EFFICIENT WIRELESS SENSOR NETWORK (WSN) IMPLEMENTATION FOR STRUCTURAL HEALTH MONITORING (SHM) APPLICATIONS

Abstract

Sensing and analyzing the structures to detect and evaluate damage and degradation for characterization of structure’s health is referred to as Structural Health Monitoring(SHM). The strategic civil infrastructure is ageing worldwide, the ageing phenomenon is more pronounced in developing world due to non-availability of effective SHM techniques. The existing structures especially bridges were constructed decades ago and require effectual modes of SHM for dependable health characterization. The efficient evaluation of structures and subsequent implementation of appropriate maintenance actions result in considerable cost saving to agencies that manage civil infrastructures. In this doctoral study, a real-time vibrational analysis based SHM system grounded on low power Wireless Sensor Network (WSN) using advanced signal processing techniques is proposed for diagnostic as well as prognostic purposes. Tiny MEMS based accelerometers offering high accuracy rate duly integrated with energy-efficient WSN nodes are used for sensing the live data from an ageing pre-stressed bridge that has been in service for 25 years, located in the harbor area. Discriminatory features from in-situ acquired actual data are extracted. Promising results acquired through application of diagnostic and prognostic algorithms on recorded data using highly efficient WSN based SHM system showcase the novel contribution and usefulness of Abstract vi proposed research work. This thesis has addressed the major issues related to degradation quantification of in-service civil structures and has also provided detail solution to these problems through implementation of effective diagnostic and prognostic techniques. The subject area will surely benefit the readers in SHM community. The methods detailed in these thesis for data acquisition and adaptive signal processing and the proposed prognostic scheme can be used for effective health monitoring of other in-service civil infrastructures. The novelty of proposed work include a real-time analysis of actual SHM signals with respect to loading with very less communication payload. The improvement in terms of power by the design of highly power efficient SHM system through implementation of star topology of WSN and use of devised S-MAC mac layer protocol for periodic sleep. The use of Hilbert Huang Transform (HHT) for showing the efficacy of HHT technique in SHM and its real time deployment on a live civil structure with prominent results for discriminating with health and degraded civil infrastructures. The prognosis method also contains the novelty to get trained as evolve by 1-step, 2- step and 3-step prediction mechanisms. The modal energy of the vibrational modes for quantification of bridge health for Remaining Useful Life (RUL) factor prediction that leads towards calculation of residual life is also a novel quantification factor used for the very first time in this dissertation.