| 阮金华.基于有限元数值模拟的宽厚板热轧过程宽度控制ANN模型[J].,2013,53(5):678-684 |
| 基于有限元数值模拟的宽厚板热轧过程宽度控制ANN模型 |
| FE numerical simulation-based ANN model for heavy plate width control during hot rolling process |
| |
| DOI：10.7511/dllgxb201305010 |
| 中文关键词: 热轧宽厚板 立轧-平轧 宽度控制 有限元数值模拟 人工神经网络 |
| 英文关键词: hot rolling heavy plate vertical-horizontal rolling width control FE numerical simulation artificial neural network (ANN) |
| 基金项目: |
|
| 摘要点击次数: 1228 |
| 全文下载次数: 1187 |
| 中文摘要: |
| 利用商业有限元分析软件DEFORM-3D,建立宽厚板热轧过程中立辊侧压和随后平轧过程的三维热力耦合刚塑性有限元模型．模拟了不同板坯厚度、板坯宽度、立辊压下量和平辊压下率条件下320种立辊侧压和随后平轧过程,探讨了不同工艺参数对调宽效率的影响规律．模拟结果表明,调宽效率随着板坯宽度和立辊压下量的增加而升高,随着板坯厚度和平辊压下率的增加而降低．基于上述有限元数值模拟结果,借助BP人工神经网络,建立了热轧宽厚板立轧-平轧宽度控制模型,经测试,模型预测的调控效率与有限元数值模拟结果符合很好．将宽度控制模型用于宽厚板实际热轧过程的有限元数值模拟,模拟轧件宽度与实测宽度吻合很好． |
| 英文摘要: |
| Using commercial FE software DEFORM-3D, a 3D thermo-mechanical rigid-plastic FE model was developed to simulate a vertical-horizontal hot rolling process for heavy plate. The 3D FE model was used to simulate 320 kinds of different vertical-horizontal rolling processes under different rolling parameters, such as slab thickness, slab width, edging reduction and thickness reduction ratio, and the influences of different process parameters on edging efficiency were discussed. Simulation results show that edging efficiency increases with increment of edging reduction and slab width, while decreases with increment of slab thickness and thickness reduction ratio. Based on the FE numerical simulation results, a heavy plate width control model for vertical-horizontal hot rolling process was established with the aid of a BP artificial neural network (ANN). Edging efficiencies predicted by the model are in good agreement with the FE numerical simulation results. The heavy plate width control model was employed in an FE numerical simulation of a real heavy plate rolling process to control the plate width, the simulated plate width is in good agreement with the measured one. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|