전기-정유압 액추에이터(EHA)의 제어 성능 향상에 관한 연구

Abstract

Hydraulic power transmission systems play an important role in construction equipment, transportation, industrial machinery and aircraft. Conventional hydraulic power transmission systems are mainly based on valve-controlled hydraulic systems. The conventional hydraulic systems have several drawbacks; that is, they have a low energy efficiency and require a heavy/bulky oil reservoir. Pump-controlled hydraulic systems, known as hydrostatic transmission(HST) are used as an alternative for the valve-controlled system. Most HST use a variable displacement pump driven at a fixed shaft speed. The movement of the hydraulic actuator is regulated by changing displacement of the pump bi-directionally. In the HST, the pump runs irrespective of the actuator’s motion. Thereby, noise and vibration due to the pump’s ripple can arise even when the actuator is not moving. To address these issues, electric hydrostatic actuator (EHA) has been introduced since the beginning of 2000. The mathematical model of EHA contains uncertainties due to the variation in the effective bulk modulus of the oil in the cylinder, and the variation in the damping coefficient in the pump and the actuator during actuation of EHA. Also, nonlinearities in the model exist due to discontinuous friction in the moving part of EHA. These uncertainties and nonlinearities, together with the external load, act as disturbances to EHA control system. In this study, to suggest effective control strategies for EHA, the author designed and fabricated an EHA and an experimental system for the EHA research. Several control strategies have been applied to the EHA, and the control performances are analysed by simulations and experiments. The mathematical model of the EHA is described in chapter two of this dissertation. At first, the model is developed experimentally by applying a simple closed loop control to the EHA and carrying out a frequency response characteristics test. Also, the model is obtained from the governing equations for the EHA consisting of a servo motor, a hydraulic pump and a hydraulic actuator. Some physical parameters are evaluated from preliminary tests. Finally, the mathematical models of the EHA from the experimental approach and the governing equations’ approach are compared and the validity of the models is confirmed. In chapter three, to suggest a control strategy for the EHA with excellent tracking control performance, a control system using an input shaping filter(ISF) is suggested. A PI-D and a feed-forward controller are applied to obtain robust tracking control while rejecting the effects of the disturbances. The feed-forward controller is very effective in obtaining a fast response, but it leads to an increase in the control input, even over the saturation limit, as the frequency of reference input increases. We may designate this concern a high gain symptom of the feed-forward controller in the frequency response. In this study, the high gain symptom is addressed by applying an input shaping filter (ISF) to the reference tracking control system for the EHA. The efficacy of the suggested control design is experimentally verified. If EHAs with system model’s uncertainties and nonlinearities are controlled using linear controllers, which is most common so far, it is not easy to achieve satisfactory control performances, as the linear controllers have to be dimensioned conservatively to ensure stability. As a countermeasure to overcome this difficulty due to system model’s uncertainties and nonlinearities, in chapter four and chapter five, sliding mode controllers(SMC) are applied for controlling the EHA. In the application of SMC to the EHA in chapter four, an extended Luenberger observer(ELO) is adopted to estimate the disturbance and the state variables in the EHA model. In the ELO, a disturbance with a combined form of the model’s uncertainties, the model’s nonlinearities and the external load is estimated. The estimated disturbance is reflected in the control input to reject the effect of the disturbance, and eventually to improve the control performances. Also, the state variables estimated in the ELO are used in materializing the SMC. A comparative study for the application of the SMC with the ELO versus the SMC with Luenberger observer(without disturbance observer) is performed through experiments. This comparison demonstrates the performance improvements resulting from SMC with the ELO, and the added robustness under the system model’s uncertainties and nonlinearities. In the application of SMC to the EHA in chapter five, a sliding mode observer(SMO) is adopted to estimate the disturbance and the state variables in the EHA model. In the SMO, a disturbance with a combined form of the model’s uncertainties, the model’s nonlinearities and the external load is estimated. The disturbance and the state variables estimated in the SMO are used in realizing the SMC in the same way as chapter four. The robustness of the SMO under the system model’s uncertainties and nonlinearities is investigated through the simulation by providing intentional variation of system’s parameter values, and providing disturbances with the type of step signal and sinusoidal signal. The effectiveness of the SMC with the SMO is confirmed through simulation and experiment.