The Experimental Analysis And Numerical Simulation Of Technical Parameters For Steel Plate By Induction Heating

The heat source of traditional line heating process is flame (oxygen-acetylene flame), there are many defects in this form of heat source, such as low heating efficiency, difficult to control heat and more polluting, etc. The application of induction heating into lineheating process can fundamentally increase the quality of steel forming, decrease the molding time, improve the work situation and can also save lots of energy. This issue is an interdisciplinary research. Set theory, research, development, and testing in one.In the first place, the basic principles and characteristics of induction heating was introduced in this paper. Then analyzed the numerical methods of electromagnetic field, temperature field and thermal elastic-plastic of induction heating, established the finite element model of induction heating by the large-scale general-purpose finite element analysis software ANSYS. Analyzed the characteristic of steel plate temperature and deformation by numerical simulation, verified the accuracy of numerical simulation by experiment, proposed a time discontinuous induction heating method and analyzed of it.In the second place, the edge temperatures and lateral shrinkage of plate was studied by Physical experiments and numerical calculation. Then selected the FEM calculation results as the sample data, modeled the shrinkage of steel plate surface by multiple linear regression method, verified the accuracy of the model by physics experiment.In the third place, a series of experiments have been carried out by mobile induction heating, the results showed that steel plate forming better in mobile heating. Then according to the experimental data of mobile heating, the mathematical model of plate edge shrinkage has been established and the accuracy of the model has been verified.Finally, a neural network program which can predict the shrinkage and heating parameters has been made by Visual Basic 6.0. Then compared the method of neural network and regression, verified the accuracy of prediction result and have some guiding significance to the study of curved hull plate forming.