Structural damage identification in thin-shell structures using a new technique combining finite element model updating and improved Cuckoo search algorithm. (November 2022)

• Record Type: Journal Article
• Authors: Minh, Hoang-Le; Sang-To, Thanh; Abdel Wahab, Magd; Cuong-Le, Thanh
• Is Part Of: Advances in engineering software. Volume 173(2022)
• Publication Date: 2022-11
• Subjects: Structural damage identification -- Damaged thin-shell structures -- Cuckoo search algorithm, finite element model updating -- SAP2000-OAPI; Computer-aided engineering -- Periodicals; Engineering -- Computer programs -- Periodicals; Engineering -- Software -- Periodicals; Periodicals; 620.0028553
• DOI: 10.1016/j.advengsoft.2022.103206
• Languages: English
• ISSNs: 0965-9978

Abstract

Highlights: An efficient algorithm named new balance of Cuckoo search algorithm is proposed. The use of the Open Application Programming Interface (OAPI) in SAP2000 software. A new finite element model updating technique is proposed based on SAP2000-MATLAB. Three thin-shell structures having complex geometry are used for the purpose of SDI. Abstract: In this paper, the powerful advantages of Open Application Programming Interface (OAPI) in SAP2000 software, known as an excellent tool to solve structural engineering problems, are exploited to establish a new technique of finite element (FE) model updating. This technique is based on the development of a successful sub-program, which can link programming language of MATLAB with SAP2000 through the OAPI library. Therefore, the two-way data exchange between MATLAB and SAP2000 can be continuously secured, and the process of modification of the initial FE model in SAP2000 is implemented by coding instead of using the standard point-and-click procedure in SAP2000′s interface. The damage identification process in structures is then done using an optimization algorithm, named a modified version Cuckoo search (CS) algorithm, and called new balance of Cuckoo search (NB-CS). The key factor in NB-CS is to create a balance between two values: the best global and the worst global. To achieve this balance, NB-CS proposes two main contributions. The first is based on the modification of the concept of a random walk whose step lengths obey Lévy distribution. Whereas, the second is to establish a new balance vector, which is presented based on the global optimal and the worse optimal exploited at each iteration. This vector is combined with each step length to establish a new movement strategy, which is more flexible and more intelligent than that of the original CS. To examine NB-CS's reliability, the first 23 classical benchmark functions are selected to illustrate the convergence rate and level of accuracy of NB-CS compared to the original CS. To authenticate the efficiency of the proposed method, three models, including square shell structure, parabolic domes, and partial sphere shell structure, are investigated for different damage cases. NB-CS is then employed to minimize the objective function with design variables related to the extent and location of damage in thin-shell elements in the structures. The statistical results show the feasible performance of NB-CS in a widespread class of optimization problems and its great potential for application to structural damage identification.