<![CDATA[Current electricity load forecasting sectoral become simpler and easier to implement. However, these conditions are faced with a situation where the results of macro forecasts still so do not show the load centers on a smaller area (grid) and resulted in no distribution substation location can be determined with certainty. In addition, the accuracy would tend to bias in a region that has limited data and the area of land use change that fast. Therefore, in this paper will outline the problem by conducting research on the electric load forecasting smaller area. This research method is to use clustering techniques to overcome the problem of the large volume of arithmetic processes. Expected benefits of this research will be able to provide information determining the magnitude of the load, when it happens and where is the location of the burden of a higher level of accuracy, making it suitable to be used for basic planning of power distribution network development. The results showed that the processing of the correlation matrix between variables with clustering techniques gained 4 cluster of 107 villages located in Kebayoran AJ. Then the average growth of electricity demand per year (2007-2016) per sector: 6:38% (Estate: 6.4%; Industry: 6.2%; Business: 6.4%, and the social: 6.0%).]]>
Copyrights © 2013
