-
摘要: 为实现航空发动机气路故障在线预诊断, 分析了地空数据链系统中发动机气路参数报文的协议格式, 建立了基于支持向量机算法的发动机气路参数在线预测模型。以便携式地空数据链收发系统为硬件基础, 构建发动机报文并行处理系统, 获取建模所需的训练样本。利用最终误差预报准则确定样本数据嵌入维数, 实现时序样本数据的相空间重构。提出自适应网格搜索法优化支持向量机建模参数, 获得气路参数在线预测模型, 与航路飞机建立地空数据链通信, 预测气路参数趋势。预测结果表明: 参数低压转子转速、高压转子转速、尾气温度与燃油流量的相对预测误差分别为2.5%、2.1%、1.9%与2.3%, 因此, 支持向量机模型具有较高预测精度。Abstract: In order to reality the online forecast of aeroengine gas path fault, the protocol and content of the gas path parameters' report in aircraft communication addressing and reporting system(ACARS) were analyzed, and an online forecast model of the parameters based on support vector machine(SVM) algorithm was established.A real-time processing system of aeroengine report was built by using portable air-ground data link transceiver for obtaining the training samples of the model.Final prediction error(FPE) principle was suggested to optimize the embedding dimensions of the samples, and the phase spaces of the samples were reconstructed.An adaptive grid search algorithm was put forward to optimize the parameters of the model, and the model was linked to route plane by using ACARS.Forecast result shows that the relative forecast errors of low-pressure compressor rotor speed, high-pressure compressor rotor speed, exhaust gas temperature and fuel flow are 2.5%, 2.1%, 1.9% and 2.3% respectively, so the model is feasible.
-
Key words:
- aeroengine /
- fault prognosis /
- support vector machine /
- gas path parameter
-
表 1 模参数
Table 1. Modelling parameters
表 2 测性能比较
Table 2. Comparison of forecast properties
% 
-
[1] URBAN L A. Gas path analysis applied to turbine engine condition monitoring[J]. Journal of Aircraft, 1973, 10(7): 400-406. doi: 10.2514/3.60240 [2] ROEMER MJ, KACPRZYNSKI G J. Advanced diagnosticsand prognostics for gas turbine engine risk assessment[C]∥ROBERT A W. Aerospace Conference Proceedings. Manhattan: IEEE, 2000: 345-353. [3] LUOJian-hui, NAMBURU M. Model-based prognostic tech-niques[C]∥SHEPPARDJ W. AUTOTESTCON 03 Future Sustainment for Military and Aerospace. Anahei : IEEE, 2003: 7803-7837. [4] CRAWFORD J A, TANNER G F. An integrated enginehealth monitoring systemfor gas turbine aero-engines[C]∥ALLEN G, THOMPSON H A, FLEMI NG P G. Aircraft Airborne Condition Monitoring, IEE Seminar. Gloucester: Michael Faraday House, 2003: 1-12. [5] SI MON D L. An overview of the nasa aviation safety program propulsion health monitoring element[R]. Ohio: AI AA, 2000. [6] MELISSA L C, NDAONA C. Engine monitoring using neural networks[R]. Cleveland: AI AA, 1998. [7] SUYKENS J, VANDEWALLW J. Least Squares Support Vector Machines[M]. Singapore: World Scientific, 2002. [8] CHAPELLE O, VAPNIK V N. Choosing multiple parameters for SVM[J]. Mach Learnning, 2002, 46: 131-159. [9] KODIYALAM S. Multidisciplinary aerospace systems optimization computational aero sciences(CAS)project[R]. Washington: NASA/CR, 2001. [10] ARINC 620, Data link ground system standard and interface specification[S]. [11] 郝英, 孙健国, 白杰. 航空燃气涡轮发动机气路故障诊断现状与展望[J]. 航空动力学报, 2003, 18(6): 753-760. doi: 10.3969/j.issn.1000-8055.2003.06.009HAO Ying, SUN Jian-guo, BAI Jie. State-of-the-art and prospect of aircraft engine fault diagnosis using gas path parameters[J]. Journal of Aerospace Power, 2003, 18(6): 753-760. (in Chinese) doi: 10.3969/j.issn.1000-8055.2003.06.009 -
下载:
点击查看大图
图(6) / 表(2)
计量
- 文章访问数: 234
- HTML全文浏览量: 72
- PDF下载量: 273
- 被引次数: 0
