A digital twin for infrastructure monitoring

0

Building automation systems, HVAC, security instrumented systems and many other types of direct digital control are possible in large part thanks to data communications and microprocessor technologies.

Not only do devices that retain control over something work faster and more efficiently, but they can also communicate with remote locations such as a home computer or central monitoring facility.

Having data communicated faster between the many components of a system means faster control and faster configuration of the correct parameters.The solution offered is based on the digitization of infrastructures through the creation of a “digital twin”. The digital twin is a relatively new term in the world of civil infrastructure.

A digital twin is not the same as the asset information model created through Building Information Modeling (BIM); it is a digital model that completely describes a real physical asset such as a bridge, tunnel or viaduct.A digital twin offers a multitude of opportunities to truly promote improvements and efficiency of the entire infrastructure control chain.

In this scenario, the digital twin contains detailed design information on the layout and construction of the road for both the superstructure and the substructure; the first consists of a road surface, deck, beams and drainage systems, the second of piers (or abutments) and foundations that support the piers (or pylons).monitoring A solution based on neural networks

In particular, convolutional neural networks (CNNs) trained with a deep learning approach are used to provide an in-depth “vulnerability analysis” obtained by checking and verifying damage to infrastructures thanks to the extraction of visual descriptors from the analyzed images.

Each type of vulnerability is identified by a grouping of elements and / or characteristics of a structure and its systems whose attributes satisfy a particular set of selection rules called relationship-association classifiers (RAC).

The artificial intelligence solution, therefore, through the visual analysis of images contained in the inspection reports already present in the company and possibly from the UAV photogrammetry – and / or from any other source such as smartphones, tablets – identifies the defects of the infrastructure and highlights future risks and evolving structural problems.

Risk analysis in real time

The technology provides real-time, cloud-based risk analysis of any bridge or tunnel and automatically alerts you when changes in conditions are detected, before they turn into large-scale failures.

As new images are loaded, they are compared with the previous ones, in this way the platform creates a real “visual medical record” for each resource, based on images.

“The system allows any operator, inspector or maintenance technician to have actionable information at hand to decide if, when and how medium / long-term maintenance projects and daily maintenance should be conducted, providing instant alerts on anomalies” , explains Saar Dickman, CEO of Dynamic Infrastructure.

LEAVE A REPLY

Please enter your comment!
Please enter your name here