Modeling And Control Of Hydraulic Servo Systems With Unknown Dynamic

With the rapid development of control theory,computer technology and micro-electronic technology,and the popularity of microprocessor and microcomputer,the hydraulic servo systems have played an important role in many industrial applications.Then motion control of hydraulic servo systems becomes one of the important topics in the industrial areas.As the key part in the hydraulic servo systems,the strong nonlinearities of hydraulic actuators,cylinder and control elements(e.g.,electro-hydraulic servo valve)seriously affect the control performance.Therefore,how to suppress or even eliminate the effect of these unknown dynamics in the control system,such as parametric uncertainties,unmodeled dynamic,modeling errors,unknown varying load,disturbances and so on,and improve the control performance is an important and emerging issue for the control of hydraulic servo systems.However,the traditional control strategies,e.g.,PID and backstepping control,are difficult to fully handle the unknown dynamics in the hydraulic system.On the other hand,backstepping control suffers from the "explosion of complexity" issue,when it is used for high-order systems.Moreover,it requires all the system states,which are difficult to measure in the practical applications.Thus,the design of output-feedback controller for hydraulic servo systems,not only needs to solve the aforementioned problems,but also needs to ensure fast system response,better transient and steady-state performance.In this thesis,we study the output-feedback control designs for hydraulic servo systems with unknown dynamics.The main work can be summarized as:(1)Modeling of hydraulic servo systems.Consider the dynamics of hydraulic actuator and servo valve in the hydraulic servo system,and derive the system dynamic model based on the flow equation of servo valve,the continuity equation of hydraulic cylinder and force equation of load.Finally,appropriate state variables are chosen to formulate the state-space equation of hydraulic servo system.(2)Adaptive output-feedback control for hydraulic actuating servo systems.First,a coordinate transformation is introduced to reformulate the strict-feedback system model into a canonical form to avoid the backstepping technique.Then we use a high-order sliding mode differentiator to observe the unknown system states.Moreover,two adaptive output feedback control methods with different parameter estimation methods are designed to achieve motion tracking control.In the first scheme,a new adaptive law driven by both the tracking error and estimation error is introduced to design an output feedback controller,which can ensure tracking performance and accurate estimation simultaneously.In the second scheme,the adaptive law is further tailored for estimating the time-varying parameter estimation,which includes two leakage terms to improve the convergence rate of tracking error and estimation error.Finally,the system stability is theoretically exemplified and simulation results are given to verify the effectiveness of the proposed methods,respectively.(3)Prescribed performance control based on unknown system dynamic estimator.In order to accommodate the unknown dynamics and improve the practicability in the industrial applications,an unknown system dynamic estimator(USDE)with only one tuning parameter is developed to estimate the lumped unknown dynamics,which retains exponential error convergence and thus helps to achieve precise position tracking control.A prescribed performance function(PPF)characterizing the error convergence rate,maximum overshoot and steady-state error is utilized in the control design to guarantee both the transient and steady-state performance.Finally,simulation results are given to verify the effectiveness of the presented method.Apart from the above theoretical studies,we also carried out practical experiments based on a hydraulic servo system to verify the effectiveness of the addressed algorithms.Two different adaptive output-feedback control methods are tested.Finally,experiments on the hydraulic servo test-rig are given to compare the performance of generic adaptive control based on the gradient descent method with these two presented control algorithms.