A Two-Stage Approach to Locate and Size PV Sources in Distribution Networks for Annual Grid Operative Costs Minimization

This paper contributes with a new two-stage optimization methodology to solve the problem of optimal placement and sizing solar photovoltaic (PV) generation units in medium-voltage distribution networks. The is formulated mixed-integer nonlinear programming (MINLP) model, where it combines binary variables regarding nodes PV generators will be located continuous associated power flow solution. To MINLP model decoupled used firstly solved quadratic approximation; once sources are known, dimensioning secondly through an interior point method applied classical multi-period formulation. Numerical results IEEE 33-bus 85-bus systems demonstrate that proposed approach improves current literature reached combinatorial methods such as Chu Beasley genetic algorithm, vortex search Newton-metaheuristic algorithm well exact solution GAMS software BONMIN solver. All numerical simulations implemented MATLAB environment convex equivalent models CVX tool.