Dissertation > Excellent graduate degree dissertation topics show

Based on near -infrared spectroscopy method for rapid identification of swill

Author: LiangJing
Tutor: ZhuShiPing
School: Southwestern University
Course: Agricultural Electrification and Automation
Keywords: Swill Near Infrared Spectroscopy DPLS Support Vector Machine BP neural network
CLC: TS227
Type: Master's thesis
Year: 2011
Downloads: 230
Quote: 2
Read: Download Dissertation

Abstract


Swill collected from the food and beverage industry Shaoshui After dehydration, slag, bleaching and deodorization process extracted oil. Swill oil used to be processed into industrial grease or organic fuel, but never used as a cooking oil, in recent years, edible oil adulteration repeated also makes security problems increasingly become the focus of public attention. Swill of current market there is an urgent demand for rapid identification, in this context, this paper based on near-infrared spectroscopy method for rapid identification of swill. The main contents are as follows: 1, a collection of 82 oil samples using conventional methods identify swill oil samples with 37 copies, 45 copies of qualified edible oil samples. BRUKER the German company MATRIX-F FT-NIR spectrometer for 82 samples were full spectrum segment spectral acquisition. 2, the swill studied based on DPLS Near Infrared Spectroscopy method, and through data normalization preprocessing method optimization DPLS model. This method can accurately determine whether the sample is swill oil. Three studied SVM based swill Near Infrared Spectroscopy method, using S-kernel function SVM support vector machine to establish identification model and exhaustive method for parameter selection in the SVM defects on the use of genetic algorithm support vector machine parameters are selected to obtain better parameters, so that the model can quickly and accurately determine the sample category. 4, studied based on PCA-BP neural network swill Near Infrared Spectroscopy method. As BP network input for the spectral problem of too many data points, using principal component analysis to extract the main component of spectral data, thereby reducing the number of input neurons; hidden layer nodes for the determination of questions were used six kinds of BP neural network training algorithm built network is trained to identify and compare the final results to determine the final training algorithm to obtain the optimal neural network structure, the establishment of BP neural network model to identify swill oil. 5, compared DPLS, SVM and PCA-BP swill established three methods to identify model performance. In the selection of the same spectrum, instruments, the same calibration set (training set) and test sets conditions BP-ANN method is superior to the established model and SVM methods DPLS model established, the correct identification rate of 94.8%. Research shows that based on near-infrared spectroscopy swill rapid identification mechanism is feasible, the method is effective.

Related Dissertations

  1. Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
  2. Research on Autamatic Music Structrue Analysis,TN912.3
  3. Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
  4. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  5. Research on Visual Servo System of Mechanical ARM,TP242.6
  6. Fault Diagnosis Method Based on Support Vector Machine,TP18
  7. Process Support Vector Machine and Its Application to Satellite Thermal Equilibrium Temperature Prediction,TP183
  8. Municipal tourism land use planning environmental impact assessment,X820.3
  9. Study on Taste Characteristic of Taste Peptide Enzymatic Production from Oyster Base on A Neural Network Method,TS254.4
  10. Monitoring Leaf Nitrogen Status in Rice with Near Infrared Spectroscopy,S511
  11. The Research on Evaluation of Living Status Systems of Expressway Relocated People,D523
  12. Research for Infrared Image Target Identification and Tracking Technology,TP391.41
  13. Study on the Road Condition Monitoring Based on Vehicular 3D Acceleration Sensor,TP274
  14. Research of Diagnosing Cucumber Diseases Based on Hyperspectral Imaging,S436.421
  15. Mine Risk Information Integration and Intelligent Early Warning,X936
  16. Research of Orange Quality Classification Technology Based on Computer Vision,TP391.41
  17. The Research on Intrusion Detection System Based on Machine Learning,TP393.08
  18. Optimization Study on Gating System and Molding Process Parameters of Injection Mold Based on Simulation,TQ320.662
  19. Research on Improved K Neighbor Support Vector Machine Algorithm Faced Text Classification,TP391.1
  20. Study on Luohe Technical Supervision Bureau of Food Safety Early Warrning System Based on Neural Network,F203
  21. Research of Adaptive Active Noise Control Based on Neural Network,TP183

CLC: > Industrial Technology > Light industry,handicrafts > Food Industry > Edible oils and fats processing industry > Product standards and testing
© 2012 www.DissertationTopic.Net  Mobile