Return to search

Modeling of ECM Controlled Series Fan-powered VAV Terminal Units

Semi-empirical models for series fan-powered variable air volume terminal units
(FPTUs) were developed based on models of the primary, plenum, fan airflow and the
fan power consumption. The experimental setups and test procedures were developed
respectively for primary, plenum and fan airflow to test each component of the FPTUs at
typical design pressures and airflows. Two sizes of the terminal units from three
manufacturers were used in this study. All of the FPTUs were equipped with
electronically commutated motors (ECM). Data provided by the models were compared
against the data from previous experiments to prove the models’ validity. Regression
modeling was performed by using SigmaStat.
The model of primary airflow had an R2 above 0.948 for all the terminal units
evaluated while the plenum airflow model had an R2 above 0.99. For all the terminal
units, the R2 of the fan airflow model was ranged from 0.973 to 0.998. Except for one fan, the fan power consumption model was able to characterize the power performance
and had an R2 above 0.986.
By combining the airflow and power models, the model for series FPTU was
developed. Verification was made to prove the FPTU model’s validity by comparing the
measured and predicted data of airflow and power consumption. Correction factors were
used in the primary airflow model to compensate for the difference caused by large
measurement errors and the system effects. The predicted values were consistent with
measurements and no offset was needed in the primary airflow model. Generally, the
newly established model was able to describe the airflow performance as well as power
consumption of series FPTUs without adding complexity.

Identiferoai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/ETD-TAMU-2010-08-8465
Date2010 August 1900
CreatorsYin, Peng
ContributorsO'Neal, Dennis
Source SetsTexas A and M University
Languageen_US
Detected LanguageEnglish
Typethesis, text
Formatapplication/pdf

Page generated in 0.0016 seconds