Short Term Load Forecasting of Urban Electric Load Profiles Using Relevance Vector Machine

This work employs supervised learning approaches, such as support vector and relevance vector machines to anticipate short-term loads for urban electric load profiles. The Holt-Winters approach, Support vector machines, and relevance vector machines are all used to determine accuracy. For the network to learn, it needs a sufficient amount of big historical data, most of which will be utilized for training and the remainder for testing. Cross-validation is carried out before learning the network with a given data set which finally leads to the increase in the accuracy of the system. Key factors like weather data, special holidays and weekend data are also included in the analysis to increase the accuracy. In short term load forecasting with SVM and RVM which are supervised learning techniques the accuracy level is quite high enough so the mean absolute percentage error resulted is very small.