<![CDATA[The increasing penetration of photovoltaic (PV) generation in modern distribution networks introduces considerable uncertainty due to the inherently fluctuating nature of solar irradiance. These fluctuations directly affect PV output power, resulting in significant variations in bus voltages, feeder currents, and power losses. Traditional deterministic load flow (DLF) analysis, which assumes fixed PV generation, is unable to capture this stochastic behavior and therefore may lead to inaccurate estimation of system conditions. This study presents a comparative assessment between deterministic load flow using the Backward–Forward Sweep (BFS) method and probabilistic load flow (PLF) based on Hong’s Two-Point Estimate Method (PEM) to evaluate the impact of solar irradiance variability on the performance of PV-integrated distribution networks. Solar irradiance data are statistically characterized to obtain the mean and variance, which are then propagated into PV output power through a linear irradiance–power conversion model. The IEEE 34-bus radial distribution system is used as the test network, with multiple PV units installed at selected buses. The results show that deterministic analysis underestimates voltage deviations and fails to capture the range of power losses induced by PV uncertainty. In particular, the deterministic BFS solution yields a single operating point with real and reactive losses of 0.1582 MW and 0.0479 MVar, whereas the probabilistic 2PEM produces mean losses of 0.105 MW and 0.031 MVar with standard deviations of 0.057 MW and 0.016 MVar, respectively. In contrast to the fixed deterministic voltage curve, probabilistic voltage profiles form a bounded envelope around it, indicating non-negligible downstream voltage variability driven by irradiance fluctuations. Overall, the findings confirm that solar irradiance variability substantially influences distribution system performance, and incorporating probabilistic assessment is essential for more realistic, risk-informed planning and operation of PV-integrated distribution systems.]]>
