Is AI a niche market

AI for Quality Control (Industry 4.0)

The wide diffusion of composite laminates in a large number of applications in the structural field pointed out their pros and peculiar characteristics, but also their cons. Any non-visible defects, generated during the manufacturing process and located within the composite material package, can only be detected by non-destructive inspection with the ultrasonic method. Non-destructive ultrasound control is now used in many fields such as the control of metals, concretes and composite materials but also in the control of thicknesses of all kinds such as that of oil and gas pipelines, plating, plasters of ancient frescoes, etc.
The main advantages of the ultrasound method range from the high sensitivity of control to the ability to identify the position of the defect. In the field of aerostructures and in the case of composite materials, the defects that can be detected by ultrasound can be mainly of 3 types: delaminations, inclusions, porosities.
In this scenario Reco 3.26 is developing an industrial research project focused on artificial intelligence and computer vision techniques for the improvement and automation of non-destructive inspection and for the defects detection in carbon fiber fuselage sections. The idea is to transform raw data, coming from the ultrasonic probes, into images to be post processed and then analyzed using computer vision algorithms based on convolutional neural networks (CNN).

Product groups

  • Basic research in materials technology
  • Basic research in business administration, factory management
  • Basic research in artificial intelligence