AR双谱在减压阀故障诊断中的应用

摘要

双谱可以描述系统的非线性耦合,抑制高斯噪声和保留相位信息.当减压阀系统受到零均值的非高斯噪声干扰时,利用减压阀输出的位移信号建立AR时间序列模型,通过AR双谱进行减压阀故障诊断.结果表明在相同的液压压力下,减压阀在正常工作和出现故障时的双谱存在明显差异,双谱表现出了不同的特征；对于同一实验组,在不同的液压压力下减压阀的双谱特征有略微变化,液压压力对减压阀的故障诊断影响不大,这样有助于在实际应用中对减压阀做出快速诊断.理论和实验说明了双谱分析是实现液压元件故障检测的有效方法之一.%The cigarette case detection system is a special device, which is installed on the production line in front of packing machine to detect the appearance defect.The system can detect cigarette case in real-time and remove the defective products in time with characteristics of high-speed, efficient and accurate, which is an auto-detection system covering mechanic, optical, electric, computer and communications as well as other technologies.The application of the system greatly improves the automation of the tobacco company;In addition, it also can reduce the labor intensity,improve the working environment and ensure the product quality.The algorithm for common detection items of cigarette case was designed and the open source computer vision library OpenCV of Intel Company was used to improve the detection speed and accuracy in it.