The Temperature Field Of H-beams During Cooling Process Was Analyzed And Numerically Simulated And Mechanics Properties Prediction

The technology of H-beam steal controlled cooling after rolling is an on-line heat treatment which utilizes the rolling heat for quenching and self-tempering. The technology will significantly be expanded in the future because of its improvements in H-beam steal mechanical properties and its cost saving.The subject takes hot-rolled Q235 H-beam steel as research object, making sure to controlled cooling project for first naturally cooling 5s then water cooling 5s again naturally cooling 5s. With the finite element analysis software, the thesis simulates and analyses H-beam steel undercooling the temperature field. I find that exerting the different fluent density to the different nodes, for example w1=2.2 (waist), w3=5(leg1), w4=3 (leg2), w5=5.5 (bevel), has well controlled cooling effect. When naturally cooling has been finished after 15s, (namely tempering ended), the maximum temperature is 696.9℃, the lowest temperature is 661.1℃, the maximum temperature difference only has 35.8℃.Mechanics properties in H-beam steel cooling process and gathering data from the Lai Wu iron and group corperation has been analysed, I find that it may enhance obviously H-beam steel mechanics properties (the yield strengthσ_s , the tensile strengthσ_b, the elongation ratioδ_s) through adjusting Q235 H-beam steel material chemical composition (C, Si, Mn) the percentage and parameter of controlled cooling (open cold temperature, cooling rate, end cold temperature).σ_s andσ_b enhance approximately 40 Mpa,δ_s reduces approximately 1%.In this thesis, I construct prediction model of the artificial neural network technique which predicts H-beam steel mechanics properties through the chemical composition of the Q235 H-beam steel material and the controlled cooling processing parameter. To the network which has been trained on the test, I finally discover that the forecast value of the network model and the actual value are basically tallies. The maximum error is approximately 5.1%, it is in reasonable erroneous scope. The forecast model is reliable.This topic can provide the scientific method and the theoretic basis for H-beam steel controlled cooling technology.