ANN BASED NOVEL APPROACH FOR IMPROVEMENT OF POWER SYSTEM SECURITY
Keywords:
Newton- Raphson Load Flow, Contingency Analysis, Security Assessment, Feed Forward Back-Propagation Neural NetworkAbstract
This paper describes what is power system security, how it can be carried out in large networks, and new improvement to the
existing defense . Completely reliable system against contingency is an objective, however, not possible. There will always be
vulnerable in the network. This paper describes briefly some to methods to make system security more powerful . We also describe
limitations of existing method to detect and prevention of these contingency in power system networks. Feed Forward Back
Propagation Neural Network is used to classify the security condition of test bus system. The input data of ANN are derived from
offline Newton Raphson load flow analysis. The result obtained from the ANN method is compared with the Newton Raphson load
flow analysis in terms of accuracy to predict the security level of test bus system. The accuracy of 14 hidden neurons feed forward
back propagation neural network to predict the security level of test bus system is 99.99%. In conclusion, ANN is found to be
reliable to evaluate the security level of test bus system.
