Contouring Error Estimation And Cross-Coupling Control For Biaxial Servo System

Contouring error is a key criterion in high-speed and high-precision manufacturing of multi-axes feed system. The contouring motion control is also the crucial problem in high-speed and high-precision manufacturing. As such, how to achieve the required contouring tracking accuracy especially during high-speed and large-curvature contouring tasks, has always been an important problem to address in industrial applications. In this dissertation, we pay attention to the contouring error estimation methods and contouring error control methods.The conventional tangent line approximation is defined as the first-order approximation based on the local geometry properties, which can achieve satisfied contouring error estimation performance with linear contour or low-speed and small-curvature contour. That means come across with high-speed and large-curvature contour, the contouring error estimation performance deteriorate seriously. The circular approximation is the quasi-second-order approximation based on the desired contour. But it is limited to the planar contour or small torsion spatial ones. So a novel contouring error estimation method named natural local approximation is proposed in this dissertation, which is suitable for not only 2-dimensional planar contours but also 3-dimensional spatial ones. Ellipse contour simulation based on Simulink demonstrated that the contouring performance is improved by 12.6%(compared with circular approximation based method) when the feed angular was at ! = 2:0? rad=s. Meanwhile, the ellipse contour following experiment based on linear motor servo motion stage indicated that the proposed method can decrease 16.17% contouring error(compared with circular approximation based method) when the feed angular was at ! = 2:0? rad=s.Taking the uncertainties of actual system model and external disturbances into account, a modified cross-coupling adaptive control scheme by combining the cross-coupling control based on position-loop(PLCCC) and adaptive contour strategy together is proposed in this dissertation. Ellipse contour simulation based on Simulink demonstrated that the contouring performance is reduced by 33.99%(compared with PLCCC) when the feed angular was at ! = 2:0? rad=s. The ellipse contour following experiment based on linear motor servo motion stage showed that the proposed scheme can decrease 28.58% contouring error(compared with circular approximation based method) when the feed angular was at ! = 1:4? rad=s.