Asier Artega of Sidenor tells us about his work in the project
Tell us about your work in the HypeCOG project so far
Our work can mainly be divided in two parts – firstly, the work at SIDENOR, where we work with other partners and, secondly, project development, which is focused on implementation. So, as well as working on the steel use case at my company SIDENOR, I also participate in two other use cases, mostly as an interested observer.
Can you elaborate on the solutions you have been working on at SIDENOR and how they are currently being implemented?
The main purpose of our work at SIDENOR is to optimise the process planning in the steelmaking shop. This planning is based on several existing rules and the knowledge held by experienced people who work here. Our challenge is to digitise all this in a clever but practical way. There are also other complementary use cases for SIDENOR in HyperCOG. One of them is related to the management of the ladle, where we are trying to get useful data from thermal images of it, to offer improvements in its tracking. Another complementary use case is related to slag, considering it as an output of the process and in this way it relates well to the cement use case in the project where we are developing an innovative measuring system along with a blockchain pilot for possible connection with “slag users” as cement producers.
The prototype now being installed at SIDENOR aims to reduce the number of production planning problems arising from failures that occur during the steel-making process. How much progress has been made so far in this first implementation?
This is one of the most advanced developments in the project. On the one hand, there is an already working prototype, developed by DFKI, being used to optimise the production planning. This prototype needs more testing but it is already quite interesting for us. At the same time, and thanks to the work by LORTEK and other partners on the cyber physical factory and the node architecture, the general HyperCOG architecture is being tested at lab scale at MSI, where data from SIDENOR is being imitated in preparation for implementation later in the project.
How does this work tie in with other work being done in the project?
My work acts as a bridge to a great extent: I need to explain the process to the RTO and University partners, so that they can do a good job in explaining possible outputs to the industrial people who will be using this technology for process improvements. For me, of course, this is an iterative process as the project advances, which will help in testing HyperCOG in the chemical and cement pilot sites as well as here at SIDENOR.
What does SIDENOR hope to gain from its participation in HyperCOG?
I think that a tool SIDENOR can use to optimise its planning processes would be a great outcome, while the outcomes we expect from the complimentary use cases would be greatly appreciated, too. The general HyperCOG architecture is also interesting in itself for data acquisition and for applying models to these data to get valuable results. On the other hand, in a project of this magnitude there are also other types of gains to be had for a company like SIDENOR, such as contact with research centres, greater awareness of technological solutions and concepts and contact with other projects, for example.
Are you confident that HyperCOG will achieve its ambitious aims?
I think there will be some interesting outcomes from the HyperCOG project and, at the same time, I think it can act as a valuable starting point for future advances and developments in cognitive manufacturing, helping industry go forward in its ongoing work to become more sustainable and efficient. Of course, the evaluation of results and of the work done throughout the project, is something we will have to face at the end of the project – but things are looking good at this half-way stage.