Real time hybrid simulation with online model updating: An analysis of accuracy. (1st February 2017)

• Record Type: Journal Article
• Authors: Ou, Ge; Dyke, Shirley J.; Prakash, Arun
• Is Part Of: Mechanical systems and signal processing. Volume 84:Part B(2017)
• Page Start: 223
• Page End: 240
• Publication Date: 2017-02-01
• Subjects: Constrained unscented Kalman filter -- Model updating -- Nonlinear system identification -- Real time -- Hybrid simulation; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2016.06.015
• Languages: English
• ISSNs: 0888-3270

Abstract

Abstract: In conventional hybrid simulation (HS) and real time hybrid simulation (RTHS) applications, the information exchanged between the experimental substructure and numerical substructure is typically restricted to the interface boundary conditions (force, displacement, acceleration, etc.). With additional demands being placed on RTHS and recent advances in recursive system identification techniques, an opportunity arises to improve the fidelity by extracting information from the experimental substructure. Online model updating algorithms enable the numerical model of components (herein named the target model), that are similar to the physical specimen to be modified accordingly. This manuscript demonstrates the power of integrating a model updating algorithm into RTHS (RTHSMU) and explores the possible challenges of this approach through a practical simulation. Two Bouc–Wen models with varying levels of complexity are used as target models to validate the concept and evaluate the performance of this approach. The constrained unscented Kalman filter (CUKF) is selected for using in the model updating algorithm. The accuracy of RTHSMU is evaluated through an estimation output error indicator, a model updating output error indicator, and a system identification error indicator. The results illustrate that, under applicable constraints, by integrating model updating into RTHS, the global response accuracy can be improved when the target model is unknown. A discussion on model updating parameter sensitivity to updating accuracy is also presented to provide guidance for potential users. Abstract : Highlights: The paper demonstrated the power of hybrid simulation with model updating (HSMU). The proposed HSMU method is used to identify two different nonlinear models. The method discussed here yields improvements in the global behavior in the HSMU. Model complexity, parameter sensitivity, and excitation completeness are assessed. Provided guidelines for further implementation of HSMU.