发动机飞行任务剖面的主成份聚类法

本文提出了利用主成份分析对航空发动机飞行任务剖面进行分类的方法 ,并对某战斗机发动机 1 8个飞行任务剖面进行了聚类分析。选取了飞行高度、飞行马赫数、发动机转速以及发动机重心法向过载等 4个参数作为分类的原始依据参数。对上述 4个参数进行主成份分析 ,得到 4个独立的主成份 ,其中第一、二主成份的累积贡献率可达 81 .1 %。因此 ,可以根据主平面内各飞行任务剖面的第一、二主成分的分布情况直观地进行定性地分类。最后 ,本文利用重心法进行了定量的聚类 ,得到了分类的树状图。研究结果表明本文提出的方法是合理可行的

The Principal Component Analysis Method for Engine Flight Mission Profiles Categorization

In this paper,the principal component analysis method is proposed for engine flight mission profiles categorization.Eighteen flight mission profiles of a fighter engine are categorized.The categorization is based on 4 parameters,namely,flight height,flight Mach number,engine rotational speed and overload on the engine center of gravity.4 principal components of the mentioned 4 parameters are obtained,the accumulated contribution ratio of the first two principal components arrives at 81 1%.The qualitative categorization can be done from the first and the second principal components distribution range in the principal plain.Finally,a tree shaped category graph is acquired by quantitative categorization using center of gravity method.The results show that the proposed method is feasible.