Inspection Space

Where the Inspection and NDT community meets
Inspection Space Aquila Logo

Dr. Antonio Caballero Features on Tech in Asia 

 

 

Tech in Asia, an online digital publication that serves a highly engaged tech and business community of over 3.5 million monthly users, asked our Chief Technology Officer to answers questions from the internet about InspectionTech and NDT. Watch the interview or take a look at the script below and test your knowledge!

 

 

 

Hello, my name is Antonio Caballero and I am the Chief Technology Officer at Screening Eagle Technologies. I’m here today to answer some InspectionTech questions from the Internet.

Let’s take a look at the first one…

 

What is NDT?

NDT stands for Non-Destructive Testing, and as the name says, it’s the process of inspecting, testing or evaluating material properties, components or larger elements such as infrastructure without causing damage to the original part.

In other words, you don’t need to make a hole into the concrete wall to test it.

 

What are the types of NDT?

There are many types of NDT. It depends on the materials, the physic principle behind the method and the application. Visual inspection is the first type of NDT, which is very powerful and useful to collect the first pieces of information. Other types of NDT are for instance

-        rebound technology for concrete, paper or rocks

-        ultrasound for concrete, metal or composites

-        electromagnetic testing… and many more.

 

Are there any advanced non-destructive methods?

Yes, there are several advanced NDT methods. We could mention for instance, Ultrasonic Phased Array, Time of Flight Diffraction, Eddy Current Testing or Step-Frequency Continuous Wave GPR.

Advanced methods are also less understood, and their procedure and data interpretation might be very complicated. You may need a Ph.D. to understand some of them.

That is why at Screening Eagle Technologies, we put special focus on developing intelligent software to process the data and simplify the interpretation and user experience.

Our goal is to democratize technology and make those advanced NDT methods easy and accessible to everyone.

 

How accurate is ground-penetrating radar (GPR)?

Tricky! This really depends on the GPR technology, the application and also the condition of the asset.

For instance, our GPR technology is based on a step-frequency approach which gives an enormous advantage compared to pulsed GPRs. In this technology, the sensor is able to modulate the transmitted signal in a wide range of frequencies, leading to high-resolution images in deep areas.

However, for those who like to keep some numbers, I would say that GPR technology can achieve sub-centimeter accuracy in concrete within the first 60 to 80cm and sub-decimeter accuracy in soil within the first 5 to 10m.

 

How data from inspections is collected and managed?

Believe it or not, most inspections are still managed with paper and camera!! Still, traditional methods which reduce the data quality and availability.

This is very critical as the health and safety of the infrastructure and the users depend on it.

This is why we have developed a platform called INSPECT, that delivers all necessary tools in one app. Managing inspection data is easy when you can automatically pin your data and pictures to the exact location, have 3D visualization, collaboration possibility and generate reports in a matter of seconds!

 

How can machine learning be used for automated visual inspection?

Great question!

Machine learning opens an entire world of possibilities in regard to visual inspection. It could help to overcome challenges like high costs, lack of objectivity and poor traceability.

Let’s take the following example. If two technicians are sent to inspect the same bridge, their inspection report and evaluation might be very different. Probably they report different defects, or the same defects are reported with different dimensions or levels of severity.

This problem could be avoided with Machine Learning.

For instance, our machine learning engine, DEFECT, solves the problems mentioned above. All users will obtain the same results when detecting and digitizing cracks, it is faster and fully traceable.

In the future, machine learning models will not only identify defects but will also provide an indication of the cause. All very exciting, at least for me.

 

Can Artificial Intelligence predict future defects?

Absolutely yes! In fact, machine learning in combination with big data analysis is already being used to predict future defects or to optimize maintenance processes in industrial machinery, aerospace, mining equipment and many others.

The adoption of AI and big data analysis in construction or inspection of infrastructure is still lagging. However, the opportunity is there, and at Screening Eagle Technologies, we are working very hard in that R&D line to bridge that technological gap and be able to anticipate defects early enough.

Somehow, be able to provide engineers and asset owners a crystal ball to predict future defects.

 

What is structural health monitoring?

Another interesting question…

Structural health monitoring refers to the method of evaluating and controlling the structural health of the asset.

There is the general misconception that structural health monitoring consists only of installing sensors and collecting the data from them. However, it’s more than that.

A proper structural health monitoring system should also consider the data collected from visual and NDT inspections in order to generate a holistic overview of the structural condition of the asset.

To understand this better, let us think of the sensors like our nervous system… Our nervous system helps us to sense things that we cannot access or see. But if you only had a nervous system, you may not detect something wrong happening at your skin. Something that you would easily see with your eyes.

Therefore, you need all levels of information, from sensor data to visual and NDT inspections to really have the full picture. 

 

What are the applications of machine learning and data science in structural health monitoring?

Well… The future is heading towards data-driven models and with that more possibilities to leverage machine learning and data science in structural health monitoring.

The possibilities are limitless.

For example, it will make possible predictive maintenance thanks to the continuous and automatic verification of the condition of the asset…

….With that it will come to a budget optimization for the owner, as he or she will have all necessary information to make the right decision at the right time, without putting in danger the safety of the people using it…

….overall we can all benefit from healthier assets that will last longer and preserve our natural environment.

 

Are robots used for autonomous inspection?

For this question, I would like to introduce you to Max. Hello Max!!!

Robots like Max can play an important role in the future of inspection. The use of autonomous drones or robots, will transform the entire inspection industry, increasing quality, safety and bringing a tremendous boost in productivity.

 

Well, that’s it all from me today! I hope I have sparked in you more curiosity about the world of InspectionTech. Thanks for watching!

 

Want to know more about non-destructive testing with intelligent InspectionTech? Contact us to speak with one of our experts around the world and have your questions answered - we are present in over 100 countries.