A metaheuristic evolutionary method for electrical performance estimation of distribution networks

This paper proposes a new probabilistic method to evaluate power distribution electrical performance. Electrical performance is assumed to be the evaluation of network congestion parameters, losses and voltage level. The proposed method is suitable for off-line applications and it considers steady state performance only. The electrical performance estimation is formulated as a multiobjective optimization problem where the objective functions correspond to an evaluation of occurrence probability, and also correspond to a proximity evaluation of calculated voltage levels with values obtained by measurement (goal programming). Load profile values are discretized according to its occurrence probabilities (universal discrete probability density function), so that formulation results in a multiobjective combinatorial optimization. The optimization formulation considers feeder output power and voltage level measurement in any point of the network as restriction parameters. Network reduction procedures to substantially reduce decision domain and network expansion procedures to rebuild it are proposed. Specific heuristics applicable to radial networks - were also developed to generate a-priori feasible candidate solutions combining unbalanced and diversification load values. A metaheuristic evolutionary method is proposed and applied in order to improve feasible solutions and adequately apply the heuristics. feasible solutions are classified according to dominance ranking algorithm, so that electrical performance is obtained for each dominance rank frontier.