Modelling and Control of a Multi-Zone HVAC System for an Unmanned Transformer Platform
Translated title
CFD Flow Field Study of a Hydrocyclone Operating under Non-Optimal Conditions
Authors
Pedersen, Sandra Lindberg ; Ntourai Or Duraj, Ioannis Or Joanis
Term
4. semester
Education
Publication year
2015
Submitted on
2015-06-09
Pages
180
Abstract
Varme-, ventilation- og airconditionanlæg (HVAC) holder indendørs temperatur, luftfugtighed og tryk på ønskede niveauer. Dette speciale udvikler og afprøver styring af et multizone HVAC-system på en ubemandet offshore-platform. En eksisterende enkeltzone-model udvides til en multizone-model, som beskriver temperatur, fugt og tryk/pressurisering i hver zone og medtager varmeoverførsel gennem vægge. Centrale komponenter – køleflade, varmeflade og befugter – indgår i den samlede model. Der designes to typer regulatorer: PI-regulatorer (proportional–integral), som er standard feedback-regulatorer, og Model Predictive Control (MPC), som bruger en model til at forudsige og optimere kommende styreindgreb. PI-regulatorerne styrer fladerne, befugteren, ventilatoren, indløbsspjæld til zonerne og variable rumvarmere. MPC sammenlignes med PI-opsætningen ved at påføre en forstyrrelse i form af en døråbning. Til sidst implementeres PI-regulatorerne på et lille ventilationsforsøg i NI LabView for at observere ydeevnen. MPC blev ikke implementeret på grund af begrænset softwarelicens.
Heating, ventilation, and air conditioning (HVAC) systems keep indoor temperature, humidity, and pressure at desired levels. This thesis develops and tests control strategies for a multi-zone HVAC system on an unmanned offshore platform. An existing single-zone model is extended to a multi-zone model that tracks temperature, humidity, and pressurization in each zone and includes heat transfer through walls. Key components—the cooling coil, heating coil, and humidifier—are included in the overall model. Two types of controllers are designed: PI controllers (proportional–integral), which are standard feedback controllers that correct errors, and a Model Predictive Controller (MPC), which uses a model to predict and optimize future control actions. The PI controllers regulate the coils, humidifier, fan, inlet dampers to the zones, and variable heaters. The MPC is compared with the PI setup by applying a door-opening disturbance. Finally, the PI controllers are implemented on a small ventilation test setup in NI LabView to observe their performance. The MPC was not implemented due to software license restrictions.
[This abstract was generated with the help of AI]
Documents