EV Prioritization and Power Allocation During Outages: A Lexicographic Method-Based Multiobjective Optimization Approach

The growth in number of electric vehicles (EVs) has resulted in increased dependence of transportation on the power sector. During power outages, especially for elongated time spans, the locally available energy may not be sufficient to fulfill the energy needs of all the EVs. Therefore, a multicriteria EV prioritization scheme is proposed in this study to fairly allocate available energy among EVs during power outages. The five major factors considered for EV prioritization are trip purpose, EV occupants, energy gap, departure time, and customer behavior. With different combinations of the five prioritization factors, three indices are formulated, one each for social welfare, community wellbeing, and individual satisfaction of the EV owners. To this end, a multiobjective optimization problem is formulated, based on the three indices, to allocate the available power among EVs with higher index values. The formulated multiobjective optimization problem is solved using the lexicographic method, which has superior performance over the conventionally used weighted-sum method and $\varepsilon $ -constraint (EC) method. The proposed method is not sensitive to the weights of individual functions and has the ability to handle multiple priority levels. In order to quantify the results, percent served and unserved indices are formulated for each of the three parameters (social welfare, community wellbeing, and individual satisfaction) and the results of the proposed method are compared with those of the weighted-sum method and the EC method. Sensitivity analysis of different uncertain factors, such as number of EVs, uncertainty in EV demand, uncertainty in renewable power, and error in battery state-of-charge estimation, is also carried out. Simulation results have shown the superiority of the proposed method in allocating power to EVs during outages.