In this paper, we present the results of two methods developed to forecast cutset (intra-regional transmission line) power flows. The first method involving neural network load forecasting and simultaneous equations can only be applied to radial electricity networks. This method was applied to forecast the cutset flows in a simplified radial formulation of the Queensland electricity network with an error of approximately 5%. The second method of forecasting the cutset flows directly from the data set utilising a neural network can be applied to any radial or meshed electricity network. This direct method was tested on a more complex model of Queenslandu27s electricity network and gave cutset forecasts with an error of approximately 10% MAE. Results indicate that the cutset flow forecasting tool is accurate, robust and adaptable to suit the requirements of other utilities.
展开▼
