Study On Magnetic Leakage Detection And Quantitative Identification Of Broken Wire In Spiral Climbing Mode

The operation status of the cable-stayed bridge bears on the national economic development and public safety.Therefore,it is necessary to inspect and maintain the cable on a regular basis.Meanwhile,it is of great significance to study damage detection and evaluation methods of cable-stayed bridges.The cable is built of high carbon steel with strong magnetic conductivity,making it appropriate for detecting damage to its internal steel wire utilizing Magnetic Flux Leakage Inspection technology.Based on the analysis of the existing research status of Magnetic Flux Leakage Inspection technology of cable-stayed bridges at home and abroad,this paper proposes a Magnetic Flux Leakage Inspection technique to inspect cable’s damaged wire in the spiral ascending mode.The main contents of the research are as follows:Firstly,the magnetic flux leakage testing technique is determined to carry out the nondestructive testing of the cable’s broken wire based on the analysis of the cable’s damage and the nondestructive testing methods and their limits.This paper will analyze the magnetic excitation circuit of magnetic flux leakage detection and optimize the design of magnetic circuit parameters.Meanwhile,the influence of cable strand gap during spiral climbing will be explored,the magnetic dipole model of cable strand gap will be developed,and the axial and circumferential leakage magnetic fields of air gap model will be simulated and investigated in this paper.Utilizing Maxwell finite element modeling software,the magnetic flux leakage detection model of cable broken wire in spiral ascending mode is constructed.The magnetic detection model of wire strand gap with cable is simulated and assessed.Secondly,the optimal exciter design strategy is investigated in order to produce saturation stimulation within the cable.To optimize the size of exciter,the finite element simulation approach is used to analyze the impact of permanent magnet size on excitation intensity,and the magnetic dipole model is utilized to research the influence of defect width,depth,and lifting distance on the detecting structure.An experimental platform for magnetic flux leakage detection of damaged wires of cable in spiral climbing mode is created using theoretical and simulation analyses and includes cable excitation,signal acquisition,filtering,display,and storage functionalities.For quantitative identification of broken wire numbers,22 various types of broken wire damage of steel wire rope are developed,each having a different number and breadth of broken wire.Then,the magnetic detection under spiral climbing is next investigated experimentally in order to understand better the law of equidistant spiral motion and the impact of alternative detection pathways on Hall element detection findings.The three-dimensional model of cable broken wire defect is established,the distribution of leakage magnetic field in the circumferential direction is analyzed,the detection results in spiral climbing are classified into four categories,and the detection experiments are carried out.The experimental findings reveal that four distinct magnetic flux leakage signal waveforms would be generated under different detection pathways.The optimal detection path is chosen after the comparison analysis.The magnetic detection outcomes in the cases of spiral climb and vertical climb are experimentally compared to examine the causes for the differences as well as the benefits and drawbacks of the two detection approaches.Finally,the eigenvalues are extracted from the detection waveforms of magnetic flux leakage signals from 22 different defects in the spiral climbing mode of exciter.The characteristic data set of magnetic detection broken wire in the spiral climbing mode is obtained for the first time.The quantitative detection of cable damaged wire is studied using a genetic algorithm and a random forest method.The total recognition rate of the genetic algorithm is 94.086 percent.In contrast,the overall recognition rate of the random forest is 96.618 percent,indicating that the random forest has a more significant recognition accuracy.In the spiral ascending mode,the maximum identification errors of magnetic detection data in width,length,and depth are 0.21 mm,0.3mm,and 0.14 mm,respectively.