Using SVM based method for equipment fault detection in a thermal power plant Kai-Ying Chen a , Long-Sheng Chen b, *, Mu-Chen Chen c , Chia-Lung Lee d a Department of Industrial Engineering and Management, National Taipei University of Technology, Taipei, Taiwan, ROC b Department of Information Management, Chaoyang University of Technology, Taichung, Taiwan, ROC c Institute of Traffic and Transportation, National Chiao Tung University, Taipei, Taiwan, ROC d Shanghai FirstTech Company Limited, China 1. Introduction Thermal power plants fired by fossil fuels are one of the primary sources of noxious greenhouse gas emissions, producing, carbon dioxide. Even so, they are still the major source of supplying electricity in Taiwan. According to the annual report of the Taiwan Power Company (TPC), the total power generation of their eight thermal power plants exceeds 70% of the total energy generated nowadays [3] in Taiwan. Consequently, due to growing demands on electricity, how to improve the efficiency of equipment in a thermal power plant has become a critical issue. Huang et al. [6] indicated that the efficiency and availability depend heavily on high reliability and maintainability. In order to raise efficiency, the equipment of thermal power plants is becoming larger and more complex. However, due to lack of manpower and information resources, the diagnosis and repair of failed equipment cannot usually be performed immediately. From lots of published articles [58–62], we can find that to identify the failure types of steam turbines and their root causes is time consuming. It needs professional knowledge regarding materials and mechanical engineering. Generally speaking, thermal power plant engineers can merely handle routine or uncomplicated maintenance tasks. Additional tests and expert advice are additionally required from the technical support of original equipment manufacturers for complex fault diagnosis and maintenance, although these additional tests are often costly and involve some risk to equipment [8]. Hence it leads to long downtimes for equipment and causes significant production losses [11]. In order to reduce the cost of maintenance and risky experiments, the concept of e-maintenance has been introduced to identify the root cause of component failure, to reduce the failures of production systems, to eliminate costly unscheduled shutdown maintenances, and to improve productivity [12]. In an e-maintenance system, the intelligent fault detection system plays a crucial role for identifying failures. Data mining techniques are the core of such intelligent systems and can greatly enhance their performance [6,8,9]. Applying these techniques to fault detection makes it possible to eliminate additional tests or experiments which usually involve high expense and highly risk [8]. Recently, several data mining techniques such as artificial neural networks, fuzzy logic systems, genetic algorithms, and rough set theory have all been employed to assist the detection and condition monitoring tasks [4,10]. For example, Yang and Liu [8] presented a hybrid-intelligence data mining framework which involves an attribute reduction technique and rough set theory to diagnose the faults of boilers. Shu [45] established an interactive data mining approach based inference system to solve the basic technical challenge and speed up the discovery of knowledge in nuclear power plant. Besides, some related works designed data Computers in Industry 62 (2011) 42–50 ARTICLE INFO Article history: Received 19 February 2009 Received in revised form 25 May 2010 Accepted 31 May 2010 Keywords: Thermal power Maintenance Data mining Support vector machines Classification ABSTRACT Due to the growing demand on electricity, how to improve the efficiency of equipment in a thermal power plant has become one of the critical issues. Reports indicate that efficiency and availability are heavily dependant upon high reliability and maintainability. Recently, the concept of e-maintenance has been introduced to reduce the cost of maintenance. In e-maintenance systems, the intelligent fault detection system plays a crucial role for identifying failures. Data mining techniques are at the core of such intelligent systems and can greatly influence their performance. Applying these techniques to fault detection makes it possible to shorten shutdown maintenance and thus increase the capacity utilization rates of equipment. Therefore, this work proposes a support vector machines (SVM) based model which integrates a dimension reduction scheme to analyze the failures of turbines in thermal power facilities. Finally, a real case from a thermal power plant is provided to evaluate the effectiveness of the proposed SVM based model. Experimental results show that SVM outperforms linear discriminant analysis (LDA) and back-propagation neural networks (BPN) in classification performance. ß 2010 Elsevier B.V. All rights reserved. * Corresponding author. E-mail addresses: kychen@ntut.edu.tw (K.-Y. Chen), lschen@cyut.edu.tw (L.-S. Chen), ittchen@mail.nctu.edu.tw (M.-C. Chen). Contents lists available at ScienceDirect Computers in Industry journal homepage: www.elsevier.com/locate/compind 0166-3615/$ – see front matter ß 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.compind.2010.05.013