Improving magnitude and phase comparison metrics for frequency response functions using cross-correlation and log-frequency shifting

This paper provides a new method for matching dominant features of Frequency Response Functions (FRFs). In particular, this paper proposes a slicing and shifting method where the key features (in this case, resonant amplitudes) of a baseline FRF are compared with a similar FRF using cross-correlation and a Log-Frequency Shift (LFS). The goal is to provide an alternative to classic point-to-point methods for FRF comparison and instead to match the dominant features of two FRFs in a way similar to visual inspection. This enables existing FRF comparison metrics to then be applied with greater fidelity. Additionally, this paper introduces the Phase Similarity Metric (PSM) for comparing phases of two FRFs and illustrates the improvement of using LFS for phase comparisons as well. This paper uses a cantilever beam experiment in multiple configurations to provide a benchmark case for FRF comparison improvements.

Automated summary

This paper presents a new method for matching the dominant features of Frequency Response Functions (FRFs). The method involves slicing and shifting the baseline FRF to compare its key features with a similar FRF using cross-correlation and a Log-Frequency Shift (LFS). The paper also introduces a Phase Similarity Metric (PSM) for comparing the phases of two FRFs. Experimental results with a cantilever beam demonstrate the effectiveness of the proposed method.