Dissertation > Excellent graduate degree dissertation topics show

Study on the Vibration Fault Diagnosis of Centrifugal Fan

Author: WangSuoBin
Tutor: ZhouYunLong
School: Tohoku Electric Power University
Course: Thermal Power Engineering
Keywords: k-means clustering algorithm Centrifugal fan Vibration Fault Empirical Mode Decomposition Blind Source Separation
CLC: TH442
Type: Master's thesis
Year: 2011
Downloads: 104
Quote: 0
Read: Download Dissertation

Abstract


Modern large generating units single unit capacity is growing , and the safe operation of the unit more and more attention by the industry , vigorously develop the vibration of rotating machinery fault diagnosis technology is particularly important fault feature extraction and classification vibration fault diagnosis technology the key problem . Describe the system chaotic features the largest Lyapunov exponent , Hurst index , and the description of the signal complexity of approximate entropy to describe the failure characteristics of the vibration signal . The same time , some of the theories for brain waves and ECG denoising is used to analyze the signal preprocessing , combined with the improved k-means algorithm , we carried out a diagnostic test centrifugal fan vibration fault signal . Using autoregressive method of combining and EMD centrifugal fan coupling misalignment fault analysis, has been the size of the fan load , the relationship with the fault frequency diagram, as well as misalignment of the respectively follow the trend of machine load , fan speed between relationship. Based the the the Gaussian Moment of FASTICA algorithm blind signal separation and independent component contained in the observed signals , the frequency of the vibration source corresponding independent component analysis to determine the fault component ; independent component other than the fault component as interference noise is set to zero, and then reconstruct the respective components to obtain a new signal matrix , again using the inverse matrix of the separating matrix and the signal matrix obtained by multiplying the vibration acceleration signals after denoising according to the the class distance criterion , its denoising effect with several other the kinds of methods were compared . Introduced the original idea of the k-means algorithm , its easy to fall into the local optimum drawback cited the results of four cluster centers in two separate space to the cluster center mobile guidelines be optimized combining the fan vibration time domain signal extracted three characteristic parameters , the fault classification. Use of the the EMD theory of pretreatment of raw vibration signal , then the reconstructed signal Empirical Mode Decomposition , extracted under different fault IMF energy characteristics , and energy characteristics of the normalization process , composed of eigenvectors , enter to the inside of the k-means algorithm improved classification.

Related Dissertations

  1. The Recognization of License Plate Based on EMD and Its Application,TP391.41
  2. Research on Clustering Algorithm Based on Mutation Particle Swarm Optimization,TP18
  3. Research on Fuzzy C-Mean Clustering Algorithm Based on Particle Swarm Optimization and Shuffled Frog Leaping Algorithm,TP18
  4. Research on Clustering Algorithm Based on Genetic Algorithm and Rough Set Theory,TP18
  5. Segmentation of cDNA Microarray Image Using Fuzzy C-means Algorithm Optimized by Particle Swarm,TP391.41
  6. Development and Experimental Study about Virtual Instruments of Vibration Test and Analysis System and Combustion Analyzer,TK407
  7. High efficiency and low noise centrifugal fan Numerical Simulation and Optimization,TH432
  8. Weak sparse underdetermined blind signal separation technology research,TN911.7
  9. Sub -band Blind Separation of Speech,TN912.3
  10. EMD-based medical image fusion algorithm,TP391.41
  11. Radio Monitoring Signal Separation Study Based on Cyclic Statistics,TN911.7
  12. Research on Diagnosis of Sudden Cardiac Death Based on ECG Signals and TWA,TN911.7
  13. Design and implementation of energy management systems based on EMD-BP neural network prediction model,TP315
  14. Research of Image Edge Detection Based on Wavelet Transform and EMD,TP391.41
  15. Study on Independent Component Analysis-Based Seismic Blind Deconvolution Method and Its Application,P631.4
  16. Blind Source Separation of Vibration Signals of AC Motor Speed Control System,TM921.51
  17. Research on the Vibration Performance of Test-bed and Validation of Impeller Strength for Small Centrifugal Fans,TK83
  18. Product Platform Construction for the Development of Equipments,TJ08
  19. The Simulation of Low Concentration of Gas-Solid in the Centrifugal Fan,TH442
  20. Study on Repeated Blind Separation of Independent Component Analysis and Its Application in Mechanical Fault Diagnosis,TH165.3
  21. Parametric Modeling and Design Analysis of Centrifugal Fan,TH442

CLC: > Industrial Technology > Machinery and Instrument Industry > Gas compression and transportation machinery > Blower > Centrifugal
© 2012 www.DissertationTopic.Net  Mobile