考虑数字成像的GIS设备机械振动故障控制方法

Mechanical Vibration Fault Control Method of GIS Equipment Considering Digital Imaging

GIS设备故障控制方法由于未将机械振动故障信息与信号能量结合起来进行量化分析,导致控制灵敏度不佳,为此,结合数字成像与机械振动故障信息,通过射线能量分布矩阵对GIS设备机械振动故障进行控制。首先,经X射线透照GIS设备后,根据X射线的衰减程度,通过二维平面图转换,获取DR探测器上形成的射线能量分布矩阵图,通过数模软件转换,最终得到GIS设备数字图像,得到GIS设备的裂纹、气孔和缩孔等故障信息,以此对GIS设备机械振动故障控制分析;其次,通过中值滤波去除图像中的脉冲噪声部分,应用小波滤波去除图像中的散斑噪声部分,根据模糊增强算法对GIS设备数字图像进行增强处理;然后,基于神经网络构建GIS设备机械振动故障识别函数,将处理后的GIS设备数字图像输入至故障识别函数完成GIS设备机械振动故障类型识别,得到GIS设备机械振动故障类型;最后,结合RBF神经网络与PID构建故障控制模型,利用梯度下降法获得RBF神经网络模型的控制参数,完成GIS设备机械振动故障控制。实验结果表明,所提方法的GIS设备机械振动故障控制效果好,故障控制灵敏度高,具有较高的实际应用价值。

GIS equipment fault control method does not combine mechanical vibration fault information with signal energy for quantitative analysis,resulting in poor control sensitivity.Therefore,this paper will combine digital imaging and mechanical vibration fault information to control GIS equipment mechanical vibration fault through ray energy distribution matrix.First of all,after the GIS equipment is transilluminated by X-ray,according to the attenuation degree of X-ray,the ray energy distribution matrix diagram formed on DR detector is obtained through two-dimensional plan transformation,and finally the digital image of GIS equipment is obtained through digital analog software transformation so as to obtain the fault information of GIS equipment such as cracks,air holes,shrinkage holes,and so on,by which the mechanical vibration fault of GIS equipment can be controlled and analyzed;secondly,the impulse noise in the image is removed by median filtering,and the speckle noise in the image is removed by wavelet filtering.the digital image of GIS equipment is enhanced according to the fuzzy enhancement algorithm;then,the GIS equipment mechanical vibration fault identification function is constructed based on neural network,and the processed GIS equipment digital image is input to the fault identification function to complete the identification of the GIS equipment mechanical vibration fault type,and the GIS equipment mechanical vibration fault type is obtained;finally,combining RBF neural network and PID,the fault control model is built,and the control parameters of RBF neural network model are obtained by gradient descent method to complete the mechanical vibration fault control of GIS equipment.The experimental results show that the proposed method has good performance in controlling mechanical vibration faults of GIS equipment,high sensitivity in fault control,and high practical application value.