Paolo Bendiscioli

MEMS-based surface mounted health monitoring system for composite laminates

Abstract Composite structures subjected to extreme loadings like, e.g., impacts, can undergo a reduction of their stiffness and strength properties, due to the nucleation and subsequent propagation of cracks along the interfaces between different phase materials. This phenomenon turns out to be difficult to detect through cheap monitoring procedures. Here, we discuss a methodology to monitor the state of crack-containing composite laminates through low-cost, commercial off-the-shelf MEMS accelerometers. By adopting cyclic loading conditions, we track the evolution of the cracked, or debonded region in a double cantilever beam; this is achieved by surface mounting a MEMS (along with its board) and by monitoring the drift in the compliance of the specimen induced by crack growth. The methodology is validated through an analytical model of the experimental test, which highlights the sensitivity of the monitoring scheme to the crack length.

Investigation of the Effectiveness and Robustness of an MEMS-Based Structural Health Monitoring System for Composite Laminates

Failure of layered composites can be triggered by small inner defects; such defects can be present inside the composite from the beginning, or can be incepted by external actions during the life cycle of the structure. Defects generally evolve, under repeated loadings, into a delamination or debonding between adjacent laminae, and cannot be easily detected as they are masked by the composite skin: ad hoc health monitoring systems are therefore required. To prevent any distortion of the composite microstructure, which may trigger by itself the nucleation of the aforementioned defects, sensors should not be embedded inside the composite; in addition, to avoid a mechanical interaction between the monitoring system and the structure, sensors should be as lighter as possible. To comply with these two major requirements, we have recently investigated a microelectromechanical system-based, surface-mounted health monitoring strategy for laminates. In this paper, we provide the results of an experimental campaign to show the effectiveness of this strategy, in terms of sensitivity to the length of delamination in standard specimens (independently of the loading conditions at the tip of the delamination), and its robustness, in terms of repeatability of the outcomes relevant to the monitored state. The experimental data are compared with analytical predictions based on beam bending theory, showing a good accuracy.

Sensor deployment over damage-containing plates: A topology optimization approach

Health monitoring systems for composite laminates should detect in real-time if damage, typically consisting of delamination (i.e. debonding between adjacent laminae), is incepted or growing under the external actions. To avoid detrimental effects on the overall structural strength induced by embedded sensors, we recently proposed a surface-mounted monitoring scheme adopting microelectromechanical systems (MEMS) devices. We already investigated the capability of this methodology to detect the inception/growth of delamination in a standard test geometry; in this article, we instead focus on the optimization of sensor deployment over flexible plates, so as to maximize the sensitivity of the monitoring system to damage/delamination. This is achieved through a topology optimization approach. For a square thin plate, either supported or clamped along its boundary, we present optimal distributions of MEMS sensors to detect a damage of known or unknown position. We show that, no matter what the location, size, and shape of the damaged area are, a trivial solution consisting of an array of evenly spaced sensors does not represent the optimal one to monitor the structural health.

Smart sensing of damage in flexible plates through MEMS

In this work, we present a methodology to monitor the state of damage- or crack-containing flexible plates through low-cost, commercial off-the-shelf MEMS accelerometers. Under quasi-static, time-varying loading conditions we track the evolution of the damaged zone (e.g., the length of an inner crack) accounting for the drift in the compliance of the specimens, measured by means of the magnitude of the load-induced rotation at MEMS position. We then validate this health monitoring methodology, showing that it turns out to be sensitive to the damage and provides results in good accordance with theoretical findings. Next, we propose a technique to optimally deploy the MEMS sensors over the plate. Referring to an isotropic square plate containing damaged zones of reduced bending stiffness, we numerically investigate the sensitivity of the load-induced state (in terms of out-of-plane displacement and rotation of the normal to the mid-plane) to the position of the damaged area, and we adopt a constrained topol...