Advertisement

Modelling Dynamic Processes Using System Identification Methods

  • Yuri A. W. Shardt

Abstract

This chapter introduces the reader to the topic of system identification. System identification seeks to develop a generalised framework for the development of deterministic and stochastic models for complex chemical processes for application to control. First, the most common linear models, including the prediction error model with its simplifications and the impulse response model, are examined theoretically. Next, the prediction error method is developed for open-loop system identification. It is shown that the method provides unbiased and consistent parameter estimates. Furthermore, model validation is presented for open-loop models to assess not only the standard regression results but also such concepts as linearity, time delay, and time invariance. Then, the open-loop approach is extended to closed-loop system identification, and appropriate changes in the estimation and validation approaches are noted. Three different methods are considered: indirect, direct, and joint closed-loop identification. Finally, nonlinear system identification is briefly introduced and examined. All examples in this chapter are drawn from a single experiment to determine the water level in a four-tank system. By the end of the chapter, the reader should have a good understanding of the theoretical underpinning of system identification and be able to apply its results to develop complex models for industrial applications.

Keywords

Time Delay Estimation Disturbance Model Nonlinear System Identification Process System Identification White Noise Signal 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. Ashley R (1988) On the relative worth of recent macroeconomic forecasts. Int J Forecast 4:363–376CrossRefGoogle Scholar
  2. Nelson CR (1972) The prediction performance of the FRB-MIT-PENN model of the U.S. Economy. Am Econ Rev 62(5):902–917Google Scholar
  3. Shardt YA, Huang B (2011) Closed-loop identification with routine operating data: effect of time delay and sampling time. J Process Control 21:997–1010CrossRefGoogle Scholar
  4. Shardt YA, Huang B (2014) Minimal required excitation for closed-loop identification: implications for PID control loops. In: ADCONIP conference proceedings, Hiroshima, Japan, pp 296–301. doi:http://www.nt.ntnu.no/users/skoge/prost/proceedings/adconip-2014/pdf/SUBS61TO80/0069/0069_FI.pdf
  5. Zhu Y (2001) Multivariable system identification for process control. Elsevier Science Ltd., OxfordGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Yuri A. W. Shardt
    • 1
  1. 1.Institute of Automation and Complex Systems (AKS)University of Duisburg-EssenDuisbergGermany

Personalised recommendations