【摘要】 润滑油金属含量是航空发动机关键部件出现磨损及裂纹等情况的表征。通过对其进行预测可提前发现相应部件的机械故障,能保证飞行安全并降低航空发动机维护费用。润滑油金属含量受许多复杂因素影响,传统方法难以预测其变化趋势。为此,提出了一种基于双并联过程神经网络的润滑油金属含量预测方法,并给出了基于正交基函数展开的学习算法。将该方法用于某型航空发动机润滑油中铁含量预测,结果表明其预测精度满足工程需要。更多还原

【Abstract】 The concentration of metal elements in the aeroengine lubricating oil is a reflection of abrasion and fracture of aeroengine key components.By predicting the metal elements concentration,mechanical faults in the aeroengine can be confirmed in advance,and it also can guarantee the flight safety and cut down the maintenance cost.The metal elements concentration is influenced by many complicated factors,and its change tendency is difficult to be predicted by traditional prediction method.A novel prediction method based on double parallel process neural network was proposed to solve this problem,the corresponding learning algorithm based on the expansion of the orthogonal basis functions was developed.The proposed prediction method was utilized to predict the concentration of iron in some type aeroengine’s lubricating oil,and the test results show the prediction precision can meet the require of engineering.更多还原