Abstract (Summary)
Heating, Ventilation and Air Conditioning (HVAC) systems are an integral part of contemporary life. People experience the direct effect of their operation while being in factories, hospitals, universities, homes and other facilities where people work or live. The main function of HVAC systems is to create and maintain a comfortable microclimate for the occupants of a building. The comfort is achieved by varying the temperature, humidity and cleanliness of the air that the occupants breathe and operate in. These systems have been constantly evolving and improving to become more reliable, economical and autonomous. To a large extent, the improvements have been possible due to the microprocessor technology, which introduced limited self-tuning, self-control and fault detection capabilities. These capabilities are realized with the help of special programs written by HVAC operators in specialized control languages. Unfortunately, even with the systems becoming more and more intelligent, human intervention is often still required to identify conditions that the programs are unable to detect. Building on the prior research results, this thesis seeks to demonstrate that knowledge-based fault detection and diagnosis (FDD) for HVAC systems is feasible and effective. The implementation utilizes a Java Expert System Shell (JESS) to provide the FDD reasoning. Since JESS does the reasoning based on rules written in the JESS language and each HVAC system is unique in its design and configuration, we also designed a Java-based JESS rule generator. The latter generates expert system rules for a particular HVAC system based on an XML description of the target HVAC system. To be more precise, we used the Industry Foundation Classes (IFC) dialect of XML to provide these descriptions. From the very beginning of the research, one of our goals was to provide an Internet friendly solution to the problem of HVAC FDD. That is why the implementation integrates several Internet technologies such as Java and XML. The quality of FDD can be further improved by collecting more formalized knowledge of the professionals in the HVAC maintenance and design.
Bibliographical Information:


School:University of Cincinnati

School Location:USA - Ohio

Source Type:Master's Thesis

Keywords:hvac fdd


Date of Publication:01/01/2002

© 2009 All Rights Reserved.