Study On Fault Diagnosis Of Flexible Manufacturing System Based On Neural Network

The content of this paper is the research of fault diagnosis of flexible manufacturing system(FMS) based on neural network.With the rapid economic development, some traditional manufacturing mode has been unable to meet the needs of the public to the manufacturing industry.FMS came into being in this situation.The birth of FMS, can greatly reduce the workload, greatly meet the company’s products on the market changes of the strain.The advantages of FMS in manufacturing industry are more and more obvious.But in the actual production process, because of its complex structure, frequent failure. The technology of FMS is difficult to reflect. This makes it necessary to study the fault diagnosis of FMS.This paper analyzes the basic structure and working principle of the FMS,and then combined with the fault information of a FMS in a few years.Analysis of the FMS, which is the weak link, and then the vertical machining center as the fault diagnosis research point.Aiming at the fault diagnosis of domestic and foreign research results, considering the advantages of neural network technology in fault diagnosis, this paper determined the application of neural network technology in fault diagnosis of FMS.In this paper, through the combination of genetic algorithm and neural network, the algorithm of BP neural network algorithm is improved. By comparing the advantages and disadvantages of different signal acquisition methods, the current energy method is applied to the identification and fault diagnosis of the tool condition of vertical machining center. Acquisition of the spindle motor current signal, current energy value was normalized after input to the training has been improved BP neural network should model. Through the experimental data analysis and validation, achieved faster diagnosis speed and more accurate diagnosis results.In the production practice of flexible manufacturing system, frequent failure is often the key reason for the serious decline in production efficiency. FMS, the reasons for the failure of various types of complex. Therefore through statistical analysis, identify the key parts, with an effective method of the for condition monitoring and fault diagnosis, can effectively control of FMS fault processing time, greatly improve the production efficiency. So this paper has some reference value and practical application value.