关键词: 数控机床位置伺服系统, 数据驱动控制, 无模型自适应控制, 预测控制

Abstract:

Since the nonlinearity and time-variation of CNC position servo system during manufacturing in practice, it is difficult to precisely model the position servo system. To avoid the challenge of modeling, a data-driven adaptive control method was adopted in this paper. A compact form dynamic linearization technique was considered to equivalently transform the position servo system. By virtue of the predictive data model, the data-driven predictive position servo control system was designed by applying the model-free adaptive control (MFAC) and bi-level hierarchical techniques. The simulation results demonstrated that, comparing with PID, the accuracy and stability of the adopted data-driven predictive control was improved by 6.2% and 15.94% respectively.

Key words: CNC position servo system, data-driven control, model-free adaptive control (MFAC), predictive control