Model-Based Optimal Control of Multidimensional and Multi-Tonal Frequency Varying Disturbances

Juha Orivuori

    Research output: ThesisDoctoral ThesisCollection of Articles

    Abstract

    Vibration is a phenomenon related to every physical system that has the potential to cause severe problems that range from increased structural fatigue to potential operator health hazards. The traditional approach to solve the vibration related problems is to dissipate the vibration energy through the addition of passive damping elements consisting of springs,masses and dampers. Even though these methods have been widely applied in all branches of industry, they are becoming increasingly inadequate in meeting the industrial standards of today. In the past decade the significant increase in the available computational power has given arise to another approach to tackle the vibration related problems; namely the active control of vibrations, which is capable of meeting the tightened standards. In this approach, the vibrations are suppressed through the excitation of external energy in a suitable form into the system, resulting in the compensation of the vibrations. The use of this approach has enabled the mitigation of vibrations in very complex structures in a deterministic manner. The major benefit of the method is the change of the underlying design problem. Namely, the original structural design problem is converted into a standard control design problem, enabling the designer to use the very powerful tools of control theory. In this thesis, a novel method for active vibration mitigation is presented. The proposed approach shares many similarities with the existing model-based methods while addressing many of the drawbacks and problems encountered with the current approaches. The proposed method is a generic nonlinear control law capable in the simultaneous suppression of multiple tonal disturbances in multiple dimensions. The control design procedure is simplified such that the number of free parameters is minimal and the impact of these parameters on the process performance is transparent. Such an approach enables the method to be applied by an industrial system specialist with possibly very little experience in control theory, unlike what is the case with the most of the existing methods. In addition, the essential tools for the performance and stability evaluation of the obtained control law are presented in detail with the focus being in the problems commonly encountered in the practical implementation.This thesis consists of a summary and five publications with the focus being on the control design, performance analysis and test-bed implementation in several industrial processes. An extensive comparison of the proposed method against the existing linear control approaches is also included in this work.
    Translated title of the contributionTaajuusmuuttuvien Häiriöiden Mallipohjainen Optimaalinen Vaimennus
    Original languageEnglish
    QualificationDoctor's degree
    Awarding Institution
    • Aalto University
    Supervisors/Advisors
    • Zenger, Kai, Supervising Professor
    • Zenger, Kai, Thesis Advisor
    Publisher
    Print ISBNs978-952-60-4977-9
    Electronic ISBNs978-952-60-4978-6
    Publication statusPublished - 2013
    MoE publication typeG5 Doctoral dissertation (article)

    Keywords

    • active vibration control
    • continuous gain-scheduling
    • nonlinear optimal control
    • optimal control
    • performance evaluation
    • vibration isolation

    Fingerprint Dive into the research topics of 'Model-Based Optimal Control of Multidimensional and Multi-Tonal Frequency Varying Disturbances'. Together they form a unique fingerprint.

    Cite this