Numerical experiments with an optimal power flow algorithm based on parametric techniques

This paper presents the results of numerical experiments with a new optimal power flow (OPF) algorithm based on a parametric technique, The approach c onsists of relaxing the original OFF problem by incorporating parametric te rms to the objective function, the equality and inequality constraints, Suc h relaxation assures that any arbitrary initial solution, feasible or unfea sible, be the optimal solution of the OFF problem. As the scalar parameter changes, a family of OFF problems is created, whose necessary conditions ar e solved by Newton's method, An efficient strategy is proposed for updating the parameter and the optimal set of active inequality constraints of each intermediate problem. Two applications of the methodology are reported: th e economic dispatch problem and the minimum transmission loss problem. Thes e problems were solved for an 810-bus and a 2256-bus equivalent network of the South/Southeast interconnected Brazilian power system, The results show that the parametric approach is robust and efficient when applied to large -scale OFF problems.