Process dynamics study based on multivariate AR-modelling

Seppo Rantala, Heikki Jokinen, Martti Jokipii, Olli Saarela, Risto Suoranta

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

    4 Citations (Scopus)

    Abstract

    In this paper a method for studying process dynamics will be presented. The method is based on statistical time series modelling called multivariate autoregressive modelling. The proposed method can be used for trouble-shooting diagnostics and control system modelling applications. After the multivariate autoregressive model is identified the dynamics of the system can be analyzed with the model. The model parameters are used for determination of power spectrum estimates, transfer functions with step responses and noise source contributions. Those parameters can be used also for simulation purposes. The result of the present study suggests effectiveness and usefulness of the method as a tool in process dynamics research.

    Original languageEnglish
    Title of host publicationProceedings of the 14th Annual Conference of Industrial Electronics Society
    PublisherIEEE Institute of Electrical and Electronic Engineers
    Pages320-325
    Number of pages6
    DOIs
    Publication statusPublished - 1988
    MoE publication typeA4 Article in a conference publication
    Event1988 International Conference on Industrial Electronics: Control and Simulation, IECON 1988 - Singapore, Singapore
    Duration: 24 Oct 198828 Oct 1988

    Conference

    Conference1988 International Conference on Industrial Electronics: Control and Simulation, IECON 1988
    Country/TerritorySingapore
    CitySingapore
    Period24/10/8828/10/88

    Keywords

    • Autoregressive modelling
    • Identification
    • Multivariable systems
    • Process dynamics
    • Signal and system modelling
    • Signal processing
    • Simulation
    • System analysis
    • Technical diagnostics
    • Time series analysis

    Fingerprint

    Dive into the research topics of 'Process dynamics study based on multivariate AR-modelling'. Together they form a unique fingerprint.

    Cite this