hig.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard-cite-them-right
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • sv-SE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • de-DE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Degradation Control of System of Systems with Multiple Stochastic and Deterministic Failure Modes
University of Gävle, Faculty of Engineering and Sustainable Development, Department of Electrical Engineering, Mathematics and Science, Electronics.ORCID iD: 0000-0003-2162-6996
University of Gävle, Faculty of Engineering and Sustainable Development, Department of Electrical Engineering, Mathematics and Science, Electronics.ORCID iD: 0000-0001-5429-7223
University of Gävle, Faculty of Engineering and Sustainable Development, Department of Electrical Engineering, Mathematics and Science, Electronics.ORCID iD: 0000-0003-2887-049x
(English)Manuscript (preprint) (Other academic)
Abstract [en]

Digital Twins integrating degradation models can be harnessed to enable industrial digitalization and promote sustainable manufacturing processes. Research in control has transitioned from methods ensuring stability and performance in the presence of degradation or faults to strategies that balance the performance and degradation of complex systems. This work formulates the the lifetime distribution for systems with multiple subsystems and several possible failure modes, considering both deterministic and stochastic failure modes. The proposed formulation is applied to the design of a two-layer control strategy with an model predictive control-based upper-layer controller that updates the parameters of the lower-layer PID-controlled closed-loop subsystems. The proposed method was evaluated on a numerical example using two hydraulic valves operating in parallel, a system commonly found in a steel manufacturing processes. The upper layer controller achieved equal time to failures for both valves, by varying the load on respective valves. The results suggest that heterogeneous systems, in terms of lifetime distributions and failure modes, can be controlled by balancing degradation and performance.

National Category
Control Engineering
Research subject
Intelligent Industry
Identifiers
URN: urn:nbn:se:hig:diva-46310OAI: oai:DiVA.org:hig-46310DiVA, id: diva2:1927274
Available from: 2025-01-14 Created: 2025-01-14 Last updated: 2025-10-02Bibliographically approved
In thesis
1. Digital Twins of Two Types of Machinery in the Steel Industry for Condition Monitoring and Control
Open this publication in new window or tab >>Digital Twins of Two Types of Machinery in the Steel Industry for Condition Monitoring and Control
2025 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

This thesis investigates methodologies for modelling dynamical systems in steel manufacturing processes using data collected from industrial production in real factory settings. The systems under study operate as a plant in a closedloop system, posing unique challenges. The primary objective is to develop input-output models for a two-stage proportional cartridge valve and a radialaxial ring rolling machine, facilitating the construction of digital twins for system ageing monitoring, fault diagnosis, optimization, and the exploration of degradation-performance trade-offs.

The research is grounded in five papers addressing key challenges: (1) developing and evaluating white-box and black-box models for a two-stage proportional cartridge valve and a radial-axial ring rolling machine, (2) modelling systems with limited sensor data using closed-loop data characterized by limited excitation, and (3) formulating an optimization problem to balance performance and degradation in complex systems composed of multiple subsystems with competing failure modes.

A combination of data analysis, parameter estimation techniques, simulations, and experimental evaluations using industrial data from steel manufacturing processes forms the methodological foundation of this work. Additionally, control strategies are explored to address performance degradation trade-offs in industrial systems.

This thesis validates input-output models for system ageing monitoring and diagnostics while also comparing white-box and black-box models. It also explores the applicability of these models for predictive maintenance. Furthermore, a novel formulation of the life distribution for complex systems with multiple subsystems and competing failure modes is introduced.

In conclusion, this thesis establishes a comprehensive framework for modelling systems in steel manufacturing processes using real production-time data under closed-loop conditions. By addressing challenges related to poor sensor data and closed-loop conditions, this thesis lays the foundation for digital twin development, enhancing sustainability and efficiency of modern steel manufacturing.

Abstract [sv]

Denna avhandling undersöker metoder för att modellera dynamiska system i stålindustrins tillverkningsprocesser med hjälp av data insamlad från industriell produktion i verkliga fabriksförhållanden. De studerade systemen fungerar som en del av en sluten slinga, vilket medför unika utmaningar. Det primära målet är att utveckla in- och utgångsmodeller för en tvåstegs proportionell patronventil och en radial-axiell ringvalsningmaskin, för att möjliggöra konstruktionen av digitala tvillingar för övervakning av systemets åldrande, feldiagnostik, optimering och analys av avvägningar mellan degradering och prestanda.

Forskningen är baserad på fem artiklar som behandlar centrala utmaningar: (1) utveckling och utvärdering av vitlådemodeller och svartlådemodeller för en tvåstegs proportionell patronventil och en radial-axiell ringvalsningmaskin, (2) modellering av system med begränsad sensordata från sluten slinga, kännetecknad av begränsad excitation, och (3) formulering av ett optimeringsproblem för att balansera prestanda och degradering i komplexa system bestående av flera delsystem med konkurrerande felmoder.

En kombination av dataanalys, parameterestimeringstekniker, simuleringar och experimentella utvärderingar med industriella data från stålindustrins tillverkningsprocesser utgör den metodologiska grunden för detta arbete. Dessutom undersöks styrstrategier för att hantera avvägningar mellan prestanda och degradering i industriella system.

Denna avhandling validerar in- och utgångsmodeller för övervakning av systemets åldrande och diagnostik, samt jämför vitlådemodeller och svartlådemodeller. Den utforskar också modellernas tillämpbarhet för prediktivt underhåll. Vidare introduceras en ny formulering av livslängdsfördelningen för komplexa system med flera delsystem och konkurrerande felmoder.

Sammanfattningsvis etablerar denna avhandling en omfattande ram för modellering av system i stålindustrins tillverkningsprocesser med hjälp av realtidsdata från produktion under slutna slinga-förhållanden. Genom att adressera utmaningar kopplade till begränsad sensordata och sluten slinga, lägger avhandlingen grunden för utvecklingen av digitala tvillingar och bidrar till att förbättra hållbarheten och effektiviteten i modern stålproduktion.

Place, publisher, year, edition, pages
Gävle: Gävle University Press, 2025. p. 65
Series
Doctoral thesis ; 58
Keywords
white-box modelling, black-box modelling, system identification, two-stage proportional cartridge valves, radial-axial ring rolling, steel industry, closed-loop identification, condition monitoring, fault diagnosis, model predictive control, health-aware control, digitala tvillingar, vitlådemodellering, svartlådemodellering, tvåstegs proportionell patronventil, radial-axiell ringsvalsning, stålindustri, sluten slingaidentifiering, tillståndövervaking, feldiagnostik, modellpredktiv styrning, hälsoanpassad styrning
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Intelligent Industry
Identifiers
urn:nbn:se:hig:diva-46311 (URN)978-91-89593-58-9 (ISBN)978-91-89593-59-6 (ISBN)
Public defence
2025-03-14, Lilla Jadwigasalen (12:108), Kungsbäcksvägen 47, Gävle, Gävle, 12:15 (English)
Opponent
Supervisors
Available from: 2025-02-21 Created: 2025-01-15 Last updated: 2025-10-02Bibliographically approved

Open Access in DiVA

No full text in DiVA

Authority records

Bautista Gonzalez, OscarBjörsell, NiclasRönnow, Daniel

Search in DiVA

By author/editor
Bautista Gonzalez, OscarBjörsell, NiclasRönnow, Daniel
By organisation
Electronics
Control Engineering

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 154 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard-cite-them-right
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • sv-SE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • de-DE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf