On-line determination of transient stability status using multilayer perceptron neural network
Emmanuel Asuming Frimpong – Philip Yaw Okyere – Johnson Asumadu
A scheme to predict transient stability status following a disturbance is presented. The scheme is activated upon the tripping of a line or bus and operates as follows: Two samples of frequency deviation values at all generator buses are obtained. At each generator bus, the maximum frequency deviation within the two samples is extracted. A vector is then constructed from the extracted maximum frequency deviations. The Euclidean norm of the constructed vector is calculated and then fed as input to a trained multilayer perceptron neural network which predicts the stability status of the system. The scheme was tested using data generated from the New England test system. The scheme successfully predicted the stability status of all two hundred and five disturbance test cases.
Keywords: power system stability, stability prediction, transient stability, out-of-step, neural network, Euclidean norm