Durham School of Architectural Engineering and Construction


Date of this Version

Spring 5-20-2011

Document Type



A THESIS Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Master of Science, Major: Architectural Engineering, Under the Supervision of Professor Haorong Li. Lincoln, Nebraska: April 2011
Copyright 2011 Daihong Yu


Virtual sensing technology aims to estimate difficult to measure, expensive, or new quantities by using multifarious mathematical models along with non-invasive and low-cost measurements. Such embedded intelligence is a key to improving the performance of building systems in terms of functionality, safety, energy efficiency, environmental impacts, and costs. Considering the progress that has been achieved over many various fields (e.g., process controls, automobiles, avionics, autonomous robots, telemedicine) within the last two decades, numerous intelligent features have been incorporated and enabled that would otherwise not be possible or economical.

To identify the potential opportunities and research/development needs of virtual sensing technology in building systems,

First, this thesis reviews the major milestones of virtual sensing development in other emerging fields and its formulation of development methodologies.

Second, the state-of-the-art in virtual sensing technology in building systems is summarized as a starting point for its future developments and applications.

After that, a cost-effective virtual supply airflow (SCFM ) meter for rooftop air-conditioning units (RTUs) is created by using a first-principle model in combination with accurate measurements of virtual or virtually calibrated temperature sensors (a virtual mixed air temperature sensor and a virtually calibrated supply air temperature sensor) as a supplementary example. Modeling of the virtual meter, uncertainty analysis, and experimental evaluation are performed through a wide range of laboratory testing in the development. The study reveals that the first-principle based virtual SCFM meter could accurately predict SCFM values for RTUs (uncertainty is ± 6.9%). This innovative application is promising with a number of merits, such as high cost-effectiveness, ease-of-implementation, long-term availability after one-time development, and generic characteristics for all RTUs with gas heating.

Significant research and developments are needed before virtual sensors become commonplace within buildings. It is believed a wealth of virtual sensing derived applications would facilitate the sustainable management and optimize the advanced controls in building systems. It is hoped that this study can provide a resource for future developments.