Research And Implement Of Fuzzy-Smith Predictive Optimal Control To Time-delay Process

Temperature process exists widely in various industrial process controls in metallurgy,medicine and defense in.Time-delay process control is found to be among the industrial control problems.Conventional control methods using classical PID control are sometimes difficult to achieve the desired control effect.So far,the control field recognized J M Smith in 1957 estimated the proposed compensation scheme is the best way of solve the time-delay process control problems.But the traditional Smith predictor control is based on an accurate model of the controlled object design,with the characteristics of robustness,so the absence of an accurate model or parameter variations and has a Time-delay of controlled object,is difficult to achieve satisfactory control effect,greatly limits its wide application in industrial process control.So many scholars work on improving the robustness of Smith predictive control.However,most studies scholars confined to enhance the robustness,but rarely Optimal Control Smith Predictor(the Smith predictive optimal control),which Smith predictor for optimal control in actual industrial process applications the study,regardless of its theoretical value or practical significance,are worthy of further study.Issues in order to achieve the Time-delay process in the actual process of Smith estimated optimal control engineering applications targeted research on controlled object online identification system parameters Fuzzy-Smith estimated optimal control methods.First,the main features of the control method to analyze today's Time-delay process,explain their advantages and disadvantages of each method,and presented the results of this study are based on the analysis and description of the overall scheme of the method compared with other control methods advantages.After then proposed to delay process for the three system parameters identification method using system simulation data to establish the system parameter identification database,obtain the system output error and system failure response to quantify the extent of the relationship between the distribution and use this as the fuzzy controller design knowledge base,using fuzzy controller for the controlled object online identification system parameters,get the actual industrial control system accurate real-time time-varying parameters.Then the system simulation using MATLAB Fuzzy-Smith estimated the optimal control system performance and other control methods were compared,the subject of the proposed method to prove the correctness and effectiveness.Further directed to the actual process control due to disturbance and noise output will lead to a serious distortion of the system can not meet the accuracy requirements of the phenomenon of parameter identification is proposed signal processing method for digital filtering system output signal in the actual system output is not guaranteed minimize distortion of the case in order to ensure the accuracy of identification to eliminate noise.At the same time break the traditional digital filter design approach is proposed filter design based on optimal control theory methods,including most in the case study given passband,stopband design specifications of the filter order and the width of the transition zone excellent relationships,and analysis filter pole position distribution on the filter performance and so on.Due to research this topic in its position in the actual industrial control applications,and finally to the fiber AHRS device temperature control system hardware platform,first introduced Fuzzy-Smith estimated the optimal control in the software design of the main implementation method,then will be applied to the fiber AHRS temperature process control equipment,the subject of the proposed Fuzzy-Smith estimated the effectiveness of the optimal control method in practical engineering applications through the assessment of its performance in the actual temperature process control verification.