首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0825111-162509


URN etd-0825111-162509 Statistics This thesis had been viewed 1526 times. Download 0 times.
Author Tai-yuan Chang
Author's Email Address No Public.
Department Electrical Engineering
Year 2010 Semester 2
Degree Master Type of Document Master's Thesis
Language English Page Count 46
Title PARAMETER ESTIMATION OF PID CONTROLLER BASED ON ADAPTIVE SELF-CONSTRUCTING TYPE Ⅱ FUZZY NEURAL NETWORK
Keyword
  • adaptive law
  • robust controller
  • fuzzy neural network estimator
  • PID controller
  • self-constructing
  • self-constructing
  • PID controller
  • fuzzy neural network estimator
  • robust controller
  • adaptive law
  • Abstract The parameters of traditional Proportional-Integral-Derivative (PID) controller are fixed and cannot be trained. If process system has unknown terms such as the modeling error and external distance, it has unstable tracking and the error of system will be greater and greater. In order to overcome this problem and achieve accurate tracking performance, the parameters of PID controller are on-line estimated using the Self-Constructing Type 2 Neural Network (SCT2FNN) in this thesis.The Self-Constructing Type 2 Fuzzy Neural Network Base PID (SCT2FNN-Base PID) controller has combined the PID controller, SCT2FNN estimator and robust controller. The main controller is PID controller. The inputs of controller are composed of the error, integral of the error, and derivative of the error to control uncertain non-linear system. The structure and parameter learning of SCT2FNN estimator are done automatic and online. The Mahalanobis distance (M-distance) method in the structure learning is employed to determine if the fuzzy rules are generated/ eliminated or not. Furthermore, the parameter learning is applied to adjust the parameters of the SCT2FNN estimator via adaptive laws which is proven to be stable by Lyapunov theorem. In order to compensate the uncertainties of the system parameters and achieve robust stability of the considered system, the robust controller is adopted. Finally, the simulation results are compared with other controllers to demonstrate the performance of the proposed controller.
    Advisor Committee
  • Hung-ching Lu - advisor
  • none - co-chair
  • none - co-chair
  • Files indicate not accessible
    Date of Defense 2011-05-31 Date of Submission 2011-08-26


    Browse | Search All Available ETDs