Efficiency of the slime mold algorithm for damage detection of large‐scale structures

Abstract

Summary Optimization‐based methods are increasingly being implemented for structural damage detection problems through the minimization of the objective functions based on vibration data. The adopted optimization algorithm and objective function play an important role in the accurate detection and quantification of damages. Meanwhile, the challenge of long computational time is another aspect of structural damage identification problems, especially upon addressing large‐scale structures. In this paper, recently developed optimization techniques called slime mold algorithm (SMA) and marine predators algorithm (MPA) are applied to damage assessment of large‐scale structures for the first time. The performance of these algorithms is compared with those obtained using other well‐known optimization techniques such as ant lion optimizer (ALO), whale optimization algorithm (WOA), and grasshopper optimization algorithm (GOA). Furthermore, the sensitivity of three objective functions based on modal assurance criterion (MAC), modified total modal assurance criterion (MTMAC), and natural frequency vector assurance criterion (NFVAC) are examined. Two numerical studies, including the 53‐bar planar truss and the Guangzhou New TV tower, and a full‐scale three‐story frame as an experimental investigation are conducted to present a statistical comparison. The overall results show that the combination of SMA and objective function based on MTMAC provides an accurate tool for damage identification. However, improved SMA (ISMA) has been introduced to enhance the capability of standard SMA for damage detection, and especially finite element model updating in the experimental example. Five benchmark functions are also used to evaluate the global optimization capacity of ISMA and SMA. The results show that the ISMA has many benefits in terms of tackling global optimization problems. MPA‐MTMAC can provide promising results compared with ALO‐MTMAC, WOA‐MTMAC, and GOA‐MTMAC. However, excessive computation time is a big drawback for MPA. The overall results confirm the perfection of the objective function based on MTMAC compared with two others based on MAC and NFVAC.

Modal Identification of Train Passenger Seats Based on Dynamic Tests and Output-Only Techniques

“…Regarding the present work, pairing comfort and standard seat modes can be conducted through the MAC parameter, which, as stated, measures the linear correlation between modal vectors through their mean square deviation [48]. The MAC parameter is a scalar, varying between 0 and 1, whose unit value indicates that the modes are coincident or linearly dependent [49,50]. Its value is estimated based on the following expression: In the case of well-separated modes, only the first singular value will have significant information.…”

Developing and Testing High-Performance SHM Sensors Mounting Low-Noise MEMS Accelerometers

“…Indeed, accurate identification of the modal features (e.g., frequencies, modal shapes, and damping ratios) is fundamental for the reliable improvement of a representative structural model [9]. In recent years, thanks also to the development of optimized computational methods, vibration-based damage detection procedures have found further enhancements [10].…”

Data-Driven Structural Health Monitoring: Leveraging Amplitude-Aware Permutation Entropy of Time Series Model Residuals for Nonlinear Damage Diagnosis

“…Currently, SHM techniques are divided into two categories: data-driven and modeldriven [3][4][5]. Data-driven methods extract damage-sensitive features (DSFs) from measurement data and perform statistical decisions based on damage classifiers [6,7].…”