Afef Fekih talks with
ScienceWatch.com and answers a few questions about
this month's New Hot Paper in the field of Computer
Article Title: Neural networks based system
identification techniques for model based fault detection
of nonlinear systems
A;Xu, H;Chowdhury, FN
Journal: INT J INNOV COMPUT INF CONTRO
Year: OCT 2007
* Univ SW Louisiana, Dept Elect & Comp Engn, POB 43890,
Lafayette, LA 70504 USA.
(addresses have been truncated)
Why do you think your paper is highly
I think the paper is highly cited because it describes the development and
application of a new structure of a neural networks-based system
identification technique for nonlinear systems with the specific goal of
real-time residual generation for fault detection purposes. The technology
was tested on a Boeing 747 model but is of general interest to several
complex technological systems.
Does it describe a new discovery, methodology, or
synthesis of knowledge?
This paper describes a new structure of partially connected neural networks
for real–time residual generation in nonlinear systems. For nonlinear
systems, the task of residual generation is sometimes complicated by the
size of the problem, or by the lack of a suitable model from where the
residual can be generated.
"This research is going to change
the way we design technological systems in
This paper develops and implements a new structure of partially connected
neural networks for such systems and successfully implements it on a Boeing
747 aircraft model. It also provides a comparison between the performance
of the proposed partially connected neural networks structures and the
fully connected one.
Would you summarize the significance of your paper
in layman's terms?
Residual generation is an essential part of model-based fault detection
schemes. This paper presents a new tool for residual generation in
nonlinear systems and its application to aircraft systems.
How did you become involved in this research, and
were there any problems along the way?
I got involved in this research through my participation as a Co-PI for a
research project funded by NASA. I have faced few technical challenges
during the course of this research, as it was a relatively new research
topic for me. My prior research work was not specifically on this subject.
Where do you see your research leading in the
Fault detection and identification technology is fast becoming an issue of
primary significance in intelligent and autonomous control system design
since it provides the prerequisites for increased reliability, safety, and
system availability, automation of inspection procedures, and minimization
and maintenance activities and cost. Real-time FDI would insure high
performance of technological systems even with impairments to the
actuators, sensors, or control surface, and thus increase the system's
survivability, and probability of mission success.
Do you foresee any social or political implications
for your research?
Yes. This research is going to change the way we design technological
systems in the future. Research into FDI technology is going to grow in
importance with the increased complexity of engineering systems along with
the stringent requirements on reliability, safety, and performance.
Afef Fekih, Ph.D.
Harold Callais Memorial/BORSF Professor in Electrical & Computer
University of Louisiana at Lafayette
Electrical & Computer Engineering Department
Lafayette, LA, USA Web
Keywords: neural networks-based system identification technique
for nonlinear systems, real-time residual generation, fault detection
purposes, Boeing 747, research into FDI technology.