Nc Machine Tool Fault Diagnosis Based On Fuzzy Neural Network Research

CNC machine tool is computer integrated manufacturing, flexiblemanufacturing and realization of e-manufacturing base, has an irreplaceable rolein the modern manufacturing industry. But because of its structure and kinds ofcomplexity, their failures can be complex and often hidden features. However,due to the complex structure of CNC machine tools, a wide range of variousfactors cross both uncertainties, there are uncertainties, and thus lead to thefailure of CNC machine tools with a large gradient concealed. CNC machinetool failure will cause incalculable damage machinery and equipmentmanufacturing. In advance, fast, accurate fault diagnosis problem solvenumerical control machine tools, is particularly important, and not only forenterprise development and even mechanical equipment manufacturing industryin China has a significant role in promoting and practical significance.As is known to all, artificial intelligence technology has played a huge rolein modern CNC machine fault diagnosis technology, but the use of the singleartificial intelligence techniques to fault diagnosis of CNC machine tools thereaccuracy is not high and weak generalization. As one of the most successfulalgorithms in the artificial intelligence technology, artificial neural networks andfuzzy logic theory to each other there are similarities and complementarity.Sothe fusion of two models in a reasonable manner, form a fuzzy neural networkmodel,which has a complementary strengths, weaknesses.The fuzzy neuralnetwork model not only has the artificial neural network of distributed, parallelprocessing complex information, as well as a high degree of self-organization,self-learning ability, but also with the fuzzy logic to deal with imprecise andincomplete information. Thus, the fuzzy neural network can be an effectivemethod of research CNC machine fault diagnosis.In this paper, based on the common fault diagnosis of CNC machine toolsfor research purposes, and conducted in-depth study of the basic principles offuzzy neural network and engineering applications.34fault symptoms ofCAK6150CNC horizontal lathe with54groups cause of the malfunction isselected as the object of study. After a careful induction, the formation of the sample data for CNC machine tools based on fuzzy neural network faultdiagnosis model. Finally, using MATLAB software on the CNC machine faultdiagnosis model based on fuzzy neural network training and simulation process.The simulation results not only very close to the theoretical output, and theconvergence speed is very fast, and proves that the model has a strongfeasibility.In summary, this article CNC machine tools based on fuzzy neural networkfault diagnosis model to study the establishment of the model for the rapiddiagnosis of the common faults of CNC machine tools, CNC machine toolsdowntime reduction, keep machine stable processing capacity to ensure theCNC machine tools successful completion of the production tasks of greatsignificance.