‘Sensing skin’ could monitor the health of concrete infrastructure continually and inexpensively
June 29, 2011
By Denise Brehm
Civil & Environmental Engineering
In 2009, the American Society of Civil Engineers (ASCE) assigned a grade of “D” to the overall quality of infrastructure in the United States, saying that ongoing evaluation and maintenance of structures was necessary to improve that grade. Since then, federal stimulus funds have made it possible for communities to repair some infrastructure, but high-tech, affordable methods for continual monitoring remain in their infancy. Instead, most evaluation of bridges, dams, schools and other structures is still done by visual inspection, which is slow, expensive, cumbersome and in some cases, dangerous.
Civil engineers at MIT, working with physicists at the University of Potsdam in Germany, recently proposed a new method for continual electronic monitoring of structures. In papers appearing in Structural Control Health Monitoring (December 2010) and the Journal of Materials Chemistry (April 2011), the researchers describe a flexible fabric with electrical properties that could adhere to areas prone to cracking — such as the undersides of bridges — and detect cracks almost immediately when they occur.
Installing this “sensing skin” would be as simple as unrolling it and gluing it to the surface of a structure. The rectangular patches on the skin that detect changes in its electrical charge could be tailored in a geometric design appropriate for the type of crack likeliest to form in a particular part of a structure: for example, diagonal square patches to detect cracks caused by shear, or horizontal patches to detect the cracks caused by a sagging horizontal beam.
The formation of a crack would cause a tiny movement in the concrete under the patch, changing the capacitance, or stored energy, of the sensing skin. Once a day, a computer system attached to the sensing skin would send a current to measure the capacitance of each patch and detect any difference among neighboring patches. In this way, it could detect a flaw and its exact location within 24 hours — a task that has proved difficult for other types of sensors proposed or already in use, which tend to rely on detecting global changes in the entire structure using a few strategically placed sensors.
“The sensing skin has the remarkable advantage of being able to both sense a change in the general performance of the structure and also know the damage location at a pre-defined level of precision,” said Simon Laflamme Ph.D. ’11, who did this research as a graduate student in MIT CEE. “Such automation in the health-monitoring process could result in great cost savings and more sustainable infrastructures.” Laflamme worked with CEE Professor Jerome Connor and University of Potsdam researchers Guggi Kofod and Matthias Kollosche.
The researchers originally tested their idea using a commercially available, inexpensive stretchy silicon fabric with silver electrodes. While this worked in some lab experiments performed on both small and large concrete beams under stress, the material ultimately proved too thin and flexible for this application. The researchers have now developed a prototype sensing skin made of soft stretchy thermoplastic elastomer mixed with titanium dioxide that is highly sensitive to cracks; painted patches of black carbon measure changes in the electrical charge of the skin. A patent for the sensing method has been filed.
“The innovation of this proposed sensor design is in its use of a material that provides mechanical flexibility and serves as a capacitor,” said Professor Tzu-yang Yu of the University of Massachusetts at Lowell, a structural engineer who specializes in the mechanical analysis of structures and the design of nondestructive methods for testing infrastructure. “This design allows the sensor to overcome the difficulties associated with conventional piezoelectric sensors which have strict contact conditions between the sensor and the structure’s surface. The proposed sensor is also superior to conventional fiber-optic sensors in the way that two-dimensional readings can be collected from one sensor.