Periodic Reporting for period 1 - SonicScan (Innovative quality inspection methods for CFRP primary structural parts)
Reporting period: 2019-04-01 to 2020-09-30
The technologies developed in the project will have impact on the manufacturing of structural composite parts, in particular on their quality control in series production, and on the efficient deployment of ultrasonic inspection robots, by increasing the usability of such robots through automatic adaptation to new part designs. SonicScan aims at two main objectives:
(A) Enhanced methods for ultrasonic testing and data analysis, building upon recent advances in the sampling phased array technology including 3D reconstruction and data analysis methods based on machine learning.
(B) Automatic methods for planning of inspection tasks in robotic systems to ensure automatic and cost-efficent inspection. This will be achieved by including a physical inspection process model to generate an automatic inspection path from a CAD model of the part.
(1) Reconstruction technology for ultrasound data acquired with the full matrix capture (FMC) method will be developed. Adaptive methods are applied here that target the maximum out of the acquired data.
(2) Automatic path planning algorithms are developed that enable
(3) The use of deep learning methods for evaluation of ultrasonic scan data is investigated.
We expect that all three technologies will have a significant impact that goes beyond the specific application of ultrasonic testing and quality control for carbon fiber re-inforced polymers. Other applications where the SonicScan technology will have an impact are different ultrasound-based quality control tasks and automated analysis of robotic scanning data via machine learning in general. Path planning capabilities, as developed within the project are expected to enable more flexibility of robots to automate inspection processes and in increase flexibility in general. We expect that this will have a positive impact on the digital transformation of European economy.