Dissertation > Excellent graduate degree dissertation topics show

Pattern Recognition-Artificial Neural Network Method Study and Application to Material Field

Author: Shen
Tutor: ZuoLieXiang
School: Wuhan University of Technology
Course: Materials Science
Keywords: Materials Design Formulation Optimization Pattern Recognition Artificial Neural Networks Up competition algorithm
CLC: TB30
Type: Master's thesis
Year: 2005
Downloads: 299
Quote: 1
Read: Download Dissertation

Abstract


This paper is a material to optimize the design of the new method - pattern recognition - artificial neural network for research, and to solve new material development process frequently encountered in the test cycle is long, complex factors, and the reaction mechanism is unknown problems. The use of this new method of pattern recognition method of qualitative analysis of sample data and sample screening, the use of artificial neural network model for quantitative analysis and accurate forecasting, and intuitive two-dimensional mapping diagram shows the distribution pattern of the target value. The basic principle is: based on the same principle of topology mapping process, the multi-dimensional vector first mapped to the two-dimensional plane, and then extended to the high-dimensional space through a non-linear function, superposition and output, this can be avoided in the transfer process The relationship between the variables in shielding or covering. To show the data implied by the law, by mapping the plane image from a flat image to determine the most visually advantages or regional material to optimize the design and experimental design provides an effective decision support methods. His research work in this paper is as follows: build artificial neural network to achieve nonlinear mapping model of pattern recognition, artificial neural network learning and training is easy and the introduction of up competition algorithm with the gradient method combined with a hybrid algorithm to solve into local minima problems, a good balance between the requirements of global and local search. Up competition algorithm is a global search, while the role of the gradient method to achieve local optimization, the combination of these two methods, you can achieve a balanced and unified global search and local search. The actual calculation results show that the hybrid algorithm is obvious than the single algorithm, the mapping error is small and fast rate of learning training, mapping model can truly reflect the objective laws. Study with computer pattern recognition - artificial neural network-assisted materials design and test optimization, PR-ANN screening sample data, screening variables, qualitative analysis of pattern recognition, artificial neural network accurately forecast, and the development of the system of PR-ANN optimization software. Tested PR-ANN system on a single target, multi-objective optimization problem to obtain satisfactory results. The contour distribution pattern in the human visual judgment two-dimensional mapping plane so complex multi-objective optimization problem becomes simple. PR-ANN system for material formulation design, performance optimization case study, respectively on a single target, two goals and three goals instance, and forecast better results than other optimization methods.

Related Dissertations

  1. The Classification of High Dimsnsion Flew Field Based on Manifold Learning,V231.3
  2. Research on Text Classification Based on Biomimetic Pattern Recongnition,TP391.1
  3. Research on Identification System of Cashmere and Wool Fiber,TS101.921
  4. Intrusion detection based on the ultrasonic echo envelope in the military security patrols,E919
  5. Research on Key Technologies of Turning Expert System Based on SE300,TP182
  6. Improved Mathematical Model of Spatial Information Processing Based on Fuzzy Technology,O159
  7. The EMG Pattern Recognition System Based on Generalized Dynamic Fuzzy Neural Network,TP183
  8. Design and Implementation of System of Tables with Handwritten Content,TP391.41
  9. A Study on the Value Indictors System of Cultural Heritages in Henan,G122
  10. CUDA-based video fire detection system,TP391.41
  11. Based on DCE-MRI of computer-aided diagnosis of benign and malignant breast lesions studied,TP391.7
  12. The Research of Simulation and Operating Optimization on Oilfield Water Injection System,TE357.6
  13. Solving Multi-component Mass Exchange Network Using Line-up Competition Algorithm,TH186
  14. Lipovitellin in Rosy Barb (Puntius Conchonius) is a Pattern Recognition Receptor with an Opsonic Activity,S917.4
  15. QSAR Studies on Medicinal Activity Based on Quantum Chemical Calculations,TQ460.1
  16. The Research of Simulation on Solar-Earth Source Heat Pump System,TK519
  17. Molecular Cloning, Recombinant Expression and Functional Analysis of a Novel Scavenger Receptor from Zhikong Scallop,S917.4
  18. Quantitative and Qualitative Analysis of Drugs by Near Infrared Spectroscopy Technology,O657.33
  19. Differential Kinetic Spectrophotometric (Or Spectrofluorometric) Determination of Some Substances in Complex Systems by Chemometrics,O657.32
  20. Research on Braking Distance of Vehicles Prediction Based on Artificial Neural Networks,U461
  21. Chemometrics in some medicine or food fingerprint application,R284

CLC: > Industrial Technology > General industrial technology > Materials science and engineering > General issues of engineering materials
© 2012 www.DissertationTopic.Net  Mobile