The Graduation Fault Diagnosis Algorithm’s Study and Simulation Based on Immune Neural Network and Fuzzy Logic Applied in Complex Industrial System

Abstract

Industrial system has the characteristics of large scale and high complexity and much variable. Fault diagnosis with single theory or method is insufficient accurate. This paper presented a kind of graduation fault diagnosis algorithm based on immune neural network and fuzzy logic. As an example of the cooling system in nitric acid production process, the cooling system is divided into loop level and component level, using immune neural network to identify loop level faults, using fuzzy logic to identify component level faults. The simulation results show that the graduation fault diagnosis algorithm based on immune neural network and fuzzy logic has faster training speed and better generalization ability, and it can distinguish multi-routes faults. This algorithm can be used fault diagnosis for other complex system.