Location:Home > Engineering science > Detection Technology and Automation > The Studies of Bei Piao Wind Farm Generator Gearbox Fault Diagnosis

The Studies of Bei Piao Wind Farm Generator Gearbox Fault Diagnosis

Downloads: []
Tutor: ZhaoGuoCai
School: Liaoning Technical University
Course: Detection Technology and Automation
Keywords: Diagnosis,Genetic Algorithm(GA),Fuzzy rules,FNN
CLC: TH165.3
Type: Master's thesis
Year:  2011
Facebook Google+ Email Gmail Evernote LinkedIn Twitter Addthis

not access Image Error Other errors

The gear box is the important part of the generator, early detection and diagnosis of the failure, normal operation of the maintenance system have great significance to financial arrangements for the maintenance of equipment reasonable time and reduce the failure rate,avoiding major personal and equipment¡¯s fatal accidents .There are many diagnose ways, such as :traditional diagnose, mathematics failure diagnosis, and intelligence failure diagnosis method (fuzzy logic and nervous system, expert systems),etc.By comparison,we propose the model based on the genetic algorithm(GA) and the fuzzy nerve network(FNN). Before training the neural network, the introduction of GA chromosome of crossover and mutation operator, to find the optimal BP network initial weights, through training,NN,and bring the rules of the vague with the nerve network in the framework to enable the network to a distinct sense, and retain study mechanism of the nerve network.In the searching ,we bing in the process of the use of genetic algorithm, achieving the best convergence and global optimization. network, re-introduction of GA to optimize network parameters, can effectively avoid the premature convergence of the network.The article uses the evolution of the algorithm to the study of the value, avoiding the shortcomings of the original BP algorithm,early running into the local best value.This paper uses the fuzzy rule expert system, make gearbox fault diagnosis,obtaining diagnostic results.,this paper analyzes that this method in fault diagnosis is practical, but also has significant limitations. Since having the learning ability of neural networks, it is widely used in fault diagnosis. Used in the article to establish fault diagnosis of BP network model and genetic algorithms, neural networks, fuzzy logic combined with the fuzzy neural network model, fault diagnosis of gearbox, can get the correct fault diagnosis. Through the training time for the two methods, the relative error and the comparison shows the superiority of GA-FNN, indicating the effectiveness of the method, the feasibility to achieve the desired effect.
Related Dissertations
Last updated
Sponsored Links
Home |About Us| Contact Us| Feedback| Privacy | copyright | Back to top