| Abstrak/Abstract |
The increasing integration of photovoltaic (PV) generation into power distribution networks introduces greater uncertainties for the system. Therefore, it is crucial for power flow analyses to implement probabilistic approaches considering the unpredictability of a generation. This study introduced a model for scenario uncertainty in PV generation by integrating kernel change point detection (CPD) techniques, Monte Carlo simulation (MCS), and k-Means Clustering (KMC) with metric soft-dynamic time warping (DTW). Using a combination of methods, we derived PV profiles under five scenarios for two seasons. Further, the impact of PV profile models on probabilistic power flow (PPF) is evaluated on 69-bus distribution networks. The findings indicate that integrating PV generation can enhance the voltage profile. Simulation results show that PV integration reduces total energy losses by 21.58 % and 25.86 % in Seasons 1 and 2, respectively, demonstrating the effectiveness of seasonal PV deployment. Meanwhile, the average correlation between PV scenarios is 0.9996. This finding indicates that a rise in PV output plays a significant role in enhancing voltage support throughout the network. Implementing PPF in this study will provide insight into network planning and operations with more realistic PV generation integration. |
