基于支持向量机的发动机气路故障预诊断  

Gas path fault prognosis of aeroengine based on support vector machine

作　　者：王旭辉[1] 黄圣国[1] 施鼎豪[2] 舒平[2] 

机构地区：[1]南京航空航天大学民航学院,江苏南京210016 [2]中国民用航空局安全技术中心,北京100028

出　　处：《交通运输工程学报》2008年第5期33-37,共5页Journal of Traffic and Transportation Engineering

基　　金：国家863计划项目(2006AA12A108)

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。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. 2 tabs, 6 figs, 11 refs.

关 键 词：航空发动机 故障预诊断 支持向量机 气路参数 

分 类 号：V263.6[航空宇航科学与技术—航空宇航制造工程]