Optimal Power Flow with the Clustering-Improved Fuzzy Models for Uncertainty Sources of Wind Power and Nodal Loads

The main goal of the paper is to develop an Optimal Power Flow calculation method considering the uncertainties of the wind power and the nodal loads modelled through the improved fuzzy models defined based on a clustering process of the active and reactive powers recorded hour by hour in the database of the Supervisory Control and Data Acquisition Systems belonging to the transmission network and distribution network operators. The decision variables used in the optimisation process are represented by the positions of the on-load tap changers corresponding to the transformers from the electric substations (ultra-high/high voltage). The proposed approach has been tested in an electrical network with two voltage levels (ultra-high and high voltage) and two wind power plants integrated. The comparison with the results obtained using the real data from the analysed period (one day) indicated a mean absolute percentage error below 1% for the voltages from the high voltage nodes and the same total number of turns of the on-load tap changers which equip the power transformers from the network.