金属力学性能质量预测及其控制

Quality prediction model and control of the metallic mechanical properties

中文关键词  力学性能  人工神经网络  组织性能预测和控制  自适应逆控制

英文关键词  mechanical properties  artificial neural network  structure property prediction and control  adaptive inverse control

中文摘要

      为了在生产中能够对力学性能做出精确的预测, 在用神经网建立起由工艺参数预测力学性能的质量模型后, 又提出一种新的建模方法—-逐层逼近法, 从测试的结果看, 后者预测精度明显高于前者. 然后, 对利用自适应逆控制方法实现对力学性能的在线控制进行了研究, 仿真结果证明了该方法的有效性.

英文摘要

      In order to precisely predict the mechanical properties, the model, that could predict the mechanical properties of hot-rolled steel strip with the technological parameter, is established based on artificial neural network. A new method by applying the technology of layer-to-layer prediction and impending is also proposed. The test result shows that the precision of the latter prediction is higher than that of the former. Then, the real time control method of the mechanical properties is researched and simulated with adaptive inverse control. Simulation result shows that this control method is effective.