Spatial electric load forecasting model for Sri Lanka

Abstract

With the high level of city expansion observed during the last few decades, distribution utilities currently face new challenges when planning network expansion with profitable operations. Thus distribution utilities should consider spatial electric load forecasting as the basis for the planning of the electricity distribution networks. Spatial electric load forecasting helps in determining how the increase in demand of electrical energy will be distributed geographically in the service area. In Sri Lanka, the load forecasting in distribution planning is mainly based on trending methods which lacks the accuracy needed for present dynamic consumer market. The objective of this research is to prepare a simple yet accurate and effective spatial electric load forecasting model which can be used in the local context. This research deals with a new method for spatial electric load forecasting using artificial neural networks. The electric load growth inside the service area of an electric utility can be expected for two reasons, natural load growth of existing consumers and addition of new loads because of new consumers. In the study, the addition of new consumers in future is regarded as the new load additions in the vacant areas. This is forecasted using the spatial electric load forecasting model implemented using artificial neural network. The growth of existing consumers is addressed with a constant growth. The implemented model is presented and tested with data from two real midsized cities. The outcome is compared with the ones obtained from the utility planning department existing software. The results illustrate that the proposed model provides an accurate and user-friendly technique to predict yearly residential electrical load in Sri Lanka