Principal component analysis (PCA) models are implemented for monitoring and fault detection and diagnosis (FDD) of air-handling units (AHUs), which is modified to solve the problems caused by the nonlinearity and variations in the environment. Independent heat balance and pressure-‘ ow balance models are developed to reduce the effects of the system nonlinearity and to make the PCA method valid in different control modes. Sensor faults are detected and partly isolated using Q-statistic (square prediction error) and Q-contribution plot. Simulation tests are conducted to demonstrate the use of the PCA method for automatic commissioning of AHU monitoring instrumentations and to validate the PCA method in detecting and diagnosing the AHU sensor faults under various typical operating conditions.
Get full access to this article
View all access options for this article.
References
1.
Mah RSH, Tamhane AC. Detection of gross errors in process data. AIChE Journal1982; 28: 828–830.
2.
Staroswiecki M, Comete-Varga G. Analytical redundancy relations for fault detection and isolation in algebraic dynamic systems. Automatica2001; 37: 687–699.
3.
Kumar S, Sinha S, Kojima T, Yoshida H. Development of parameter based fault detection and diagnosis technique for energy efficient building management system. Energy Conversion and Management2001; 42: 833–854.
4.
Dexter AL, Ngo D. Fault diagnosis in airconditioning system: a multi-step fuzzy modelbased approach. HVAC&R Journal2001; 7: 83–102.
5.
McGhee J, Henderson IA, Baird A. Neural networks applied for the identification and fault diagnosis for the process valves and actuators. Measurement1997; 20: 267–275.
6.
Kaszkurewicz E, Bhaya A, Ebecken NFF. A fault detection and diagnosis module for oil production plants in offshore platforms. Expert Systems with Applications1997; 12: 189–194.
7.
Kano M, Nagao K, Hasebe S, Hashimoto I, Ohno H, Strauss R, Bakshi R. Comparison of multivariate statistical process monitoring methods with applications to the Eastman Challenge Problem. Comp. Chem. Eng.2001; 26: 161–174.
8.
Dorr R, Kratz F, Ragot J, Loisy F, Germain J L. Detection, isolation, and identification of sensor faults in nuclear power plants. IEEE Trans. on Control System Tech. 1997; 5: 42–60.
9.
Usoro PB, Schichk LC, Negahdaripour S. An innovation based methodology for HVAC system fault detection. J. Dyn. Syst. Meas. Control1985; 107: 284–285.
10.
Stylianou M, Nikanour D. Performance monitoring, fault detection, and diagnosis of reciprocating chillers. ASHRAE Trans1996; 102: 615–627.
11.
Lee WY, House JM, Shin DR. Fault diagnosis and temperature sensor recovery for an airhandling unit. ASHRAE Trans1997; 103: 621–633.
12.
Wang SW, Wang JB. Automatic sensor evaluation in BMS commissioning of building refrigeration systems. Automation in Construction2002; 11: 59–73.
13.
Wang SW, Wang JB. Law-based sensor fault diagnosis and validation for building airconditioning systems. Int. J. HVAC&R Res. 1999; 5: 353–380.
14.
Pranatyasto TN, Qin SJ. Sensor validation and process fault diagnosis for FCC units under MPC feedback. Control Eng. Prac.2001; 9: 877–888.
15.
Qin SJ, Li W. Detection, identification, and reconstruction of faulty sensors with maximized sensitivity. AIChE Journal1999; 45: 1963–1976.
16.
Dunia R, Qin SJ. Joint diagnosis of process and sensor faults using principal component analysis. Control Eng. Prac.1998; 6: 457–469.
17.
Dunia R, Qin SJ. A unified geometric approach to process and sensor fault identification and reconstruction: the unidimensional fault case. Comp. Chem. Eng.1998; 22: 927–943.
18.
Jolliffe IT. Principal component analysis. New York: Springer, 1986.
19.
Jackson JE. A user's guide to principal components. New York: John Wiley & Sons, 1991.
20.
Wang SW. Dynamic simulation of building VAV air-conditioning system and evaluation of EMCS on-line control strategies. Building and Environment1999; 34: 681–705.
21.
Dunia R, Qin SJ, Edgar TF, McAvoy TJ. Identification of faulty sensors using principal component analysis. AIChE Journal1996; 42: 2797–2812.
22.
Yoon S, MacGregor JF. Fault diagnosis with multivariate statistical methods part I: using steady state fault signatures. J. Process Control2001; 11: 387–400.
23.
Gertler J, Li W, Huang W, McAvoy T. Isolation enhanced principal component analysis. AIChE Journal1999; 45: 323–334.
