Performance of swarm intelligence based chaotic meta-heuristic algorithms in civil structural health monitoring

• Published in: Measurement : journal of the International Measurement Confederation Vol. 169; p. 108533
• Main Authors: Das, Swagato, Saha, Purnachandra
• Format: Journal Article
• Language: English
• Published: London Elsevier Ltd 01.02.2021; Elsevier Science Ltd
• Subjects: Algorithms; ASCE benchmark structure; Benchmarks; Chaotic maps; Chaotic meta-heuristic optimization; Damage assessment; Entrapment; Heuristic methods; Optimization; Stiffness; Structural health monitoring; Swarm intelligence; Swarm intelligence; Structural health monitoring; ASCE benchmark structure; Chaotic meta-heuristic optimization; Chaotic maps
• ISSN: 0263-2241; 1873-412X
• DOI: 10.1016/j.measurement.2020.108533

•Application of meta-heuristic algorithms for civil structural health monitoring.•Proposal of chaotic nature in Firefly Algorithm (FA) and Bird Swarm Algorithm (BSA).•Study the performance of the optimization algorithms for damage analysis.•Calculate standard deviation and 95% confidence level range of algorithms.•Chaotic FA and BSA shows better accuracy in identifying structural damage results. Whale Optimization Algorithm, Eagle Perching Optimization, Dragonfly Algorithm, Flower Pollination Algorithm, Bird Swarm Algorithm (BSA) and Firefly Algorithm (FA), are few of the Swarm-Intelligence based optimization techniques that have been developed by researchers and tested on benchmark functions only and have not been explored for real-life structural health monitoring (SHM). This paper deals with the use of these six algorithms for SHM on real-life quarter-scaled ASCE-Benchmark structure using stiffness-based objective function. It is observed that the performances of all the algorithms are smooth except for BSA and FA due to increased randomness and entrapment in local optima. Hence to improve their performances, modification has been introduced using the chaotic maps in the foraging behaviour of BSA and randomized movement of FA. With the proposed chaotic modifications, it is observed that Chaotic BSA and FA shows good accuracy in damage analysis, with 95% of damage results falling well-within the acceptable range.