Monday, 26 September 2016

Detecting Damage on Bridge Structures Using Inclinometers

Hello Everybody,

I have been developing a new bridge Condition Assessment (CA) concept recently that I would like to share with you through this post. The idea is about detecting any possible anomalies involved in bridge structural condition using rotation as a main parameter. To achieve so, I conducted the below mentioned studies.

I developed an algorithm, which employs the Influence Line and the Moment Area theorems, to study the sensitivity of rotation to detect change in structural stiffness of a bridge structure under moving load. The algorithm calculates the rotation recorded along the length of a simply supported beam structure while load is moving. It also simulates damage at any point across the structure in terms of reduction in stiffness. The figures provided below depicts the results obtained from the aforementioned study.

Figure 1. Rotation recorded at the left end of a beam structure under moving load

Figure 2. Difference in rotation between healthy and damaged cases

Figure 1 shows the rotation recorded at left end (x=0) of a beam structure while load is moving from one end to another. 15% damage was also simulated in the study at midspan of the beam and blue plot shows the healthy and red plot the damaged cases. Figure 2 shows the difference between two plots where peak of the graph matches with the damage location and its magnitude represents the severity of the damage. From this study the required resolution of a sensor (inclinometer) was determined as 10^-3 radians to detect 15% stiffness loss on a beam structure. Following this, I made an extensive research about inclinometer sensors available in the market. It was a challenging task to find a high resolution sensor which is also cost efficient. Eventually, I came up with an idea of using accelerometers to record rotation. The idea was tested on a bridge structure available in the Structures Laboratory at the University of Exeter. The inclinometers were placed horizontally on the structure and recorded accelerations while loaded trolley were manually pulled (8 runs) from one end to another. Because the output of the accelerometer obeys a sinusoidal relationship as it is rotated through gravity, the inverse sine function of it converted recorded acceleration to rotation (angle). The idea worked well and 10-6 radians resolution was obtained.

Figure 3. 15m long bridge structure available in the Structures Lab at The University of Exeter

Figure 4. Rotation time history obtained using accelerometers under loaded trolley

The findings obtained from the above mention numerical analysis are promising to detect any possible damage involved in structural condition of a bridge structure. The idea is still being developed. I built up a 3m long simply supported beam structure in the lab and will test the concept experimentally. I look forward to sharing the results through upcoming posts.

Figure 5. 3m long, simply supported beam structure built in the lab.
Stay tuned!