Transmission loss allocation in power systems using artificial neural network

Cost allocation and determining the part of the operators in a power system from the total costs are among the most important issues appeared along with the restructuring in the power industry. One of these imposed costs in the utilization domain is the cost of the power losses which should be fairly distributed between the participants in the electric power market. In this paper, using the load flow calculations, the part of each bus in the power losses is determined by the Z-Bus method. These results are compared with those obtained from the load flow calculations using the artificial neural network. It is shown that the artificial neural network is an efficient tool for power loss allocation in the large and complicated power systems which may have a nonlinear nature. The proposed method is then applied to two test bench systems, the IEEE 5-bus and 30-bus test benches, and the results from two approaches are compared and the differences in term of error are reported. A real case study including a 400kV transmission system is also studied and the annual peak-load power loss allocation assuming the peak hour in a month is determined and the related errors are computed.