基于相关峭度反卷积逆滤波器的薄膜生产线故障监测

Fault detection of film production line monitoring system based on correlation kurtosis

摘要:

在薄膜生产的过程中，由于模块或者部件的失效会产生随机故障。为了减少故障的发生，保证设备能够安全高效运行，通过对关键的机组振动进行采样分析，采用基于相关峭度的故障检测方法，设计最大三阶相关峭度(M3CKD)反卷积逆滤波器，以采集到的实时数据序列的三阶相关峭度最大作为代价函数，提取并识别机组振动信号峰值，进行故障信号识别与定位。算法性能验证结果表明该算法的可行性及可操作性，并且算法有较快的收敛速度和较好地稳定性，能有效地识别故障信号，适用于实际的薄膜生产线工业生产过程。

Abstract:

In the process of thin film production, random faults will occur due to the failure of modules or components. In order to reduce the occurrence of faults and ensure the safe and efficient operation of the equipment, through sampling and analyzing the key unit vibration, using the fault detection method based on correlation kurtosis, the maximum third-order correlation kurtosis (m3ckd) deconvolution inverse filter is designed. The maximum third-order correlation kurtosis of real-time data sequence is taken as the cost function to extract and identify the vibration signal peak value of the unit To identify and locate the fault signal. The algorithm performance verification results show that the algorithm is feasible and operable, and the algorithm has fast convergence speed and good stability, and can effectively identify fault signals, which is suitable for the actual thin film production line industrial production process.