Periodic Reporting for period 3 - COGNITWIN (COGNITIVE PLANTS THROUGH PROACTIVE SELF-LEARNING HYBRID DIGITAL TWINS)
Reporting period: 2022-03-01 to 2023-02-28
The COGNITWIN project has been supporting the digitalization of the European process industries. A main ambition of the COGNITWIN project has been to develop digital, hybrid and cognitive digital twins that can support a significant improvement in industrial operation.
The final results of the project has met the overall objectives as follows:
COGNITWIN for Industry Process Excellence, COGNITWIN for SPIRE, and Cognitive Hybrid Digital Twins for Retrofitting and Optimised Process Performance has been provided through 6 different cognitive production plants.
This has showed improved performance in the SPIRE process industry areas of non-ferrous (by Hydro and Elkem), steel (by Saarstahl, Noksel and Sidenor) and engineering (by Sumitomo).
The results have boosted the industries through the use of cognitive and hybrid digital twins. This has also been supported through use of new sensors fore retrofitting and digital transformation in the involved process industries.
The COGNITWIN Toolbox provides access to the project results, including the industrial digital twin pipelines and the various components and services for the creation of new digital twins. Further dissemination is being done through associated European Digital Innovation Hubs.
The final results of the project has met the overall objectives as follows:
COGNITWIN for Industry Process Excellence, COGNITWIN for SPIRE, and Cognitive Hybrid Digital Twins for Retrofitting and Optimised Process Performance has been provided through 6 different cognitive production plants.
This has showed improved performance in the SPIRE process industry areas of non-ferrous (by Hydro and Elkem), steel (by Saarstahl, Noksel and Sidenor) and engineering (by Sumitomo).
The results have boosted the industries through the use of cognitive and hybrid digital twins. This has also been supported through use of new sensors fore retrofitting and digital transformation in the involved process industries.
The COGNITWIN Toolbox provides access to the project results, including the industrial digital twin pipelines and the various components and services for the creation of new digital twins. Further dissemination is being done through associated European Digital Innovation Hubs.
The COGNITWIN project has provided several enabling technologies that collectively supports the vision of cognitive production plants. Technology components are being provided through the COGNITWIN toolbox, supported by 28 scientific papers and book chapters.
COGNITWIN Toolbox provides access to the following Key Exploitable Results:
1. CoTwins: A Micro Service Based Open Architecture to Develop Multi Model, Collaborative and Cognitive Digital Twins (by TEKNOPAR).
2. Cognitive CENIT: Digital twins for estimation, optimisation, and control of industrial processes – with nonlinear model predictive control (by Cybernetica)
3. Cognitive tool-wear monitoring tools (by Nissatech).
4. Fraunhofer FA³ST - Fraunhofer Advanced Asset Administration Shell Tools (for Digital Twins)
5. SINTEF – COGNITWIN Digital Twin Toolbox portal – and SINTEF components
6. Optimization and optimal control of GTC in aluminium production – with Hydro Digital Twin pipeline
7. Cognitive digital twin of steel ladle – with Sidenor Digital Twin pipeline
8. Optimization of silicon process – with Elkem Digital Twin pipeline
9. Fouling management in energy boilers with fuel characterization – with Sumitomo Digital Twin pipeline
10. Tracking system for rolled bars in the rolling mill - with Saarstahl Digital Twin pipeline
11. Cognitive digital twin powered condition monitoring (and control) in steel pipe manufacturing industry – with Noksel Digital Twin pipeline.
The COGNITWIN project has shown how to achieve improved performance in Cognitive Production plants in the SPIRE Process Industry sectors of Aluminium, Silicon (Non Ferrous), Steel and Engineering through the use of Hybrid and Cognitive Digital Twins.
1. Hydro Cognitive Digital Twin for Aluminum Gas Emission Control in the Gas Treatment Center (GTC). The result is a reduction of the overall suction rate by 10%, with 1500 MWh/y of saved fan work, and increased available recovered thermal energy of 13500 MWh/y. The reduction of energy consumption in th eGTC is 10% and decreased process disturbance by preventive maintenance is 4%.
2. Elkem Cognitive Digital Twin for temperature in tapping stream and slag in silicon refining ladle has shown a calculated improvement in increase of post taphole output up to by 120 MT/per year for an equivalent to an industry potentsial of 0.5 MEuro per 100 000 metric to produced, with a possible for further increase, pending further deployment and evolution of the solution.
3. Saarstahl Cognitive Digital Twin for a steel rolling mill tracking system shows a potential to improve rolling line efficiency by 15% and to reduce energy consumption and process emissions by 15%, and automatic error detection, pending the further deployment and operational of a complete tracking system with fully interacting subsystems.
4. Noksel Cognitive Digital Twin for Predictive Maintenance for machinery in steel production, with real-time online monitoring of process efficiency shows a minimal latency of less than 1ms, 99% Accuracy of data analytics, and AI algorithms, 62% reduction in machine downtime due to conducted predictive maintenance and 4.8% reduction in energy consumption.
5. Sidenor Cognitive Digital Twin for steel ladle refractory wear, with a wear predictive model developed take into account both production parameters and knowledge from the operators who take the decision. The model has been run in parallell with the current operational system. The improvement on the refractory life could reach 14% what means saving around 54.000€ per year.
6. Sumitomo Cognitive Digital Twin for Boiler fouling management. The measurable improvements has shown improved boiler operating efficiency of 0.09 %, with a lower operating costs of 119 k€/year and a decrease of emission of 28%, and an improved reliability and availability of 1.3%.
The six pilots has evolved to provide good "Best practices" examples for various aspects of Digital Twins.
The overall goal of COGNITWIN’s exploitation, dissemination, communication and standardisation activities was to ensure that the COGNITWIN results will have a maximum impact and that the consortium reaches its impact objectives. This has been done through production of online video demonstrators for the pilots and toolbox components, and through participation in more than 3 exhibitions, 4 workshops and 15 conferences, including workshops and webinars with the SPIRE Process Industry community. The project has published 10 journal articles, 2 book chapters and 16 articles in conference proceedings.
COGNITWIN Toolbox provides access to the following Key Exploitable Results:
1. CoTwins: A Micro Service Based Open Architecture to Develop Multi Model, Collaborative and Cognitive Digital Twins (by TEKNOPAR).
2. Cognitive CENIT: Digital twins for estimation, optimisation, and control of industrial processes – with nonlinear model predictive control (by Cybernetica)
3. Cognitive tool-wear monitoring tools (by Nissatech).
4. Fraunhofer FA³ST - Fraunhofer Advanced Asset Administration Shell Tools (for Digital Twins)
5. SINTEF – COGNITWIN Digital Twin Toolbox portal – and SINTEF components
6. Optimization and optimal control of GTC in aluminium production – with Hydro Digital Twin pipeline
7. Cognitive digital twin of steel ladle – with Sidenor Digital Twin pipeline
8. Optimization of silicon process – with Elkem Digital Twin pipeline
9. Fouling management in energy boilers with fuel characterization – with Sumitomo Digital Twin pipeline
10. Tracking system for rolled bars in the rolling mill - with Saarstahl Digital Twin pipeline
11. Cognitive digital twin powered condition monitoring (and control) in steel pipe manufacturing industry – with Noksel Digital Twin pipeline.
The COGNITWIN project has shown how to achieve improved performance in Cognitive Production plants in the SPIRE Process Industry sectors of Aluminium, Silicon (Non Ferrous), Steel and Engineering through the use of Hybrid and Cognitive Digital Twins.
1. Hydro Cognitive Digital Twin for Aluminum Gas Emission Control in the Gas Treatment Center (GTC). The result is a reduction of the overall suction rate by 10%, with 1500 MWh/y of saved fan work, and increased available recovered thermal energy of 13500 MWh/y. The reduction of energy consumption in th eGTC is 10% and decreased process disturbance by preventive maintenance is 4%.
2. Elkem Cognitive Digital Twin for temperature in tapping stream and slag in silicon refining ladle has shown a calculated improvement in increase of post taphole output up to by 120 MT/per year for an equivalent to an industry potentsial of 0.5 MEuro per 100 000 metric to produced, with a possible for further increase, pending further deployment and evolution of the solution.
3. Saarstahl Cognitive Digital Twin for a steel rolling mill tracking system shows a potential to improve rolling line efficiency by 15% and to reduce energy consumption and process emissions by 15%, and automatic error detection, pending the further deployment and operational of a complete tracking system with fully interacting subsystems.
4. Noksel Cognitive Digital Twin for Predictive Maintenance for machinery in steel production, with real-time online monitoring of process efficiency shows a minimal latency of less than 1ms, 99% Accuracy of data analytics, and AI algorithms, 62% reduction in machine downtime due to conducted predictive maintenance and 4.8% reduction in energy consumption.
5. Sidenor Cognitive Digital Twin for steel ladle refractory wear, with a wear predictive model developed take into account both production parameters and knowledge from the operators who take the decision. The model has been run in parallell with the current operational system. The improvement on the refractory life could reach 14% what means saving around 54.000€ per year.
6. Sumitomo Cognitive Digital Twin for Boiler fouling management. The measurable improvements has shown improved boiler operating efficiency of 0.09 %, with a lower operating costs of 119 k€/year and a decrease of emission of 28%, and an improved reliability and availability of 1.3%.
The six pilots has evolved to provide good "Best practices" examples for various aspects of Digital Twins.
The overall goal of COGNITWIN’s exploitation, dissemination, communication and standardisation activities was to ensure that the COGNITWIN results will have a maximum impact and that the consortium reaches its impact objectives. This has been done through production of online video demonstrators for the pilots and toolbox components, and through participation in more than 3 exhibitions, 4 workshops and 15 conferences, including workshops and webinars with the SPIRE Process Industry community. The project has published 10 journal articles, 2 book chapters and 16 articles in conference proceedings.
The progress beyond the state of the art has in particular focused on the development of the concept of Cognitive Digital Twins and the application of this to support Cognitive Plants in the Process industry.
The results at the end of the project includes the final version of the COGNITWIN Toolbox with components that can support the development of Hybrid and Cognitive Digital Twins for the Process industry, supported by the technology partners, as applied for the 6 pilot users.
The COGNITWIN consortium has had an active involvement in relevant standardisation activities, in particular in the IDTA – Industrial Digital Twin Association (with AAS), the Digital Twin Consortium (DTC), the Industrial Internet Consortium Digital Twin Interoperability group and with ISO SC41 IoT and Digital Twins.
The results at the end of the project includes the final version of the COGNITWIN Toolbox with components that can support the development of Hybrid and Cognitive Digital Twins for the Process industry, supported by the technology partners, as applied for the 6 pilot users.
The COGNITWIN consortium has had an active involvement in relevant standardisation activities, in particular in the IDTA – Industrial Digital Twin Association (with AAS), the Digital Twin Consortium (DTC), the Industrial Internet Consortium Digital Twin Interoperability group and with ISO SC41 IoT and Digital Twins.