Information Fusion Fault Diagnosis System Of Hydroelectric Generating Set

Hydropower plays an important role in our country. The relevant departments take serious to the security and stability of the production of hydropower. Hydroelectric generating sets(HGS) are key units in hydroelectric power station, whose running state is relate to not only whether the hydroelectric power station can provide safety and economy electricity, but also the security of hydroelectric power station.Information fusion rises in the military fields, dealing with the multi-source information. As the fruit of the traditional subjects and the infant projects, it is a new frontal subject, which has exceeded the application in military fields and got broad application in many fields. It is the nature of information fusion to take advantage of the subsistent information as far as possible, and to achieve optimal control of the system or have the best evaluation about present state of the system through analyzing and handling multi-source information.There is much useful information in the HGS fault diagnosis procedure. We cannot improve the accuracy and reliability of fault diagnosis unless we fully utilize that useful information. HGS fault diagnosis is an information fusion procedure at this time.This paper analyzes information fusion applying to HGS fault diagnosis system. On the basis of deeply analyzing the key fault and the cause, the paper analyzes the feasibility and availability of D-S evidence fusion applying to the HGS fault diagnosis system. Pointing to that the fundamental confidence assigns of the evidence fusion is difficult to ascertain specifically, the information fusion method, using the neural networks and D-S evidence theory, is put forward to diagnose the faults of HGS. The diagnostic tests prove that the system is good to improve the reliability of the diagnosis. At last, this paper develops D-S evidence fusion decision-making fault diagnosis software.