The major challenges are represented by:1. Real-time responseComputational costs of the artificial intelligence algorithms are constrained by the necessity of detecting defects on a layer-by-layer basis.To tackle this problem, we will consider software/hardware acceleration strategies.2. Need for large number of training images and datasetsML techniques are built on top of a pre-annotated training sets. This set should be:Massive ? in case of deep learning, huge amount of experimental dataBalanced ? same number of examples of different experimental conditions (faulty/not faulty, different categories of defects, different geometries of manufacts, different materials, etc.) 3. Need for heterogeneous data integrationThe algorithms need to integrate data from different sources (cameras, sensors, CAD, post-processing tests, user manuals, etc.), which may be structured (e.g. images) or unstructured (e.g. manuals) and have different formats and granularities
Quality assurance for additive manufacturing, (2020-2021) - Responsabile Scientifico
Ricerca da Enti privati e Fondazioni
PE6_11 - Machine learning, statistical data processing and applications using signal processing (e.g. speech, image, video)PE6_2 - Computer systems, parallel/distributed systems, sensor networks, embedded systems, cyber-physical systems
Obiettivo 12. Garantire modelli sostenibili di produzione e di consumo
Ensuring quality and standardization of mechanical properties of the manufacts through implementation of robust on-line process monitoring.The project concept relies on the development and integration of state-of-the-art quality monitoring systems and artificial intelligence into the AM process, to avoid uncontrolled defects, improve process robustness,stability and repeatability.Improving AM processing time and costs To this date, AM is mainly a trial-and-error process, where most of the faulty artifacts are detected only AFTER the end of a job. This impacts on the processing time and overall costs of the process. Layer-wise detection of faults would reduce the overall processing time and costs.
The growing complexity of modern engineering systems and manufacturing processes is an obstacle to concept and implement Intelligent Manufacturing Systems (IMS) and keep these systems operating at high levels of reliability. Additionally, the number of sensors and the amount of data gathered on the factory floor constantly increases. This opens the vision of truly connected production processes where all machinery data are accessible allowing easier maintenance of them in case of unexpected events. SERENA project will build upon these needs for saving time and money, minimizing the costly production downtimes. The proposed solutions are covering the requirements for versatility, transferability, remote monitoring and control by a) a plug-and-play cloud based communication platform for managing the data and data processing remotely, b) advanced IoT system and smart devices for data collection and monitoring of machinery conditions, c) artificial intelligence methods for predictive maintenance (data analytics, machine learning) and planning of maintenance and production activities, d) AR based technologies for supporting the human operator for maintenance activities and monitoring of the production machinery status. SERENA represents a powerful platform to aid manufacturers in easing their maintenance burdens and for this purpose will be applied in different applications. More specifically, SERENA project will focus on advancing the TRL of the existing developments into levels TRL5 to TRL7. For this purpose, SERENA consortium will fully demonstrate the proposed approach in different industrial areas (white goods, metrological engineering and elevators production) and investigate applicability in steel parts production industry (extended-demonstration activities) checking the link to other industries (automotive, aerospace etc.) showing the versatile character of the project.
TEKNOLOGIAN TUTKIMUSKESKUS VTT OY
FRAUNHOFER GESELLSCHAFT ZUR FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V.