Dissertation > Excellent graduate degree dissertation topics show

Application Research on Fault Diagnosis of Power Electronic Circuits Based on BP Neutral Network Optimized by QGA

Author: FengAnHui
Tutor: SuHongSheng
School: Lanzhou Jiaotong University
Course: Power Electronics and Power Drives
Keywords: Power electronic circuit fault diagnosis Spectrum Analysis Technology BP neural network Quantum Genetic Algorithm
CLC: TN710
Type: Master's thesis
Year: 2011
Downloads: 85
Quote: 0
Read: Download Dissertation

Abstract


Power electronics technology range of applications in the industrial production of a wide range of areas such as power, transportation, machinery, chemicals, mining, metallurgy, textile, aerospace, laser, communications, robotics, and energy efficient use. Power electronic devices generally acts as a power supply or controller, once the fault and can not receive timely diagnosis, it will damage the equipment, the destruction of continuous production, and even lead to system failures, casualties, and these will cause huge economic losses, and therefore in power electronic circuits in fault diagnosis technology has enormous economic significance and necessity. Integration and complexity of today's power electronics technology trends, which rely solely on experience makes the traditional the artificial diagnostic methods has been difficult to meet the requirements. Country, often used inside and outside the fault diagnosis method for power electronics, such as fault dictionary method can only solve the single fault the fault tree achievements heavy workload, prone to error and difficult to achieve direct detection hardware circuit, investment, residual method need to establish precise mathematical model, expert system knowledge acquisition \The fault diagnosis method based on artificial neural networks due to the good nonlinear mapping features and the ability to self-learning, well adapted to this trend has become a hot research field of fault diagnosis. BP network is widely used in neural network fault diagnosis of power electronic circuits slow convergence, easy to fall into the local extremum defects, which seriously affects the performance of fault diagnosis. Quantum genetic algorithm combines the characteristics of quantum computation and genetic algorithms, the ability of global optimization, convergence speed, but speed slower, chromosome effective utilization rate is low, this proposed improvement measures in the later stage of evolution, and then improved quantum genetic algorithm to optimize BP neural network, and also its position in BP network gives a good initial connection coefficient smaller search space for a global optimal solution, which effectively improve the defects of BP network to accelerate the the learning speed of the network, in order to achieve a higher diagnostic performance. Subject to improvement and development of power electronic circuit fault diagnosis method based on BP neural network for the purpose, specifically to do the following work: For double-stranded quantum genetic algorithm and coded quantum genetic algorithm based on the Bloch spherical coordinates in late optimizing the speed of evolution slow, weak local optimization, chromosomal defects in the effective use of low, made two improvements that improved the population evolutionary mechanisms and join the optimization process of the comparison population. By multimodal function optimization instance, verify that the improved algorithm effectively improve the defects, while the diversity of the population is stronger, and faster convergence. Finally select more excellent performance BP neural network optimization algorithm based on Bloch spherical coordinates encoded quantum genetic algorithm program design. Rectifier circuit, for example, the combination of spectral analysis-based the fault information extraction method and MATLAB software simulation of the circuit failure. Combined with simulation software simulation of the circuit failure analysis to determine the failure mode, spectrum analysis theory to extract fault, design fault samples. Combined with improved quantum genetic algorithm, BP neural networks, spectrum analysis, the fault information extraction technology of power electronic circuit fault diagnosis research, and other algorithms, the simulation results are compared. The improved algorithm has the network learning fast diagnosis of the advantages of high while for nonstandard samples tested also has a certain generalization ability. Verify the effectiveness and superiority of the improved quantum genetic algorithm and BP neural network-based fault diagnosis method for power electronic circuits.

Related Dissertations

  1. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  2. Research on Visual Servo System of Mechanical ARM,TP242.6
  3. Municipal tourism land use planning environmental impact assessment,X820.3
  4. Study on Taste Characteristic of Taste Peptide Enzymatic Production from Oyster Base on A Neural Network Method,TS254.4
  5. The Research on Evaluation of Living Status Systems of Expressway Relocated People,D523
  6. Mine Risk Information Integration and Intelligent Early Warning,X936
  7. Research of Orange Quality Classification Technology Based on Computer Vision,TP391.41
  8. Research and Application of Single-Stage Multi-Product Batch Scheduling Based on Quantum Genetic Algorithm,TP18
  9. Quantum Genetic Algorithm and Its Application in the Scheduling Problem,TP18
  10. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  11. Research on State Diagnosis on Fan Based on Factor Analysis and BP Neural Network,F426.61
  12. In-furnace Temperature Information Included Combustion Optimization of a Utility Boiler,TK227.1
  13. Research on DNA Encoding Based on Quantum Computing,Q75
  14. Study on Agricultural Machinery Injury in North of China and Related Risk Factors,R139
  15. TD-SCDMA Base Station Fault Alarm Expert System Based on Neural Network,TP277;TN929.5
  16. Research on Credit Evaluation and Promotion of Real Estate Development Enterprises,F293.33
  17. GPS Technique and Data Processing for Deformation Monitoring,P228.4
  18. Research on the Inspection and the Application of GPS Kenimatic Surveying Technology in the Project of FAST,P228.4
  19. Study on Optimization Modeling and Intelligent Control of Alcohol-based Fuel Boiler Combustion System,TK223.2
  20. Research of QGA on the Mechanical Optimization Problems,TP18
  21. Research on Intelligent Optimization Algorithm and Its Application in Communication,TN92

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Basic electronic circuits > Electronic circuit
© 2012 www.DissertationTopic.Net  Mobile