Application Of Genetic Algorithm And Neural Network In Steam Generator Fault Diagnosing

The Steam Generator is one important of The Nuclear Power Plant (NPP). Therefore it is very significant that monitoring the SG running condition, which can ensure the SG running safely and reduce the rate of accident.The NPP is a very complex system, and on actual operation course the parameter relatively heavy to change range, which make traditional fault diagnose method is limited in use of this field. Developing rapidly with the theory such as artificial intelligence, pattern-recognition, etc., intelligent-diagnose technology have already become a new research direction in the field of fault diagnosing.A lot of national and international condition montoring and fault dagnosis systems for the set of the power plant are study by the author and let as the basis. In this paper, the object of study is the. A fault simulation model of the Steam Generator in an atomic power installation was built up to dynamically simulate three kinds of faults frequently occur in the Steam Generator operation. It is known that the genetic algorithms (GAs) are good in global searching, and the artificial neural networks (ANNs) are effective on accurate local searching. We proposed a method to join the genetic algorithm and BP algorithm together and supplementing mutually by optimizing the initial weights of ANNs with GAs. In thesis we use the instance of sequence XOR verify the exactness of the Genetic Neural Network. Finally the traditional BP network and the Genetic Neural Network were applied to the fault diagnosis of steam generator respectively, the result proved that the Genetic Neural Network could accurate and effective realize the fault diagnosis quickly.The thesis finally utilizes Visual C ++ platform to set upthe Fault Diagnose System of Steam Generator based on Genetic Neural Network. It adopt human-computer interaction interface and has the characteristic of friendship interface and simple operat ion.