RobuSinter research project at GKN Sinter Metals
RobuSinter-Team at kick-off workshop in GKN Powder Metallurgy plant in Bruneck (April 2019)
Our RobuSinter research project focuses on the optimization of the robustness of the compression process through adaptive manufacturing. New approaches for advancing this technology are being developed together in a joint effort.
What is the content of the RobuSinter research project?
We aim at developing new methods and tools for optimizing the robustness of the compaction process by means of adaptive manufacturing, which is expected to result in significantly reduced costs due to a better use of natural and human resources. This will ultimately allow maintaining international competitiveness of GKN Sinter Metals and with this preserving work places.
Specific focus will hereby be given to identify key parameters while modelling the compaction process. Therefore, physical (white box) models and machine learning (black box) models will be combined and used to predict product quality.
Based on this model we will implement an adaptive control system in our production environment to continuously optimally update our process parameters to ultimately guarantee the targeted process robustness.
Visualitzation of the RobuSinter approach with GKN Sinter Metals
During the development of our automatically adapting compaction press we will benefit from an iterative and incremental work approach:
- We start with a relatively simple part (one upper layer, one lower layer, and a pin) while coping with the high complexity of the whole production process. Thereby the main challenges are originating from the highly nonlinear behavior of the powder, the deflection of the compaction tools, the compaction press itself with up to 10 axes which needs to be moved in a synchronized way, as well as from the changing environmental conditions.
- During this first iteration we will set up sensor systems to collect relevant data while at the same time deducing physical models of our system from scratch. We will compare the output of the models with real measurement data and refine the models using machine learning and artificial intelligence. This will help us to predict the quality of the produced parts. Based on this we will optimize both our production process and the newly developed models.
- After that we will develop adaptive controllers which will finally lead to our adaptive manufacturing system. In doing so we will cope with the high complexity of the whole system while demonstrating the advantages on a real (but simple) part.
- In the following two iterations we will focus on the increased complexity of the produced parts (more upper layers, more lower layers, and a pin) and continuously improve the whole circle ranging from sensor systems, over models predicting quality, to the control algorithm of the compaction press. The parts of the second and third iteration are closer to real customer parts and will help to transfer the research progress to competitive business advantages.
All this challenging work will be done in a close collaboration between industry and university partners. Since this project is partly supported by an Italian EFRE-grand, most of the involved people are based in the Bruneck GKN plant as well at the University of Bozen. However, to benefit from the huge knowledge of the GKN network and to directly transfer the new findings to all plants also GKN colleagues from the United Kingdom, from Germany, form the USA, as well as from China are part of the international and interdisciplinary consortium.
With this project we aim to enhance our compaction process and to explore further research directions. Thus, RobuSinter will be a big step towards our Digital and Adaptive Value Stream.
RobuSinter Research Project at GKN Sinter Metals (ENG SUB)
RobuSinter Research Project at GKN Sinter Metals (IT SUB)
The project partners
The consortium consists of four major partners.
GKN is the leader and will provide his expertise in powder metallurgy, supply compaction presses and sensor systems for experiments, as well as skilled people from several countries.
INTEC will contribute with his long-standing experience in compaction press software and low-level hydraulic press control.
UNIBZ (Free University of Bozen), the faculty of Science and Technology represented by the group of Prof. Angelika Peer, will enhance the physical (white/gray box) model of the compaction process and develop a robust and adaptive high-level controller for the compaction process.
UNIBZ (Free University of Bozen), the faculty of Computer Science, represented by the group of Prof. Francesco Ricci, will investigate correlations between process parameters and quality characteristics using Machine Learning techniques to derive a black box model. This information will inform the modelling of the white box model.
RobuSinter is an EFRE (European Fond for Regional Development) supported research project.
Related sub-research project FESR1113
To achieve a high process robustness of sintered work pieces manual quality checks and manual adjustments of production parameters are currently required in regular intervals. This work represents one of the major bottlenecks in achieving a more cost-efficient production, which will be required to maintain international competitiveness of companies like GKN Sinter Metals.
The project aims at targeting this challenge and at developing new methods and tools for achieving such a higher process robustness. In doing so, the project will develop control and machine learning methods for modelling the process of creating sintered work pieces as function of a large series of inputs (machine data, CAD data, material and environmental data) and for identifying driving production parameters that mainly define the finally achieved quality. Using this information and with the help of methods for the active control of machine parameters the project aims at influencing the production process to attain the required process robustness.
Project Leader: Thomas Villgrattner – GKN Sinter Metals
Principal Investigator: Angelika Peer — Faculty of Science and Technology – Free University of Bozen
Co-Investigator: Francesco Ricci - Faculty of Computer Science – Free University of Bozen
Project Duration: 01/10/2018 - 30/09/2021
Public funding:
- € 635.517,50 (overall)
- € 318.367,18 (GKN)
- € 317.150,32 (unibz)
