Improved forecasting compensatory control to guarantee the remaining wall thickness for pocket milling of a large thin-walled part

Abstract

Pocket milling is widely used in the machining of large thin-walled parts. The remaining wall thickness of the pocket is critical to compromise the weight reduction and the strength, and becomes a crucial dimension requirement during machining. However, the cutting deformation of large thin-walled parts will greatly decrease the accuracy of the remaining wall thickness and needs to be compensated online due to the stochastic characteristics of the deformation. For online compensation methods, the time-delay is usually the critical factor to reduce the control accuracy. The forecasting compensatory control (FCC) provides an effective way to compensate the stochastic errors and solve the time-delay problem. The key of the FCC is the prediction accuracy of the modeling technique. An improved FCC system with an accurate cutting deformation prediction model is developed to guarantee the remaining wall thickness for the pocket milling of a large thin-walled part. The improved FCC system makes use of the advanced online measurement system, the deformation prediction modeling, and the real-time compensation. The proposed prediction model considers the deterministic and stochastic cutting deformations to improve the prediction accuracy. The Kalman filtering is also applied to further enhance the prediction accuracy because of its correctable ability. The cutting simulation and the experiment of machining a rectangular pocket on a large thin-walled plate are both carried out to validate the effectiveness of the proposed method. The accuracy of the remaining wall thickness of the pocket is finally improved.