Control Design Based on FMI: A Diesel Engine Control Case Study

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding


Modelica allows systems to be described with reuseable components and with a high precision. To be able to use such complex models efficiently, high demands are set on tools that allow the user to extract the information needed from the models in a straight-forward manner. For this purpose, design-of-experiments techniques can be used to systematically analyze the complex models.
In this paper, it is demonstrated how a Modelica model of a diesel engine can be used for control design. The engine model has multiple inputs and outputs, it is nonlinear, has many parameters, and has a higher order than most control design algorithms are able to handle in a numerically robust way.

It is shown how the features for dynamic design-of-experiments analysis in the FMI Toolbox for MATLAB can be used to analyze the variation in system dynamics across the engine operating range. A gain scheduling of nine multivariable linear-quadratic-gaussian (LQG) controllers, is designed based on linearization and model reduction of the original nonlinear FMU model.


External organisations
  • Modelon AB
  • Lund University
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Control Engineering
Original languageEnglish
Title of host publication8th IFAC Symposium on Advances in Automotive Control AAC 2016
EditorsPer Tunestål, Lars Eriksson
Publication statusPublished - 2016 Jun 20
Publication categoryResearch
Event8th IFAC Symposium Advances in Automotive Control, AAC 2016 - Kolmården Wildlife Resort, Sweden
Duration: 2016 Jun 202016 Jun 23

Publication series

Name IFAC-PapersOnLine


Conference8th IFAC Symposium Advances in Automotive Control, AAC 2016
Internet address

Related projects

Per Tunestål, Maria Henningsson & Rolf Johansson

Project: ResearchCollaboration with industry

View all (1)