Determination of Loss of Load Probability for Stand-Alone Photovoltaic Power System

Abstract

In this paper, the loss of load probability for stand-alone photovoltaic (SAPV) power system was determined for an ICT Center with total daily energy demand of 346480 Wh/day. However, the different electrical appliances are classified into four (4) different load priority levels depending on the acceptable loss of load probability of the appliance in the data center. The ICT Center has annual averaged daily solar radiation of 4.7kWh/m 2 . day and minimum (worst case) daily solar radiation of 0.574 kWh/m 2 day which occurred on 17 th of June. The SAPV system is expected to satisfy with zero loss of load probability the critical load (server, switches, routers, Vsat) estimated at about 81210 Wh/day. In this wise, dynamic load shading approach can be employed to switch off certain loads based on their priority level and available solar irradiation. A cubic regression model is derived to enable the load scheduler to determine the possible LLOP for any give load level. The approach presented in this paper provides the relevant mechanism to determine at what point the dynamic load shading unit can turn off or turn on appliances in any load priority level in response to the temporal variation in solar radiation at the Data Center.