approach to recover high-frequency data.
• This technique significantly enhances the analysis of material relaxation spectra from stress relaxation tests.
• The findings support the broader application of deconvolution in materials characterization.