A Bio-inspired Framework for Highly Efficient Structural Health Monitoring and Vibration Analysis

Abstract

Civil engineering structures are continuously exposed to the risk of damage whether due to ageing effects, excessive live loads or extreme events, such as earthquakes, blasts and cyclones. If not readily identified, damage will inevitably compromise the structural integrity, leading the system to stop operating and undergo in-depth interventions. The economic and social impacts associated with such an adverse condition can be significant, therefore effective methods able to early identify structural vulnerabilities are needed for these systems to keep meeting the required life-safety standards and avoid the impairment of their normal function. In this context, vibration-based analysis approaches play a leading role as they allow to detect structural faults which lie beneath the surface of the structure by identifying and quantifying anomalous changes in the system’s inherent vibration characteristics. However, although the considerable degree of maturity attained within the fields of experimental vibration analysis (EVA) and structural health monitoring (SHM), several technical issues still need to be addressed in order to ensure the successful implementation of these powerful tools for damage identification purposes. The scope of this paper is to present a bio-inspired framework for optimal structural health monitoring and vibration analysis. After a critical overview on current methods and tools, three main sources of bio-inspiration are described along with the relative algorithms derived for SHM applications. It is shown how uncovering the general principles behind the functioning of selected biological systems can foster the development of efficient solutions to the technical conflicts of actual SHM architectures and lead to new sensing paradigms for optimal network topology and sensors location. A compatibility-matrix is proposed to help compare biological and SHM systems and discriminate desired from unwanted features. Such a framework will ultimately assist in seeking for the most suitable nature-inspired solutions for more accurate condition screening and robust vibration analysis.