基于深度强化学习的数控铣削加工参数优化研究

Research on machining parameter optimization of NC milling based on deep reinforcement learning

铣削加工参数设定决定了生产效能与设备使用寿命。为此,文章针对参数设定多采用经验算法对其进行优化,以深度强化学习算法为基本工具,引入铣削合力、材料去除率等指标对其进行多目标优化构建;并在仿真的条件下进行多轮迭代,最终确定最优解的近似值,以此为具体的参数进行加工生产实践,发现在效能上具有较大的提升效果,证实了此种方式可以作为数控铣削加工参数优化的方式予以应用及推广。

The setting of milling parameters determines the production efficiency and the service life of equipment.Therefore,this paper mainly uses empirical algorithm to optimize the parameter setting,takes the deep reinforcement learning algorithm as the basic tool,introduces the milling force,material removal rate and other indicators to construct the multi-objective optimization.After several rounds of iteration under simulation conditions,the approximate value of the optimal solution is finally determined,and it is found that it has a great improvement effect on the efficiency of machining production practice.It proves that this method can be used as a way to optimize the machining parameters of CNC milling.