Visit the Energy Systems Laboratory Homepage.
A Qualitive Modeling Approach for Fault Detection and Diagnosis on HVAC Systems
MetadataShow full item record
This paper describes the basics and first test results of a model based approach using qualitative modeling to perform Fault Detection and Diagnostics (FDD) on HVAC and R systems. A quantized system describing the qualitative behavior of a dynamical system is established by transforming numerical inputs into qualitative values or states. Then, the qualitative model is used to determine system-states or outputs that may occur in the future. The qualitative model determines the probability that a subsequent condition might occur. The model can then be used for FDD purposes by comparing the expected states of the faultless system with the occurring states of the real process. The paper presents the first results of the model, trained with measurement data of an air handling unit (AHU) heating coil. The authors plan to extend the model to further AHU components and to test them against real data to assess their performance for FDD and their economic viability in terms of engineering efforts and costs by comparing them with a rule-based FDD system. It is then planned to implement and test the models on several large HVAC and R systems operating at two major European airports in the framework of the FP7 European project CASCADE ICT for Energy Efficient Airports.
Muller, T.; Rehault, N.; Rist, T. (2013). A Qualitive Modeling Approach for Fault Detection and Diagnosis on HVAC Systems. Energy Systems Laboratory (http://esl.tamu.edu); Texas A&M University (http://www.tamu.edu). Available electronically from