The Spacecraft Diagnostics Expert System Based-on the Flexible Knowledge Modeling
The knowledge representation of spacecraft has many forms, leading to the establishment of the spacecraft fault knowledge base imperfect. This paper proposed a method of the spacecraft flexible knowledge modeling, making the internal conversion between di
- PDF / 2,112,100 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 17 Downloads / 174 Views
The Spacecraft Diagnostics Expert System Based-on the Flexible Knowledge Modeling Hongzheng Fang, Jian Sun, Liming Han and Haobo Yang
Abstract The knowledge representation of spacecraft has many forms, leading to the establishment of the spacecraft fault knowledge base imperfect. This paper proposed a method of the spacecraft flexible knowledge modeling, making the internal conversion between different kinds of knowledge, such as model, case and rule using the unified knowledge modeling platform. Besides, an integrated diagnostic inference method is proposed. Furthermore, a spacecraft power system diagnostic expert system is implemented. The method and expert system has been applied and verified in the process of the spacecraft operation and maintenance, and has strong practical significance in the field of the spacecraft engineering. Keywords Flexible knowledge
Diagnostics Expert system Knowledge
86.1 Introduction For the spacecraft fault diagnosis system, the diagnostic inference engine capabilities and the correctness and completeness of the knowledge base directly determine the accuracy and efficiency of the spacecraft fault diagnosis. Especially the spacecraft failure knowledge base is a core component of the fault diagnosis system, if the lack of knowledge in database or knowledge error, whatever the
H. Fang (&) J. Sun L. Han H. Yang Beijing Key Laboratory of High-speed Transport Intelligent Diagnostic and Health Management, 100041 Beijing, China e-mail: [email protected] H. Fang J. Sun L. Han H. Yang Beijing Aerospace Measure and Control Corp. Ltd, 100041 Beijing, China
W. Lu et al. (eds.), Proceedings of the 2012 International Conference on Information Technology and Software Engineering, Lecture Notes in Electrical Engineering 211, DOI: 10.1007/978-3-642-34522-7_86, Ó Springer-Verlag Berlin Heidelberg 2013
813
814
H. Fang et al.
diagnosis inference engine cannot arrive at a correct diagnosis conclusions. Therefore establishing reliable and complete fault knowledge is very important. There are some related references. Jet Propulsion Laboratory (JPL) developed diagnostic expert system, SHINE, which is a high speed inference system based on the rules [1]. Besides, the JPL has also developed fault diagnostic software, BEAM, which can learn the normal behavior of the system and make an independent diagnostic for the system failure [2]. The Stennis Research Center diagnosed the J-2X rocket engine failure using the fault tree [3]. The Ames Research Center for NASA developed the diagnosis software Livingstone, which build based on the qualitative behavioral models, and has been used for the Deep Space 1 and other spacecrafts [4, 5]. In 2003 ARC researched the HyDE system, using the mixed model describing the system dynamics [6]. However, the flexible knowledge representations are still need to be researched in-depth.
86.2 The Spacecraft Flexible Knowledge Modeling Technology 86.2.1 Models Knowledge Modeling Method The spacecraft model knowledge representation can make full use of the
Data Loading...
