Leakage Detection of a Boiler Tube Using a Genetic Algorithm-like Method and Support Vector Machines
In this paper, we propose a method to detect boiler tube leakage using a genetic algorithm (GA)-like method and support vector machines (SVM). The GA-like method allows for selection of significant features, and the SVM detects a leak in boiler tubes using the selected features. Experimental results indicate that the proposed method outperforms a state-of-the-art principle component analysis (PCA) method in leakage detection.
KeywordsBoiler tube Genetic algorithm Support vector machine Tube leakage detection
This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20161120100350, No. 20181510102160, No. 20162220100050).
- 1.Lee, S.B., Roh, S.M.: Developing an early leakage detection system for thermal power plant boiler tubes by using acoustic emission technology. J. Korean Soc. Nondestr. Test. 38(2), 181–187 (2005)Google Scholar
- 2.Kim, J., Kim, J.: Methods and devices for diagnosing facility conditions, Patent Registration No. 10-1818394 (2018)Google Scholar
- 3.Kim, J., Kim, J.: Methods and devices for diagnosing machine faults, Patent Registration No. 10-1797402 (2018)Google Scholar
- 4.Kim, J., Kim, J.: Apparatus and method for machine fault diagnosis, Patent Registration No. 10-1745805 (2017)Google Scholar
- 5.Kim, J., Kim, J.: Machine fault diagnosis method, Patent Registration No. 10-1808390 (2017)Google Scholar
- 6.Kim, J., Kim, J.: Apparatus and method for monitoring machine condition, Patent Registration No. 10-1745805 (2017)Google Scholar
- 7.Kim, J., Kim, J.: Method and apparatus for predicting remaining life of a machine, Patent Registration No. 10-1808461 (2017)Google Scholar