Stage Fault Diagnosis Based On Data Fusion

With the improvement of the people's life, the audience's require to civilization and art is simultaneously increased, which brings forward even more requests to the development of the stage electronic technology. However, as the complex control system of stage, the expense of maintenance is high. It is necessary to find effective and practical stage fault diagnosis method.The paper is written under the premise of reading much information about fault diagnosis and combines stage fault characteristics. Trying to use data fusion technology which is popular in the field of control system in recent years to develop a fault diagnosis system used in the stage. This paper gives a new diagnostic method based on BP neural network and D-S evidence theory, it can achieve their complementary advantages and increase efficiency of diagnosis.This paper introduces firstly basic theory of artificial neural network, according to characteristic of BP neural network, it points out that the fault diagnosis using neural network feasibility and analyzes the basic principles of BP neural network fault diagnosis.Secondly, because BP neural network has some disadvantage in fault diagnosis field, and D-S evidence theory has some relational advantage, such as no prior probability and distinction between "unknow" and "uncertain ",we choose the method of BP neural network and D-S evidence theory. Then BP neural network model in fault diagnosis system is given, which provides theoretical direction for diagnostic system design.Finally, this paper discuss mechanical fault and characteristic of revolving stage, the integrated diagnostic model is brought forward. the noise and vibration signal is measured from revolving stage. The measured data is local diagnosed by BP neural network, and it is together with by D-S theory, and the results is satisfied. The accuracy of stage fault diagnosis proves the effectiveness this diagnostic method based on BP neural network and D-S evidence theory.