一种涡轮泵故障阈值检测算法

A Turbo-pump Fault Threshold Detection Algorithm

涡轮泵是火箭发动机的心脏,是火箭发动机中的高速旋转组件,也是故障率较高的部件之一。涡轮泵的可靠性在整个发动机的可靠性中占有重要地位。传统的故障检测有时会出现虚警或漏警的现象,而且主要用于试车后的状态分析与评估;在实时故障检测中存在很大的不足。涡轮泵发生故障的主要原因是由于振动,因此振动加速信号研究对涡轮泵的故障检测显得尤为重要。通过一种阈值检测算法对振动加速信号进行研究,并结合已有涡轮泵故障信息对该算法进行验证,验证了该算法的准确性、实时性等要求。

The turbo pump is the heart of the rocket engine, whitch is a high-speed rotating componet, and is also one of the higher parts of the engine failure rate. The reliability of the turbo pump in the engine plays an im- portant role. Traditional health monitoring system can sometimes be a false alarm or missed alarm phenomenon, there is a great lack of state analysis and evaluation, and is mainly used to test in real-time fault detection. Tur- bopump failure is maily due to vibration, the vibration signal of the turbopump fault detection is particularly impor- tant. An adaptive fault detection algorithm to study the vibration on signal and combined with exiting turbopump failure of the algorithm to validate and verify the accuracy of the detection algorithm, real-time requiremengts.