Dissertation > Excellent graduate degree dissertation topics show

The Research of Optimize Numerical Blending Based on Artificial Neural Networks

Author: WangHui
Tutor: LiuMeiHong
School: Kunming University of Science and Technology
Course: Mechanical Manufacturing and Automation
Keywords: Computational Intelligence All recipes Computer Aided Design Recipe leaf group Genetic Algorithms
CLC: TS41
Type: Master's thesis
Year: 2010
Downloads: 38
Quote: 0
Read: Download Dissertation

Abstract


Computational intelligence as a means of artificial intelligence research, has been in the areas of health workers and peasants has been widely used, and to provide a new formulation of cigarette design ideas. The proportion of tobacco leaf group formulation and sensory characteristics between showing a complex and difficult to determine the nonlinear relationship. Based on computational intelligence principle, designed on the basis of previous studies, we propose using neural network technology presents a digital cigarette formula design ideas. Firstly, from the recipe, silk craft, cigarette parameters introduced three fully formulated tobacco cigarette design. Tobacco for the current sample of the unknown is no good scientific classification method is proposed clustering method using two classical idea of ??combining both take into account the experience of experts and combined with neural networks and fuzzy mathematics method, using a sample of expertise to support vector machine, \material, the replacement of the tobacco material between the proposed new solution. Tobacco is currently a hot research designers is to find a more accurate expression of physical and chemical indicators leaf group formulation and sensory quality of comprehensive evaluation index, cigarette use this indicator to predict whether a product has been reached to requirements. In this paper, on the basis of experimental data to the physical and chemical properties of raw tobacco as an input to the sensory characteristics of tobacco as an output, the establishment of physicochemical properties and sensory characteristics reflect mapping relationship between BP neural network model and training, thereby establishing neural networks mapping model. Excluding France and paper establishes quadratic polynomial regression equation comparison results show that BP neural network is more accurate in fitting the error smaller parallel processing ability. Can effectively solve the actual production of cigarettes sensory indicators unpredictable problems. Meanwhile this paper, neural network model and Evolutionary Computation Methods leaf group program to optimize the combination formula is proposed to evaluate the absorption, combined with chemical analysis and computer-aided design of integrated solutions leaf group formula, first select the tobacco material according to the design requirements; utilization uniform design to the selected ratio, and staff through the smoking formula feeling for quality assessment, and then use BP neural network modeling; BP neural network algorithm for the S-function value in the limited predictive ability of generalization is difficult to find extreme points propose solutions. That are based on expert experience appropriate to reduce the amount of raw materials in low-grade tobacco to narrow interval; finally completed through genetic algorithm optimization formulation design, based on expert assessment after absorbing the improvement amendments designed to meet the requirements and take into account the economic optimum formulations to expect good market efficiency. In this paper, leaf group formulation design simulation, and compare several formulations sensory difference between the methods, the results prove the correctness of the design ideas in the experiments and actual production has a guiding role.

Related Dissertations

  1. Development of the on-line Training and Examination System of Army,TP311.52
  2. Designs and Applications of Fuzzy Synthetic Evaluation Models Based on Parallel Algorithms,TP18
  3. Application of Numerical Designing Cigarette Blending Formula on Computational Intelligence,TS44
  4. Based on Genetic Algorithm Pishihang irrigation canal water allocation marshalling model of,S274
  5. Genetic Algorithm in logistics and warehousing Optimization Research,F259.2
  6. Mining resources based on genetic algorithm optimization model of,O224
  7. Research and Development of Computer Aided Design System ramp,TP391.72
  8. The Research and Application of Modified Algorithms About Fuzzy Predictive Functional Control,TP273
  9. Optimal Control of Emulsion System in Cold Rolling,TP273
  10. Research on the Marshalling-scheduling Model and Algorithms of Freight Trains Based on Game Theory,O225
  11. Multi-directional Mutation Genetic Algorithm and Research on Neural Network Optimization,TP18
  12. The Application of Using Genetic Algorithms on Universities Course-arranging System,TP18
  13. Research on Mobile Robot Path Planning and Simulation Realization,TP242
  14. Research on Routing Algorithmin Sensor Networks Based on Cluster with Mobile Sink,TP212.9
  15. Research and Implement of the Theme Crawler for Automotive Industry,TP391.3
  16. Research on Co-Design and Management System Based on AutoCAD,TP311.52
  17. The Research and Implement on Camera Calibration Technology Based on Trifocal Tensor,TP391.41
  18. Decision Support System of Vehicle Scheduling in Double Level Garage,TP242
  19. The Studies on Some Improvements of the GA and Their Applications in SVM,TP18
  20. A Reduction Method for Artificial Neural Network Inputs Based on An Improved Genetic Algorithm,TP18
  21. The Key Technology Research of Finite Element Analysis Pre-processing System Based on AutoCAD,O241.82

CLC: > Industrial Technology > Light industry,handicrafts > Tobacco Industry > Basic science
© 2012 www.DissertationTopic.Net  Mobile