An approach to fault detection and isolation for control components in the aircraft environmental control system

Abstract

Control components of the aircraft environmental control system (AECS), which is fast becoming an increasingly complex system, are of significant importance from the viewpoint of safety. However, few studies have focused on fault diagnosis of the AECS. This study proposes a method based on adaptive threshold and parameter extraction (ATPE) to realize fault detection and isolation for control components in the AECS. To overcome the drawback of a fixed threshold for fault detection, a practical approach is employed by combining a radial basis function (RBF)-based observer with an RBF-based adaptive threshold producer. The RBF neural network observer is used to generate a residual error signal. By comparing the residual error signal with the adaptive threshold, fault occurrence can be detected. To improve the fault isolation accuracy, an RBF fault tracker is used; the parameters of this tracker are extracted for fault isolation along with the residual error, unlike in the case of conventional fault diagnosis methods that are based on a single residual error signal. Finally, an RBF-based fault isolator is adopted to realize fault isolation and classification. Two commonly occurring faults in the control components of the AECS are simulated to verify the performance and effectiveness of the proposed method. The experimental results demonstrate that the proposed method based on ATPE is effective for fault detection and isolation for the control components in the AECS.

Keywords

Fault detection fault isolation adaptive threshold control components aircraft environmental control system radial basis function neural network

Get full access to this article

View all access options for this article.

References

Jenab

Rashidi

. Operational reliability assessment of an aircraft environmental control system. Reliab Eng Syst Safety 2009; 94: 456–462.

Lin SQ and Zhao JQ. Simulation research in fault diagnosis of environmental control system for fighter aircraft. In: IEEE circuits and systems international conference on testing and diagnosis, Chengdu, China, 28–29 April 2009, pp. 1–4.

Spengler

Wilson

. Air quality in aircraft. Proc IMechE, Part E: J Process Mechanical Engineering 2003; 217: 323–335.

Zheng YA and Shi ZK. Research on fault diagnosis of temperature control system in aircraft electronic equipment cabin. In: 14th international conference on mechatronics and machine vision in practice, Xiamen, China, 4–6 December 2007, pp. 229–233.

Shah

Liu

Greatrix

. Modelling, simulation, and experimental evaluation of a crossflow heat exchanger for an aircraft environmental control system. Proc IMechE, Part G: J Aerospace Engineering 2010; 224: 613–623.

Hajiyev

Okatan

. Innovation approach to detect the faults in multidimensional dynamic systems. Kybernetes 2010; 39: 127–139.

Venkatasubramanian

Rengaswamy

Yin

. A review of process fault detection and diagnosis, Part I: Quantitative model-based methods. Comput Chem Eng 2003; 27: 293–311.

Garcia

Frank

. Deterministic nonlinear observer-based approaches to fault diagnosis: a survey. Control Eng Pract 1997; 5: 633–670.

Liu HM, Lu C, Hou WK, et al. An adaptive threshold based on support vector machine for fault diagnosis. In: 8th international conference on reliability, maintainability and safety, Chengdu, China, 20–24 July 2009, pp. 907–911.

10.

Wang ZB and Lu C. An adaptive threshold based on RBF neural network for fault detection of a nonlinear system. In: 2011 international conference on computer, communication, control and automation, Zhuhai, China, 19–20 November 2012, LNEE 121, pp. 495–502.

11.

Frank

. Residual evaluation for fault diagnosis based on adaptive fuzzy thresholds. IEEE Colloq Qual Quant Model Meth Fault Diag 1995; 79: 401–411.

12.

Napolitano

Seanor

. A fault tolerant flight control system for sensor and actuator failures using neural networks. Aircraft Des 2000; 3: 103–128.

13.

Napolitano

Windon

Casanova

. Kalman filters and neural network schemes for sensor validation in flight control systems. IEEE Trans Control Syst Technol 1998; 6: 596–611.

14.

Perhinschi

Napolitano

Campa

. An adaptive threshold approach for the design of an actuator failure detection and identification scheme. IEEE Trans Control Syst Technol 2006; 14: 519–525.

15.

Sagoo

Gururajan

Seanor

. Evaluation of a fault-tolerant scheme in a six-degree-of-freedom motion flight simulator. J Aerosp Comput Inform Commun 2010; 7: 47–67.

16.

Campa G, Fravolini ML, Naplilano M, et al. Neural networks-based sensor validation for the flight control system of a B777 research model. In: American control conference, Anchorage, Alaska, USA, 8–10 May 2002, pp. 412–417.

17.

Zhang

Ran

. Research of fault diagnosis in flight control system based on fuzzy neural network. Procedia Eng 2011; 15: 70–74.

18.

Guo L, Tian Y and Fang M. Fault detection and diagnosis for nonlinear system based on neural network on-line approximator. In: Proceedings of the 6th world congress on intelligent control and automation, Dalian, China, 21–23 June 2006, pp. 5530–5534.

19.

Noman

Shamsuddin

Hassanien

. Hybrid learning enhancement of RBF network with particle swarm optimization. Adv Neural Networks 2009; 5553: 19–29.

20.

Samy

Postlethwaite

. Sensor fault detection and accommodation using neural networks with application to a non-linear unmanned air vehicle model. Proc IMechE, Part G: J Aerospace Engineering 2010; 224: 437–447.

21.

Lee

Chung

Tsai

. Robust radial basis function neural networks. IEEE Trans Syst Man Cybern B: Cybern 1999; 29: 674–685.

22.

Shi

Lennox

. The development of an adaptive threshold for model-based fault detection of a nonlinear electro-hydraulic system. Control Eng Pract 2005; 13: 1357–1367.

23.

Le K, Huang Z, Moon CW, et al. Adaptive thresholding - A robust fault detection approach. In: Proceedings of the 36th conference on decision & control, San Diego, California, USA, 1997, pp.4490–4495.

24.

Zhang

Tan

. An RBF neural network classifier with centers, variances and weights directly determined. Comput Technol Autom 2009; 28: 5–9.

25.

Schwenker

Kestler

Palm

. Three learning phases for radial-basis-function networks. Neural Networks 2001; 14: 439–458.

26.

Taghi

Baghmisheh

Pavesic

. Training RBF networks with selective backpropagation. Neurocomputing 2004; 62: 39–64.

27.

Ince

Kiranyaz

Gabbouj

. Evolutionary RBF classifier for polarimetric SAR images. Exp Syst Appl 2012; 39: 4710–4717.