Time-variant random interval natural frequency analysis of structures. (3rd February 2018)

• Record Type: Journal Article
• Authors: Wu, Binhua; Wu, Di; Gao, Wei; Song, Chongmin
• Is Part Of: Journal of sound and vibration. Volume 414(2018)
• Page Start: 284
• Page End: 298
• Publication Date: 2018-02-03
• Subjects: Hybrid random interval analysis -- Time-variant natural frequency -- Chebyshev surrogate model -- Perturbation method -- CFST structure; Sound -- Periodicals; Vibration -- Periodicals; Son -- Périodiques; Vibration -- Périodiques; Sound; Vibration; Periodicals; Electronic journals; 620.205
• DOI: 10.1016/j.jsv.2017.11.009
• Languages: English
• ISSNs: 0022-460X

Abstract

Abstract: This paper presents a new robust method namely, unified interval Chebyshev-based random perturbation method, to tackle hybrid random interval structural natural frequency problem. In the proposed approach, random perturbation method is implemented to furnish the statistical features (i.e., mean and standard deviation) and Chebyshev surrogate model strategy is incorporated to formulate the statistical information of natural frequency with regards to the interval inputs. The comprehensive analysis framework combines the superiority of both methods in a way that computational cost is dramatically reduced. This presented method is thus capable of investigating the day-to-day based time-variant natural frequency of structures accurately and efficiently under concrete intrinsic creep effect with probabilistic and interval uncertain variables. The extreme bounds of the mean and standard deviation of natural frequency are captured through the embedded optimization strategy within the analysis procedure. Three particularly motivated numerical examples with progressive relationship in perspective of both structure type and uncertainty variables are demonstrated to justify the computational applicability, accuracy and efficiency of the proposed method. Highlights: Time-variant hybrid random interval natural frequency analysis is conducted. Unified interval Chebyshev-based random perturbation method is proposed. Efficient algorithms are devised to determine the random interval natural frequency. Extreme bounds of probabilistic features of natural frequency can be captured. Day-to-day based natural frequency assessment in practical structures is provided.