Load flow analysis using traditional methods for power flow is becoming complex (reverse power flow and voltage volatility) due to the configuration complexity brought about by renewable energy resource (RER) integration. The variable and intermittent nature of RER integration also contributes to the power flow complexity. Power system operators should be aware of the state of the operation. An alternative to traditional power flow methods could be an artificial intelligence technique. Therefore, in this study, the node voltage estimation of a distribution system using an artificial neural network (ANN) has been proposed. Since a significant portion of residential load and RER generation are dependent on weather conditions, load flow analysis including weather data in GridLAB-D has been carried out. Typical meteorological year (TMY) information has been used as the weather data. Results show that node voltage estimation using the ANN technique is robust on different photovoltaic (PV) and wind power penetration levels as well as the significant loss of load measurement data and/or PV and wind generation data.