| Pressure bearing components are an important cornerstone of current industrial development and are widely used in key fields such as petrochemicals and rail transit.However,due to the harsh working environment,structural damage such as fatigue cracks and corrosion defects can easily form on pressure bearing components,which is one of the key reasons for failure accidents of pressure bearing components.Therefore,in order to ensure the safety of pressure bearing components during service,it is of great significance to study the detection methods for pressure bearing components.However,most mainstream non-destructive testing methods have certain limitations when testing pressure components,while magneto-optical imaging methods have become a hot research topic in non-destructive testing due to their advantages such as intuitive testing results,high detection rate,and high resolution.At present,research on magneto-optical imaging methods mostly focuses on enhancing magneto-optical images or identifying and classifying defects,with little indepth exploration of the magnetic field distribution mapped by magneto-optical images.Therefore,this article takes the transformation of various information quantities in the process of magneto optical imaging as the starting point,conducts in-depth analysis of the magneto optical sensing process and image sensing process,and conducts research on how to establish a mapping relationship between grayscale and magnetic induction intensity,as well as the optimization of magnetic field distribution results.The main research content and innovation points of this article are as follows:(1)characteristics of signal transformation in magneto-optical imaging in principle are analyzed,and a finite element simulation model of defect leakage magnetic field is established.the distribution patterns of defect leakage magnetic field under different influencing factors are summarized.Subsequently,based on simulation results and detection requirements,the structure and device parameters of the portable magnetooptical imaging defect detection system are designed and optimized,and the development work of the detection system is completed.(2)the influence of non-ideal factors on the results in the magneto optical sensing process is analyzed,and a magneto optical conversion equation is established to characterize the conversion relationship between magnetic signals and optical signals.Then,a camera response curve calibration method based on Debevec algorithm is designed to obtain the mapping curve between grayscale and light intensity.Subsequently,the LM nonlinear optimization algorithm is used to achieve nonlinear fitting and unknown parameter solution of the magnetic field grayscale calibration equation.Analysis shows that the magnetic field resolution of the calibration equation output reaches 0.2157 m T.(3)Zhang Zhengyou calibration method is used to correct the distortion of magnetooptical images and extract effective detection areas from the correction results.Then,the composition of interference information in magnetic field distribution images is analyzed,and an interference calibration method is designed to achieve background interference filtering.In addition,an improved non-local mean filtering algorithm is proposed to achieve denoising of magneto-optical images,and the denoising effect of the improved NLM algorithm is compared with that of the original NLM algorithm.The results show that the improved NLM algorithm can effectively denoise while better preserving defect imaging details,and the relative error between the calibration result and the measured magnetic field value is within 3%... |
PDF Full Text Request