Strategic Organizing of Distributed Generation Resources to Enhance Reliability and Resilience of the Distribution System

Two key research topics that aim to safeguard the system against unforeseen malfunctions or disasters and lessen their effects by reducing the resulting outages are distribution system reliability and resilience. There is a gap in the simultaneous optimization of power system resilience and reliability, particularly in distribution networks, even though are many researches devoted to their assessment and enhancement. This study proposes a new optimization paradigm for distributed generation (DG) placement-based reliability and resilience evaluation and improvement in distribution networks. Using the network's integrated remote-control switches, an optimum service restoration approach and optimal DG unit allocation are employed in this stochastic multi-objective optimization model. The methodology keeps DG investment costs low while minimizing distribution network outage costs brought on by resilience events and reliability contingencies. A mixed-integer linear programming (MILP) model that complies with network technical restrictions is used to describe the optimal service restoration issue. Two distinct scenario sets are created to represent the unpredictable nature of fault situations. Reliability and resilience scenarios are based on historical data of the network's fault rates and the failure probability functions of network components derived from Monte Carlo Simulation (MCS), respectively. A Pareto-optimal solution pool is obtained by solving the model using the non-dominated sorting genetic algorithm (NSGA-II) technique. To help the network planners choose the best option from the Pareto front, a fuzzy decision-making logic tool is then used. The suggested model is evaluated on an IEEE ۳۳-bus system, and the simulation results demonstrate the model's efficacy.

کلیدواژه ها:

Reliability ، Resilience ، Service Restoration ، Multi-Objective Optimization ، Monte Carlo simulation ، non-dominated sorting genetic algorithm