Options
A computational steering approach towards sensor placement optimization for structural health monitoring using multi-agent technology and evolutionary algorithms
Publikationstyp
Conference Paper
Date Issued
2007-09
Sprache
English
Author(s)
Volume
1
Start Page
877
End Page
886
Citation
Structural Health Monitoring 2007: Quantification, Validation, and Implementation - Proceedings of the 6th International Workshop on Structural Health Monitoring, IWSHM 2007 1: 877-886 (2007-01-01)
Contribution to Conference
Scopus ID
Publisher
DEStech Publ.
This paper introduces a hybrid approach towards sensor placement optimization combining both Multi-Agent Technology and Evolutionary Algorithms (EA). Multi-Agent Technology, on the one hand, allows for an autonomous solving even of heterogeneous and distributed problems by applying interacting and cooperating software agents as specialized "problem solving entities". Evolutionary Algorithms comprise on the other hand various nature-inspired optimization technologies such as Genetic Algorithms (GA), Evolution Strategies (ES), or Memetic Algorithms (MA). Consequently, effective optimization strategies for complex and nonlinear as well as multimodal problems are provided, based on principles of biological evolution such as reproduction, mutation, recombination, natural selection and survival of the fittest. In order to manage the complexity of holistic sensor placement optimization strategies appropriately, a simulation and experimentation platform, called JASOE (Java Agent Simulation and Optimization Environment), has been developed at the Institute for Computational Engineering (ICE). As a Computational Steering application, JASOE applies Multi-Agent Technology as well as Evolutionary Algorithms for an interactive steering, modeling and visualizing of distributed function and structural optimization simulations. It provides in particular pre-and post-processing tools for managing the infrastructure of sensor placement optimizations, presented in this paper. The capability of the developed application is finally demonstrated by the example of a selected structure and by a distributed EA-based simulation model.
DDC Class
004: Informatik
600: Technik