Short-Term Nodal Congestion Price Forecasting in Large-Scale Power Market Using ANN with Genetic Optimization Training

In a daily power market, price and load forecasting is the most important signal for the market participants. In this paper, an accurate feed-forward neural network model with a genetic optimization levenberg-marquardt back propagation (LMBP) training algorithm is employed for short-term nodal congestion price forecasting in different zones of a large-scale power market. The use of genetic algorithms for neural network training optimization has had a remarkable effect on the accuracy of price forecasting in a large-scale power market. The necessary data for neural network training are obtained by solving optimal power flow equations that take into account all effective constraints at any hour of the day in a single month. The structure of the neural network has two input signals of active and reactive powers for every load busbar in every hour of the programming model. These two signals are always available. In this study, an IEEE 118-bus power system is to test the proposed method authenticity. This system is divided into three zones, and a neural network with a genetic algorithm training optimization is employed for every zone. Simulation results show the ability of the proposed method in forecasting the nodal congestion price and its severity in a large-scale power market with a rather low and acceptable error, especially at points of price spikes.