Assessment of hybrid machine learning models for non-linear system identification of fatigue test rigs
This work addresses the problem of non-linear system identification for fatigue test rigs, which is incremental as it builds on existing linear models by adding LSTM networks.
The paper tackled the challenge of predicting system responses for fatigue test rigs by proposing a novel hybrid model that augments existing linear approaches with Long Short-Term Memory networks to account for non-linear phenomena, and it was tested using experimental data from a servo-hydraulic test rig with evaluation based on various metrics in time and frequency domains as well as fatigue strength under variable amplitudes.
The prediction of system responses for a given fatigue test bench drive signal is a challenging task, for which linear frequency response function models are commonly used. To account for non-linear phenomena, a novel hybrid model is suggested, which augments existing approaches using Long Short-Term Memory networks. Additional virtual sensing applications of this method are demonstrated. The approach is tested using non-linear experimental data from a servo-hydraulic test rig and this dataset is made publicly available. A variety of metrics in time and frequency domains, as well as fatigue strength under variable amplitudes, are employed in the evaluation.