元胞自动机模型在航空发动机故障诊断中的应用

Application of Cellular Automata Model in Aeroengine Fault Diagnosis

航空发动机一般在高温、高压和高速转动的状态下工作,因此很难获取其全生命周期试验数据。针对无完整生命周期数据的小样本集合进行设计,提出一种基于元胞自动机的航空发动机故障诊断方法,该方法在获取发动机故障特征信息之后,利用元胞的扩散机制获取故障模式的分类边界。其优势在于:在给定的数据集前提下,可以利用较少的运行时间来约减给定的规则样本;可以利用积累或迭代的方式来分步获得原给定样本集的一致性子集。同时,算法的可积累性、运算时间可控等特点,使得该算法能连续应用在航空发动机试验样本数据集由小样本持续增加到大样本的过程中。该方法的应用对发动机的故障诊断的研究具有重要的指导意义。

Aeroengine usually works under high temperature,high pressure and high speed,so it is difficult to obtain its life cycle test data. A fault diagnosis method of aeroengine based on cellular automata is proposed,which is designed for a small samples set without complete life cycle data. After acquiring engine fault feature information,the classification boundary of fault patterns is obtained by using the cellular diffusion mechanism.Its advantages are: under the premise of a given data set,it can not only use less running time to reduce the given rule samples,but also can use accumulation or iteration ways to obtain the consistency subset of the original given sample set step by step. At the same time,the characteristics of the algorithm,such as accumulation and controllable operation time,make the algorithm can be used in the process of aeroengine test sample data set from small sample to large sample continuously. The application of this method has important guiding significance for the research of engine fault diagnosis.