基于LS-SVM的军用航空发电机寿命预测研究

为解决航空发电机寿命有效预测的难题,以期实现视情维修,提升飞机的综合作战效能,提出一种基于LS-SVM（Least Squares Support Vector Machines）模型的航空发电机寿命预测方法。采用某型真实航空发电机,运用航空发电机专用寿命试验平台,对其进行加速寿命试验,获取输入转速、注油压力、负载、频率、电流、进油温度、出油温度、进口压力、出口压力等多种寿命相关参数,深入分析这些参数间的内在联系与寿命变化规律,设计LS-SVM寿命预测模型,并运用该寿命预测模型对航空发电机寿命表征参数进行预测。研究表明,所设计的LS-SVM寿命预测模型能较好实现对航空发电机的寿命预测效能,具有广阔的应用前景和较大的实际应用价值。

Military aero-generator life prediction based on LS-SVM

To solve aero generator life prediction problem in order to realize condition maintenance,to improve the overall combat effectiveness of aircraft,a new methods,based on LS-SVM（least squares support vector machines,LS-SVM） model for life prediction of aero generators is proposed in this paper.Using a certain type of real aero-generator,and special life test platform for aero-generator,the accelerated life testing was conducted.And the input speed,fuel injection pressure,load,frequency,current,inlet oil temperature,outlet oil temperature,inlet pressure,outlet pressure and other relevant parameters is obtained,by in-depth analysis of the intrinsic link between these parameters and changes of lifecycle was carried out.A life prediction model is designed by LS-SVM,and model parameters of aero-generator to predict the life characterization is used.The results show that the LS-SVM model can realize the aero-generator life prediction and is of a wide application prospect and great practical value.