Many-objective interactive optimization and decision making for distribution network expansion planning

摘要

This paper establishes a mixed integer linear programming model with many-objective for distribution network expansion planning (DNEP) problem, in which the distributed generation is considered and the original expressions of three commonly used reliability indices are improved into a form of anticipation functions to participate in the optimization process along with economic indices. An interactive optimization and decision making method that named preference-based e-constraint method is developed to solve the DNEP with many-objective. To reduce the dimensionality of problem, only one of the many-objective functions is optimized and the remaining objective functions are transformed into equality constraints with slack variables. To save on the computing cost, the search direction is guided along the preference of decision makers and the process of optimization and decision making can be simultaneously carried out. Simulation studies are conducted on a fictitious distribution system consisting of 24 load nodes, 3 substation nodes and 39 branches to verify the effectiveness of the improved model and the preference-based e-constraint method.