冷带轧机液压自动厚度控制系统失稳机理及智能控制策略研究

As the core control system to ensure the thickness accuracy of cold rolled plate, the hydraulic automatic gauge control (HAGC) system is critical for the rolling process to obtain high precision, high speed, continuousness and steadiness. Meanwhile, the improvement of its intelligent degree is also the key to enhance the intelligent level of cold rolling mill. In this project, the HAGC system of cold rolling mill is taken as the research object. Combining with nonlinear dynamics theory, nonlinear vibration theory, data mining and artificial intelligence, the instability mechanism and intelligent control strategy for HAGC system are studied. Firstly, the electromechanical-hydraulic coupling nonlinear dynamics model of HAGC system is established. Then, the instability conditions of the basic closed-loop subsystem are deduced. Secondly, the influence laws of structural parameters and nonlinear factors on system dynamics behaviors are explored. And the instability mechanism and inducement of HAGC system are revealed. Finally, an intelligent control strategy based on reinforcement learning is constructed to adaptively control the instability vibration of HAGC system. In terms of the above, some problems in HAGC system can be solved such as exploring the instability mechanism, process optimization decision and intelligent control, etc. The expected results will lay a theoretical foundation for vibration traceability and restrain of HAGC system. Moreover, the research results help to ensure the safe and stable operation of cold rolling production, and contribute to upgrade the intellectualization of cold rolling mill. Furthermore, the achievements will provide an innovative and key technology foundation for the research and development of high-precision and intelligent thickness control system with domestic intellectual property.

液压自动厚度控制系统作为保证冷轧板厚精度的核心控制系统，其工作的可靠性是保证高精度、高速、连续稳定轧制的关键，同时，其智能化程度的提高亦是提升冷带轧机智能化水平的关键。本项目以冷带轧机液压自动厚度控制系统为研究对象，结合非线性动力学理论、非线性振动理论、数据挖掘和人工智能等手段，开展该系统失稳机理及智能控制策略研究。首先，建立系统机电液耦合非线性动力学模型，推导系统关键基础闭环子系统的失稳条件；其次，探究结构参数和非线性因素对系统动力学行为的影响规律，揭示系统的失稳机理和诱因；然后，构建基于强化学习的智能控制策略，对系统失稳振动进行自适应控制。研究成果可以破解制约液压自动厚度控制系统失稳机理探究困难、过程优化决策与智能调控等难题，为该系统的振动溯源与抑制奠定理论基础；而且将有助于保障冷轧生产的安全稳定运行、助力冷带轧机智能化升级，为国产高精度智能板厚控制系统的研发提供创新性关键技术基础。

冷带轧机液压自动厚度控制系统是保证冷轧板厚精度的核心控制系统，其工作的可靠性是保证高精度、高速、连续稳定轧制的关键，同时，其智能化程度的提高亦是提升冷带轧机智能化水平的关键。本项目以冷带轧机液压自动厚度控制系统为研究对象，以提高轧制过程运行的稳定性、提升冷带轧机的智能化水平为主要目标，建立了液压自动厚度控制系统机电液耦合非线性动力学模型，推导出了关键位置闭环子系统和压力闭环子系统的失稳条件；探究了控制腔容积、刚度系数、阻尼系数、增益系数、非线性弹性力、非线性阻尼力、非线性激励力等结构参数和非线性因素对系统动力学行为的影响规律，揭示了液压自动厚度控制系统失稳机理和诱因；构建了基于强化学习的智能控制策略。研究成果为液压自动厚度控制系统的振动溯源与抑制奠定了理论基础，有助于保障冷轧生产的安全稳定运行、提升成材率和产品质量、助力冷带轧机智能化升级，为高精度智能板厚控制系统的研发提供了关键技术基础。培养博士研究生4名、硕士研究生3名；出版学术专著1部；发表学术论文11篇（SCI收录10篇）；授权发明专利4件。