Exploration of control strategies for air side design in active chilled beams

Heating Ventilating and Air Conditioning systems can consume up to 30% of the overall energy consumed by a building. Profound emphasis is being laid on constructing green buildings conforming to Greenmark and LEED certifications or making the existing building more energy efficient. Energy...

Full description

Saved in:
Bibliographic Details
Main Author: Kapur Pranav Rakeshkumar
Other Authors: Cai Wenjian
Format: Theses and Dissertations
Language:English
Published: 2015
Subjects:
Online Access:http://hdl.handle.net/10356/65171
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Heating Ventilating and Air Conditioning systems can consume up to 30% of the overall energy consumed by a building. Profound emphasis is being laid on constructing green buildings conforming to Greenmark and LEED certifications or making the existing building more energy efficient. Energy savings by making HV AC systems more efficient has become feasible since the advent of Active Chilled Beams which have become a popular alternative to generic VA V systems over the recent years due to their energy saving potential and minimal space requirements. The physical design characteristics of chilled beams cannot be credited entirely for operational efficiencies and low power consumption in HV AC systems. Without the assistance of high and low level, robust and stable control system it is difficult to maintain the desired thermal load with acceptable levels of operational efficiencies. This master's thesis aims to explore multiple control strategies for air side design utilizing active chilled beams. The air side design considerations include the control of fan that supplies the primary air to the duct and a damper that controls the air-flow rate to the ACB terminal unit. A model of the system has been developed using an intelligent identification algorithm based on empirical system modelling established on the method of linear least squares. A comparison of controller and system performance was conducted on a pilot plant between PID controller using Internal Model Control structure and Generalized Predictive Controller based N-Star tuning algorithm and the latter has shown promising performance than a PID controller by displaying less overshoot, faster settling time and better robustness.