Data for zero-defect manufacturing
For the past decades, the use of feedback and feedforward control systems has been essential in every automated process stage in advanced production systems for fulfilling tight and rigid product specifications at each stage.
Integrating models and real-time data for zero-defect manufacturing control systems
Yet, such systems cannot handle well natural variations in material properties and process conditions that can lead to waste due to nonconformity with the specifications which, in turn, can cost millions of euros a year in product, labor and energy waste for manufacturers. With the increasingly strict requirements for modern high-tech products, a new intelligent control strategy is needed that can take into account material variations in the complete control system, allow for flexible tolerances at each stage and meet the final product specification.
The increasing use of ICT and internet-of-thing sensors in modern manufacturing has opened a new way for the development of novel control systems design that can lead to drastic improvements in production accuracy and will be the focus of this project.
Developing data analytic tools
The proposed research program aims at developing novel data analytics and control design methods for zero-defect manufacturing systems based on the complementary use of real-time product, process, as well as, material data and existing process models.
In particular, we will firstly develop data analytic tools which will enable us to translate product and material data into process information based also on the use of existing high-fidelity (such as finite-element) process model. Such process information will contain important information on the stage-to-stage variation and product-to-product variation due to material variation and environmental conditions. Based on the process information, we will then develop new control reconfiguration strategies that are able to fine-tune/adapt the existing process control systems, in terms of both feedback and feedforward control systems, in order to compensate for the material and process variations. By using real-time monitoring of the process variation in the control systems, we will be able to adapt all process stages to natural material variation effectively and efficiently.
If successful, the proposed project will provide effective data analytics and control design methods to realize zero-defect manufacturing systems with minimal cost through the intelligent use of ICT and internet-of-thing sensors, recouping millions of euro loss in waste.
Looptijd: Lopend (t/m 2021)
Mede-projectleiders: Prof. dr. B. Jayawardhana
Partners: Rijksuniversiteit Groningen, Faculty of Science and Engineering (FSE), ENgineering and TEchnology institute Groningen (ENTEG)
Projectmiddelen: € 700.592
Schrijf u in voor onze nieuwsbrief en blijf op de hoogte van het laatste nieuws over Commit2data