Designing ANN for load flow in electrical power systems using EBP algorithm

Abstract EN

Load flow studies are important in power system planning, control and operating strategies when applied in real time mode.

Comparative study between different load flow solution methods [the (Newton Rap son (NR), Fast De-coupled (FD)) conventional methods arid 'Artificial-Neural Network (ANN) method] is carried out.

Feed Forward model of the neural network based on Error Back propagation (EBP) algorithm has been used for load flow problem and, a method tested using one of the Iraqi National Grid Systems.

A computer program was performed using MA.TLAB software program in order to design the system that is used to enter the sample values of the power system units.

The result of the study shows an efficient technique to be used for getting the difference in time rate of the final processing reveal that ANN showed good efficiency in the process of deriving results from the two conventional methods to the same input data.

The simulations of the proposed approach are presented.