Distribution networks reinforcement planning: a dynamic multi-criteria approach

In this paper, the problem of distribution network reinforcement planning is formulated as a multi-criteria dynamic optimisation and decision-making task. To deal with dynamic nature of the planning problem, the paper suggests an algorithm, which combines the genetic algorithm and dynamic programming in one iterative procedure. The algorithm reduces the drawbacks of both algorithms and results in faster and more stable convergence. In order to find the best trade-off among the accepted criteria, the two-phase preference-order optimisation procedure is used. Still, it remains the task of the planner to define the preferred planning direction.