An Improved Negative Selection Algorithm-Based Fault Detection Method

Abstract

Fault detection has been an active research field and increasingly important for the safety of technical processes and systems. A variety of fault detection methods have been developed that are generally designed for a specific system; therefore, they target limited types of fault. An efficient approach that could monitor the degradation conditions and be less problem-specific is necessary for improved reliability and efficiency. This paper presents an improved negative selection algorithm using specialized detectors that is model-free and independent of prior knowledge about fault types. An artificial immune system employs a negative selection algorithm and requires only normal (self) patterns for detector generation. In the training phase, uncovered gaps are identified and covered with new detectors to improve nonself (faulty) space coverage. Moreover, to alleviate the online detection, cost reshaping process of nonself space with detector clusters is performed. Theoretical analysis shows that the nonself space coverage with a specialized detector is effectively improved. In experimental analysis, three data sets are used for training and testing, including two-dimensional patterns (cross-mid, ring-mid), Fisher Iris, and KDD CUP99. The experimental results show that the accuracy of the developed method is better than that of the standard negative selection algorithm and other established machine learning algorithms with reduced online anomaly detection time.

KEYWORDS:

• Artificial immune system
• and Negative selection algorithm
• Artificial intelligence
• Clustering
• Detectors
• Fault detection

Additional information

Notes on contributors

A. Abid

Anam Abid received her bachelor's degree in electrical engineering from the University of Engineering and Technology, Peshawar, Pakistan, her master's degree in electrical engineering from the National University of Sciences and Technology, Pakistan and her PhD degree in mechatronics from the Department of Mechatronics Engineering at the University of Engineering and Technology, Peshawar. She also holds faculty position with the Department of Mechatronics Engineering at the University of Engineering and Technology, Peshawar. Her research interests include fault detection, control systems, biomedical engineering systems, machine learning and signal processing. E-mail: [email protected]

M. T. Khan

Muhammad T Khan received his bachelor's degree in mechanical engineering from NWFP University of Engineering and Technology, Peshawar, Pakistan, his master's degree in mechatronics from the University of New South Wales, Sydney, Australia and PhD degree from the University of British Columbia, Vancouver, BC, Canada, in 1997, 1999, and 2010, respectively. He was a postdoctoral fellow with the Industrial Automation Laboratory at the University of British Columbia, Vancouver, BC, Canada for 2 years until January 2012. Currently, he is an associate professor with the Department of Mechatronics at the University of Engineering and Technology, Peshawar, Pakistan. His research interests include robotics and intelligent control systems.

I. U. Haq

Izhar Ul Haq received his PhD degree in manufacturing automation from Loughborough University, UK. Currently, he is working as an assistant professor at Institute of Mechatronics Engineering, University of Engineering and Technology, Peshawar. His research mainly focuses on the design and development of more agile and reconfigurable automation systems using component based technologies. Previously, he was involved in European Commission research project titled “Radically Innovative Mechatronics and Advanced Control Systems”. He also worked on project “Business Driven Automation” funded by Ford Motor Company, UK and Krause Bremen (Germany) and their supply chain partners. E-mail: [email protected]

S. Anwar

Shahzad Anwar is currently serving as a coordinator at Institute of Mechatronics Engineering, University of Engineering and Technology, Peshawar, Pakistan Previously, he served as the Head of Research, Department of Computer Science, at IQRA National University Peshawar. He obtained PhD degree from UWE Bristol (Frenchay Campus) UK. His work focuses on computer vision and artificial intelligence with particular attention into innovative intelligent system techniques. E-mail: [email protected]

J. Iqbal

Javaid Iqbal received the BE degree in mechanical engineering from the University of Engineering and Technology Lahore in 1990, the MS degree from the University of New South Wales, Sydney, Australia in 1998 under the scholarship from NUST for split program with UNSW and PhD degree from University of New South Wales, Sydney, Australia in 2001 under the scholarship from Deans of Faculty of Engineering UNSW, Sydney. He is currently working as a dean in the College of Electrical & Mechanical Engineering (CE&ME), National University of Sciences and Technology (NUST), Rawalpindi, Pakistan. He has extensive experience in the area of mobile robots, vision system and artificial intelligence. He was declared “Best Teacher of NUST – 2005” and received President of Pakistan's Gold Medal. E-mail: [email protected]