论文部分内容阅读
采用RBF神经网络对204组X70管线钢生产数据进行训练,建立了管线钢成分与力学性能的预测模型,经检验该模型预报精度高,网络预报值与实际值较吻合。利用此模型预报了C、Mn、Mo、Nb、V、Ti等元素含量对管线钢性能的影响规律,并在此基础上确定了X80管线钢的成分范围。对试制生产的X80管线钢进行组织性能检测,结果表明,X80钢的显微组织主要由针状铁素体和粒状贝氏体组成,晶粒细小,力学性能指标达到X80管线钢应用要求。
The production data of 204 sets of X70 pipeline steel were trained by RBF neural network, and the prediction model of composition and mechanical properties of pipeline steel was established. The forecasting accuracy of the model was good and the predicted value of network was in good agreement with the actual value. The model was used to predict the influence of C, Mn, Mo, Nb, V, Ti and other elements on the properties of pipeline steel. Based on this, the composition range of X80 pipeline steel was determined. The microstructure and mechanical properties of X80 pipeline steel were tested. The results show that the microstructure of X80 steel is mainly composed of acicular ferrite and granular bainite, and the grain size is small. The mechanical properties of X80 steel reach the requirements of X80 pipeline steel.