Bastian Ritter, Home

Bastian Ritter, M.Sc.

Arbeitsgebiet(e)

  • Physikalische Modellbildung und Systemanalyse von Windenergieanlagen
  • Entwurf von nichtlinearen Zustandsschätzern und Störgrößenbeobachtern
  • Mehrgrößenregelung von (schwimmenden) Windenergieanlagen
  • Parameterschätzung und Systemidentifikation von nichtlinearen Systemen

Lehre

  • Betreuung von Studierenden im Rahmen des Forschungsseminars „Weiterführende Methoden der Regelungstechnik“

zur Liste

Research Project

Introduction

To meet the steadily increasing industrial requirements, wind turbine controllers have reached a mature level of closed-loop performance with respect to energy production and load reduction. These controllers are nonlinear, complex and build heavily upon design heuristics gained from practical experiences. Unfortunately, the latter significantly complicates direct extension of field-proven controllers to new advanced turbine designs without major modifications.

Therefore the institute and the companies Industrial Science and IAV have teamed up to investigate and develop modern control schemes for wind turbine closed-loop operation. Nonlinear model predictive control (NMPC) is considered as the most promising among them (Schlipf et al. 2013, Gros & Schild 2016). Yet, more advanced control algorithms demand for more information about the inner state of the turbine which must be provided by additional sensors and/or observers. Since load sensors like strain gauges are often expensive and error-prone, nonlinear state estimation algorithms as inexpensive and powerful alternative come into play to reconstruct desired, but unmeasured, quantities.

Nonlinear state estimation and distributed architecture

State estimation is a central topic for numerous technical and biological applications. Especially advanced state-feedback controllers need the full state information to be accessible in real-time. A major obstacle for the reliable utilization is the requirement not only for accurately known full dynamic state, but also for external disturbances and critical system parameters.

As highlighted in Ritter et al. 2016 the full-scope estimation problem for wind turbines consists of multiple sub-problems including state, parameter and load estimation. Moreover, the disturbance inputs like rotor-effective wind speed and wind shear components have to be observed as well. If real-time application is critical, these sub-problems can be tackled by a tailored observer architecture. This architecture allows to address further requirements like selective filter parameter adaptation and situational by-passing triggered by observability measures.

The numerical investigation is conducted with the high-fidelity simulator FASTv8 and with an internal complex wind turbine simulator. Both confirm the validity of the chosen approach and the accurate estimation of turbine states and relevant mechanical loads based on an advanced control-oriented design model.

Future work will focus on adaptation rules for the different observer components to manage changing wind and site conditions by automated filter parameter update. These include multiple model adaptive estimation and covariance matching techniques as well as master-slave Kalman filtering schemes.

Moreover, nonlinear Moving Horizon estimators (NMHE) as optimization based batch processing algorithms will be reviewed for wind turbine application for instance to tackle the parameter estimation problem. Additionally, practical issues for design and implementation in real-world wind turbines like robustness, numerical stability and constraints need to be addressed explicitly.

Summary and Conclusions

Nonlinear observers for state and parameter estimation offer substantial improvements for future wind turbine operation and control. The employed state estimation algorithms have proven to be real-time capable and feasible on industrial controllers. The incorporation of (hidden) state estimates as additional information for the closed-loop controller show significant benefits for power generation and turbine load reduction. Yet, a thorough understanding of nonlinear filters and their specific properties is essential to harvest this potential and to design improved closed-loop state-feedback controllers.

+

+ +

+ + +

Publications in Journals

(1) Bastian Ritter, Axel Schild, Matthias Feldt, Ulrich Konigorski: The design of nonlinear observers for wind turbine dynamic state and parameter estimation, Journal of Physics: Conference Series, Vol. 753 (2016) 052029, pp. 1-12, Sept. 2016, DOI: 10.1088/1742-6596/753/5/052029

+

Conference contributions

(5) Bastian Ritter, Axel Schild, Matthias Feldt, Ulrich Konigorski: The design of nonlinear observers for wind turbine dynamic state and parameter estimation, 6th edition of the conference „The Science of Making Torque from Wind“ (TORQUE2016), October 5-7, 2016, Munich, Germany, http://www.torque2016.org

(4) Bastian Ritter, Axel Schild, Ulrich Konigorski: „Making nonlinear state estimation techniques ready for use in industrial wind turbine control systems“, Proc. of WindEurope Summit 2016, September 27-29, Hamburg, Germany, https://windeurope.org/summit2016/conference/programme/

(3) Bastian Ritter, Ulrich Konigorski: Entwurf nichtlinearer Beobachter zur Zustands- und Störgrößenschätzung für Windenergieanlagen, Karlsruher Regel-Kreis 2016, June 19-20, Institute of Automatic Control and Mechatronics, Technische Universität Darmstadt, Germany

(2) Bastian Ritter, Ulrich Konigorski: Advanced multivariable control design for modern multi-MW wind turbines, Proc. of the 11th EAWE PhD Seminar on Wind Energy in Europe, 23.-25.9.2015, Stuttgarter Lehrstuhl für Windenergie (SWE), Stuttgart, Germany, http://eawephd2015.sciencesconf.org/

(1) Bastian Ritter, Holger Fürst, Ulrich Konigorski, Mike Eichhorn: Multivariable model for simulation and control design of wind turbines, Proc. of the 12th German Wind Energy Conference (DEWEK 2015), 19.-20.5.2015, Bremen, Germany, http://www.dewek.de

+

Contributions to lectures

Titel: „Nichtlineare Zustands- und Störgrößenbeobachter“, Beitrag zum Forschungsseminar „Weiterführende Methoden der Regelungstechnik“ am Institut für Regelungstechnik und Mechatronik, Technische Universität Darmstadt, 11.5.2017, 13:30-15:00 Uhr in S3|10-406A, Weitere Informationen: http://www.rtm.tu-darmstadt.de/rtm_lehre/

Titel: „Regelung und Modellierung von Windenergieanlagen“, Beitrag zur Vorlesungsreihe Wind-, Wasser- und Wellenkraft – Optimierung und Skalierung von Fluidkraftsystemen am Institut für Fluidsystemtechnik (FST), Technische Universität Darmstadt, 16.11.2016, 9:50-11:20 Uhr in L1|01-K466, Weitere Informationen: http://www.fst.tu-darmstadt.de/

Vortragstitel: „Nichtlineare Beobachter zur Zustands- und Störgrößenschätzung“, Beitrag zum Forschungsseminar „Weiterführende Methoden der Regelungstechnik“ am Institut für Regelungstechnik und Mechatronik, Technische Universität Darmstadt, 23.6.2016

Vortragstitel: „Regelung von Windenergieanlagen – eine systemtheoretische Einführung“, Beitrag zur Vorlesungsreihe „Wind, Wasser, Wellenkraft“ am Institut für Fluidsystemtechnik, Technische Universität Darmstadt, 11.11.2015

+

Literature and references for student work

Interesting books, conference and journal articles, as well as internet links with special emphasis on state and parameter estimation:

The Kalman Filter and its derivatives

Least-Squares Estimation

Moving Horizon Estimation

Adaptive Kalman Filters

  • Article by Raman Mehra entitled „On the identification of variances and adaptive Kalman filtering“ (1970) +++ http://ieeexplore.ieee.org […]
  • Book by Peter S. Maybeck entitled „Stochastic models, estimation, and control“ (1979)

+

Open and assigned student topics

Title / Topic Type Status
   
Future topics   -
Nichtlineare Mehrgrößenregelung von Betriebsfestigkeitsprüfständen mit hydraulischen Aktoren Master-Thesis now
Optimal wind farm control for power maximization and fatigue load reduction Master-Thesis 10/2017
Modeling and control of floating offshore wind turbines Master-Thesis 11/2017
Implementierung einer Methode zur integrierten Auslegung einer elektrisch angetriebenen Pumpe (at Institute of Fluid Systems http://www.fst.tu-darmstadt.de/) Master-Thesis (external) now
   
Assigned topics and work in progress   
Adaptive filter design and numerical performance analysis for wind turbine control systems Master-Thesis in progress
Nonlinear state estimation, modeling and observability analysis for wind turbines Master-Thesis in progress
Literature research on wind farm power and wake control Proseminar in progress
   

Completed theses and assignments

Title of the thesis Type Finalized
   
   
Nonlinear Observability Analysis for Wind Turbines using Empirical Gramian Matrices Forschungsseminar 09/2017
Development of a test bench for real-time evaluation of nonlinear state observers Bachelor-Thesis 09/2017
Modeling and simulation of wind farms in dynamic wakes Bachelor-Thesis 08/2017
Development of a test platform for the main control system of
MW-scale wind turbines
Master-Thesis (external) 08/2017
Ensemble Based Model Predictive Control for the Artificial Pancreas Master-Thesis (external) 07/2017
Literature research on empirical measures for practical nonlinear observability and identifiability Proseminar 06/2017
Systematischer Entwurf adaptiver Kalman Filter für Windturbinen Master-Thesis 06/2017
Entwicklung und Aufbau eines Demonstrators zur modellbasierten Windschätzung mit Raspberry Pi Bachelor-Thesis 04/2017
Literature research on nonlinear state and parameter estimation for wind turbines Proseminar 04/2017
Literature research on adaptive Kalman filtering techniques for wind turbines Proseminar 04/2017
Wind turbine system identification using least-squares and moving horizon estimation techniques Bachelor-Thesis 03/2017
Implementierung der DTU-10MW Windturbine in FAST Bachelor-Thesis 12/2016
Beobachtbarkeitsanalyse für Windenergieanlagen Bachelor-Thesis 11/2016
Modelling and Control of Floating Wind Turbines Master-Thesis 10/2016
Bewertung der Leistungsfähigkeit nichtlinearer Filteralgorithmen zur Zustands- und Störgrößenschätzung Bachelor-Thesis 10/2016
Exploring sequential Monte Carlo and unscented particle filters for nonlinear and non-Gaussian estimation Forschungsseminar 09/2016
A survey on linear adaptive Kalman filtering and suitable adaptation rules for filter parameter update Forschungsseminar 09/2016
Implementation and testing of Sigma-Point Kalman Filters in Simulink for nonlinear estimation Bachelor-Thesis 08/2016
Modellierung und Regelung von schwimmenden Windenergieanlagen Projektseminar 03/2016
Nonlinear approaches to estimate the state of a wind turbine Master-Thesis 03/2016
Modellbildung, Simulation und Regelung von WEA zur Gier- und Nickmomentkompensation Master-Thesis 01/2016
Linear model predictive control of wind turbines with constraints Bachelor-Thesis 01/2016
Theoretische Untersuchung und Bewertung von aktiven und passiven Methoden zur Turmschwingungsdämpfung Bachelor-Thesis 12/2015
Entwurf einer Zustandsrückführung mit reduziertem Beobachter für eine 5 MW-Windturbine Bachelor-Thesis 09/2015
Implementation and testing of estimators for rotor effective wind speed Bachelor-Thesis 06/2015
Using Park's-Transformation for wind turbine control Proseminar 06/2015
Overview of model predictive control (MPC) for wind turbine application Proseminar 05/2015
Estimation of rotor effective wind speed Proseminar 03/2015
Modellierung und Regelung einer Windkraftanlage zur Reduzierung dynamischer Lagerlasten Master-Thesis 11/2014
Analyse und Vergleich zweier Simulationswerkzeuge zur aero- und elastodynamischen Berechnung von Windkraftanlagen Master-Thesis 10/2014
Optimization of floating wind turbines by modeling a control prototype Master-Thesis 10/2014

Internships and student employees

The company Industrial Science offers interesting research and advanced industrial topics to students (and also graduates). These include among others wind turbine and tidal turbine applications, industrial process control as well as medical technologies.

Please contact to submit your application or your questions. Visit http://www.industrial-science.de for more details.

+

Publications & Conference Contributions

Gruppiere nach: Datum | Typ des Eintrags | Keine Gruppierung
Springe zu: 2016 | 2015 | 2013 | 2012
Anzahl der Einträge: 11.

2016

Ritter, Bastian ; Schild, Axel ; Feldt, Matthias ; Konigorski, Ulrich :
The design of nonlinear observers for wind turbine state and parameter estimation.
EAWE Torque Conference 2016 - The Science of Making Torque from Wind
[Konferenz- oder Workshop-Beitrag], (2016)

Ritter, Bastian ; Schild, Axel ; Konigorski, Ulrich :
Making nonlinear state estimation techniques ready for use in industrial wind turbine control systems.
[Online-Edition: https://windeurope.org/summit2016/conference/programme/info....]
In: Wind Europe Summit 2016, 27.-29.9.2016, Hamburg.
[Konferenz- oder Workshop-Beitrag], (2016)

Ritter, Bastian ; Konigorski, Ulrich :
Entwurf nichtlinearer Beobachter zur Zustands- und Störgrößenschätzung für Windenergieanlagen.
Karlsruher Regel-Kreis 2016
[Konferenz- oder Workshop-Beitrag], (2016)

Schild, Axel ; Gros, Sebastien ; Ritter, Bastian :
High performance nonlinear model predictive control for wind turbines.
[Online-Edition: https://windeurope.org/summit2016/conference/programme/info....]
In: Wind Europe Summit 2016, 27.-29.9.2016, Hamburg.
[Konferenz- oder Workshop-Beitrag], (2016)

Ritter, Bastian ; Chrysalidis, Niko ; Konigorski, Ulrich :
Application of nonlinear Kalman filters to wind turbine state observation.
Proceedings of 12th EAWE PhD Seminar on Wind Energy in Europe
[Konferenz- oder Workshop-Beitrag], (2016)

Ritter, Bastian ; Schild, Axel ; Feldt, Matthias ; Konigorski, Ulrich :
The design of nonlinear observers for wind turbine dynamic state and parameter estimation.
[Online-Edition: http://iopscience.iop.org/article/10.1088/1742-6596/753/5/05...]
In: Journal of Physics: Conference Series, 753 (5) 052029.
[Artikel], (2016)

2015

Ritter, Bastian ; Konigorski, Ulrich ; Eichhorn, Mike :
Overview of advanced control design for optimal wind turbine operation.
Energy, Science & Technology - International Conference and Exhibition (EST 2015) Karlsruhe, Germany
[Konferenz- oder Workshop-Beitrag], (2015)

Ritter, Bastian ; Fürst, Holger ; Konigorski, Ulrich ; Eichhorn, Mike :
Multivariable model for simulation and control design of wind turbines.
Proceedings of the 12th German Wind Energy Conference (DEWEK) Bremen, Germany
[Konferenz- oder Workshop-Beitrag], (2015)

Ritter, Bastian ; Konigorski, Ulrich :
Advanced multivariable control design for modern multi-MW wind turbines.
In: Proceedings of the 11th EAWE PhD Seminar on Wind Energy in Europe Stuttgart, Germany
[Konferenz- oder Workshop-Beitrag], (2015)

2013

Kubasiak, Nicolai ; Ritter, Bastian :
Verfahren zur Identifikation des Drehschwingungsverhaltens von Bogenoffsetdruckmaschinen.
In: 8. VDI-Fachtachtagung Schwingungen in Antrieben 2013. In: VDI-Berichte, 2197 .
[Konferenz- oder Workshop-Beitrag], (2013)

2012

Mink, Fabian ; Kubasiak, Nicolai ; Ritter, Bastian ; Binder, Andreas :
Parametric model and identification of PMSM considering the influence of magnetic saturation.
In: Proceedings of the 13th IEEE International Conference on Optimization of Electrical and Electronic Equipment (OPTIM) 2012.
[Konferenz- oder Workshop-Beitrag], (2012)

Diese Liste wurde am Fri Sep 22 09:03:11 2017 CEST generiert.