Fault diagnosis in HVAC chillers

被引：47

作者：

Choi, K ^{[1
]}

Namburu, SM ^{[1
]}

Azam, MS ^{[1
]}

Luo, JH ^{[1
]}

Pattipati, KR ^{[1
]}

Patterson-Hine, A ^{[1
]}

机构：

[1] Univ Connecticut, Dept Elect & Comp Engn, Storrs, CT 06269 USA

来源：

IEEE INSTRUMENTATION & MEASUREMENT MAGAZINE | 2005年 / 8卷 / 03期

关键词：

D O I：

10.1109/MIM.2005.1502443

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

A data-driven technique was used for fault detection and isolation (FDI) of HVAC chillers employing generalized likelihood ratio (GLR) test for multiway dynamic principal component analysis (MPCA), multiway partial least square (MPLS), and support vector machines (SVM). The classifications via MPCA, MPLS and SVM including severity estimation process provided accurate results of fault severity levels. These techniques were applied successfully to real chiller data provided by ASHRAE.

引用

页码：24 / 32

页数：9

共 14 条

[1]

Alpaydin E., 2004, Introduction to Machine Learning, V2nd

[2]

[Anonymous], 2004, MULTIWAY ANAL APPL C

[3]

Basseville M., 1993, PRENTICE HALL INFORM

[4]

BENDAPUDI S, 20028 HL PURD U

[5] Fault diagnosis using support vector machine with an application in sheet metal stamping operations [J].

Ge, M ;

Du, R ;

Zhang, GC ;

Xu, YS .

MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2004, 18 (01) :143-159

[6] A comparison of multiway regression and scaling methods [J].

Gurden, SP ;

Westerhuis, JA ;

Bro, R ;

Smilde, AK .

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2001, 59 (1-2) :121-136

[7] A comparison of methods for multiclass support vector machines [J].

Hsu, CW ;

Lin, CJ .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (02) :415-425

[8]

Jackson JE, 1991, A user's guide to principal components

[9]

JUNGHUI C, 2002, CHEM ENG SCI, V57, P63

[10]

NAMBURU M, 2005, P SPIE C ORL FL MAR

← 1 2 →