An algorithmic framework for reconstruction of time-delayed and incomplete binary signals from an energy-lean structural health monitoring system. (1st February 2019)

• Record Type: Journal Article
• Authors: Salehi, Hadi; Das, Saptarshi; Chakrabartty, Shantanu; Biswas, Subir; Burgueño, Rigoberto
• Is Part Of: Engineering structures. Volume 180(2019)
• Page Start: 603
• Page End: 620
• Publication Date: 2019-02-01
• Subjects: Structural health monitoring -- Matrix completion -- Pattern recognition -- Self-powered sensor network -- Time-delayed binary signals; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2018.11.072
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: A new structural health monitoring strategy based on time-delayed binary signals is presented. A machine learning (ML) framework merging matrix completion and pattern recognition is developed. A data fusion model is introduced to pre-process the incomplete signals for the ML framework. A statistical approach is employed to detect damage with time-delayed signals. The approach is evaluated for plates through numerical and experimental tests. Abstract: Recent advances in energy harvesting technologies have led to the development of self-powered structural health monitoring (SHM) techniques that are power-efficient. Energy-aware data transmission protocols, on the other hand, have evolved due to the emergence of self-powered sensing. The pulse switching architecture is among such protocols employing ultrasonic pulses for event reporting through the substrate material. However, the noted protocol raises the necessity for new types of signal/data interpretation methods for SHM purposes. This is because a system using such technology demands dealing with power budgets for sensing and communication of binary signals that leads to unique time delay constraints. This study presents a novel computational approach to reconstruct delayed and incomplete binary signals provided by a through-substrate ultrasonic self-powered sensor network for SHM of plate-like structures. An algorithmic framework incorporating low-rank matrix completion, a data fusion model, and a statistical approach is proposed for damage identification. Performance and effectiveness of the proposed method for the case of dynamically loaded plates was evaluated using finite element simulations and experimental vibration tests. Results demonstrate that the energy-lean damage identification methodology employing the proposed algorithmic framework enables dependable detection of damage using reconstructed time-delayed binary signals.