Research on Correlation Mining of Compressor Failure Factors based on Apriori Algorithm
In order to reduce the influence of uncertainty factors during the operation of centrifugal compressors, text data mining method is used to extract text data from 942 fault records of centrifugal compressors of a pipeline. The 942 data were classified by 5M1E analysis, and the data set was scanned several times based on Apriori algorithm, and the support, confidence and boost of the itemsets were calculated to discover all the frequent itemsets, so as to generate association rules. The results show that by setting the thresholds of support and confidence, 22 strong association rules are obtained, and "the degree of aging of compressor components" and "the degree of performance degradation of compressor components" are found to have higher support and more association rules, which provides a good solution for the evaluation and control of the factors affecting compressor failure. This provides an effective reference for evaluating and controlling the factors affecting compressor failure.
References
- Zhao Jingyan, Ge Kai, Chu Chenkeng, et al. Current application and prospects of China-made natural gas compressors[J]. Natural gas industry, 2015, 35(10): 151-156.
- Yang Lan, Feng Siyang. Current status and development trend of centrifugal compressor fault diagnosis research[J]. China Science and Technology Journal Database Industry A, 2022, (4): 58-60.
- Xing Jianwen, Liang Xi, Li Wei. Research on compressor fault location method based on HAZOP analysis and Bayesian network[J]. Automation Instrumentation, 2022, 43 (3): 11-14,19.
- Zhu Yongren, Cai Jie, Shan Yingying. A fault diagnosis method for centrifugal compressors based on ant colony clustering algorithm[J]. Oil and Gas Storage and Transportation, 2019, 38(4): 424-428.
- Xu Ye, Huang Wenjun, Mi Junpeng, et al. Surge diagnosis method of centrifugal compressor based on multi-source data fusion[J]. CIESC Journal, 2023, 74(7): 2979-2987.
- Jun HB, Kim D. A Bayesian network-based approach for fault analysis[J]. Expert Systems with Applications, 2017, 81: 332-348.
- Spuntrup F.S, Londono J.G, Skourup C, et al. Reliability improvement of compressors based on asset fleet reliability data[J]. IFAC-Papers Online, 2018, 51(8): 217-224.
- Golmoradi M, Ebrahimi E, Javidan M. Compressor fault diagnosis based on SVM and GA[J]. Vibroengineering Procedia, 2017, 12: 49–53.
- AGRAWAL R, SRIKANT R. Fast algorithms for mining association rules[C]//Proceedings of the 20th International Conference on Very Large Data Bases. Santiago: Morgan Kaufmann, 1994: 487-499.
- Liu Wenya, Xu Yongneng. Association Rule Mining of Metro Failures Based on lmproved Apriori Algorithm[J]. Journal of Weapons and Equipment Engineering, 2021, 42(12): 210-215.
- Arshiya. Trends on Data Mining Concepts and its Techniques[J]. Test Engineering and Management, 2020,Vol.83.
- Tobback, E.;Naudts, H. Belgian economic policy uncertainty index: Improvement through text mining[J]. International Journal of Forecasting, 2018, Vol.34(2): 355-365.
- Zheng Binbin, Feng Tingting, Wang Jiahe, et al. Causes and correlation analysis of urban gas accidents based on text mining[J]. Chinese Journal of Safety Science, 2023, 33(7): 190-195.
- Li Tangzhenhao, You Xiaoyue. Analysis of quality influencing factors of assembly building based on Apriori[J]. Journal of Tongji University (Natural Science Edition), 2022, 50 (2): 147-152.
- Cheng Congcong, Zhao Yicheng, Jiag Linjing, et al. Correlation Analysis on the Hidden Risk Factors of Tailings Pond Based on Text Mining[J]. Mining Research and Development, 2021, 41(11): 26-33.