부산대학교 도서관

In order to improve the control accuracy of the multi-cylinder hydraulic press machine (MCHPM), a mathematical model considering the characteristics of MCHPM is established in this paper. A design method of MCHPM optimization controller considering the characteristics of different operation phases is proposed. This method separates the optimization controller design into three steps: the design of high-level motion controller, the design of control allocation based on optimization and the design of actuators control laws. For achieving the control goal, the highlevel motion controller uses the way of the matrix singular value decomposition and the sliding mode control method to obtain the vector of virtual. Then, in the light of the different operation phases of the MCHPM, different optimization goals are designed. The vector of virtual is used to calculate the expected driving force of each hydraulic cylinder by control allocation theory. Finally, the expected driving force is used as the expected value to get actuators control laws. Compared with the simulation results without considering different operation phases, it shows that the optimization controller can achieve better position tracking and leveling of the slider under the variable load force at different positions and angles. It can also greatly reduce the allocation errors.