In this paper, a new mathematical formulation for the electric vehicle routing problem (EVRP) is proposed. This formulation extends the Green Vehicle Routing Problem (GVRP) considering time-of-use energy prices, and including a detailed model for the EVs' energy consumption. The main decisions for the considered EVRP are relevant to the choice among different types of charging modes at recharging stations, the speed of EVs, the loaded cargo and the battery charge. The model objective consists of minimizing the cost for the total travel distance and that for energy purchase, which depends on the selected recharging mode. A preprocessing algorithm used to reduce the problem dimension is presented. The experimental analysis performed on a large set of benchmark instances is reported.

This paper presents an energy management platform based on a receding-horizon scheme for the optimal control of active and reactive power flows in microgrids including small-size PV, CHP, wind generation, mini-hydro and energy storage. The objective function to be minimized can be set both on the daily operational costs (economic indicator) and on the global CO2 emissions of the system (environmental indicator), while decision variables are the production schedules of the generators and the power flows across the grid. The tool includes the technical constraints characterizing LV/MV microgrids and give the user the possibility to select different models for the electrical network (nonlinear power flow equations, linear approximation and single bus-bar) and different optimization ranges. The EMS has been validated through an experimental campaign on the Smart Polygeneration Microgrid (SPM) of the University of Genoa, which provides electricity and thermal energy to the Savona Campus, an “open-air” demo-site of an environmentally sustainable urban district with a population of about 2.200 people. The results of the tests on the field highlight the robustness of the developed platform and the capability of the receding-horizon algorithm at the core of it of successfully treating data uncertainties.

In this article, we propose a new bi-level approach for the location and sizing of charging stations (CSs) based on the integration of power distribution and traffic networks. At the lower level, the user choices and the energy demands are modeled through the user equilibrium (UE) traffic assignment conditions that are derived for the case of electric vehicles (EVs). At the higher level, an optimization problem is formalized for the optimal planning of locations, sizes, and unit prices of a set of new charging stations in a territory, which embeds the UE conditions and the power flow model as constraints. The main novelties lie in the integration of traffic and electric networks, the optimization problem formalization, and the inclusion of detailed electrical issues [such as the complete alternating current (AC) power flow equations and the modeling of CSs with different phases]. The developed model has been applied to a case study in the Genoa Municipality. Note to Practitioners —In the past decade, particular attention has been posed on EVs and renewable resources to reduce pollutants’ emissions. Critical issues regard the setting of adequate recharging infrastructure and the integration of huge distributed and intermittent loads and production into the electrical grid. This article aims to provide a bi-level optimization model that can be used by companies or public administrations that want to install charging stations in a territory. In particular, the transportation and the electrical networks are integrated: at a lower level, the UE traffic assignment conditions are derived to represent users’ choices and to assess the energy demand. At the upper level, an optimization problem is formalized, to minimize costs of installation and power losses and to maximize benefits, which embeds as constraints the UE conditions for the case of EVs and the model of the electrical grid (load flow equations, bounds over-voltage, apparent power, etc.).