Energy-efficient fuzzy model-based multivariable predictive control of a HVAC system

Aleksander Preglej, Jakob Rehrl, Daniel Schwingshackl, Igor Steiner, Martin Horn, Igor Škrjanc

Publikation: Beitrag in einer FachzeitschriftArtikelBegutachtung

Abstract

In this paper the novel approach of a fuzzy model-based multivariable predictive functional control (FMBMPC) of a heating ventilating and air conditioning (HVAC) system is presented, which is implemented on a real-world test plant. The control law is derived in the state-space domain and is given in an analytical form without an optimization algorithm. The basic principles of the predictive control were extended in a fuzzy multivariable manner and the suggested tuning rules for the proposed control algorithm were depicted, which normally gives satisfactory results. The proposed approach introduces a compact and relatively simple design in the case of higher-order and nonminimal phase plants, but it is limited to open-loop stable plants. For the comparison a classical optimal proportional–integral (PI) controller was also designed and applied. The results show that the FMBMPC approach performs better due to the HVACs’ nonlinear dynamics. In case of interactions influence rejection by the HVAC system, the FMBMPC algorithm outperforms the classical PI approach. The results also show that the proposed approach exhibits better reference-model tracking across a wider operating range. The energy consumption comparison shows that the FMBMPC approach is also more energy-efficient. A shortened literature review of applications of energy-efficient and MPC control for HVAC systems is also presented. FMBMPC control is interesting in the case of batch reactors, furnaces, pressure vessels, HVAC systems and any processes that have strong nonlinear dynamics, multivariable natures and long transport delays.
Originalspracheenglisch
Seiten (von - bis)520-533
FachzeitschriftEnergy and Buildings
Jahrgang82
DOIs
PublikationsstatusVeröffentlicht - 2014

Fields of Expertise

  • Sustainable Systems

Treatment code (Nähere Zuordnung)

  • Basic - Fundamental (Grundlagenforschung)
  • Application

Fingerprint

Untersuchen Sie die Forschungsthemen von „Energy-efficient fuzzy model-based multivariable predictive control of a HVAC system“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren