Research On Monitoring Temperature Field Of Laser 3D Printing Using Color CCD

Temperature field monitoring plays an important role in the study and control of melt pool state,thermal deformation and residual stress during laser metal 3D printing.The CCD-based thermometry is a non-contact temperature measurement method that is not susceptible to interference from the measured object and allows for fast and effective monitoring of temperature field.The method is suitable for high temperature measurement environments.This thesis presents an in-depth analysis and experimental study on the subject of temperature prediction using colorimetric thermometry,based on the use of color CCDs to monitor the temperature field of 3D printing.The details are as follows.A data acquisition system for color CCD temperature measurement is designed and established to obtain the relationship between temperature and image data.The commonly used 304 and 316 stainless steels are selected as the test samples and heated by a voltage-controlled ceramic heating rod.Images are collected by two color CCDs with filters and the temperature data is measured by a thermometer.In order to study the effect of metal oxidation on the experimental results,the above experiments are carried out in oxygen and argon atmospheres respectively.For some uncertainties in the temperature acquisition process,such as the asymmetry of the angle of the dual cameras,the possession of different planes between cameras and the object,and the uncertainty of the coordinate positions of the pyrometer measurement points,etc.,image processing methods such as filtering,projection transformation,edge detection and Hough circle detection are used to correct the images to reconstruct the coordinate correspondence between the two CCD images and to obtain the coordinates of the temperature measurement points,thus minimising experimental errors.To address the uncertainty of the emissivity in the temperature measurement model,a modified emissivity ratio model is proposed based on the MRT(Mathematical multispectral radiation thermometry)emissivity model and the characteristics of the colorimetric temperature method.The model parameters are determined by calibration data and the prediction error of the model is judged by test data.Experimental results show that the new model has a low error of about 1.25% for temperature prediction,reducing the influence of metal emissivity changes on temperature predictions at different temperatures.In addition,for the same metal,the emissivity ratios in oxygen and argon environments have a high similarity in trends and little numerical error,which allows the same set of model parameters as far as the error allows.