Dissertation > Excellent graduate degree dissertation topics show

Research on the Soft Sensing of Aviation Kerosene Parameters Based on KPCA and Mixed Kernel LS-SVM

Author: DuQun
Tutor: LiZuo
School: Dalian University of Technology
Course: Control Theory and Control Engineering
Keywords: Atmospheric tower Soft Measurement Mixed kernel least squares support vector machine The nuclear principal component analysis Multigroup Genetic Algorithm
CLC: TP274
Type: Master's thesis
Year: 2011
Downloads: 29
Quote: 0
Read: Download Dissertation

Abstract


Online real-time in the refining enterprises distillation device , due to the constraints of the existing detection technology and process conditions , quality indicators parameter atmospheric tower aviation kerosene dry point and flash point , it is difficult to line detection , how to achieve the quality of aviation kerosene parameters measurement is one of the hot and difficult for the field of process control . Online testing for aviation kerosene quality indicators are difficult problems to the distillation unit of a petrochemical company as the research object , master crude oil distillation principle based on the initially selected 23 process measurable variables as required for modeling auxiliary variables . Secondly , the process required to read through the company's DCS system can be measured variable , dry point and flash point of data collection through laboratory tests collected , these data constitute the the modeling required sample set . Consider the impact of noise on the collected data , and proposed using the Savitzky-Golay filtering method for smoothing and filtering of the original data and the algorithm in the removal of most of the noise at the same time , than the conventional smoothing filter better preservation of the real information of the original data . Correlation may exist between the collected raw data vector contains too much redundant information , the proposed use of Kernel Principal Component Analysis (Kernel Principal Component Analysis, KPCA) the input data feature extraction and dimensionality reduction processing , to eliminate the collinear sex . Finally, a modeling method based on genetic algorithm optimization MKLS-SVM , sample data with the acquisition of the industrial process , the establishment of aviation kerosene dry point and flash point multiple input multiple output soft measurement model . Simulation results show that the Savitzky-Golay filter based on the data , using the the KPCA-MKLS-SVM establish the soft measurement model has good learning and generalization ability to meet the requirements of aviation kerosene dry point and flash point soft sensor modeling .

Related Dissertations

  1. A Preliminary Research on Soft-sensing Technology in Fatigue Test in the Transmission Test System of the Cooling Fan, Hydraulic Pump and Hydraulic Motor,TH137.5
  2. Soft-sensor Method of Circulating Ash Utilization in CFB-FGD Process Based on RBF Neural Network,X701.3
  3. Research on Soft-sensor Method of Component Content in Hydrometallurgy Copper Extraction,TF811
  4. The Electrolyte Components’ Prediction Based on Support Vector Machine,TQ151
  5. Approach Research of Sensor Fault Detection and Diagnosis in HVAC System,TP212
  6. Research on Soft Measurement for Arc Current of Electric Arc Furnace,TM924.4
  7. Sensor Fault Diagnosis Method Based on KPCA in Constant Air-volume Air-conditioning Unit,TB657.2
  8. Research of the System to Soft Measurement of High-temperature Field Based on CCD Image Sensor,TP274
  9. Study on the Soft Measurement for Moisture Content of Acrylonitrile Polymer,TP273
  10. Research and Application in Particle Size Soft-Sensor of Radial Basis Function Neural Network,TP183
  11. Research of a Soft Sensor for Cobalt Ion Concentration in the Inlet of Zinc Smelt Purification Process,TP274
  12. On-Line Measurement of Ceramic Paste Inner Stress Based on the Soft-Measurement Technology,TQ174.6
  13. Research on Soft Sensor Tenchnique for Rare Earth Extraction,TP274
  14. The Detection of Extraction Rate in Ultrasonic Extraction Process of Chinese Traditional Medicine,TP274.53
  15. Research on Multiple Support Vector Machines Soft Sensor Modeling Based on Particle Bee Colony Optimization Algorithm,TP274
  16. The Soft Sensor Based on Predictive Control Algorithm,TP274
  17. Research on the Soft-Sensing of Vinyl Acetate Polymerization Rate Based on Neural Networks Optimized by Genetic Algorithm,TP18
  18. Intelligent Modeling and Optimization for Polymerization System of High Density Polyethylene Unit,TQ325.12
  19. Prediction model based on the plant thermal parameters of the real - time learning strategies and applied research,TM621.4
  20. Research on Soft-sensing Method of Ball Mill Load,TD453

CLC: > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Automation systems > Data processing, data processing system
© 2012 www.DissertationTopic.Net  Mobile