Application Of BP Neural Network And Fuzzy Logic In Intelligent Fault Diagnosis

Industrial production has been the various countries to enhance their strength, to maintain technological leadership in the areas of most concern. In industrial production, machinery and equipment is a top priority. Therefore, all countries have devoted a lot of manpower and material resources for the improvement of equipment. Then fault diagnosis became a hot field. As we all know, in the life cycle of a machine, production costs account for less than one-third of the cost. If we have advanced troubleshooting techniques, undoubtedly greatly reduce production costs.Since the structure of the device before the production is simple, produced easily, so lead companies do not attach importance to inspection and maintenance of equipment failure. But with the rapid advance of modern machinery automation process, cost of production equipment has accounted for more than 50% of the production costs. So companies hesitate to spend great efforts to research equipment failure, researchers also timely use of the latest technology, invented a lot of techniques and methods of intelligence testing equipment failures. Especially with significantly improved computer performance and large-scale applications, change the case of the conventional diagnostic methods to detect and diagnose those not sufficiently high complexity, high intelligence, modern large-scale production equipment. Thus, more intelligent and humane, intelligent detection technology is the future trend of fault diagnosis.Through a comprehensive technology development trends in recent years, by studying comparative think fuzzy theory based on mature and BP neural network combined with new technology, can play an important role in technological innovation fault detection field. It is able to provide new algorithms and ideas for troubleshooting. Through the introduction of fault diagnosis research background, analyzes the current development process and troubleshooting, explained the significance of fault diagnosis and elaborated on BP neural network and fuzzy logic theory. Finally, with engine repair for example, to prove the rationality and practicality of the method.