Fault Diagnosis Of Power SCADA System Based On Artificial Neural Network

Urban power network has been an indispensable part of people’s life,home,school,factory and other industries are not separated from the electricity,which maintains the normal life of people.And the power failure and other power failures have been around people,but also affects the normal operation and stability of the power,how to reduce the power fault and restore power supply has been the problem of power,which should be solved.In power system,the SCADA system is the most widely used and its technology development is more mature.When the power is in trouble,the protection device can cause the protective action to protect the action and the corresponding circuit breaker is tripped,the protective action and circuit breaker tripping signal is transmitted to the SCADA system,then make the diagnosis decision and the SCADA system can record all the data of the power.In this paper,a power SCADA system based on.NET platform is constructed and the SCADA system is programmed with VB under WINDOWS operating system.The system mainly includes the personnel management,the system working condition,the graph index,the data acquisition interface,the fault diagnosis and so on.In this paper,the artificial neural network is used to diagnose the fault,using two kinds of diagnosis methods,respectively,with a tutor neural network and a non tutor neural network,in the neural network,K-means clustering was used to optimize neural network hidden layer center,the weights of the network are optimized by using the Genetic algorithm method,and the two methods can be used in the simulation.Finally,the use of SCADA system to demonstrate the operation of the various application modules.