超薄带钢强耦合隐性缺陷机理及无忧微控理论

Super wide ultra-thin strip has been more and more widely used in industry, especially in the high-end areas with high quality requirements. With the trend of wider and thinner strip with larger width-thickness ratio, the multidirectional metal flow becomes more complex. Fine adjustment on control means may lead to instability and recessive frequent defects of strip, even to dominant defects. Therefore, the defective rate of this type of strip is high. This paper starts from the lateral flow mechanism of metal and process control theory to build the recessive coupling defects mechanistic simulation model and the worry-free micro-control theory model of the large width to thickness ratio ultra-thin strip. The contents are as follows: (1) Based on the hot deformation theory of the strip rolling, the critical flattening phenomenon of the roll system and the state of critical instability of strip has been studied. The effects of lateral flow of metal strip on internal defects have also been recovered. Besides, an online compensation method has been given to fix the lower-level technique signal error from its mechanism. (2) Based on the dynamic effect matrix of process control, a micro control theory model for large width-thickness ratio strips has been established. Properties of the relative gain matrix have been utilized to weaken the strong coupling between them. Then, a fine adjustment to fix hidden defects has been verified. The result of this paper has important theoretical significance on studying of hidden defects formation mechanism, the optimization of rolling process, improving the shape index and realizing fine control.

超宽超薄冷轧带钢在工业中的应用日愈广泛，尤其在高端领域，质量要求极高。随着冷轧带钢趋向更宽更薄，宽厚比很大，内部金属流动复杂多向，控制手段的微小调整，都可能导致带钢失稳，诱使隐性缺陷频发，严重时转变为恶劣的显性缺陷，废次品率居高不下。本项目拟从金属横向流动机理和过程控制理论出发，建立模拟大宽厚比超薄带钢强耦合隐性缺陷的机理模型及其无忧微控模型。创新点：(1) 考虑大宽厚比超薄带钢轧制时的辊系极限弹性压扁和带钢失稳等因素，建立热效应金属变形机理模型，揭示带钢内部金属的横向流动状态及其对复合隐性缺陷的影响，并从机理上研究底层信号的工艺误差补偿方法；(2)基于矩阵控制理论，建立多变量动态无忧微控模型，判断并弱化复合隐性缺陷的强耦合关系，简化大宽厚比超薄带钢复杂的综合调控过程。本项目旨在揭示大宽厚比超薄带钢复合隐性缺陷的演变机理和强耦合关系，为研制数值测试平台和过程控制系统提供理论支撑和科学依据。

超宽超薄冷轧带钢应用日愈广泛，尤其在高端领域，但是同时存在强耦合的隐性板形缺陷，带钢易失稳、废次品率居高不下的问题。本项目针对大宽厚比超薄带钢强耦合隐性缺陷，展开以下深入研究：（1）基于板形控制矩阵和DE-ELM神经网络，建立冷轧带钢板形控制机理-智能协同调控模型。根据带钢金属模型和辊系弹性模型建立板形控制的机理仿真模型，构建静态板形控制矩阵；同时利用DE-ELM神经网络形成动态板形控制矩阵，并利用加权的方法协调板形控制矩阵的影响度，降低板形误差，提高模型的稳定性和板形精度，使不良带钢板形快速调整至良好状态。（2）针对冷轧带钢板形测控信号不稳定、失效等问题，建立板形工艺误差综合补偿模型。通过深入分析零点残存电压、带钢边部板形误差、检测辊挠度、安装精度等因素对原始板形信号的影响，解决板形信号不稳定的问题，防止板形误调事故，提高板形测控精度。（3）基于板形测控数学模型和虚拟仪器技术，研制新型冷轧带钢板形测控系统，实现无忧微控。利用虚拟仪器软件和OPC技术组建虚拟平台，搭建多功能开放式智能板形测控系统，实现板形信号采集、处理、通讯和控制等功能，替代传统的分散布置多机模式，减少通讯环节和硬件配置，便于组态和调试，提高通讯效率和计算精度，便于数据共享和拓展外部模型或硬件，避免传统设备连接的繁琐通讯协议、特定驱动程序和硬件访问冲突等问题，改善板形测控系统的通用性和开放性，满足冷轧带钢的不同板形测控要求。工业实例表明，机理智能协同调控模型能够更快速获得有效板形控制系数，提高板形调控效率，新系统有利于改善板形测控系统的通用性和开放性，满足冷轧带钢的不同板形测控要求安装方便，易于操作，稳定性好，适应恶劣工况，满足在线板形调控要求。