基于机器视觉技术的气动系统状态监控与故障诊断

针对传统的对气动系统状态监控效率较低以及成本过高的问题，提出了一种基于机器视觉的状态监控和故障诊断方法。该方法利用CCD摄像头为传感器，对气动系统的动态序列图像进行分析，通过图像分割和识别等方法获得系统运行状态，与事先进行的自学习所获得的标准状态比较，判断系统是否运转正常，并在故障时给出故障分析和处理。实验结果表明，该方法能很好地完成对气动系统的状态分析和故障诊断。

State Monitoring and Fault Diagnosis for Pneumatic System Based on Machine Vision

Aimed at the shortcomings of the traditional pneumatic system monitoring such as low efficiency and hign cost, a method based on machine vision was proposed. In this method, a CCD camera is utilized as the sensor to acquire the dynamic sequence images of the pneumatic system. After the analysis of the images via image segmatation and recognition, the system status is known. When compared with the normal status self-learned beforehand, it can be decided whether the system works normally or not,and a fault diagnosis is given if faults occur. The result of experiments shows this method is practical and accurate.