Nondestructive Testing Of Wire Rope Under Unsaturated Magnetic Excitation

As a special structural component,wire rope is widely used in various fields.The safety of wire rope in service is directly related to the normal operation of production and the life safety of personnel.Therefore,it is of great significance to detect and realize quantitative identification of wire rope in service.In this paper,the positioning and quantitative detection of wire rope defects were researched,the main works are as follows:In view of the existing wire rope magnetic flux leakage(MFL)detection equipment is bulky,inconvenient to use,unable to circumferential positioning of defects,uneven magnetization,low SNR and other problems,a nondestructive testing system of wire rope under unsaturated magnetic excitation was designed.The system includes a wire rope MFL acquisition module,a host computer and a 4G wireless transmission module.The experimental results show that the system can effectively locate and quantify the defects of the wire rope,and realize the function of remote monitoring.In order to solve the problems of various noises in the collected MFL signals,firstly,the denoising algorithm of MFL data based on wavelet theory was proposed,which can reduce the noise of MFL data,but there is a problem of partial signal loss due to the close frequency of partial signal and noise.Therefore,the wavelet denoising algorithm based on ensemble empirical mode decomposition(EEMD)was proposed,which avoids the problem of signal loss,but has the problem of adding normal white noise in the EEMD and the number of decomposed modes is uncontrollable.Therefore,the wavelet denoising algorithm based on variational mode decomposition(VMD)was proposed.The experimental results indicate that these three algorithms can effectively suppress the noise of MFL data,and the wavelet denoising algorithm based on variational mode decomposition has the best effect.In order to further deal with wire rope defects with the help of more mature methods in the field of image processing,the gray normalization method was used to transform MFL data into MFL images.In order to improve the resolution of defects,the wavelet super-resolution image enhancement algorithm was designed.In order to describe defects with the help of color image features,the pseudo-color image enhancement algorithm was designed.The experimental results manifest that the designed wavelet super-resolution image enhancement algorithm can double the resolution of defect images,and the designed pseudo-color image enhancement algorithm can transform the MFL grayscale image into pseudo-color image,making it possible to describe defect by using color features of the image.Aiming at the problem of quantitative identification of wire rope defects,firstly,the BP neural network-based defect quantitative identification method was proposed.In order to shorten the operation time,the support vector machine-based defect quantitative identification method was proposed.Finally,in order to avoid the blindness of manually selecting features,the k-nearest neighbor algorithm-based defect quantitative identification method was proposed.The experimental results of quantitative recognition demonstrate that these three methods can realize the quantitative identification of defects.Moreover,the method with the highest recognition accuracy and precision is k-nearest neighbor algorithm-based defect quantitative identification method.