Machining force control with intelligent compensation

Force control is an effective means of improving the quality and efficiency of machining operations, so various approaches for force control have been proposed. However, due to the nonlinear, time-varying and uncertain characteristics of machining processes, it is difficult to develop force control systems that are stable and robust over the full range of operating conditions. This study proposed two control schemes to address such difficulties in the field of nonlinear force control by using a linear feedback proportional-derivate (PD) controller respectively with two different nonlinear intelligent compensators: fuzzy logic compensator (FLC) and neural network compensator (NNC). The PD controller is used to improve the transient response while maintaining the stability of the process system, and the FLC or NNC is employed to eliminate the steady-state error and compensate for the system nonlinearity (or uncertainty). The applications of the proposed schemes in machining processes show that the controllers adapt well to nonlinearity under time-varying cutting conditions in comparison to PID, PD, and FLC. The online updating of the NNC parameters through the Feedback-Error Learning can further improve the system performance.