Fault Detection and Diagnosis of Air Handling Unit in HVAC System Using Cloud-based Data Logging System

Abstract

This project encompasses the engineering principles of design and testing of the implementation of a fault detection and diagnosis (FDD) using cloud-based data logging system on Heating, Ventilation, and Air Conditioning (HVAC) system. The data acquisition system is a compact measurement system made by Paragon Robotics that handles data from sensors, measurement hardware, and a computer with appropriate software to record four different categories of measurements: (1) temperature and humidity of outside air, (2) temperature of return, mixed, and supply air, (3) cooling oil valve operation measurements, and (4) supply air fan operation measurements. The project includes the design process of the equipment and installation of electrical and mechanical components needed for a functioning system on an Air Handling Unit (AHU) that serves a building in the University of Oklahoma. Tests of baseline operation are conducted to collect weeks of data after calibration. The system is verified with engineering fundamental Models to ensure robustness for fault detection and diagnosis. Common faults of AHU are evaluated and selected to be simulated into the system for days with faulty operation data. Qaulitative rule-based or physics-based FDD technique is used to detect and diagnose the system with data collected during operations of faults.