Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.

Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Fault detection and diagnostics for centrifugal chillers

Xinzhi Zhao, University of Nebraska - Lincoln

Abstract

Centrifugal chillers are large contributors of energy usage and maintenance costs for Heating, Ventilating, and Air Conditioning (HVAC) systems. The performance of centrifugal chiller degrades gradually due to faults introduced during initial installation or developed in routine operation. Fault detection and diagnostic (FDD) technology has potential to guard against faults to reduce energy penalty, equipment downtime, and maintenance costs. This research aimed to: 1) evaluate the effectiveness of three promising fault detection and diagnostics (FDD) methods for centrifugal chillers online in laboratory and field environments; and (2) identify a specification for one or more algorithms that can be incorporated within commercial FDD products to monitor the health of the chiller's operation. ^ Based on a comprehensive understanding of the chiller system, cause of the faults, impact of the faults, and associated implementation cost considerations, several decoupling features for common chiller faults were developed. The performance of the proposed decoupling features was evaluated with the historical laboratory data. The validation results showed that the proposed decoupling features could be implemented together to handle multiple simultaneous faults. ^ Three FDD methods were tested in a laboratory environment. Based on a FDD evaluation methodology, the overall FDD performance of the three FDD methods was evaluated. Based on the laboratory tests results, two of the three methods were identified for the subsequent field implement and evaluation. ^ The two selected FDD methods were fully implemented online on two field chillers produced by Manufacturer A. Several occurrences of multiple simultaneous faults were found on the two chillers. The decoupling-based FDD method was then identified as the better of the two evaluated methods. A few of the decoupling features were modified and implemented on four field chillers from Manufacturers B and C. One condenser fouling fault was detected in one chiller. The field test results showed that the proposed decoupling features could be combined together as a chiller FDD tool or be integrated into the onboard controllers to monitor and diagnose common chiller faults.^

Subject Area

Engineering, Industrial|Engineering, Mechanical

Recommended Citation

Zhao, Xinzhi, "Fault detection and diagnostics for centrifugal chillers" (2011). ETD collection for University of Nebraska - Lincoln. AAI3482809.
http://digitalcommons.unl.edu/dissertations/AAI3482809

Share

COinS