A Wireless Enabled IOT Nondestructive Fatigue Damage Sensor for Estimation of Remaining Useful Service (RUL) of Structures Through Artificial Intelligence-Machine Learning Algorithms

Abstract

In this study, a wireless enabled novel non-destructive evaluation (NDE) fatigue damage sensor for the structural fatigue health monitoring and prediction of fatigue life of structural members of large engineering structures is presented. The smart predictive IOT (Internet of Things) fatigue damage sensor has special designed smart breakable sacrificial sensor beams for early fatigue damage detection and prediction of fatigue sensitive structural and mechanical parts of fatigue sensitive structures. The IOT non-destructive fatigue damage sensor can measure directly the state of fatigue damage accumulation levels and predict the remaining useful life (RUL) of structural components or locations for the life cycle management and predictive maintenance of structures without using any fatigue damage algorithm. The non-destructive evaluation (NDE) fatigue sensor consists of mechanical and electronic units. The mechanical part of the fatigue damage sensor has special designed smart and predictive sacrificial beams with different levels of lifetimes (10% N, 20% N, 30% N, 40% N, 50% N, 60% N, 70% N, 80% N, 90% N, 95-100% N) normalized to the total lifetime of a real structure. The mechanical sensor beams designed with special geometry using stress magnifying effects to measure the fatigue damage accumulating level and early fatigue damage detection and fatigue state health monitoring of structures. The smart fatigue damage sensors attached onto the surface on a real structure is expected to have the same cyclic stress-strain loading behaviour and the same cyclic stress-strain history for during the operational service life. The sensor having multiple parallel oriented and mini-micro non-destructive fatigue damage measurement sensing beams are designed to fail earlier than a real structure in the different predetermined fatigue lifetimes like a mechanical fatigue fusing system. Due to its predictive feature the fatigue sensor will extent not only the service life of fatigue sensitive mechanical components but also increase the safety and reliability of structures. Since the wireless enabled fatigue damage sensor network through Internet of Things (IOT) technologies collect the real operational fatigue data remotely, the statistical fatigue damage sensor network data can be used for the prediction of lifetimes of structural or mechanical members by using Artificial Intelligence (AI) Machine Learning (ML) Algorithms. The distributed fatigue sensor network also provides a real operational-experimental statistical data for condition based predictive maintenance, maintenance management, end of service life indicator (ESLI) and development of new fatigue design tools for fatigue sensitive parts or locations of fatigue critic metallic and composite structures.