Modeling Photovoltaic Generation Uncertainties for Monte Carlo Method based Probabilistic Load Flow Analysis of Distribution Network

Distributed generation such as photovoltaic is more encouraged by government policies to exploit the renewable energy resource and for sustainable energy production. The integration of photovoltaic generation to the distribution network introduces large uncertainties, the inclusion of these uncertainties in load flow analysis used for planning, operation, and expansion of the power system networks is done by using the probabilistic load flow studies approach. With this, the risk identification and its mitigation associated with the power network can be easily performed. In this work, the modeling of photovoltaic generation based on historical data that is obtained by power system measurements is presented. The distribution of photovoltaic generation obtained is complex and does not match any standard distributions. A numerical method is used to generate the samples that are used for Monte Carlo simulation to include photovoltaic generation with its original characteristics and uncertainties in probabilistic analysis. The analysis is done by integrating a 1000 kW photovoltaic generation system in IEEE 13 Node Test Feeder, which is highly loaded and unbalanced. As a function of a photovoltaic generation uncertainty, statistical distribution of variables such as currents and voltage of each network elements are calculated. Using them, the network health and the power loss associated with the network are determined.