Abstract
Blade fracturing is a catastrophic failure to gas turbine. Condition monitoring is one of the effective methods to improve the reliability during operation. However, there are always existing two problems with blade monitoring: 1) difficult to catch the precise blade characteristic frequencies under complex conditions and 2) challenging to give early warning of blade failure. In this article, we attempt to solve these problems by means of reconstructing blade passing frequency waveform. Firstly, a novel method, high-resolution harmonic product spectrum (HR-HPS), is proposed to accurately extract the fundamental rotating frequency and to calculate blade characteristic frequencies from gas turbine casing vibration. Along with Vold-Kalman filter, the blade passing frequency waveform can be reconstructed. Numerical simulation shows that the proposed HR-HPS method can be well employed under varying operating conditions, and the accuracy of the frequency extraction plays a vital role in waveform reconstruction. Finally, blade fault diagnosis frame based on reconstruction waveform is proposed. A blade fracturing fault experiment is conducted and the result shows that the features extracted from the reconstructed waveform are more sensitive to fault severity than traditional features which can accurately identity the moment of failure and provide early warning for gas turbine blade fracturing faults.
Original language | English |
---|---|
Title of host publication | Proceedings of TEPEN 2022 |
Subtitle of host publication | Efficiency and Performance Engineering Network |
Editors | Hao Zhang, Yongjian Ji, Tongtong Liu, Xiuquan Sun, Andrew David Ball |
Publisher | Springer, Cham |
Pages | 1006-1017 |
Number of pages | 12 |
Volume | 129 |
ISBN (Electronic) | 9783031261930 |
ISBN (Print) | 9783031261923, 9783031261954 |
DOIs | |
Publication status | Published - 4 Mar 2023 |
Event | International Conference of The Efficiency and Performance Engineering Network 2022 - Baotou, China Duration: 18 Aug 2022 → 21 Aug 2022 https://tepen.net/ https://tepen.net/conference/tepen2022/ |
Publication series
Name | Mechanisms and Machine Science |
---|---|
Publisher | Springer |
Volume | 129 MMS |
ISSN (Print) | 2211-0984 |
ISSN (Electronic) | 2211-0992 |
Conference
Conference | International Conference of The Efficiency and Performance Engineering Network 2022 |
---|---|
Abbreviated title | TEPEN 2022 |
Country/Territory | China |
City | Baotou |
Period | 18/08/22 → 21/08/22 |
Internet address |