Due to the recent environmental concerns and long-term challenges in energy security, the Global energy scenarios are shifting more towards sustainable and renewable energy resources. Brunei has planned to increase the use of cleaner energy technologies by contributing 10 percent or 954 GWh of renewable energy in its power generation mix by 2035. Out of the available renewable options, solar is the most promising one for Brunei, for example, the daily average solar installation is around 5kWh per day [1]. Though solar energy is an abundant resource, for optimally designing and successfully managing solar power projects, its availability in different time scales are to be analyzed and understood in a local context. In this paper, we present models for estimating the output of six different solar PV technologies using machine learning methods Performance data from the solar PV systems installed at the Tenaga Suria PV plant in Brunei are used to develop the models. Influence of relevant environmental parameters, such as irradiance, relative humidity, ambient temperature and wind speed on the power output has been analyzed for optimal feature selection. Performance models based on Support Vector Machine (SVM) and K-Nearest Neighbour (KNN) were developed and tested for predicting the PV system performance. Both the models could estimate the system performance with reasonably high level of accuracy.