Frequent changes in primary voltage can be a significant power quality concern on distribution systems. The main cause of these voltage fluctuations is rapid changes in loads on a system. However, there are other phenomena apart from load variation which can also create voltage fluctuations. One of these is the variation in output level of distributed energy resources (DERs). Photovoltaic based DERs for instance, have an inherent uncertainty in their output level due to the naturally occurring intermittency of irradiance levels. This uncertain injection pattern can create a significant impact on the number of voltage fluctuation events experienced on a distribution feeder. A study of this phenomenon is of utmost importance to help distribution engineers understand the potential effects of solar irradiance ramp events which can cause voltage fluctuation and further to consider possible mitigation steps. In this paper, a high-resolution irradiance dataset is analyzed with a probabilistic approach to find the extremes of these uncertain events. In addition, voltage sensitive load modeling is used for higher accuracy. Moreover, the analysis is performed on each phase separately to capture the varied effects due to load imbalance. The analysis is then repeated using reduced primary voltage to observe the impact on the system when it is operated under a conservation by voltage reduction (CVR) scheme at peak demand. The entire analysis is performed in a quasi-static time series simulation platform using OpenDSS and MATLAB. Finally, smart inverter Q-V droop functionality is used to assess its effectiveness in reducing the number of voltage fluctuation events.