Dynamic observability method for durability assessment considering measurement noise
Document typeConference report
Rights accessOpen Access
Due to the inevitable degradation of material properties in structures in daily use, such as stiffness degradation due to cracking in concrete elements, their durability will definitely be influenced, and their serviceability and safety could be in danger. Thus, understanding and identifying the change in the structural parameters provides new approaches to evaluate their durability. Structural system identification by dynamic observability method, which is using subsets of masses, natural frequencies and modal shapes, is a powerful tool to detect the change of structural parameters. Taking into account the presence of noise in the measurement data in real world structures, this method establishes the relative dynamic equation with the error separation items. The equation is solved by error minimization of an objective function combining the measured frequencies and mode shapes through the parameter MAC (Modal Assurance Criterion). Additionally, the algorithms and the steps are introduced based on the dynamic eigenvalue equation, which can fully demonstrate the performance of observability techniques. The present paper provides an example on how to successfully identify structural parameters. Its suitability for practical applications is demonstrated in a large frame structure. The result is a much more accurate identification of the parameters involved in the durability of the structure even in the case of noise-corrupted measurement signals.
CitationPeng, T.; Casas, J.; Turmo, J. Dynamic observability method for durability assessment considering measurement noise. A: International Conference on Durability of Building Materials and Components. "Current Topics and Trends on Durability of Building Materials and Components: proceedings of the XV edition of the International Conference on Durability of Building Materials and Components (DBMC 2020), Barcelona, Spain, 20-23 October 2020". 2020, p. 811-818. ISBN 978-84-121101-8-0. DOI 10.23967/dbmc.2020.047.
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder