Dissertation > Excellent graduate degree dissertation topics show

Study of Reactor Temperature Control Based on Compensatory Fuzzy Neural Network

Author: ZhengXiaoZuo
Tutor: FengDongQing
School: Zhengzhou University
Course: Detection Technology and Automation
Keywords: Fuzzy Neural Network Compensation operator Clustering algorithm
CLC: TP183
Type: Master's thesis
Year: 2009
Downloads: 53
Quote: 1
Read: Download Dissertation

Abstract


Omethoate synthesis process reaction kettle temperature control of a direct impact on the quality of the product. Due Omethoate production of synthetic reaction process has varying delay and nonlinear characteristics by a methylamine flow, a-methylamine total feeding amount, reaction time, cooling brine temperature and other factors, it is difficult to establish a precise mathematical model, the conventional control method based on accurate mathematical model is difficult to achieve satisfactory control effect. In recent years, fuzzy neural network has both advantages of fuzzy logic and neural networks, have unique advantages, has been widely used in dealing with nonlinear problems. Conventional fuzzy neural network control, fuzzy operations generally use static, local optimization computing method, compensation fuzzy neural network control, fuzzy operation using a dynamic, global optimization arithmetic. Omethoate synthesis reaction is a dynamic process, taking into account the dynamic characteristics of omethoate synthesis reaction, with output the order delay feedback neural network Omethoate object recognition; compensation and fuzzy neurons introduced fuzzy neural network compensation fuzzy neural network controller design. Prior to the design of the controller in order to get the controller more reasonable initial parameters and fuzzy rules to avoid the subjectivity and blindness of man-made, analysis and comparison of several typical clustering algorithm using Fuzzy C-Means since adapt clustering algorithms to extract features and clustering point, the initial rule base and initial parameters of the system. Using data collected from the actual production process, compensation fuzzy neural network controller offline training, to determine the initial structure and parameters of the controller, and then trained compensation fuzzy neural network controller and omethoate synthesis reaction object model connected together, the controller simulation. Reasonable fuzzy rule-based compensation fuzzy neural network controller can adjust the input and output fuzzy membership functions, and can dynamically optimize fuzzy inference rules, the compensation fuzzy neural network operators established from experts or less reasonable fuzzy rules mutual compensation calculation, to overcome the disadvantage of artificially selected fuzzy rules with greater subjectivity. Compensation fuzzy neural network parameters preset by the heuristic algorithm, thereby speeding up the training speed. Simulation results show that: The controller has a good self-learning ability, can better meet the of Omethoate synthesis reaction temperature control requirements.

Related Dissertations

  1. The Application of Ant Colony Algorithm in Meteorological Satellite Cloud Pictures Segmentation,TP391.41
  2. Research on Clustering Algorithm Based on Mutation Particle Swarm Optimization,TP18
  3. Research on K-means Optimization Clustering Algorithm,TP311.13
  4. Research on Fuzzy C-Mean Clustering Algorithm Based on Particle Swarm Optimization and Shuffled Frog Leaping Algorithm,TP18
  5. Research on Clustering Algorithm Based on Genetic Algorithm and Rough Set Theory,TP18
  6. The Research on Routing Protocol of Agricultural Environmental Monitoring System Based on Wir Eless Sensor Networks,TN915.04
  7. SAR interferometric method for optimal selection,P225.2
  8. Segmentation of cDNA Microarray Image Using Fuzzy C-means Algorithm Optimized by Particle Swarm,TP391.41
  9. Theory and Practice of Ant Clustering and Partitioning-based DBSCAN Clustering,TP311.13
  10. Study on the Technique of Information Fusionapplied to Enbedded Driver Fatigue Detection,TP368.12
  11. Improved Mathematical Model of Spatial Information Processing Based on Fuzzy Technology,O159
  12. Study on Application of Solar Energy Street Lamp Intelligent Controller,TM923.5
  13. Research of Clustering Routing Protocol in Ad Hoc Network,TN929.5
  14. Research on Routing Algorithmin Sensor Networks Based on Cluster with Mobile Sink,TP212.9
  15. The Research and Design of Pressure Cooker System Based on Fuzzy Neural Network,TM925.5
  16. The EMG Pattern Recognition System Based on Generalized Dynamic Fuzzy Neural Network,TP183
  17. Modulation Classification Algorithms of Digital Communication Signals,TN914.3
  18. A sample based on clustering neural network self- learning system,TP391.6
  19. University Human Resources Information Management System Design and Implementation,TP311.52
  20. Ad Hoc network clustering routing protocol research and Optimization,TN929.5
  21. Based on Ant Colony Clustering Algorithm,TP311.13

CLC: > Industrial Technology > Automation technology,computer technology > Automated basic theory > Artificial intelligence theory > Artificial Neural Networks and Computing
© 2012 www.DissertationTopic.Net  Mobile