Subsystem level fault diagnosis of a building's air-handling unit using general regression neural networks
John M. House and
Applied Energy, 2004, vol. 77, issue 2, 153-170
This paper describes a scheme for on-line fault detection and diagnosis (FDD) at the subsystem level in an Air-Handling Unit (AHU). The approach consists of process estimation, residual generation, and fault detection and diagnosis. Residuals are generated using general regression neural-network (GRNN) models. The GRNN is a regression technique and uses a memory-based feed forward network to produce estimates of continuous variables. The main advantage of a GRNN is that no mathematical model is needed to estimate the system. Also, the inherent parallel structure of the GRNN algorithm makes it attractive for real-time fault detection and diagnosis. Several abrupt and performance degradation faults were considered. Because performance degradations are difficult to introduce artificially in real or experimental systems, simulation data are used to evaluate the method. The simulation results show that the GRNN models are accurate and reliable estimators of highly non-linear and complex AHU processes, and demonstrate the effectiveness of the proposed method for detecting and diagnosing faults in an AHU.
Keywords: Fault; detection; and; diagnosis; Air-handling; unit; General; regression; neural-network (search for similar items in EconPapers)
References: View complete reference list from CitEc
Citations View citations in EconPapers (9) Track citations by RSS feed
Downloads: (external link)
Full text for ScienceDirect subscribers only
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:eee:appene:v:77:y:2004:i:2:p:153-170
Ordering information: This journal article can be ordered from
http://www.elsevier. ... 405891/bibliographic
Access Statistics for this article
Applied Energy is currently edited by J. Yan
More articles in Applied Energy from Elsevier
Series data maintained by Dana Niculescu ().