Funding: DOD and CEC
Electrification of non-tactical vehicle fleets represents a key efficiency and energy security objective for the U.S. Department of Defense (DoD). To achieve electrification, the DoD has targeted vehicle-to-grid (V2G-Sim) services as a way to decrease the overall cost of operating the vehicle fleet and achieve rough parity with traditional internal combustion vehicle fleets. Among other planned demonstrations, a mixed purpose and duty 29-vehicle 100% plug-in electrical vehicle (PEV) pilot test fleet will be deployed at the Los Angeles Air Force base (LA AFB). The LA AFB fleet will provide a V2G service, frequency regulation, to the California Independent System Operator (CAISO) wholesale electricity market in order recoup some of the additional costs of procuring electric vehicles and their supporting infrastructure. Lawrence Berkeley National Laboratory (LBNL), with their partners Kisensum and LLC, have developed the fleet scheduling, optimization, and control software to enable the vehicle fleet at LA AFB to participate in CAISO’s ancillary services markets. This project is the first of its kind, attempting to provide financially binding market participation with bi-directional charging and discharging of an operational vehicle fleet to provide the most technologically demanding service that is procured in wholesale electricity markets. The final project goal is to analyze the potential to use these electric vehicles to support critical infrastructure on the base in the event of an emergency. The project is funded jointly between the California Energy Commission and the U.S. DoD’s Environmental Security Technology Certification Program (ESTCP). The vehicle fleet and charging infrastructure, largely procured by the DoD, consists of sedans, vans, pickup trucks, box trucks, and a bus. The vehicles are a mix of plug-in hybrid electric vehicles and pure battery electric vehicles capable of charging and discharging via both AC level 2 and DC fast charging interfaces. The charging infrastructure is a mix of AC and DC charging, in which the AC level 2 charging is limited to 15 kW and DC fast charging is between 15 and 50 kW. The technical approach to fleet participation in regulation markets includes gathering the travel requirements of the fleet through a fleet operations management system, developing schedules for jointly optimal electric vehicle charging and regulation bid capacities, communicating those bids and the resulting awards and dispatches to/from the CAISO using open standard communications, and disaggregating electricity dispatches in real-time to command individual vehicles to charge or discharge through an optimal hierarchical control framework. These systems are described below:
- Kisensum’s fleet operations management system, known as On-Base Electric Vehicle Infrastructure (OB-EVI), allows base personnel to reserve and check-out electric vehicles for mobility tasks. This system collects information on planned trip departure and return times, vehicle preferences, expected distance traveled, etc. It stores this along with actual trip information upon the vehicle’s return to provide the system with an expectation of vehicle energy needs and times in which the vehicles may be available to participate in grid services.
- Optimization capability based on Berkeley Lab’s Distributed Energy Resources Customer Adoption Model (DER-CAM) will be extended to deliver optimal scheduling for the fleet. DER-CAM is a mixed integer linear programming optimization that minimizes the cost of vehicle operations subject to the physical, travel, and market constraints inherent to the system and the CAISO context. The cost of vehicle operations includes both the cost of vehicle charging under the LA AFB’s retail tariff, as well as the potential revenue that could be generated through regulation market participation. The optimal schedules will be passed back to OB-EVI to be bid through the scheduling coordinator’s (Southern California Edison) bidding mechanism into CAISO markets. These schedules are also used by OB-EVI’s implementation of the optimal control algorithms developed for disaggregation of dispatch signals DER-CAM will additionally collect other necessary input data, such as weather forecasts and historical market prices, for optimal scheduling.
- Communications between OB-EVI and electric vehicle charging infrastructure utilize two standard data formats, Open Charge Point Protocol (OCPP) and Smart Energy Protocol 2.0 (SEP2).
- Lastly, the real-time charging control algorithm disaggregates CAISO dispatch signals into individual charging and discharging commands for the electric vehicles plugged-in at the base. This algorithm attempts to minimize the norm of the deviation from vehicular optimal energy schedules predetermined by DER-CAM as uncertain frequency regulation dispatches are received at four second intervals by CAISO.
A system, as demonstrated at LA AFB can provide a number of economic and environmental benefits. The overall security and environmental benefits of reducing fossil-fuel powered non-tactical military vehicle fleets are clear; however, wisely managing PEV fleets can minimize costs and create non-transportation benefits. Vehicle charging can be costly if not scheduled well relative to the prevailing utility tariff and other constraints, while the fast responding energy storage capability of vehicle batteries can provide valuable services to help satisfy building and local base energy requirements. Further, while vehicles individually are not large electricity loads or sources, when aggregated or when integrated with the buildings at which they are interconnected, they can become a controlled entity able to ameliorate the adverse effects of variability in renewable resources and loads, and provide demand response and ancillary service to the local utility and the wider power system around it. These power system services can provide additional revenue to offset the costs of electric vehicle ownership.