航空发动机温度过热故障挖掘模型仿真

Aviation Engine Temperature Overheating Fault Mining Model Simulation

航空发动机温度过热故障准确检测,可保证航空安全。航空发动机的温度变化过程是一套复杂的流体热动力演变过程,伴随着高温、高压、高能量释放,使得温度曲线呈现较强非线性。传统的温度过热故障挖掘算法为了描述上述非线性变化过程,需要运用高阶、非线性模型完成温度异常挖掘,模型过于复杂,尤其是非线性干扰导致数据分类过程存在较大困难。提出采用萤火虫优化支持向量机算法的航空发动机温度过热故障挖掘方法。通过对采集的待测发动机温度数据进行数值化和归一化操作,采用萤火虫优化支持向量机算法构建航空发动机温度过热故障挖掘模型,根据萤火虫优化算法获取航空发动机温度数据最佳分类,分辨出高于正常温度数据范围的故障数据,完成对航空发动机温度过热故障挖掘。实验结果表明,利用改进算法进行航空发动机温度过热故障挖掘,能够提高温度过热故障数据挖掘的准确性和挖掘效率。

Aircraft engine overheating fault accurately detect temperature, is of great significance to aviation safe- ty. Temperature variation process of aircraft engine is a complex fluid dynamic evolution process, with high tempera- ture and high pressure, high energy release, the temperature curve has a strongly nonlinear. Traditional mining algo- rithm of temperature overheating fault in order to describe the nonlinear change process, needs to be done using the high -order, nonlinear model temperature anomaly mining, model is too complex, especially the nonlinear interfer- ence cause data classification process is difficult. Support vector machine was optimized based on the firefly algorithm of aviation engine temperature overheating fault mining method. Through the acquisition of engine temperature under test data for numerical value and normalized operation, support vector machine was optimized by using the firefly al- gorithm to build the aircraft engine temperature overheat fault mining model, according to the fireflies optimization al- gorithm for obtaining the best classification of aircraft engine temperature data tell a higher than normal temperature data range of failure data of aircraft engine temperature overheating fault mining. Experimental results show that the improved algorithm to the aircraft engine temperature overheating fault mining, can improve the accuracy of tempera- ture overheating fault data mining and the mining efficiency.