Estimation Of Critical Clearing Time Using Artificial Neural Network For Based Transient Stability Mode

This study represents simulations for estimating transient stability using a neural multi-layer perceptron (MLP) network to obtain the neural network training set to evaluate the transient stability of the Iraqi Super Grid (400 KV) to measure the critical clearing time. Whether the actual and active load varies on each bus when the system fails It will show the actual loading sequence of the system to use as an input to the neural network, while the critical clearing time (CCT) is the basis for calculating the critical clearance time target data used in the neural network from the Rate of Change Kinetic Energy (RACKE) principle, Multimachine power system using the Runge-Kutta fourth-order combination of methods. A three-phase ground error was conducted for transient stability analysis and was not altered during simulation. In the result of the test, The proposed approach applied to Iraq super grid 400kv system has been shown to be more appropriate for estimating transient stability when actual and active loads shift over critical clearing time With a minimum 0.0016% error and a maximum 0.0419% error relative to multimachine CCT (RACKE).