McC BarsMcCormick School of Engineering HomeMcCormick School of Engineering HomeNorthwestern University Homepage

Infrastructure Technology Institute

Frontiers in ISHM: an interview with
Sridhar Krishnaswamy


Post-doctoral fellow Bradley Regez and Professor Sridhar Krishnaswamy.

Sridhar Krishnaswamy is a professor of mechanical engineering and director of the Center for Quality Engineering and Failure Prevention at Northwestern University. He is also principal investigator for the collaborative NSF-sponsored Partnerships in International Research and Education - Structural Health Management program (PIRE-SHM).

What is the mission of PIRE-ISHM?

The mission of the PIRE-ISHM program is to form substantive collaborative interactions with partners from the US and abroad in the general area of Intelligent Structural Health Management (ISHM). The basic idea underlying ISHM is to instrument safety-critical structures with several diagnostic sensors that can provide timely information that can be used to predict structural reliability going forward. Diagnostics is an essential component for ensuring structural reliability, and this has typically relied mostly on visual inspections. In the future, most structures will be instrumented with several sensors that can report on the condition of the structures. A second very important component of ISHM is the ability to use the information from the sensors to assess whether the structure is currently operationally safe. ISHM therefore requires a cross-disciplinary approach bringing together several emerging and some mature sub-fields of science and engineering, including smart structures and materials, structural health monitoring and nondestructive evaluation, damage and failure mechanics, materials science, and probabilistic reliability analysis.

The PIRE-ISHM program is funded by the National Science Foundation for a five-year period. It is led by Northwestern University and it involves the University of Illinois-Chicago as well as universities from China, India, Korea and Hong Kong. Honeywell, GE, and Boeing are among the industry partners.

What can we expect next from intelligent sensors?

For ISHM, we need sensors that can provide information about the structural, environmental and material states. Structural and environmental sensors such as strain gauges, load cells, temperature gauges, accelerometers, tilt gauges, etc are fairly well developed. We can expect continued performance improvements in these types of sensors, and, more importantly, we are also likely to see a definite move towards wireless sensors as they are much easier to install than bulky wired sensors. Wireless sensing would also require self-powering through energy-harvesting so that we are not faced with having to change sensor batteries often. Some of these types of wireless sensors are beginning to enter the market, but there is still a lot of work to be done in this area.

Going forward, we would also like to see new sensors developed that can provide information about the material state and not just the structural state. Sensors that can provide early indication of fatigue damage, stiffness degradation, incipient corrosion, etc., would all be very valuable diagnostic tools.

Is there a crisis in infrastructure?

The large portfolio of aging infrastructure that we have in the US will obviously have to be fixed or replaced over the next couple of decades. While this is a big economic challenge, it also presents a great opportunity to upgrade the infrastructure so that they can be more efficiently managed and maintained in the future. Manual inspection is always going to be more expensive and perhaps more prone to error. On the other hand, structural health monitoring using remote sensing might involve added initial costs during installation, but the case for reduced total cost of ownership using the ISHM approach is very strong.

To give a parallel example, the electrical meters in the town where I live are read by a person from the electric company having to trudge through rain or snow from house to house. Since this is time consuming and expensive to do, the electric company often chooses to read the meters only once every couple of months or so, and they estimate the usage for the months that they do not have an actual meter reading. By contrast, our water meter can be remotely accessed by the city through a wireless radio connection, with significant cost savings for the city. It is easy to imagine that the wireless water meter sensor could also be used to automatically detect water leaks, etc., which is not currently done.

Current infrastructure inspection process is more like the electric company’s approach in that most of the time we estimate the structural state based on relatively few manual inspections. This is both inefficient and prone to error. Structural health monitoring using remote sensors can provide information as often as necessary at a fraction of the cost of manual inspection. The resulting savings from the diagnostics process can be utilized towards better prognostics and rehabilitation.

 


Thermographic imaging apparatus used to detect structural deficiencies in concrete


Have there been prominent failures where structural health monitoring could have made a difference?

In the book ‘Outliers’ by Malcolm Gladwell, there is a chapter on the root causes of airplane crashes. To paraphrase Gladwell’s point, catastrophic failures (especially of systems with built-in redundancy for safety) are often the result of cascading string of errors. The I-35W bridge collapse, I think, is a classic case of that. My understanding is that improper design, combined with unexpected overload, and structural aging (fatigue cracking) were all contributing factors. In hindsight, it is conceivable that if we had sensors to monitor overload and aging, perhaps the catastrophic collapse could have been avoided.

What is the international component of your program?

The PIRE part of the PIRE-ISHM acronym stands for ‘Partnerships for International Research and Education.’ Our partners currently are from China, India, Korea and Hong Kong. As part of the PIRE-ISHM experience, our graduate and undergraduate students go on extended research visits to the partner labs. The graduate students, especially, have continued collaborations with their international counterparts extending through the duration of their studies.

The international component of the program serves several purposes. First, there is a lot of ISHM activity in our partner countries, which is not surprising considering that there is a lot of infrastructure construction going on there as well. There are several bridges, off-shore platforms, television towers, etc., that have been extensively instrumented in these countries which can serve as test beds for new ISHM systems that we develop in this program, and we can learn a lot from our partners in this regard. As an example, one of our PIRE-ISHM faculty affiliates, Professor Pablo Durango-Cohen is currently in discussions with our Hong Kong Polytechnic University partner regarding the possibility of mining a twelve-year long database from the Tsing-Ma bridge in Hong Kong in order to test some of his prognostics algorithms.

A larger purpose of the international component is to provide our students an opportunity to learn how to successfully engage in complex technological endeavors with global partners. No matter what technical or management career paths they choose in the U.S., our students will most certainly have to deal with foreign clients, customers, colleagues or students, and the PIRE-ISHM program gives them an early glimpse of the emerging global work environment. It is also my hope that at least some of the PIRE-ISHM students will see that they do not have to let geography limit their career options. Opportunities for research careers exist not only at GE Niskayuna in New York, but also at GE Global Research in Bangalore or Shanghai. Or, perhaps one or two of our students might team up with their counterparts in India or China to create startup companies that can leverage the strengths of the U.S. and the international partner countries.

Are new infrastructure materials such as composites easier or more difficult to monitor?

New materials almost always increase the challenge for ISHM. Obviously, the use of new materials implies that there is some gain in terms of function or cost. However, we will need to understand their failure mechanisms and long term degradation behavior, and we will need to identify or develop appropriate inspection tools for monitoring their condition. This keeps those of us involved in ISHM research in business!

Should we deploy structural health monitoring technology more broadly to the public infrastructure?

Ultimately, we would want to deploy ISHM technology across the board for enhanced safety and potentially lower total cost of ownership. There are of course several barriers to overcome, both technical and economic. On the technical side, standardization and codification of ISHM approaches will be required. Standardization can be readily achieved for structures such as airplanes that are built to common specifications in a factory, but it is quite hard to enforce for civil infrastructure, where every structure is somewhat unique and is built on site. Organizations such as the American Society of Civil Engineers (ASCE) will obviously have to codify emerging ISHM methodologies.

On the economic side, the barriers for adoption of ISHM primarily relate to the diverging interests of the builder, operator, owner and users of the infrastructure. Aligning their interests over the long term so that everybody has a stake in the total cost of ownership of the infrastructure will be a key step in overcoming economic barriers. In the aerospace industry, for example, Boeing not only builds aircraft, but also offers maintenance services for a fixed period that the airlines can purchase. This provides an incentive to the builder to build in efficient diagnostic tools. In the automobile industry, competitive pressure leading to extended manufacturers’ warranties as well as regulations for safety and emissions control have resulted in computerized diagnostic systems that can automatically flag incipient maintenance issues. Perhaps infrastructure owners should look into models such as these to provide incentives to all the stakeholders to take into consideration the total cost of ownership in the design, construction and operational phases of infrastructure.