小波网络平均影响值的航空发动机自变量筛选

doi:10.13196/j.cims.2013.12.cuizhiquan.3062.6.20131217

摘要/Abstract

摘要： 为了快速准确地实现发动机参数非线性自变量筛选,基于平均影响值的思想和小波神经网络学习能力强、收敛速度快、具有自适应性和容错性等优点,提出小波神经网络平均影响值的发动机自变量筛选方法。根据参数之间的关系特点,建立多参数连续小波逼近网络模型,并给出学习算法。仿真实例表明,该方法不但能够实现复杂的非线性变量筛选,而且对比其他非线性变量筛选方法,具有精度更高、速度更快的特点。

关键词: 航空发动机, 小波网络, 平均影响值, 自变量筛选

Abstract: To achieve the non-linear variables selection rapidly and accurately,the engine arguments parameters selection method for wavelet neural network’s Mean Impact Value(MIV) was proposed based on the ideological of MIV and the advantages such as learning ability,fast convergence with adaptive and fault tolerance of wavelet neural network.According to the relationship characteristics of the engine parameters,the continuous multi-parameter approximation wavelet network model was established,and the learning algorithm was given.Simulation results showed that the proposed method could achieve complex nonlinear variable selection and have higher accuracy and faster features by comparing to other non-linear variable selection method.

Key words: aero-engine, wavelet network, mean impact value, arguments selection

中图分类号:

崔智全,付旭云,钟诗胜,王体春. 小波网络平均影响值的航空发动机自变量筛选[J]. , DOI: 10.13196/j.cims.2013.12.cuizhiquan.3062.6.20131217.

CUI Zhi-quan,FU Xu-yun,ZHONG Shi-sheng,WANG Ti-chun. Aero-engine arguments selection based on wavelet network mean impact value[J]. , DOI: 10.13196/j.cims.2013.12.cuizhiquan.3062.6.20131217.

参考文献

［1］LI Yibo,ZHANG Guangming,JIANG Liying.Research status of gas-path fault diagnostics for aero-engine［J］.Gas Yurbine Technology,2009,22(3):10-15 (in Chinese) .［李一波,张光明,蒋丽英.航空发动机气路故障诊断技术研究现状［J］.燃气轮机技术,2009,22(3):10-15.］
[2]MATHIOUDAKIS K,STAMATIS A,TSALAVOUTAS A,et al.Performance analysis of industrial gas turbines for engine condition monitoring［J］.Proceedings of the Institution of Mechanical Engineers,Part A:Journal of Power Energy,2001,215(2):173-184.
[3]STAMATICS A G.Evaluation of gas path analysis methods for gas turbine diagnosis［J］.Journal of Mechanical Science and Technology,2011,25(2):469-477.
[4]PU Xingxing,LIU Shangming,JIANG Hongde,et al.Sparse Bayesian learning for gas path diagnostics［J］.Journal of Engineering for Gas Turbines and Power,2013,135(7):1-8.
[5]LIU Zhirong,ZHU Rui,LIANG Zhongsheng,et al.Research of engine health baselines and evaluation criterion［J］.Journal of Xiamen University:Natural Science,2010,49(4):520-525 (in Chinese) .［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.］
[6]VOLPONI A J.Gas turbine parameter corrections［C］//Proceedings of American Society of Mechanical Engineers.New York,N.Y.,USA:ASME,1998:613-621.
[7]KURZKE J.Model based gas turbine parameter corrections［C］//Proceedings of ASME International Gas Turbine Institute Publishing.New York,N.Y.,USA:ASME,2003:91-99.
[8]CASONI A,NUNCIO,COLITTO.Corrected parameter control for two shaft gas turbine［C］// Proceedings of the ASME Turbo Expo.New York,N.Y.,USA:ASME,2004:741-748.
[9]ZHANG Yuan,ZHU Eryi,LI Jing,et al.Variable selection by use of combining the genetic algorithms with simulated annealing algorithm［J］.Chinese Journal of Analytical Chemistry,1999,27(10):1131-1135(in Chinese).［章元,朱尔一,李静,等.模拟退火算法与遗传算法结合用于变量筛选［J］.分析化学,1999,27(10):1131-1135.］
[10]ZHANG Yuan,ZHU Eryi,ZHUANG Zhixia,et al.Variable selection by genetic algorithms［J］.Chemical Journal of Chinese Universities,1999,20(9):1371-1375 (in Chinese) .［章元,朱尔一,庄峙厦,等.遗传算法用于变量筛选［J］.高等学校化学学报,1999,20(9):1371-1375.］
[11]WANG Ziwei,YE Qiwang.Cancer informative gene identification based on MIV method of neural network［J］.Mathematics in Practice and Theory,2011,41(14):47-58 (in Chinese).［王紫微,叶奇旺.基于神经网络MIV值分析的肿瘤基因信息提取［J］.数学的实践与认识,2011,41(14):47-58.］
[12]LI Wansha,TANG Xiaofang,HE Jian,et al.Prediction of coronary artery disease and evaluation of relative risk factors using artificial neural networks［J］.Chinese Heart Journal,2011,23(4):530-534 (in Chinese).［李万莎,唐小芳,何健,等.应用BP神经网络模型分析冠心病相关因素及其临床意义［J］.心脏杂志,2011,23(4):530-534.］
[13]XU Fuqiang,LIU Xiangguo.Variables screening methods based on the optimization of RBF neural network［J］.Computer Systems and Application,2012,21(3):206-208 (in Chinese).［徐富强,刘相国.基于优化的RBF神经网络的变量筛选方法［J］.计算机系统应用,2012,21(3):206-208.］
[14]BALABIN R M,SAFIEVA R Z,LOMAKINA E I.Lomakina.Wavelet neural network(WNN) approach for calibration model building based on gasoline near infrared(NIR) spectra［J］.Chemometrics and Intelligent Laboratory System,2008,4(3):58-62.
[15]PINDORIYA N M,SINGH S N,SINGH S K.An adaptive wavelet neural network-based energy price forecasting in electricity markets［J］.IEEE Transactions on Power Systems,2008,23(3):1423-1432.
[16]LING S H,IU H H C,LEUNG F H F,et al.Improved hybrid particle swarm optimized wavelet neural network for modeling the development of fluid dispensing for electronic packaging［J］.IEEE Transactions on Industrial Electronics,2008,55(9):3447-3460.

[1]	张玉金,黄博,廖文和. 面向场景的航空发动机基于模型的系统工程设计[J]. 计算机集成制造系统, 2021, 27(11): 3093-3102.
[2]	付宇,殷逸冰,冒慧杰,左洪福,冯正兴. 基于静电监测和神经网络的航发典型故障诊断[J]. 计算机集成制造系统, 2019, 25(第11): 2852-2862.
[3]	聂兆伟,熊丹丹. 航空发动机叶片自适应修复目标曲面重构[J]. 计算机集成制造系统, 2019, 25(第1): 53-60.
[4]	杜党党,贾晓亮,王卓. 基于故障彩色图谱的航空发动机故障模式识别[J]. 计算机集成制造系统, 2018, 24(第9): 2297-2305.
[5]	陈志英,刘勇,周平,刘广通. 基于改进田口试验法的装配公差分析[J]. 计算机集成制造系统, 2018, 24(第5): 1200-1206.
[6]	柳强,焦国帅. 基于改进NSGA-Ⅱ的航空发动机管路多目标布局优化[J]. 计算机集成制造系统, 2018, 24(第5): 1217-1227.
[7]	陈振,贾晓亮. 不确定条件下航空发动机大修周期预测方法[J]. 计算机集成制造系统, 2018, 24(第2): 281-289.
[8]	周金华,任军学,蔡菊. 基于RBF神经网络的航空叶片铣削残余应力预测[J]. 计算机集成制造系统, 2018, 24(第2): 361-370.
[9]	李联辉,尹冠飞,莫蓉. 面向航空发动机装配过程的信息追溯与过程监控[J]. 计算机集成制造系统, 2018, 24(第12): 2986-3000.
[10]	张军锋,史耀耀,蔺小军,段继豪,张宏基,李志山. 基于灰色关联分析的叶片砂带抛光参数优化[J]. 计算机集成制造系统, 2017, 23(第4期): 806-814.
[11]	淮文博,史耀耀,唐虹,蔺小军,张溦. 砂布轮抛光叶片表面粗糙度预测与参数优化[J]. 计算机集成制造系统, 2017, 23(第12): 2708-2718.
[12]	王辉,周明星,余杰,王文宇,黄博浩,融亦鸣. 航空发动机叶片保形加工中的定位误差数值建模与分析[J]. 计算机集成制造系统, 2016, 22(第9期): 2118-2126.
[13]	姜天华,严洪森. 面向航空发动机装配车间知识化制造系统自进化[J]. 计算机集成制造系统, 2015, 21(第12期): 3222-3230.
[14]	杜党党,贾晓亮,郝超博. 基于遗传过程神经网络算法的航空发动机健康状态图谱化预测方法[J]. 计算机集成制造系统, 2015, 21(第11期): 2980-2987.
[15]	付旭云,陕振勇,李臻,钟诗胜. 时变模糊神经网络及其在航空发动机排气温度预测中的应用[J]. 计算机集成制造系统, 2014, 20(4): 919-.