The Application Research of Support Vector Regression in Aero-engine's Baseline Mining
-
摘要: 针对航空发动机基线难以获取的问题,利用支持向量回归机(support vector regression,SVR)算法,采用厂家监控系统数据和飞机快速存储记录器(quick access recorder,QAR)数据两种方式对基线进行挖掘分析,提供了获取基线的多种途径和方法,取得了比较可靠的结果。支持向量回归机在处理非线性回归分析时具有快速、准确的优点,能够进行单参数及多参数的基线回归分析,通过计算结果比较分析,多参数基线回归与单参数基线回归、一元线性基线拟合相比具有偏差小、精度高的优势,能够有效提高发动机基线监控的准确性。Abstract: As the aero-engine's baseline is difficult to get, Support Vector Regression (SVR) is used with data from monitoring systems of manufacturers and Quick Access Recorder (QAR) to do baseline mining analysis, providing a variety of ways to get baseline and the results turned out to have a good reliability. SVR is fast and precise in dealing with nonlinear regression analysis. Single parameter regression and multi-parameter regression can be processed by using SVR. The conclusion can be drawn from analysis of calculation results that multi-parameter regression of SVR has a lot of advantages such as small deviation and high accuracy compared with linear regression and single parameter regression of SVR. As a result, the accuracy of the aero-engine's monitoring can be improved effectively.
-
[1] 钟诗胜,崔智全,付旭云.Rolls-Royce发动机基线挖掘方法[J].计算机集成制造系统,2010,16(10):2265-2270 Zhong S S, Cui Z Q, Fu X Y. Baseline mining method of RR's engine[J]. Computer Integrated Manufacturing Systems, 2010,16(10):2265-2270(in Chinese) [2] 钟诗胜,周志波,张永,等.基于三次回归分析的试车台基线库的建立[J].计算机集成制造系统,2005,11(2):270-274 Zhong S S, Zhou Z B, Zhang Y, et al. Construction of test cell baselines based on thrice regression analysis[J]. Computer Integrated Manufacturing Systems, 2005,11(2):270-274(in Chinese) [3] Luan S G, Zhong S S, Li Y. Hybrid recurrent process neural network for aero engine condition monitoring[J]. Neural Network World, 2008,18(2):133-145 [4] 北京飞机维修工程有限公司.发动机机群的科学管理[Z].北京:北京飞机维修工程有限公司,2000:121-130 AMECO, Beijing. Engine scientific management[Z]. Beijing:AMECO, Beijing, 2000:121-130(in Chinese) [5] 林兆福,范作民.发动机基线方程的建立和应用[J].中国民航学院学报,1992,10(4):20-32 Lin Z F, Fan Z M. Development and application of engine baseline equations[J]. Journal of Civil Aviation University of China, 1992,10(4):20-32(in Chinese) [6] 陕振勇.航空发动机气路参数偏差值模型研究与应用[D].哈尔滨:哈尔滨工业大学,2012 Shan Z Y. Research on aeroengine gas path parameter deviation model and its application[D]. Harbin:Harbin Institute of Technology, 2012(in Chinese) [7] 钟诗胜,崔智全,王体春,等.基于偏差值的航空发动机参数标准化修正模型[J].航空动力学报,2012,27(11):2592-2598 Zhong S S, Cui Z Q, Wang T C, et al. Corrected standardization model of aero-engine parameter based on deviation[J]. Journal of Aerospace Power, 2012,27(11):2592-2598(in Chinese) [8] 崔智全.民航发动机气路参数偏差值挖掘方法及其应用研究[D].哈尔滨:哈尔滨工业大学,2013 Cui Z Q. Civil aeroengine gas path parameter deviation mining method with application[D]. Harbin:Harbin Institute of Technology, 2013(in Chinese) [9] Vapnik V N.统计学习理论的本质[M].张学工,译.北京:清华大学出版社,2000 Vapnik V N. The essence of statistical learning theory[M]. Zhang X G, translate. Beijing:Tsinghua University Press, 2000(in Chinese) [10] Cortes C, Vapnik V. Support-vector networks[J]. Machine Learning, 1995,20(3):273-297 [11] Suykens J A K, Van Gestel T, De Brabanter J, et al. Least squares support vector machines[M]. Singapore:World Scientific, 2002 [12] 邓乃扬,田英杰.数据挖掘中的新方法-支持向量机[M].北京:科学出版社,2004 Deng N Y, Tian Y J. New method in data mining[M]. Beijing:Science Press, 2004(in Chinese) [13] 瞿红春,林兆福.民用航空燃气涡轮发动机原理[M].北京:兵器工业出版社,2006 Qu H C, Lin Z F. Principle of civil aviation gas turbine engine[M]. Beijing:The Publishing House of Ordnance Industry, 2006(in Chinese) [14] 彭云飞.航空发动机状态参数处理技术及其应用研究[D].哈尔滨:哈尔滨工业大学,2009 Peng Y F. Research on processing technique of aeroengine condition parameters and its application[D]. Harbin:Harbin Institute of Technology, 2009(in Chinese) [15] 孔成安,李文华,尹湛.利用QAR数据实施飞机性能监控[J].中国民用航空,2008,(10):54,56 Kong C A, Li W H, Yin Z. Monitoring aircraft performance with QAR data[J]. China Civil Aviation, 2008,(10):54,56(in Chinese) [16] 刘志荣,朱睿,梁忠生,等.发动机健康基线及评估准则研究[J].厦门大学学报(自然科学版),2010,49(4):520-525 Liu Z R, Zhu R, Liang Z S, et al. Research of engine health baselines and evaluation criterion[J]. Journal of Xiamen University (Natural Science), 2010,49(4):520-525(in Chinese)
点击查看大图
计量
- 文章访问数: 249
- HTML全文浏览量: 31
- PDF下载量: 11
- 被引次数: 0