Adaptive sensors and sensor networks for structural health monitoring

Conflicting levels of spatial and temporal scales often hamper using sensor systems to monitor the health of large structures. Some structures, such as dams, bridges and pipelines can be huge, with spans often measured in kilometers. These structures also have lifetimes that can be measured in terms of decades and occasionally even centuries. However, damage to the structure is often localized both spatially and temporally. Cracks are very local events. The critical loading on the structure and/or the occurrence of critical damage may occur on time scales that are very short compared to the lifetime of the structure. Detecting and determining the extent of damage in a structure under these circumstances is often difficult. It is usually uneconomical to cover a large structure with a dense array of sensors that sample at high speed continuously. One possible solution is to have the sensor system be adaptable to changes in the structural health and to key events. This paper will discuss several strategies that can be used in adaptive structural sensing systems. One approach is to use an array of localized data processors with sophisticated trigger and data preprocessing algorithms that only send pertinent data to a central data logger/processor. Another approach is to use imaging systems, such as visible light images or those obtained from ground penetrating radar, to identify potential damage sites that require closer inspection, or squinting, of the imaging system. These could be coupled with a robotic inspection system that changes its inspection route based on the condition of the structure, or the occurrence of a possible damage-causing event, such as an earthquake.