Optimization and control of supply air temperature in air handling unit
This dissertation deals with improving the ACMV system efficiency with an advanced optimization technique, ELM (Extreme Learning Machine). Objective of this project is to achieve an AHU model and to resolve the problems associated with obtaining an optimal control setting at every instant. With...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Theses and Dissertations |
Language: | English |
Published: |
2016
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/65899 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | This dissertation deals with improving the ACMV system efficiency with an advanced
optimization technique, ELM (Extreme Learning Machine). Objective of this project is to
achieve an AHU model and to resolve the problems associated with obtaining an optimal
control setting at every instant.
With the advancements in technology of VFD and Automation Systems for buildings,
VAV AHU, saves a considerable volume of energy and also maintains the comfort level
of indoor environment. The VAV systems have two control variables, which is
temperature & airflow rate of the supplied air. Therefore by identifying an optimized
airflow rate & temperature of the supply air, consumption of energy is minimized even
with the constraints like meeting a certain comfort level in each zone of operation.
A steady state model for the energy consumption is being established under the
economizer for the AHU systems, along with an analytical method for optimization for
obtaining a set point of the temperature of the supplied air as well as to minimize cost and
energy consumption. This is because of the fast dynamics of ACMV system, at the time
constant of a minute, compared to the dynamics of systems like building thermal loads
which is in the range of an hour [2]. Hence, the energy consumption of the AHU during
transient time will be inconsequential throughout the cycle of operation.
By using a Kernel extreme learning machine, the airflow ratios & the temperature of
outside air are trained for the analytically obtained model. Using which, the future set
points for supply air temperature is determined for an optimal operation and high energy
saving. |
---|