Volume 12, Issue 4 (8-2022)
IJOCE 2022, 12(4): 517-543 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Mahdavi S H, Azimbeik K. A MODIFIED GENETIC ALGORITHM STRATEGY FOR OPTIMAL SENSOR EXCITER PLACEMENT CAPABLE OF TIME DOMAIN STRUCTURAL IDENTIFICATION. IJOCE 2022; 12 (4) :517-543
URL: http://ijoce.iust.ac.ir/article-1-532-en.html
URL: http://ijoce.iust.ac.ir/article-1-532-en.html
Abstract: (6143 Views)
This paper presents an efficient wavelet-based genetic algorithm strategy for optimal sensorexciter placement (OSPOEP) in large-scaled structures suitable for time-domain structural identification. For this purpose, a wavelet-based scheme is introduced in order to improve the fitness evaluation of GA-based individuals capable of using adaptive wavelets. A search domain reduction (SDR) strategy is proposed to reduce the wide space of initial unknowns corresponding to enormous degrees-of-freedom in large systems. The proposed reduction strategy is carried out at three stages according to the use of different wavelet functions. Furthermore, a multi-species decimal GA coding system is modified for a competent search around the local optima. In this regards, a local operation of mutation is presented in addition with regeneration and reintroduction operators. It is deduced that, the reliable OSPOEP strategy prior to the time-domain identification will be achieved by those procedures dealing with minimizing the distance of simulated responses for the entire system and condensed system considering the excitation effects. The numerical assessment on the appropriateness and capability of the proposed approach demonstrates the substantially high computational performance and fast convergence of the proposed OSPOEP strategy, especially in large-scaled structural systems. It is concluded that, the robustness of the proposed OSPOEP procedure lies on the precise and fast fitness evaluation at larger sampling rates which resulting in the optimum evaluation of the GA-based exploration and exploitation phases towards the global optimum solution.
Keywords: optimal sensor placement, optimal exciter placement, time-domain identification, wavelet analysis, genetic algorithm
Type of Study: Research |
Subject:
Applications
Received: 2022/08/25 | Accepted: 2022/08/19 | Published: 2022/08/19
Received: 2022/08/25 | Accepted: 2022/08/19 | Published: 2022/08/19
| Rights and permissions | |
 | This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
