This paper presents a predictive probabilistic approach (PPA) for the optimal sizing of new distributed generation capacities in support of the main grid to respond to a fraction of the total load during the supply current interruption duration defined in using renewable-based microgrid assets. The model is used to simulate network failures by disabling the network for a certain time step. The load profile can be changed during network interruptions to represent a critical load. The generation capacity of the microgrid must be able to generate the value in grid-connected mode while supporting a critical load during a grid outage. The probability of loss of load characterizes the adequacy and energy not supplied represents the estimated reliability and the associated cost. The approach allows all technologies in the model to be evaluated, both during grid-connected mode and during grid outages when technologies can continue to power critical loads as part of the microgrid.
Keywords distribution, renewable energy resources, reliability, outage cost, reliability