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.
Generic rating-data based (GRDB) modeling and validation for DX unitary HVAC equipment
This dissertation presents an effective and easy to implement cooling modeling methodology for DX coils. The proposed model was created using publicly available HVAC manufacturer rating data and a reconstruction of a traditional cooling model. Specifically, a wet curve was constructed based on the manufacturer's rating data for the wet coil condition. Extrapolating a dry line from the critical point of this wet curve gave the dry coil conditions. Multiple second-order polynomials as a regression method estimated the wet curves, and then the critical points were solved through a quadratic equation. Lab data were compared with manufacturers' rating data to first validate the GRDB's effectiveness, and it indicated that the method accurately predicts both wet-coil and dry-coil conditions (e.g., the average absolute relative error is below 4.4%). Then, the DX air cooling coil modeling method in EnergyPlus was summarized and compared with the GRDB using six rooftop units of two manufacturers. It demonstrated that the relative errors in sensible cooling capacity prediction ranged from −14.9% to 15.1% for EnergyPlus and from −7.4% to 4.2% for the GRDB method. In addition to higher accuracy and precision, the GRDB method is more robust against the variations in parameter selections, has a wider application range, requires less computation power, and is more straightforward. Next, a new validation method, i.e., the self-validation by manufacturer's data, is developed based on existence theorem. The manufacturer's data is divided into training part and validation part by each independent variable's extreme, and a model is then validated by the two parts. With case studies on four split systems of four manufacturers, it showed that the self-validation matches the lab validation, and the quality of manufacturer's data determines the GRDB's performance. The self-validation method can also be extended to other conditions where there are insufficient or even no lab data as validation data. Finally, simplification of the GRDB modeling is studied based on the analysis of partial differentials and using the less dimensional original manufacturer's data.
Yang, Huojun, "Generic rating-data based (GRDB) modeling and validation for DX unitary HVAC equipment" (2012). ETD collection for University of Nebraska - Lincoln. AAI3504215.