Academic Journal of Engineering and Technology Science, 2024, 7(6); doi: 10.25236/AJETS.2024.070604.
Chen Yonggang1, Kong Jing2, Jiang Haiming2
1Hudian Co. LTD Shanghai Branch, Shanghai, China
2Huadian Electric Power Research Institute Co. LTD, Hangzhou, China
Gas turbine generator units operate in high-risk environments with a high incidence of failures, necessitating an effective and timely alarm system to ensure their safe operation. However, existing alarm systems commonly use fixed-threshold alarms and lack correlation analysis between variables, leading to false negatives or false positives at certain normal operating points. This paper proposes applying the Dynamic Multivariate State Estimation Technique (DMSET) for real-time assessment of thermal power equipment operational states. The intelligent warning algorithm based on DMSET is designed to evaluate the health status of systems and equipment. Testing results indicate that, under normal operation, the Euclidean distance between the estimated vector and the real-time observation vector is short, resulting in high prediction accuracy; whereas, in abnormal conditions, the Euclidean distance is significantly larger than that in normal states. The study demonstrates that the proposed DMSET intelligent warning algorithm can sensitively detect abnormal information in equipment.
Gas Turbine; Intelligent Alarm; Dynamic Multivariate State Estimation Technique; Euclidean Distance
Chen Yonggang, Kong Jing, Jiang Haiming. Gas turbine generator unit intelligent alarm system. Academic Journal of Engineering and Technology Science (2024) Vol. 7, Issue 6: 23-28. https://doi.org/10.25236/AJETS.2024.070604.
[1] GE. Accelerating Growth in Natural Gas Power towards a Zero-Carbon Future [R]. GE General Electric Company, 2022.
[2] Wang Jiandong. Intelligent Industrial Alarm System (First Edition) [M]. China Machine Press, 2021.
[3] Xue Xiaodan. Research on Intelligent Alarm Technology in Intelligent Power Generation Operation Control [D]. North China Electric Power University (Beijing), 2021.
[4] Zhang Mingjun, Han Changlin, Wang Guiping. Research on "Integrated Control" Intelligent Alarm Technology for the Lower Cascade of Jinsha River [J]. Hydropower Station Electromechanical Technology, 2017, 40(07): 9-12.
[5] Ji Hongjin. Multivariate Methods for Geochemical Background and Anomaly Division [J]. Journal of Changchun College of Geology, 1988(03): 311-320.