Research on the Knowledge Storage Methods of SPF Tree

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 217)

Abstract

To use the historical records in fault diagnosis efficiently, we have put forward out a method of SPF (system-phenomenon-fault tree), and the method gives us a novel way to solve the problem of fault diagnosis. To improve the efficiency of SPF, we study the knowledge storage method of SPF in the paper. Reviewed the concept and characteristics firstly, came out an complex component storage method of “linear lists + Linked list,” solved the storage method of components and their relationships; used random storage method to storage component and failure phenomenon, fault and solution considering their relationships of peer-to-peer; proposed “Breadth-first + Strength-first” to enhance the retrieval hit rate and reduce the time cost; we defined the data structures and implemented the storage methods, which proved its correctness and efficiency.

Keywords

Knowledge storage Fault diagnosis Knowledge engineering SPF tree 

Notes

Acknowledgments

This research is supported by National Science Fund of China (51275547), Chongqing Science and Technology Research Projects (cstc2011pt-gc70007, cstc2012gg-yyjs40019, cstc2012jjA1481, cstc2012cx-rkxA006), Transformation Project of Chongqing Municipal Education Commission (kjzh11221, KJ121413, 201028), and Research Foundation of Chongqing University of Science & Technology (CK2010Z10, CK2011B01).

References

  1. 1.
    Ge Z, Yang Y, Hu Z et al (2007) Unscented particle filter and log likelihood ration based fault diagnosis of nonlinear system in non-Gaussian noises. Chin J Mech Eng 67:151–156 Google Scholar
  2. 2.
    Liu Z, Dou W, Wang D et al (2007) Rotating machinery fault diagnosis combination of method based on genetic algorithm. J Mech Eng 21(07):287–291Google Scholar
  3. 3.
    Xu X, Zhou X (2008) Excavator’s fault diagnosis expert system based on decision tree and rule engine. Hydraulic Pneumatic 14:52–55Google Scholar
  4. 4.
    Chen G, Yan P, Yi R, Liu F (2011) Fault diagnosis method based on system-phenomenon-fault tree. Chin J Mech Eng 03:21–24Google Scholar
  5. 5.
    Karen AR, John DA (2002) A fault tree analysis strategy using binary decision diagrams. Reliab Eng Syst Saf 07:203–206Google Scholar
  6. 6.
    Vince A (2002) A framework for the greedy algorithm. Discrete Appl Math 73:26–27MathSciNetGoogle Scholar
  7. 7.
    Goulermas JY, Liatisis P, Zeng X et al (2007) Density-driven generalized regression neural networks (DD-GRNN) for function ap-proximation. IEEE Trans on Neural Netw 32(06):332–337Google Scholar
  8. 8.
    Spielman SE, Thill JC (2008) Social area analysis, data mining, and GIS. Comput Environ Urban Syst 05:12–17Google Scholar
  9. 9.
    Beaubouef T, Petry FE, Ladner R (2007) Spatial data methods and vague regions: A rough set approach. Appl Soft Comput 11(03):34–37Google Scholar
  10. 10.
    Wang CH (2008) Recognition of semiconductor defect patterns using spatial filtering and spectral clustering. Expert Syst Appl 37:142–146Google Scholar
  11. 11.
    Ping Y, Guorong C, Yan P, Chen GR (2009) A production schedule obtaining method based on information spreading relationship network in manufacturing process 62:156–162 Google Scholar

Copyright information

© Springer-Verlag London 2013

Authors and Affiliations

  • Guorong Chen
    • 1
  • Juli Deng
    • 1
  • Jinliang Shi
    • 1
  • Jun Zhou
    • 1
  1. 1.Information EngineeringChongqing University of Science and TechnologyChongqingChina

Personalised recommendations