Dissertation > Excellent graduate degree dissertation topics show

Discussion and Research on Adaptive Neural Network Control of Discrete-Time Nonlinear System

Author: WenGuoXing
Tutor: LiuYanJun
School: Liaoning University of Technology
Course: Applied Mathematics
Keywords: Nonlinear systems Adaptive control Neural Networks Uncertainty
CLC: TP273.2
Type: Master's thesis
Year: 2011
Downloads: 67
Quote: 0
Read: Download Dissertation

Abstract


The neural network has been proved in a compact set can approach any nonlinear continuous function to the desired accuracy. Adaptive neural network control based on the nature of the neural network, many scholars have actively explored. Currently, most of the findings focused on nonlinear continuous systems. Nonlinear discrete system than the the continuous system more accurately describe the actual controlled object, study the discrete system control method has important theoretical and practical significance. The main shortcomings of the existing neural network adaptive control method for nonlinear discrete system is much adaptive parameters need to be adjusted online, especially nonlinear MIMO discrete high-end systems, and inevitably will spend a lot of learning time In practical engineering unacceptable. Taking into account this shortcoming, based on Lyapunov stability theory, the paper studies the content of the following aspects: First of all, for the two types of single-input single-output nonlinear strict feedback discrete system, an adaptive neural network state feedback control . Transformed through the system, to avoid the adaptive control may occur in the controller design process controller singular problems. By selecting the appropriate design parameters, the closed-loop system is proved to be semi-globally uniformly ultimately sector (SGUUB). This method requires only a small number of adjustable parameters, the online online computation burden can be reduced. A simulation example demonstrates the validity of this algorithm. Secondly, for a class of second-order nonlinear discrete systems, a neural network adaptive output feedback control. Backstapping technology, design output feedback controller. The neural network is used for approaching the desired controller, to overcome the non-causal problem encountered in the control process. The adaptive output feedback controller only need to adjust to less adaptive parameters, it is clear that the method can reduce online computation. Simulation confirmed the feasibility of this theory. Finally, for a class of multi-input, multiple-output (MIMO) nonlinear discrete system, a neural network adaptive output feedback tracking control. Diffeomorphism mapping, the MIMO system is converted to the form of input and output. Backstapping technology to achieve the effective output feedback control. Based on Lyapunov analysis method, this algorithm is proved to be stable, and to ensure that all signals in the closed-loop system is semi-globally uniformly ultimately sector (SGUUB). Compared to the existing control program, the advantage of this approach is only need a small amount of design parameters, greatly easing the computational burden. Simulation examples illustrate the feasibility of this scheme.

Related Dissertations

  1. High Speed Frequency Measurment and Non-Linearity Correction of Frequency Modulated Capacitive Displacement Sensor,TH822
  2. Design and Application of Scope Anti-Shake System,TH743
  3. Design and Research on Stabilization Loop of Gyro Stabilized Pod Control System,V241.5
  4. Robust Adaptive Control Algorithms Research for the System of Spacecraft Attitude Dynamics,V448.22
  5. Stability Analysis and Controller Design for Discrete-Time Switched System,TP13
  6. Optimal Synchronization Control of a Class of Underactuated Lagrangian Systems,TP13
  7. Adaptive Control of Post-Boost Vehicle Using Multiple Models,TP273.2
  8. Stability Analysis of Systems with Time Delays,TP13
  9. Study on Adaptive Fuzzy Control Algorithm and Its Implementation,TP273.4
  10. Research on Anti-periodic Solutions of Delayed Cellular Neural Networks without Assuming Global Lipschitz Conditions,TP183
  11. Designs and Applications of Fuzzy Synthetic Evaluation Models Based on Parallel Algorithms,TP18
  12. Health-based and Ecological Risk Assessment of Contaminated Sites,X820.4
  13. Research on Adaptive Control of Recovery of UUV with Near Wall Constraint,TP273
  14. The Evaluation and Forecasting Research on Uncertainty Systems Method of Student Achievement,G642.4
  15. Effects of the Therapeutic Communication System Intervention on the Illness Uncertainty of Preoperative Colorectal Cancer Patients,R473.73
  16. Spectrum Variation of Vegetation in Yanzhou Coal Mine Area and Heavy Metal Stress Characteristic,X173
  17. Uncertainty and metafiction : \,I712.074
  18. Information Aversion,G201
  19. Algorithm Research for Robot Trajectory Tracking Control,TP242
  20. The Research and Application of Modified Algorithms About Fuzzy Predictive Functional Control,TP273
  21. Robust Passive Control for Uncertain Singular Systems,TP13

CLC: > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Automation systems > Automatic control,automatic control system > Adaptive ( self- tuning ) control,adaptive control ( self-tuning ) system
© 2012 www.DissertationTopic.Net  Mobile