融合动态仿真与智能识别的高压涡轮热腐蚀故障预警播报研究

许喆; 陈彦木; 赵海心; 陈旭东; 鲁业明

doi:10.21656/1000-0887.450270

融合动态仿真与智能识别的高压涡轮热腐蚀故障预警播报研究

doi: 10.21656/1000-0887.450270

1. 中国人民解放军92942部队，北京 100161;
2. 大连理工大学能源与动力学院，辽宁大连 116024;
3. 西安热工研究院有限公司电站材料技术部，西安 710054

基金项目:

中央高校基本科研业务费(DUT25LAB110；DUT24LK008)

国家自然科学基金(52475241)

详细信息

作者简介:
许喆(1994—)，男，助理工程师，硕士(E-mail: xuzhe0520@foxmail.com)；鲁业明(1991—)，男，副教授，博士(通讯作者. E-mail: luyeming@dlut.edu.cn).

通讯作者:
鲁业明(1991—)，男，副教授，博士(通讯作者. E-mail: luyeming@dlut.edu.cn).

中图分类号: TK47
计量
- 文章访问数: 16
- HTML全文浏览量: 1
- PDF下载量: 1
- 被引次数: 0
出版历程
- 收稿日期: 2024-10-08
- 修回日期: 2025-06-22
- 网络出版日期: 2025-09-10

Research on the Early Warning and Broadcast of High-Pressure Turbine Thermal Corrosion Faults by Integrating Dynamic Simulation and Intelligent Identification

1. The 92942 Unit of PLA, Beijing 100161, P.R.China;
2. School of Energy and Power Engineering, Dalian University of Technology, Dalian, Liaoning 116024, P.R.China;
3. Power Station Material Technology Department, Xi’an Thermal Power Research Institute Co., Ltd., Xi’an 710054, P.R.China

Funds:

The National Science Foundation of China(52475241)

摘要

摘要: 燃气轮机是空天和舰船装备的重要动力来源，涡轮作为燃燃联合动力系统的关键部件，长时间工作在高温高压环境下，严苛的工作环境导致涡轮叶片易于遭受热腐蚀，从而可能引发系统级别的故障.因此，对涡轮进行热腐蚀故障诊断技术研究具有重要的工程意义.针对涡轮热腐蚀问题，提出了一种融合动态仿真和智能诊断算法的涡轮热腐蚀故障识别方法，利用模块化设计思路，在燃机运行机理数字化模型的基础上建立了整机动态仿真模型，通过标准差法检测所提取数据集中的异常值，使用KNN算法填补空缺值后，采用的小波包Bayes降噪使信号和数据更精准，然后根据人工智能算法构建了表征叶片热腐蚀受损的识别模型.最后，通过使用历史健康数据训练人工智能算法，依靠监测预警模型输出的预测值与实际测量值之间的偏差变化，实现了对涡轮热腐蚀故障的预警播报：在120台机组不同部件运行故障定位测试中，该方法故障精准识别率达95%；在24台机组不同数据特征下的高压涡轮热腐蚀故障预警测试中，故障预警准确率达91.7%以上.该研究拟为动力装备的数字化诊断提供技术参考.
- 动态仿真 /
- 高压涡轮 /
- 热腐蚀 /
- 数字化 /
- 预警故障
Abstract: Gas turbines are important power sources for aerospace and marine equipment. As a key part of the combined combustion system, turbines operating in high-temperature, high-pressure environments for extended periods are prone to blade thermal corrosion, which can lead to system failures. Therefore, researching the thermal corrosion fault diagnosis technology for turbines is of great engineering importance. To address turbine thermal corrosion, a method integrating dynamic simulation and intelligent diagnostic algorithms was proposed. A dynamic simulation model was established based on the engine's operating mechanism with a modular design approach. The outliers were detected with the standard deviation method and the missing values were filled with the KNN algorithm. The wavelet packet Bayesian denoising was then used to obtain precise signals and data, to enable the construction of an identification model for thermal corrosion damage of the blades with artificial intelligence algorithms. By means of training artificial intelligence algorithms with historical health data and monitoring the deviations between predicted values and actual measurements from the early warning model, the early detection of turbine thermal corrosion faults was realized. The testing results of different components of 120 units show that, the proposed method has a precise fault-localizing identification rate of 95%. The high-pressure turbine thermal corrosion fault early warning tests with different data characteristics in 24 units give an accuracy above 91.7%. This study provides technical references for the digitized diagnosis of power equipment.
- dynamic simulation /
- high-pressure turbine /
- hot corrosion /
- digitization /
- early warning of failure

HTML全文

参考文献(17)

[2]SEYBOLT A U, BELTRAN A. Hot corrosion problems associated with gas turbines[C]//American Society for Testing and Materials STP. 1967.

赵德孜. 海洋环境下燃气轮机涡轮叶片的热腐蚀与防护[J]. 装备环境工程, 2011,8(5): 100-103.

(ZHAO Dezi. Hot corrosion and protection of gas turbine blade in marine environment[J]. Equipment Environmental Engineering,2011,8(5): 100-103. (in Chinese))

[3]SALEHNASAB B, POURSAEIDI E, MORTAZAVI S A, et al. Hot corrosion failure in the first stage nozzle of a gas turbine engine[J]. Engineering Failure Analysis,2016,60: 316-325.

[4]TARADA F, SUZUKI M. External heat transfer enhancement to turbine blading due to surface roughness[C]//ASME 〖STBX〗1993 International Gas Turbine and Aeroengine Congress and Exposition. 1993.

[5]BONS J P, TAYLOR R P, MCCLAIN S T, et al. The many faces of turbine surface roughness[J]. Journal of Turbomachinery,2001,123(4): 739-748.

[6]GARG S. Controls and health management technologies for intelligent aerospace propulsion systems[C]//〖STBX〗42nd AIAA Aerospace Sciences Meeting and Exhibit. Reno, NV, 2004: 949．

[7]贺惠新. 燃机异常检测系统的关键技术研究[D]. 哈尔滨: 哈尔滨工业大学, 2013. (HE Huixin. Esearch on the key problems of anomaly detection system of gas turbine[D]. Harbin: Harbin Institute of Technology, 2013. (in Chinese))

[8]BOTROS K K, KIBRYA G, GLOVER A. A demonstration of artificial neural-networks-based data mining for gas-turbine-driven compressor stations[J]. Journal of Engineering for Gas Turbines and Power,2002,124(2): 284-297.

[9]TAN H S. Fourier neural networks and generalized single hidden layer networks in aircraft engine fault diagnostics[J]. Journal of Engineering for Gas Turbines and Power,2006,128(4): 773-782.

[10]VERMA R, ROY N, GANGULI R. Gas turbine diagnostics using a soft computing approach[J]. Applied Mathematics and Computation,2006,172(2): 1342-1363.

[11]KANG Y S, YOO J C, KANG S H. Numerical predictions of roughness effects on the performance degradation of an axial-turbine stage[J]. Journal of Mechanical Science and Technology,2006,20(7): 1077-1088.

[12]SUBRAMANIAN A, GAMANNOSSI A, MAZZEI L, et al. Numerical investigation of surface roughness effects on heat transfer in a turbine cascade[C]//Second International Conference on Material Science,Smart Structures and Applications: ICMSS-2019. 2019: 2191.

[13]SHUKLA A, HARSHA S P. An experimental and FEM modal analysis of cracked and normal steam turbine blade[J]. Materials Today: Proceedings,2015,2(4/5): 2056-2063.

[14]DE BONA F, LANZUTTI A, LUCACCI G, et al. Turbine blade distortion after heat treatment: preliminary experimental investigation and FEM analysis[J]. Key Engineering Materials,2019,827: 98-103.

[15]李淑英. 船舶动力装置仿真技术[M]. 哈尔滨: 哈尔滨工程大学出版社, 2013. (LI Shuying. Simulation Technology of Ship Power Plant[M]. Harbin: Harbin Engineering University Press,2013.(in Chinese))

[16]林新智. 双轴燃气轮机性能仿真及故障特征研究[D]. 北京: 北京化工大学, 2021. (LIN Xinzhi. Research on performance simulation and fault characteristics of dual shaft gas turbine[D]. Beijing: Beijing University of Chemical Technology, 2021. (in Chinese))

[17]胡杨. 涡轮叶片温度特征提取及其故障判别研究[D]. 哈尔滨: 哈尔滨工程大学, 2015. (HU Yang. Research of turbine blades temperature feature extraction and fault discriminant[D]. Harbin: Harbin Engineering University, 2015. (in Chinese))

施引文献

资源附件(0)

访问统计

计量

文章访问数: 16
HTML全文浏览量: 1
PDF下载量: 1
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

融合动态仿真与智能识别的高压涡轮热腐蚀故障预警播报研究

doi: 10.21656/1000-0887.450270

作者简介:
许喆(1994—)，男，助理工程师，硕士(E-mail: xuzhe0520@foxmail.com)；鲁业明(1991—)，男，副教授，博士(通讯作者. E-mail: luyeming@dlut.edu.cn).

通讯作者:
鲁业明(1991—)，男，副教授，博士(通讯作者. E-mail: luyeming@dlut.edu.cn).

计量

Research on the Early Warning and Broadcast of High-Pressure Turbine Thermal Corrosion Faults by Integrating Dynamic Simulation and Intelligent Identification

计量

目录

留言板

融合动态仿真与智能识别的高压涡轮热腐蚀故障预警播报研究

doi: 10.21656/1000-0887.450270

作者简介: 许喆(1994—)，男，助理工程师，硕士(E-mail: xuzhe0520@foxmail.com)；鲁业明(1991—)，男，副教授，博士(通讯作者. E-mail: luyeming@dlut.edu.cn).

通讯作者: 鲁业明(1991—)，男，副教授，博士(通讯作者. E-mail: luyeming@dlut.edu.cn).

计量

出版历程

Research on the Early Warning and Broadcast of High-Pressure Turbine Thermal Corrosion Faults by Integrating Dynamic Simulation and Intelligent Identification

计量

出版历程

目录

作者简介:
许喆(1994—)，男，助理工程师，硕士(E-mail: xuzhe0520@foxmail.com)；鲁业明(1991—)，男，副教授，博士(通讯作者. E-mail: luyeming@dlut.edu.cn).

通讯作者:
鲁业明(1991—)，男，副教授，博士(通讯作者. E-mail: luyeming@dlut.edu.cn).