The calculation of the maximum photovoltaic (PV) hosting capacity plays a crucial role in guiding power grid planning and operation. A method for calculating the maximum access capacity of PV systems based on an analytically optimized bisection approach is investigated. The Latin hypercube sampling algorithm is employed to randomly sample power data. Using these samples, the initial maximum allowable capacity is determined analytically by considering both the occurrence of power flow reversal and the requirement that the voltage at each node remains within permissible limits. This initial maximum hosting capacity is then used to define the search range for the bisection method. Subsequently, the analytically optimized bisection method is applied to compute the final time-varying maximum hosting capacity for the tracking network. Experimental results show that the area under the curve of this method is close to 1, indicating high accuracy in real-time calculation of the maximum allowable capacity for grid-connected PV systems. The deviation between the calculated and actual values is small, with a maximum error of 1 MW and an average error of 0.923 MW.
Yang et al. (Sun,) studied this question.