Bi-objective robust optimization for reliability-oriented power network planning by considering distributed generation effects: A case study in Iran

Abstract

In the present study, a new mixed-integer linear programming is presented for a reliability-oriented power network. The reliability approach is conducted using the Average Energy Not Supplied (AENS) index and System Average Interruption Duration Index (SAIDI) based on the feeder’s failure rate. In this formulation, the constraints include standard ranges for electric power and current magnitude, acceptable voltage drop, power law, and power balance considering loss power. Moreover, the relation between feeder failure rate and electric power level and subsequently SAIDI for outage duration control is added in constraints to improve network reliability. The model is studied on a real case study in Iran in two scenarios with or without considering DG. The obtained results indicate that optimization for DG location and its capacity in the power system decrease AENS cost. Furthermore, DG incorporation in the grid reduces load congestion in substations and system’ total cost. Augmented Epsilon Constraint (AEC) method is utilized to solve two-objective mixed-integer linear programming. To overcome the uncertain environment, the robust optimization is applied, and the effect of Γ parameters is investigated on Pareto sets and system’s total costs. As confirmed by numerical results, by increasing demand constraint conservative level, the objective function gets away from the nominal value, requiring that Decision-Makers create an appropriate balance between conservative levels and system costs. The obtained results reveal the efficiency and effectiveness of the proposed mathematical model in the real world. • New reliability-oriented multi stage power network planning. • Determination the feeders failure rates via power storage and AENS cost. • Considering uncertainty in demand forecasting and facility installation costs. • DG incorporation as a facility to costs reduction and reliability reinforcement.