Research on Fault Diagnosis for Air Drilling Based on an Improved PSO for Optimization of Fuzzy Neural Network

The air drilling technology, which is a new kind of drilling technology, has got a wide range of applications at home and abroad. Compared with conventional drilling technology, it has a large number of advantages, such as lower cost, faster drilling speed and less pollution. However, lots of questions like drilling tools out of operation frequently were exposed during the gas drilling which restricted the development of air drilling technology. Thus, it has great significance to hammer at studying the air drilling accident diagnostic techniques. The paper has a comprehensive introduction about air drilling principle and the cause of the accident in the process of air drilling. Meanwhile, the paper has proposed an improved particle swarm algorithm for optimization of fuzzy neural network based on new development in automation and intelligent technology and has carried out a positive analysis by using it. The new model in the convergence speed, adaptive value and diagnostic error, etc. is the optimal by comparing with BP neural network and PSO neural network. It can improve the subjective shortcoming of traditional methods that the staff in the field operation analyses data through the real-time monitoring system and the air drilling experts identify some characteristics of air drilling accident causes to reduce errors, improve the accuracy of diagnostics and become more intelligent.