1 of 1
khaskhabar.com : Thursday, September 14, 2023 2:44 PM
New Delhi. IIT Mandi has developed a machine learning and AI data-driven approach to estimate the condition of bridges and predict their life. These AI algorithms can identify structural damage without human intervention.
It can provide real-time and critical information, especially on prestressed concrete and cable-stayed bridges.
On this, researchers at IIT Mandi say that AI-based algorithms can be used widely. These are not limited to bridges only but can also be used in structures like ropeways, buildings, aerospace structures, transmission towers and various infrastructure requiring periodic condition assessment and safety measures.
IIT says bridges play a vital role in India’s infrastructure and their number is around 13,500 across the country. These structures age naturally due to environmental factors such as changes in temperature, water and wind. Which has been further increased by heavy road traffic.
Traditionally the condition of bridges has been assessed through visual inspection although this method has been considered inadequate by experts. It fails to detect all structural issues and is a more time-consuming process, requiring manual analysis of many photographs.
The findings of these studies have been recently published in the journals Mechanical Systems and Signal Processing and Neural Computing and Applications.
This research was prepared in collaboration with Dr. Subhamoy Sen, Associate Professor, School of Civil and Environmental Engineering and his researchers Dr. Smriti Sharma, Ishwar Kuncham and Neha Aswal of IIT Mandi as well as Dr. Laurent Mevel of INRIA Rennes, France. Has gone.
Elaborating in this regard, Dr. Subhamoy Sen of IIT Mandi said, “We used data-driven methods like machine learning, AI and Bayesian statistical inference to estimate the condition of a bridge and predict its remaining useful life. The results have the potential to reduce operational and infrastructure risks under adverse loading conditions.”
Especially in prestressed concrete and cable-stayed bridges, the dynamic capabilities of the bridge are greatly affected by temperature fluctuations. Therefore, it is important to consider the effects of these temperatures in both real-time and AI-based SHM.
This algorithm has been rigorously tested at IIT Mandi on a bridge in a cold region with high fluctuations in annual and daily temperatures.
Initially, researchers at IIT Mandi tested the algorithm on a damaged bridge to assess its capabilities in damage detection. They then deliberately pinpointed the damage in the computer model to evaluate the algorithm’s accuracy in pinpointing the location of the damage. The algorithm’s effectiveness in identifying structural damage was subsequently confirmed through testing.
Also read – Click to read the news of your state/city before the newspaper.