Simulation and optimisation for future industrial applications (SOFIA)
The project will enable the anticipated learning machines for the process industry with potential significantly to reduce economic and environmental costs.
Vinnova, Processindustriell IT och Automation
Mälarenergi, ABB, Sigholm, EvoThings
VSICS, Mälarenergi, ABB, Sigholm, EvoThings
Project manager at MDH
Description of the project
To be in the forefront, Swedish process industry need to make digitization efforts that matters. Many traditional automation tasks in processes industry have been refined over long periods of time and to make a leap in efficiency and competitiveness will require novel and innovative methods. Machine learning is a promising technology that have been applied to many challenging tasks. Moreover, by using IoT and cloud technologies it is now cost efficient to transfer large amounts of information to data centers and analyze them at a massive scale. It is also possible to mix data from different sources and add data sources that has not been practically available in the traditional automation environment.
The industrial process flows are constantly evolving, e.g., in a mine with progress of new drifts. Thus, industrial application of machine learning will require continuous and automated adaption of the algorithms. The Smart Flows project will enable the anticipated learning machines for the process industry with potential significantly to reduce economic and environmental costs. The project partners, MDH, SICS, ABB, Mälarenergi, Sigholm, and Evothings, together with the broad consortia and reference group from Swedish process industry, cover the wide span of the needed competence. We believe that an industrial demonstrator showing
the feasibility and value of innovative applications developed on top of future industrial IoT platforms will be of great value to Sweden.