A Fault Diagnosis Technique for Water-Steam Chemistry Process in Power Plant

Abstract:

Carrying out the fault diagnosis of water-steam chemistry process in power plant has an important significance to ensuring high qualified rates of water and steam quality as well as maintaining safe operation of units. This paper proposed a fault diagnosis method based on improved credibility theory which is used to construct fuzzy diagnosis rules and data mining technique used to determine symptom weights and rule limens of reliability rules, and also improved the setting method of rule confidence. The adoption of data mining technique and new setting method of rule confidence can solve the problem that credibility reasoning results are influenced by person’s subjective factors, and make the reasoning process more scientific. Example results prove that this diagnosis model has a high accuracy, which indicates the significant practical value of the model.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 6467-6472
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.6467
• Online since: January 2012
• Authors: Shun An Cao, Jia Yuan Hu, Yan Huang, Jian Li Xie
• Keywords: Data Mining (DM), Fault Diagnosis, Improved Credibility Reasoning, Power Plant, Water-Steam Chemistry Process

References

[1] Li YR, The Research on the Application of Fault Diagnosis Based on Fuzzy Expert System in Thermal Power Plant. Beijing: North China Electric Power University, (2006).

[2] Peng G, A review of several inaccurate inference models in common use, Journal of Guang Zhou Normal University, v19, n7, pp.71-77, (1998).

[3] Wright Edward J, Laskey Kathryn Blackmond, Credibility models for multi-source fusion, 9th International Conference on Information Fusion, Florence, pp.1-7, July (2006).

DOI: 10.1109/icif.2006.301693

[4] Yin CQ, Yin H, Artificial intelligence and expert system. Beijing: Water and Electricity press, (2001).

[5] Agrawal R, Imielinski T, Swami A, Mining Association Rules between Sets of Items in Large Databases, Proc of the ACM SIGMOD International conference on Management of Data, Washington, v22, n2, pp.207-216, Jun (1993).

DOI: 10.1145/170036.170072

[6] Hu XQ, Study of Multi-Level Association Rules Mining Alorithm Based on FP-Tree. Chongqing University, (2005).

[7] Ao ZG, Artificial intelligence and expert system introduction Hefei: University of Science and Technology of China Press, (2002).

[8] GB/T12145-1999, Quality criterion of water and steam for generating unit and steam power equipment.