An Improved ANN–BPN to Radial Distribution System Load Flow

Abstract

This paper shows an application of artificial neural networks (ANNs) to determine the bus voltages and phase angles of a radial distribution system, without executing the load flow algorithm, for any given load. The performance of the conventional load flow methods such as Newtoh—Raphson load flow, Fast decoupled load flow is found to be very poor under critical conditions, such as high R/X ratio, heavily loading condition, etc. To overcome the limitations of these regularly used methods a simple and reliable ladder iterative technique is used for solving the power balance equations of radial distribution system (RDS). The proposed method makes use of a multilayer feed forward ANN with error back propagation learning algorithm for calculation of bus voltages and its angles. A sample IEEE 33-bus is extensively tested with the proposed ANN-based approach indicating its viability for RDS load flow assessment and results are presented.KeywordsLadder iterative techniqueR/X ratioArtificial neural networks (ANNs)Backpropagation network (BPN)Radial distribution system (RDS)Forward sweepBackward sweep