IEEE Digital Library
Faculty of Computing, Health and Science
School of Engineering (SOE) / Centre for Communications Engineering Research
In this paper, a Tuning Fuzzy System (TFS) is used to improve the energy demand forecasting for a medium-size microgrid. As a case study, the energy demand of the Joondalup Campus of Edith Cowan University (ECU) in Western Australia is modelled. The developed model is required to perform economic dispatch for the ECU microgrid in islanding mode. To achieve an active economic dispatch demand prediction model, actual load readings are considered. A fuzzy tuning mechanism is added to the prediction model to enhance the prediction accuracy based on actual load changes. The demand prediction is modelled by a Fuzzy Subtractive Clustering Method (FSCM) based Adaptive Neuro Fuzzy Inference System (ANFIS). Three years of historical load data which includes timing information is used to develop and verify the prediction model. The TFS is developed from the knowledge of the error between the actual and predicted demand values to tune the prediction output. The results show that the TFS can successfully tune the prediction values and reduce the error in the subsequent prediction iterations. Simulation results show that the proposed prediction model can be used for performing economic dispatch in the microgrid.