Dissertation > Excellent graduate degree dissertation topics show

The Research on Multi-model Predictive Active Fault-tolerant Control Method Based on SVM

Author: WangFengDa
Tutor: LiZuo
School: Lanzhou University of Technology
Course: Control Theory and Control Engineering
Keywords: Nonlinear Systems Active Fault Tolerant Control Multi-model SVM Predictive Control Data-driven Robustness
CLC: TP273
Type: Master's thesis
Year: 2009
Downloads: 164
Quote: 0
Read: Download Dissertation

Abstract


In this paper, nonlinear system as the research object, in the early initiative proposed fault-tolerant control based on multiple models based on the idea, that may occur for a variety of systems known or unknown failure scenario is proposed based on LS-SVM multi-model prediction Active Fault Tolerant control programs around the proposed scheme, mainly for the following aspects of work: 1) failure for having prior knowledge of the nonlinear system is proposed based on LS-SVM multi-model prediction active fault tolerant control methods. Taking into account the multi-model based on BP neural network prediction of active fault tolerant control method, there is a sample required for modeling multi-network structure is difficult to determine, slow convergence and easy to fall into local minima and other shortcomings, the use of LS-SVM modeling of the small sample size required , the global optimal solution fast and advantages of BP neural network modeling to make up the deficiency; while taking into account the computational complexity of nonlinear predictive control, turn on SVM kernel function linear representation of the complex non-linear multi-step prediction equation is transformed into a series of simple and intuitive form of linear multi-step prediction, thereby simplifying the predictive control law to strike the complexity and nonlinear system simulation example shows that the method is feasible and effective. 2) may occur for unknown nonlinear systems failure, the dynamic data-driven technology into the LS-SVM multi-model prediction active fault tolerant control for a transitional strategy to achieve fault-tolerant system is unknown active fault tolerant. When a system failure occurs is unknown, the dynamic model library without a matching model, you need online modeling, text with dynamic data-driven techniques complement the model, improve function, dynamic data presented five steps supplementary loop algorithm, in the rapid establishment of unknown to the system failure mode LS-SVM model, while the model used Closeness control strategy to achieve a safe transition fault modeling process, and thus to create a new model predictive control law is calculated to achieve the system failure of the unknown active fault tolerant. 3) on the paper gives a multi-model based on LS-SVM active fault tolerant control system performance prediction for a simple analysis. In theory, the local linear predictive control algorithm convergence is proved. From the experimental point of view, there is a certain range of the system of outside interference or other uncertain parameter perturbation carried out simulation studies, the results show that the fault-tolerant control system has good robustness, that is, except predictive controller, the right not cause system performance does not meet the actual requirements of small faults, monitoring mechanism will not fail misjudgment and error control law group.

Related Dissertations

  1. Soft Sensor of Naphtha Dry Point on Support Vector Machines Regression,TE622.1
  2. Satellite Attitude Control Methods Using Only Magnetorquers,V448.222
  3. The Research of the Fault Diagnoses Algorithm for the Liquid Rocket Engine Testing Bed Based on PCA-SVM,V433.9
  4. ISAR Imaging Simulation of Space Targets and Target Recognition Based on ISAR Images,TN957.52
  5. Research on Autamatic Music Structrue Analysis,TN912.3
  6. Research on Feature Extraction and Classification of Pulse Waveform for Cholecystitis and Nephrotic Syndrome Diagnosis,TP391.41
  7. Research on Classification Method of Tongue Substance Color and Tongue Coating Color Based on SVM,TP391.41
  8. The Research on Paper Currency Classification Method Based on Harr-Like Feature and Minimal Ball Including Samples,TP391.41
  9. Research of Predictive Control in Water Treatment Plant Based on Neuron Network,TP273.1
  10. Robust Model Predictive Control of Networked Systems,TP273
  11. Research on Predictive Control and Simulation of pH Value Based on the Confection Process of Sodium Nitrate,TP273
  12. Research on Focused Crawler Based on SVM Classification Algorithm,TP391.3
  13. Research on Predicting Intrinsic Disorder Protein Structure Based on Supervision Manifold Learning Algorithm,Q51
  14. Study on the Road Condition Monitoring Based on Vehicular 3D Acceleration Sensor,TP274
  15. Identification and Robustness Analysis of Nonlinear Hybrid Dynamical System Concerning Glycerol Transport Mechanism,TQ223.162
  16. Research of Orange Quality Classification Technology Based on Computer Vision,TP391.41
  17. Analysis on Synoptic Climatology Characteristics and Forecast Methods of Fog in Hainan,P457
  18. Research of License Plate Recognition Based on Rough Sets and Fuzzy SVM,TP391.41
  19. The Research and Application of Modified Algorithms About Fuzzy Predictive Functional Control,TP273
  20. Study on Visual Target Detection Based on SVM,TP391.41
  21. Research into Visual Orientation in the Weeding Robot Based on COCM,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Automation systems > Automatic control,automatic control system
© 2012 www.DissertationTopic.Net  Mobile