Damping parameter estimation using topological signal processing

Energy is dissipated in engineering systems through a variety of different, typically complex, damping mechanisms. While there are several common models with physical interpretations, it often necessary to examine noisy, experimental data order identify and fit the parameters. One class methods fits parameters based on decay envelope signal’s peaks some assumed mechanism. exist results literature, especially for viscous damping, guide selection spacing between optimum ratio identification, these overlook difficult classic problem identifying “true” presence noise. Therefore, this work we utilize tools from Topological Data Analysis (TDA) address robust automatic power-law identification or free-response function. The function can be obtained initial excitation, using random decrement method thus making our approach viable tool operational modal analysis. We present one-dimensional examples, describe extensions multiple degree freedom systems. Our framework uses persistent homology zero dimensional (0D) sub-level sets, one main TDA, separate true topological features signal (i.e. valleys) measurement Using able automatically represent damped response noise-robust, two-dimensional summary peak–valley pairs called persistence diagram. diagram then analyzed two methods: (1) theoretical analysis significant (2) fitting space. that establish general terms where applies, develop specific viscous, coulomb, quadratic damping. show computationally fast, operates raw itself without requiring any pre-processing, reduces number decisions needed by user perform calculations. validated combination numerical data.