论文部分内容阅读
针对冷轧宽带钢连续热镀锌生产线上的拉矫机,为了建立可针对不同钢种、规格、来料板形的拉矫机插入深度、延伸率和工艺张力等重要工艺参数的设定模型,结合对拉矫变形过程的力学分析和现场实测数据统计与分析的结果,从板形良好的钢卷中择取有代表性的生产数据作为训练样本,将良好的操作经验转化为数学模型,利用BP网络模型的自适应和自学习优势设定带钢拉矫过程中的工艺参数。此模型可以将新的样本随时添加到网络的矩阵中,对新品种具有较好的扩展能力。经现场试用表明,它能准确地预设定拉矫机工艺参数,基本满足工厂生产要求。
In view of the straightening and straightening machine on the continuous hot-dip galvanizing line of cold-rolled wide strip, in order to establish the setting model of important process parameters such as depth of penetration, elongation and process tension for different steel grades, specifications, , Combined with the mechanical analysis of the straightening deformation process and the results of field measurement data statistics and analysis, the representative production data is taken as the training sample from the good shape of the coil and the good operating experience is transformed into the mathematical model, Adopting the advantage of self - adaptation and self - learning of BP network model to set the process parameters during strip straightening. This model can add new samples to the network matrix at any time, and has good scalability for the new breed. The field trial shows that it can accurately preset the straightening process parameters, the basic meet the factory production requirements.