论文部分内容阅读
用人工神经网络研究了化学成分及热处理工艺参数对低碳低合金钢的硬度的影响。首先设计了RBF型人工神经网络模型,用“舍一法”改进了模型,使其具有较好的预测性能。然后,用神经网络研究了化学成分和冷速对低碳低合金钢的硬度的定量影响。结果表明,碳的质量分数为0.11%～0.15%时,硬度随碳含量的增加而增大;硅的质量分数为0.24%～0.38%、锰的质量分数为0.94%～1.02%时,硬度值基本不变;铬的质量分数为0～0.6%时,硬度值呈增加趋势;镍的质量分数为0～0.04%时,硬度值基本不变;钼的质量分数为0～0.2%时,硬度值从HV 288降至HV 282;硼的质量分数为1%～2%时,硬度随含量增加而升高;钛、铌、钒的总质量分数为0.06%～0.14%时,硬度值基本不变;冷速从10℃/m增加至170℃/m,硬度值从HV 290增至HV 420。
The effects of chemical composition and heat treatment process parameters on the hardness of low carbon and low alloy steels were studied by artificial neural network. Firstly, the model of RBF artificial neural network is designed, and the model is improved by using “a method ” to make it have better prediction performance. Then, the quantitative effects of chemical composition and cooling rate on the hardness of low carbon and low alloy steel were studied by using neural network. The results show that with the mass fraction of carbon of 0.11% ~ 0.15%, the hardness increases with the increase of carbon content; the mass fraction of silicon is 0.24% ~ 0.38% and the mass fraction of manganese is 0.94% ~ 1.02% The hardness is basically the same; when the mass fraction of chromium is 0 ~ 0.6%, the hardness value is increasing; the hardness value is basically unchanged when the mass fraction of nickel is 0 ~ 0.04%; when the mass fraction of molybdenum is 0 ~ 0.2% The value decreases from HV 288 to HV 282. The hardness increases with the content of boron when the mass fraction of boron is 1% ~ 2%. When the total mass fraction of titanium, niobium and vanadium is 0.06% ~ 0.14% The cooling rate increased from 10 ℃ / m to 170 ℃ / m, and the hardness increased from HV 290 to HV 420.