A Comprehensive Model for Zonal Short Term Load Forecasting.

Short-term load forecasting for small power systems, namely zonal STLF, is one of the issues in the load forecasting area which has vital value for electricity market competitors like retailers and ISO. This paper discusses the common problems related to short term load forecasting which are particularly raised in the small power systems. Identifying bad data which constitute a large portion of load data in these systems, choosing a proper manner for selecting ANN training data, and requirements which should be considered in the structure of zonal STLF to facilitate its employing to different power systems are investigated in this paper. Here, a comprehensive model which considers these problems as well as the generality of structure and the simplicity of implementation is proposed. This model is tested on Isfahan power system which is one of the regional power companies of Iran national power system. The results show that this structure can be implemented effectively and have an acceptable error in spite of large amount of bad data among historical data. [ABSTRACT FROM AUTHOR]