Results

European industry has recognized the importance of AI tools within the context of Cyber-Physical Systems (CPS) since they are the keys to providing excellent services, enriched customer knowledge and also exploit new business opportunities.

COGNINTEL deploys an AI-based decision-making and control platform tailored to the machine tools industry. The aim is to optimize operations by enabling a holistic and intelligent Quality, Energy and Maintenance Management (QEM) approach.

AI-based platform of COGNINTEL relies on an integrated time series analysis and unsupervised clustering approach to sharply enhance diagnostic/prognostic capabilities while ensuring a quick, confident and robust deploy-ability.

The experiments were done and the data was collected in the experimental and industrial facilities of ERREDUE and Scorta Srl in Italy as well as in the Testbed for Industry 4.0 at CIIRC CTU in Prague, Czech Republic as a part of the RICAIP research infrastructure.

Watch the video and learn how the solution contributes to increasing the flexibility, reliability, sustainability of manufacturing and production efficiency.

Data pre-processing, data exploration, training testing and deploying models of the COGNINTEL experiment have been developed and implemented by exploiting the KNIME analytics platform, which is a free and open-source data analytics, reporting and integration platform.

In Italy, experiment on the application of AI-based techniques has been conducted to early detect anomalies related to incipient failure (predictive maintenance), quality issues (workpiece errors) and energy issues (abnormal absorption). Approaches and methods have been validated in the context of grinding processes for precision thread tools production. 

In Prague, also series of experiments of carbon steel drilling by solid carbide drills were conducted as a part of the research on the topic of machining process monitoring and tool wear estimation based on the machine tool control system signals only. The acquired data were processed for the training of AI algorithm for online drill geometry and wear monitoring.