عنوان مقاله [English]
In an electric power system, peak load forecasting plays an important role, in terms of economic optimization and fuel management. Accurate forecasting methods can be helpful especially for developing countries where the demand is increased with dynamic and high growth rate. This paper presents a new hybrid regression model, which combines a self-organizing map (SOM) and a linear regression model for monthly peak load forecasting (PLF). Peak load data is used to test the model from the Iran electric power network from the last 14 years. Also, a principal component analysis (PCA) has been used to reduce input dimension. To evaluate the effectiveness of the model, forecasting has been performed by using a regression that uses the un-clustered data. Analysis and comparison of the results have shown the superiority and accuracy of the model.