Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Computational Science, Engineering & Technology Series
PARALLEL, DISTRIBUTED AND GRID COMPUTING FOR ENGINEERING
Edited by: B.H.V. Topping, P. Iványi
Integrating Sensory Data within a Structural Analysis Grid
A.I. Khan and A.H. Muhamad Amin
Clayton School of Information Technology, Monash University, Clayton, Australia
A.I. Khan, A.H. Muhamad Amin, "Integrating Sensory Data within a Structural Analysis Grid", in B.H.V. Topping, P. Iványi, (Editors), "Parallel, Distributed and Grid Computing for Engineering", Saxe-Coburg Publications, Stirlingshire, UK, Chapter 18, pp 389-412, 2009. doi:10.4203/csets.21.18
Keywords: structural health monitoring, wireless sensor network, pattern recognition, computational grid, structural engineering.
Highly-invested and sophisticated structures such as aerospace vehicles, offshore installations, maritime vessels, and critical infrastructure frequently require up-to-date information regarding their structural status. Maintenance costs for such structures are usually high and involve major technical operations to be conducted. In this paper, we propose a novel holistic integrated grid-sensor network framework for structural engineering lifecycle that incorporates an end-to-end structural analysis, design, and monitoring. This framework occupies the integration between wireless sensor network (WSN) technology and computational grid for rapid structural data acquisition and processing.
In this framework, the sensory data collected from real-time monitoring of structures can be used for less conservative analysis and design, while the information generated during analysis can be used towards real time interpretation of sensory data. The integrated grid-sensor network framework is a combination of commodity-grid based system, for large-scale structural data analysis, and sensor networks. The proposed framework utilises an associative memory algorithm known as Graph Neuron (GN) [1-5], which is an in-network processing scheme for resource constrained sensor networks, for real time monitoring. The grid incorporates a hierarchical version of the Graph Neuron known as Distributed Hierarchical Graph Neuron (DHGN) [6-9] to rapidly correlate sensory patterns of interest with all available analytical data. Within this framework, data generated through the analysis would flow through the grid and into the sensor network, thereby allowing intelligent filtering of sensory outputs in real time. The filtered event pattern data from this process would then flow into the grid, where the grid-enabled pattern recognising scheme would check for repeated and critical patterns by comparing the sensory outputs with the available analytical data. Many of the iterative processes within the analysis, design, and monitoring stages can be automated by defining work flows within the framework. The resultant grid-sensor network framework would enable the use of sensory data within structural design and it would also improve sensory data interpretation.
purchase the full-text of this chapter (price £20)