Dissertation > Excellent graduate degree dissertation topics show

Research on the AMB Control Based on Neural Networks and Improved PSO Algorithms

Author: ShenJun
Tutor: SuYiZuo
School: Wuhan University of Technology
Course: Control Science and Engineering
Keywords: Magnetic bearings PID controller Neural Networks Particle swarm optimization Dynamic mutation
CLC: TP273
Type: Master's thesis
Year: 2011
Downloads: 75
Quote: 1
Read: Download Dissertation

Abstract


Frictionless magnetic bearings have the advantages of long life, no lubrication, rotary precision, is widely used in the aerospace, machining, power transmission, energy, transportation and other fields. A magnetic bearing control system involving electromagnetics, control the discipline of science, mechanical science and rotor dynamics of complex non-linear open-loop unstable system. The object of this paper, active magnetic bearings, using the online way to adjust the parameters of PID controller based on improved particle swarm optimization neural network to achieve closed-loop control of magnetic levitation rotor. This paper mainly do the work of the following aspects: the basic structure and working principle expounded magnetic bearing control system. By the the electromagnetic relationship in the analysis of magnetic bearing systems to create a mathematical model of the magnetic bearing control system. Introduce PID control algorithm and improvement measures are proposed for the lack of algorithms. PID control scheme and design of magnetic bearings magnetic bearing PID controller settings based on the relevant parameters of the simulation study, in-depth analysis of simulation results using BP neural network online tuning PID controller parameters KP, K1, KD's value . Proposed the magnetic bearings BP neural network PID controller design. According to the degree of complexity of the system to set the structure of BP neural network, select two input neurons, 15 hidden layer neurons, and three output neurons. The BP network two inputs are the displacement of the rate of change of the deviation and the deviation of the magnetic levitation of the rotor, the three outputs respectively correspond to the three parameters KP of the PID controller, K1, KD. First offline training of BP neural network based on a set of input and output sample data, and then the BP network to adjust the network weights coefficient makes P1D controller parameters optimal system performance online. Introduction of particle swarm algorithm for the deficiencies of BP algorithm. Through the analysis of the impact of various parameters on particle swarm optimization algorithm performance improved method, which highlighted the improved particle swarm algorithm based on dynamic mutation thinking. The algorithm to improve particle swarm algorithm inertia weight w, namely w linear attenuation second amendment inertia weight w, and then according to the specific circumstances of particle convergence. Using Improved Particle Swarm alternative the BP algorithm optimization neural network weight coefficient. On this basis, the the optimized magnetic bearing based on improved particle swarm optimization neural network PID control scheme. Simulation results show that the improved PSO algorithm to optimize the design of the magnetic bearing neural network PID control scheme not only has excellent dynamic performance and steady-state performance of the maglev rotor and control system with good anti-jamming capability.

Related Dissertations

  1. High Speed Frequency Measurment and Non-Linearity Correction of Frequency Modulated Capacitive Displacement Sensor,TH822
  2. Single Neuron PID Control for Electro-Hydraulic Servo Unit of Ship Rudder,U666.152
  3. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  4. Computing Minimum Distance between Curves/Surfaces Based on PSO Algorithm,O182
  5. Research on Anti-periodic Solutions of Delayed Cellular Neural Networks without Assuming Global Lipschitz Conditions,TP183
  6. Designs and Applications of Fuzzy Synthetic Evaluation Models Based on Parallel Algorithms,TP18
  7. Active Power Filter and Its Application in Distribution Network,TN713.8
  8. Research on Fuzzy C-Mean Clustering Algorithm Based on Particle Swarm Optimization and Shuffled Frog Leaping Algorithm,TP18
  9. Spectrum Variation of Vegetation in Yanzhou Coal Mine Area and Heavy Metal Stress Characteristic,X173
  10. Research on the Improvements and Applications of Particle Swarm Optimization,TP18
  11. Segmentation of cDNA Microarray Image Using Fuzzy C-means Algorithm Optimized by Particle Swarm,TP391.41
  12. Reactive Power Optimization Based on Modified PSO,TP301.6
  13. Stage Fault Diagnosis Based on Data Fusion,TP18
  14. Study on Electromagnetic Bandgap Structure with Bow-tie Units,TN454
  15. The Research and System Design of Assessment Model for Network Learning Based on Intelligent Computation,TP18
  16. Non-linear Time-Varying System Modeling Using Time-Varying High-Order Neural Networks,TP183
  17. PID Controller Design of Linear Systems with Time Delay Based on Parametric Space and GUI Simulation,TP273
  18. Predicting Short-term Foreign Exchange Based on Wavelet Neural Network,F224
  19. Research on Modification and Application of Particle Swarm Optimization Algorithm Based on Control Methods,TP301.6
  20. Based on RFID Prison Intelligent Management System Research and Implementation,TP315
  21. Research of Path Planning for Small-Size Intelligent Soccer Robot in Complex and Dynamic Environment,TP242

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