Structural Health Monitoring: A Technologically Preparedness Solution for Infrastructure Health Management

By Ajoy Kumar Das

Living in the era where daily life is driven by fast evolving technologies, it is often hard to believe that civil and architectural infrastructures are the longest living creations by the human society. Although they are intended for certain design lives, they often exist longer. And today’s new constructions become tomorrow’s heritage structures. They need to be safe and reliable.

Structural failures cost billions of dollars every year. The economic burden of infrastructure maintenance, upgrading, and retrofitting increases daily. Industrially advanced and civilized nations including the United States have been very concerned looking at the economic consequences the nations will face if an appropriate infrastructure maintenance strategy is not developed ahead of time.

Globally, an estimated 2 percent of gross domestic product (GDP), which is around U.S. $960 billion, is spent annually on infrastructure investment and maintenance.  The  American  Society  of  Civil  Engineers  (ASCE) publishes report cards regularly on the health status of U.S. infrastructure. In its 2017 report, the ASCE wrote that the nation’s infrastructure averages a grade of “D.” This means that infrastructure conditions are mostly below standard. Most of them exhibit significant deterioration with a high risk of failure. According to an estimate, there is a total infrastructure gap of nearly $1.5 trillion needed by 2025. Looking at the infrastructure health statistics, out of about 600,000 bridges in the U.S., over 71,200 bridges are structurally deficient; 78,500 bridges are functionally obsolete. In the oil & gas industry, there are over 9,000 offshore platforms in the sea water worldwide. Most of them are built conforming to old design standards. They do not meet the standards for new design and other safety requirements.

It is practically impossible to replace all infrastructure that is still operational beyond design life. One line of approach is to extend the life of infrastructures by locating defects and their severity at the local element level and taking the appropriate actions without exposing public to out-of-ordinary risk. The engineering profession is working tirelessly on developing technologically prepared solutions for infrastructure health management. These procedures are commonly referred as Monitoring (SHM).

To conduct SHM, we need high precision smart sensors for measuring physical responses such as vibration data (displacement, velocity, and acceleration) at key locations determined by an SHM specialist based on the geometric and framing configuration of the structure. Measured vibration data is collected by high sampling-rate data acquisition system (DAS) and stored inside a computer’s memory. The final part is to process the vibration data using advanced processing and signal analysis software. Data analytics initially determine if a structure is defective or free of defects by processing just few of the measured vibration responses. If defect signature is present in the responses, then all the measured responses are post-processed in conjunction with the structural mathematical model to pinpoint defect locations, their numbers, severities, and overall structural integrity.

Looking at the technical aspect, SHM is an inverse process which requires multidisciplinary expertise. Structural design on the contrary is a forward process. However the fundamental axiom behind SHM is very simple to understand – structural dynamic responses, strains, and other physical parameters in the presence of defects shift from their baselines. The defect signature is detectable from the changes in the measurements. The philosophy is very similar to practice in medical field. To monitor the health of living people, a physician first looks at some basic parameters, such as body temperature, heartbeat, blood pressure, height, weight, and any changes in color or complexion. Then depending on the symptoms, additional lab tests are requested that can best determine the reason for physical anomaly and decide the health management plan including medication. Similarly, for structures, it is important to examine the responses first to detect the presence defects and then perform prognosis of the defects.

Many of the technologically advanced procedures, specifically based on Kalman filters and its extended versions, use detailed mathematical models of the structure. The benefit of mathematical models is that a much deeper level of health assessment. I have been involved in developing an extension of the basic Kalman filter approach, called Extended Kalman Filter with Unknown Input (EKF-UI), which is combined with Iterative Least-squares (ILS) to calculate the stiffness parameter of each and every element in a three dimensional (3D) structure. An Advanced Digital Integration Technique (ADIT) is developed and integrated with this procedure to post-process the measured data before being analyzed by ILS-EKF-UI. The stiffness parameters are the true indicator of the health of structural elements, such as beams, columns, bracings, foundations, etc. A reduction in stiffness parameter indicates the presence of defect.

Interesting to note that Kalman filter is a predictive- corrective data analysis technology and was initially developed in 1958 during space mission programs. It guided the Apollo 11 lunar module to the Moon’s surface. It now has applications in missile tracking, inertial guidance systems, GPS, and others

The most appealing feature in SHM is the integration of multi-sensor network to enable real-time analysis, assessment, and control of damages before failure happens. The detection, localization, and quantification of damages are the three essential parameters that guide the comprehensive maintenance decisions made by the engineering profession.

During the past decade, there has been significant progress on the development of integrated SHM strategies that brings multidisciplinary engineering expertise together. This helps the profession in the transition from the traditional visual inspection approach to the quantitative data-based asset maintenance strategy. Such an integrated SHM approach benefits a variety of industries, such as civil infrastructure (bridges, buildings, roads, etc.), renewable energy (such wind turbines), automotive, aerospace, oil & gas (offshore platforms, pipelines), magnets and superconductors, telescope support structures, semiconductor and biotechnology industries, industrial equipment (rotating machinery, bearings, etc.), and nuclear energy.

Practicing SHM in industrial settings has also caused significant growth as part of forensic engineering. Some industries have a developed consensus on designing structural performance monitoring plan correctly during the design stage. Nowadays, it is feasible to construct smart infrastructures with minimal maintenance cost and reduced insurance premiums through the collaborative efforts among asset owners, architects, engineers, structural monitoring consultants, and insurance companies.

In fact, when a brand-new structural configuration having strong symbolic and aesthetic significance is put into place, there can be several uncertain factors that worry the owners. Designs can be sometimes faulty and often remain undetected until failure happens. There can be defects during construction process too. Without any doubt, SHM of such structures during and after construction can significantly reduce risk as defect would be identified at an early stage without exposing public to excessive risk.

Ajoy Kumar Das, Ph.D., P.E.  is a Structural Engineer with Big Top Manufacturing, Perry, FL.