Autonomous Structural Health Monitoring Systems Research Articles

Development of data anomaly classification for structural health monitoring based on iterative trimmed loss minimization and human-in-the-loop learning

Huge amounts of data can be generated during long-term monitoring performed by structural health monitoring (SHM) and structural integrity management applications. Monitoring data can be corrupted, and the presence of abnormal data can distort information during signal processing, extract incorrect characteristics during system identification, produce false conclusions during damage detection, and ultimately lead to misjudgment of structural conditions during diagnosis and prognosis. Therefore, developing effective techniques to autonomously detect and classify anomalies becomes necessary and significant. Generally, conventional physics-based strategies can be straightforward, but their performance highly depends on prior knowledge of measurement. Recently, data-driven methods leveraging machine learning (ML) have been used to directly handle the task. This study proposes an ML-based classifier and improves it by incorporating the human-in-the-loop (HITL) learning. The classifier is built on a shallow neural network with high performance to address potential online or real-time applications for long-term monitoring. First, a field monitoring dataset is introduced, and various anomalies are defined to investigate the effectiveness. To further enhance the performance of the proposed classifier, the mislabels in the monitoring dataset are examined, and a correction technique is performed. Additionally, HITL ML is developed to overcome the disadvantages of the conventional correction technique. As a result, the proposed procedure can improve both the classifier and the field dataset, and the proposed classifier can now function as a fundamental component in achieving a continuous and autonomous SHM system.

Vibration energy harvesting of multifunctional carbon fibre composite laminate structures

A sustainable power supply for a wide range of applications, such as powering sensors for structural health monitoring and wireless sensoring nodes for data transmission and communication used in unmanned air vehicles, automobiles, renewable energy sectors, and smart city technologies, is targeted. This paper presents an experimental and numerical study that describes an innovative technique to harvest energy resulted from environmental vibrations. A piezoelectric energy harvester was integrated onto a carbon fibre reinforced polymer (CFRP) laminate structure using the co-curing method. The integrated composite with the energy harvester was lightweight, flexible and provided robust and reliable energy outcomes, which can be used to power different low-powered wireless sensing nodes. A normalised power density of 97 μW cm−3m−2s4 was obtained from resonance frequency of 46 Hz sinusoidal waves at amplitude of 0.2 g; while the representative environmental vibration waves in various applications (aerospace, automotive, machine and bridge infrastructure) were experimentally and numerically investigated to find out the energy that can be harvested by such a multifunctional composite structure. The results showed the energy harvested at different vibration input from various industrial sectors could be sufficient to power an autonomous structural health monitoring system and wireless communications by the designed composite structure.

Energy Management Systems for Energy Harvesting in Structural Health Monitoring Applications

Autonomous structural health monitoring systems with independent power sources and wireless sensor nodes are increasingly seen as the best solution for monitoring a diverse range of machines and structures including pumps, bridges and aircraft. Powering these systems using harvested energy from ambient sources provides an attractive alternative to the use of batteries which may be either inaccessible for routine maintenance or unsuitable (for example in aerospace applications). A number of techniques are currently being considered including harvesting energy from vibration and thermal gradients. Harvesting energy can however lead to a highly variable power supply in opposition to the requirements of a wireless sensor node which requires continuous standby power with an additional capacity for power peaks during transmission of data. A power management system with embedded energy storage is therefore necessary in order to match supply and demand. Due to the low levels of power harvested in a number of applications, an important factor in the design of such a system is its efficiency to ensure sufficient power reaches the sensor node. Based on the requirements for a simple power management system for thermoelectric power harvesting consisting of a rectifier, a DC/DC convertors and a battery, this paper first examines the possibilities in terms of basic components with a number of commercially available units tested and characterised. Potential designs for a management system incorporating these components are then discussed and a blueprint for an optimal system is suggested.

Structural Health Monitoring of Bridges Using Wireless Sensor Networks

Aging and degradation of transportation infrastructure pose significant safety concerns, especially in light of increased use of these structures. The economic downturn further exacerbates such concerns, especially for critical structures such as bridges, where replacement is infeasible and maintenance and repair are expensive. The US Federal Highway Administration has classified over 25% of the bridges in the United States as either structurally deficient or functionally obsolete, underscoring the importance of structural health monitoring (SHM) to ensure public safety. We give an overview of emerging wireless sensor networks (WSN) for autonomous SHM systems, their application, the power use and sources needed to support autonomy, and the type of communication that allows remote monitoring.

A feasibility study of damage tracking through the diffusive communication of wireless sensors

The objective of this paper is to investigate the feasibility of wireless sensors in the development of an autonomous structural health monitoring system. A collaborative searching algorithm is developed such that massively deployed wireless sensor nodes in a structure conveniently comprise a group and constitute a damage-surveillance perimeter. Wireless sensors in this perimeter spontaneously activate themselves for damage-tracking tasks by networking with neighboring sensors. When the damage-sensitive parameter that is measured by a sensor node exceeds a certain threshold, the process of damage-tracking begins. The proposed damage-tracking algorithm does not require any type of global control. Instead, sensor-networking and a pairwise-comparison algorithm that is implemented at each sensor node allows collaborative decision-making for tracking the changes, such as local strain, in structural properties. The extant autonomous, damage-tracking algorithms have been demonstrated through only numerical simulations for a single-damage case. Here, the study is further expanded to address the problem of simultaneously tracking multiple instances of damage in three-dimensional space by using improved algorithms for sensor networking. An event-based task-executing functionality of individual sensor nodes is successfully implemented and verified using four wireless strain sensors that are mounted on a cantilevered beam structure. Experimental results reveal that the overall capability of wireless sensor nodes is functional enough to enable a wireless-based autonomous structural health monitoring system.