定模动辊机电系统动力学智能优化方法

定模动辊成型是一种变截面渐进成型方法，依赖定模动辊成型装备实现高强度金属板材变截面渐进成型。定模动辊成型装备运动平稳性与构件成型质量密切相关，因此以加速度幅值最小为目标函数，考虑定模动辊成型装备伺服电机、传动装置、辊模机构与成型板材的机电系统动力学优化，是机电装备设计理论研究的新课题。采取深度强化学习方法，应用深度确定性策略梯度算法对定模动辊成型装备进行优化。结果表明:该算法适用于机电系统动力学优化，具有科学意义和工程应用价值。

Dynamic Intelligent Optimization Method of Roll-die Forming Electromechanical System

Roll-die forming is a variable cross-section progressive forming method, which relies on roll-die forming machine to realize variable cross-section progressive forming of high-strength metal sheet. The motion stability of the roll-die forming machine is closely related to the component forming quality, therefore, taking the minimum acceleration amplitude as the objective function and considering the electromechanical system dynamics optimization of the servo motor, transmission device, roll forming mechanism and formed plate of the roll-die forming machine is a new topic in the theoretical research of mechanical and electrical equipment design. The method of deep reinforcement learning was used, the deep deterministic policy gradient was applied to optimize the roll-die forming machine. The results show that the algorithm is suitable for the dynamic optimization of electromechanical system and has scientific significance and engineering application value.