A System and Method for Electricity Price Forecasting with a Novel Forecast Error Correction

This technical article presents the concepts for improving the electricity price forecasting which were patented (application no. IN236/MUM/2011). A novel system and method for electricity price forecasting are also disclosed to set the context for explaining the technicalities involved in the invention. The invention relates to forecasting, and particularly to price forecasting by employing error correction techniques external to the algorithm chosen to solve the forecasting problem. Existing mechanisms do not offer effective error correction mechanisms external to the forecasting engine. The proposed technique uses the stochastic formulations and golden search methods to improve the final forecast. The method initially employs a forecasting module which consists of SVM, robust regressions, and neural network based mechanisms to determine the initial forecasts. In case there is an error, it determines if the error is within the permissible values through various probability checks. Further, if the error is outside the permissible values then a correction factor is applied to the computed values and prediction is updated. The correction factor is determined using the golden search technique. The framework for the proposed approach has been tested on the Indian and Australian electricity market data to prove its efficacy.