Identification of Instantaneous Modal Parameters of a Time Varying Structure Via a Neural Network

Abstract eng:
The present work develops a novel procedure of establishing a neural network for a time varying system and estimating the instantaneous modal parameters of the system from the established neural network. The connective weights and thresholds in a neural network are assumed as functions of time and are expanded by shape functions constructing by a moving least-squares technique with polynomial basis functions. The instantaneous modal parameters of the system are directly estimated from the connective weights. The feasibility of the proposed procedure is demonstrated by processing numerically simulated dynamic responses of a time-varying linear system. The proposed procedure is also applied to process the dynamic responses of a five-story steel frame, subjected to 10% and 60% of the strength of the Kobe earthquake, in shaking table tests. The steel frame responded nonlinearly when it subjected to 60% Kobe earthquake, while it responded linearly under 10% Kobe earthquake input. This work further assesses the possible damage in the steel frame based on the instantaneous modal parameters identified from the responses corresponding to different strengths of the Kobe earthquake.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket