On-line network reconfiguration for enhancement of voltage stability in distriution systems using artificial neural networks
Network reconfiguration for maximizing voltage stability is the determination of switching-options that maximize voltage stability the most for a particular set of loads on the distribution systems, and is performed by altering the topological structure of distribution feeders. Network reconfiguration for time- varying loads is a complex and extremely nonlinear optimization problem which can be effectively solved by Artificial Neural Networks (ANNs), as ANNs are capable of learning a tremendous variety of pattern mapping relationships with- out having a prior knowledge of a mathematical function. In this paper a generalized ANN model is proposed for on-line enhancement of vo ltage stability under varying load conditions. The training se ts for the A NN are carefully se - lected to cover the entire range of input space . For the A NN model, the training data are generated from the Daily Load Curves (DLCs). A 16-bus test system is considered to demonstrate the perfo rmance of the deve loped A NN model. The proposed A NN is trained using Conjugate Gradient Descent Back-propagation A lgorithm and tested by applying arbitrary input data generated from DLCs. The te st re sults of the A NN model are found to be the same as that obtained by oŒ-line simulation. The enhancement of vo ltage stability can be achieved by the proposed method without any additional cost involved for installation of capac itors, tap-changing transformers, and the re lated switching equipment in the distribution systems. The deve loped A NN mode l can be implemented in hardware using the neural chips currently available.