Optimized coordination of Vehicle to the Microgrid (V2MG) and logistic services of medium and heavy-duty electric vehicles (MHDEV) in military bases
Abstract
Project AbstractApproved for Public ReleaseProposal Title: Optimized coordination of Vehicle to the Microgrid (V2MG) and logistic services of medium and heavy-duty electric vehicles (MHDEV) in military bases.Electrification of transportation is imminent, and several manufacturers of medium and heavy-duty electric vehicles (MHDEV) are promising to put a variety of electric trucks models on themarket. The adoption of military MHDEV fleets in Navy bases creates an opportunity to co-optimize their use for transportation and for providing Vehicle to the Microgrid (V2MG) services. Since this transition will be coupling the base transportation demand with conventional electric power demand, the decisions on how to operate both logistics and energy delivery need to be coordinated; because of the shear complexity of operating both systems, state of the art tools that can learn optimum policies can advise the military personnel and make a significant difference, if not completely automate their management. An added benefit of this solution comparedto relying solely on a fixed electric storage installation is the reconfigurability needed for highly dynamic bases# operations.This project goal is to provide software that will manage optimally the V2MG services to provide simultaneously the best logistic service possible to meet the needs of the military base. The solution will be evaluated through realistic simulations and with real experiments on the NAVFAC EXWC Microgrid Testbed (MGTB). The research activity will include: 1) the design of a digital twin of the military microgrid, the acquisition and curation of realistic data of electric load and truck use, using as case study the military base of NAVFAC EXWC San Nicolas Island, Naval Base Ventura County, California and of the MGTB available for the experiments; 2) the definition of the risks and costs trade-offs between meeting travel needs and providing V2MG services; 3) the formulation of a Markov Decision Process, to besolved using a reinforcement learning approach, to schedule travel and V2MG services, capturing the stochasticity of power generation from renewables, the random electric load and that includes contingency planning; 4) the evaluation through simulations and emulation via the on the NAVFAC EXWC MGTB. The technical idea is to define routes as loops that start and end at the base, with deadline constraints on the stops and charging and discharging (for V2MG) as set of virtual transit routes each corresponding to a possible (discrete) choice of final state of charge and corresponding duration. This approach allows to formulate the co-optimization as an equivalent routing problem.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 15, 2023
- Source ID
- N000142312349
Entities
People
- Anna Scaglione
Organizations
- Cornell University
- Office of Naval Research
- United States Navy