论文部分内容阅读
提出了一种基于多工况聚类的飞机发动机健康状态评估方法。首先对工况进行自动划分;再从大量传感器数据中选取与发动机性能衰退趋势密切相关的传感器变量进行分析并提取特征,定义逻辑回归模型;最后给出对飞机发动机的健康状态评价。该方法在NASA提供的飞机发动机传感器数据上进行验证,对其健康性能状态进行评估,结果表明该方法不仅结果可靠,而且具有一定实用性,可为飞机发动机的剩余使用寿命预测提供一种可行且有效的理论方法和手段,对提高飞机发动机的安全性和可靠性具有重要价值。
A method for aircraft health assessment based on multi-condition clustering is proposed. Firstly, the working conditions are divided automatically. Then, the sensor variables that are closely related to the declining trend of the engine performance are selected from a large number of sensor data to analyze and extract the features, and the logistic regression model is defined. Finally, the evaluation of the health status of the aircraft engine is given. The method is validated on NASA-supplied aircraft engine sensor data to evaluate its health performance status. The results show that the proposed method is not only reliable, but also has some practicality. It can provide a feasible and effective way to predict the remaining life of the aircraft engine Effective theoretical methods and means are of great value in improving the safety and reliability of aircraft engines.