Sensitivity of Hosting Capacity to Data Resolution and Uncertainty Modeling

Abstract

Integration limits of distributed generations (DGs) in distribution networks, i.e. the hosting capacity (HC), are highly dependent on uncertainties associated with the size, location and output power of DGs. Addressing these uncertainties to a great extent is reliant on the availability and resolution of the historical data. This paper investigates the effects of data resolution and uncertainty modeling on the HC calculation. To do so, a mathematical model of the HC problem is used in a Monte Carlo-based framework. Our analysis is carried out on an agricultural distribution network in Australia. It is shown that decreasing the resolution of historical data shifts the probability distribution function of the HC towards right implying an increase in the estimated HC. Further, it is illustrated that assuming a fixed capacity for DGs instead of proper modeling of the uncertainty associated with their size results in underestimation of the HC in the network.