基于局部窄带分解和分形的自动机故障诊断

Automaton Fault Diagnosis Based on Local Narrow-band Decomposition and Fractal Theory

自动机作为自动武器最核心最复杂的部分,其可靠性和故障检测与诊断成为故障诊断领域关注的热点。因此有必要发展一种快速、高效的高速自动机故障诊断方法,实现在线健康状态预测和视情维修,提高军事装备的使用和维修效率。以W85高射机枪自动机为实验对象,针对自动武器在射击时会产生一定方向和频率的冲击振动,采集其短时振动冲击信号,利用基于瞬时频率和局部窄带信号的自适应分解方法对信号进行分解并重构,计算重构后振动信号的网格维数,利用维数距离进行故障类别分析,实现了故障诊断,为高速自动机的健康状态预测奠定了基础。

Automata is the most core of the automatic weapons, its reliability and fault detection and diagnosis has become the hot spot in the field of fault diagnosis. It is necessary to develop a rapid and efficient high-speed automatic fauh diagnosis method, realize the online health condition prediction and maintenance, enhance the efficiency of the use and maintenance of military equipment., we take W85 anti-aircraft guns automata as experiment object, in view of the automatic weapon when shooting because of gunpowder gas shock and collision between components, and shock vibration frequency will produce certain direction. Its short-term impact vibration signal collected, the use of S. 1. Peng（ Peng Silong） based on the instantaneous frequency and local adaptive narrow-band signal decomposition method of signal decomposition and reconstruction, the computation of the vibration signal reconstruction grid dimension, using dimen- sion distance to fault category analysis, realizes the fault diagnosis and state prediction to the health of the high speed automatic laying the groundwork.