Geometric Programming for Optimal Hosting Capacity Allocation in Radial Grids
Sicheng Gong, Koen Kok, Sjef Cobben · 2024
Mixed load/generation profiles and non-convex optimization models pose significant computational challenges in hosting capacity allocation. On the use of constraint conversion and geometric programming, this paper presents an improved solving framework for optimal hosting capacity allocation, which advances in computational speed and solution quality. Specifically, by preliminarily calculating the feasible region in the combinational operational space of involved units, voltage and current variables from the original grid constraints can be omitted for model complexity reduction. Through subsequent model reformulation guided by geometric programming, the model convexity is retrieved to further accelerate the solving process and improve the solution quality. Relevant case studies reveal our method can significantly reduce computation time by up to 88.9% and improve the solution quality. These findings highlight the potential of this approach to enhance computation performance in capacity determination. We anticipate this work to open a new avenue for more comprehensive capacity allocation algorithms using geometric programming.