Font Size: a A A

Research On Control Method Of MDF Continuous Hot Press Hydraulic Servo System Based On Prescribed Performance

Posted on:2020-01-03Degree:MasterType:Thesis
Country:ChinaCandidate:Y G ZhouFull Text:PDF
GTID:2381330578973985Subject:Pattern Recognition and Intelligent Systems
Abstract/Summary:PDF Full Text Request
Due to its excellent material and mechanical properties and low manufacturing costs,medium density fiberboard(MDF)has quickly become the main product in the wood-based panel market.In the MDF production process,the fixed thickness stage of continuous hot press determines the quality of the product ultimately.Therefore,in order to improve the quality of medium-density fiberboard products,how to achieve stable and rapid position tracking of the continuous hot press hydraulic system has become a key issue in the medium density fiberboard production industry.However,due to the inherent mechanical characteristics of the MDF continuous hot press hydraulic system and the external load force disturbance,the steady state performance and dynamic performance of the system have difficulties to be guaranteed.In addition,the system also has problems such as input saturation and unmeasurable state,which makes the controller designing more difficult.In order to solve the above problems,this dissertation takes the prescribed performance control as the main control method,by introducing neural network and Nussbaum function,to study the position tracking control problem of the MDF continuous hot press hydraulic system.The specific researching work is given as follows:To solve the problem that the transient performance of the MDF continuous hot press hydraulic system is difficult to improve,a novel controller with prescribed performance is designed.In order to effectively constrain the dynamic performance parameters(convergence time,overshoot,etc.)of the hydraulic system,a performance function is constructed,and a new type of error transition function is designed based on the barrier Lyapunov function,so that the controller is effectively simplified in theory,which is convenient to combine with other control methods.In order to solve the problem of irregular perturbation and external unknown interference of MDF continuous hot press hydraulic servo system,an adaptive RBF neural network approximator is proposed to approximate the unknown composite disturbance.By comparing it with the prescribed performance control algorithm,a new controller is designed for achieving fast,accurate and no-overshoot position tracking of the hydraulic system.Simulation experiments designed to verify the effectiveness of the controller.Considering the input saturation problem of the MDF continuous hot press hydraulic system in engineering production,by introducing the hyperbolic tangent function and Nussbaum function,the prescribed performance anti-saturation controller is proposed.Under the premise that the steady state performance and dynamic perfonnance of the system arc guaranteed,the input saturation of the system is compensated,thereby improving the overall robustness of the system.Applying the proposed controller to the control system,and the feasibility and effectiveness of the controller are verified by theoretical derivation and simulation experiments.For the problem of the unmeasurable state quantity of the MDF continuous hot press hydraulic system,this paper designs a reduced-order state observer,which can observe the unknown unmeasurable states.The designed observer is combined with the neural network preset controller,so that the system can still achieve stable,fast and no overshoot position tracking target under the condition that the partial state quantity is known.Simulation experiments are used to confirm the effectiveness of the above control method.In summary,aims to improve the steady-state and dynamic control performance of the hydraulic servo system at the same time,and to solve the problems of unknown compound disturbance,input saturation and unmeasured states of the system,this paper designs some novel controllers based on prescribed performance control algorithm are proposed.Both theoretical results and simulation experiments can prove that the designed controllers can guarantee the system with fast and no-overshoot precision control,which makes the research results of this paper have theoretical reference value and engineering practice significance for improving the MDF products quality.
Keywords/Search Tags:Medium density fiberboard, Hydraulic servo system, Prescribed performance control, Adaptive neural network, Input saturation, Reduce order state observer
PDF Full Text Request
Related items