Static security assessment on power system using artificial neural network

Abstract

In modern industrialized society, a supply of electric energy is expected to be reliable and continuous since a high availability of secure power system is essential for its’ progress. A secure power system is expected to be free from risk or danger and to have the ability to withstand without exception to any one of the pre-selected list of credible contingencies. The objective of this research is to investigate the reliability of the Static Security Assessment (SSA) in determining the security level of power system from serious interference during operation. Therefore, back propagation Artificial Neural Network (ANN) is implemented to classify the security status in the test power system. Offline Newton-Raphson load flow is employed to gather the input data for the ANN. The large dimensionality of input data is scaled down by screening process to reduce the computational time during ANN training process. This method has been tested with 4 bus test system and IEEE 24 bus test system. Bus voltage and thermal line variables are set as a limit to the developed method. It has been discovered that error of trained ANN are within the acceptable range if compared to similar results from published works. The ANN has been found to be faster than the conventional method in predicting the security level of the tested system. It is concluded that the ANN works well in providing status of the current operating point for specific contingency of power system.