Abstract：Reliability assessment of distribution systems is a crucial foundation for maintaining their power

supply continuity. For existing distribution network photovoltaic systems，solar irradiance modeling was limited to

its exponential autocorrelation. To enhance the reliability assessment through accurate prediction of photovoltaic

power generation and considering the autocorrelation sinusoidal wave characteristics of solar irradiance influenced

by cloud events，a reliability assessment model was proposed for distribution network photovoltaic systems based

on stochastic diffusion process（SDP）. Firstly，a Monte Carlo simulation based on SDP was used to build solar

irradiance and temperature prediction models. Secondly，transient clouds were modeled using jump diffusion

processes integrated into SDP. Then，a reliability assessment model for photovoltaic system components and a

photovoltaic power generation prediction model were constructed based on failure physics reliability models and

historical data. Finally，the stability and reliability of the proposed model were verified on an actual distribution

network. Experimental results demonstrate that the model effectively addresses the uncertainties of distribution

networks，enabling a more efficient assessment of the reliability of photovoltaic systems.

      Key words：distribution network；stochastic diffusion；Monte Carlo simulation（MCS）；reliability assessment