Non-stationary random vibration modelling and analysisvia Functional Series TARMAX models

摘要

A concise overview of Functional Series Time-Dependent AuroRegressive Moving Average with eXogenousrnexcitation (FS-TARMAX) based methods for non-stationary random vibration modelling and analysis isrnpresented. Parameter and model structure estimation methods are discussed, and certain new results arernpresented. The practicality and effectiveness of the methods, as well as some of their advantages overrnalternative non-stationary methods, are demonstrated through brief presentations of three case studies: Thernstochastic modelling and analysis of an earthquake ground motion signal, a manipulator vibration signal, andrnthe vibration of a 'bridge-like" laboratory structure.