Development of Air Flow Rate Prediction Model Using Multiple Regression in VAV Terminal Unit

Abstract

Accurate measurement of air flow rate is essential in automatic building control using the variable air volume (VAV) system. In order to solve the problems of the existing air flow measurement method and improve the accuracy of air flow control, this study developed a data-based multiple regression air flow prediction model. The independent variables used in the development of the predictive model were selected as the factors used for control and monitoring when operating with variable air flow rate in the existing air conditioning system. Data collection and correlation between independent variables and air flow rate of the terminal unit were analyzed. Using the IBM SPSS statistics version 25, an air flow rate prediction model was developed using multiple regression analysis. Reliability of model was evaluated by comparing the measured airflow. The relative error of −9.3% to 10.4% is shown when comparing the estimated air flow rate by the developed model with the measured air flow rate.