Optimal planning of robust DSO grid

This master thesis proposes to model the robust planning of the DSO grid. The model includes the possibility for the DSO to invest in storage, reinforce its network and impose fuse limits which are limits on the reinjection and withdrawal of power in and from the network. The problem takes also into account the operational costs and determines the optimal power flow in the alternating current grid. A primal-dual algorithm is then used to find the optimal solution that is robust to uncertainties. The nonconvex optimal power flow equations are solved using an inexact second-order cone programming relaxation and a sequence of convex problems to recover the feasibility of the solution. Experiments are then performed on a real distribution network with large amount of renewable production and we demonstrate how our approach can help to reduce the curtailment of PV production.