Probabilistic optimal robust multistage feeder routing under load forecasting uncertainty

One of the main uncertainties in long-term distribution network planning is load forecasting error. This paper presents a probabilistic multistage expansion planning method to consider the load forecasting error as well as pseudo dynamic behavior of network parameters and geographical constraints. The optimal routes of MV feeders as backbone of distribution networks are obtained for both mid and long-term case with deterministic and probabilistic modeling. The Imperialist Competitive Algorithm (ICA) is adapted for the optimal expansion planning of MV distribution network. To check and maintain the radial configuration of the MV distribution network during ICA random operators, a novel algorithm is employed. Comparative tabular and graphical results are given for deterministic and probabilistic planning. In probabilistic case the results is also represented by histogram of statistics data, mean and standard deviation of input and output variables using Monte Carlo Simulation (MCS). Based on the results it is possible to design the MV network at the presence of uncertainties more robust and flexible with respect to deterministic planning without reasonable increase in total cost. The efficiency and capability of the method is tested on a relatively large-scale distribution network.