冷轧变形抗力智能算法模型研究  

Research on intelligent algorithm model of cold rolling deformation resistance

作　　者：肖保树[1] 沈至諴 XIAO Baoshu;SHEN Zhixian(Shanghai Baosight Software Co.,Ltd.,Shanghai 201900,China)

机构地区：[1]上海宝信软件股份有限公司,上海201900

出　　处：《宝钢技术》2023年第4期27-31,共5页Baosteel Technology

摘　　要：阐述了传统的冷轧变形抗力计算模型,研究了影响带钢变形抗力的相关因素,提出了基于TensorFlow深度学习框架和Keras神经网络算法库,建立冷轧变形抗力自学习模型,对比分析传统模型与神经网络自学习模型预测变形抗力参数的精度,有效提高了模型预测的精度,大大减少模型的人工干预操作。The authors describe the traditional calculation model of cold rolling deformation resistance,study the relevant factors affecting the deformation resistance of strip steel,propose a self-learning model of cold rolling deformation resistance based on the TensorFlow deep learning framework and Keras neural network algorithm library,compare and analyze the accuracy of the traditional model and neural network self-learning model in predicting deformation resistance parameters,effectively improve the accuracy of model prediction,and greatly reduce the manual intervention operation of the model.

关 键 词：变形抗力 自学习模型 深度神经网络 

分 类 号：TG335.12[金属学及工艺—金属压力加工] TP183[自动化与计算机技术—控制理论与控制工程]