Study On Printing Defects Of Mixed Polymer Materials Based On Machine Learning

Based on the key research and development plan of the Ministry of Science and Technology,"3D printing degradable interbody fusion cage",this paper is further developed on the basis of some completed projects.In this paper,warping deformation,a common defect when mixed polymer materials are used as printing materials in the micro extrusion 3D printing process,is taken as the research object.Through analyzing the entire printing process,a qualitative physical model is established,and the generation process and influencing factors of warping defects are discussed by means of finite element analysis of the thermal conduction of the micro extrusion module.select features,introduce algorithms and analyze reasons for printing scattered data according to machine learning methods.The main contents of this paper are:(1)The working principle and process of micro extrusion 3D printing technology are analyzed,and the adjustable parameters and their functions in the printing process are pointed out;The mechanism and influencing factors of the warping deformation of the formed parts during the processing are analyzed;The measuring method of the warpage of the formed parts in the processing is described.(2)Analyze the heat transfer process of micro extrusion 3D printing technology,build the finite element simulation model of extrusion module according to the printing equipment,and verify the relationship between nozzle temperature and temperature control through experiments.The physical model is established according to the warpage,the qualitative relationship of the model is verified by experiments,the experimental data is collected,and the measures to reduce warpage are analyzed through the experimental results.The qualified printing scaffold was partially characterized and the adhesion of fibroblasts on the scaffold was observed by SEM.(3)In order to conform to the use mode of the decision tree algorithm,preprocess the printed experimental data,analyze the concepts and principles of the decision tree and random forest algorithm,and analyze the features that have a greater impact on the printed data.Finally,the two most influential variables in the warpage characteristics are found to make efforts to greatly alleviate the warpage defects.