论文部分内容阅读
每个热力参数都可从某一角度反映航空发动机的性能状态,在分析各个性能参数变化特点基础上,选取能够最大程度反映发动机性能变化的参数组合。运用统计粗糙集理论属性重要性判定方法,确定各个参数的权值,从而进一步对所选参数组合进行信息融合,能够更加准确的判断发动机性能状况。提出了利用支持向量数据描述的发动机视情维修阈值确定方法。利用非目标数据即性能下降到限制值附近时刻的数据,来提高视情维修的准确率。仿真实验结果表明,该方法能够准确确定视情维修时机,对发动机视情维修时机选择有一定的参考价值。
Each thermal parameter can reflect the performance state of aeroengine from a certain angle. Based on the analysis of the characteristics of each performance parameter, the parameter combination that can reflect the change of engine performance to the greatest extent is selected. By using the method of statistic rough set theory to determine the importance of attributes, the weight of each parameter can be determined, so as to further fuse the information of the selected parameters, which can judge the performance of engine more accurately. A method to determine the threshold-based maintenance of engine is proposed by using support vector data. The use of non-target data that performance fell to the limit value near the time of the data, to improve the accuracy of the situation as the maintenance. Simulation results show that this method can accurately determine the timing of maintenance as appropriate and has some reference value for the timing of engine maintenance.