A Properly Ordered Analysis Can Save Money:
The structure of a comprehensive fleet electrification assessment is important to ensure that charging infrastructure needs are not overestimated, leading to unnecessarily high implementation costs. Beginning with a vehicle suitability analysis, followed by charging load forecasting, and then concluding with charging infrastructure need identification and the impacts on existing electrical infrastructure will help to avoid any missteps.

Fleet Electrification Assessment Approach

The analysis phase of fleet electrification begins with vehicle analysis and ends with infrastructure analysis; the implementation phase begins with infrastructure installation and ends with vehicle purchasing.

Data Management is Foundational:
Having robust data—specifically telematic data—on existing vehicles will give organizations the information they need to execute accurate vehicle suitability analyses. Having an accurate vehicle suitability analysis and detailed understanding of which EV models are a good fit for an organization’s fleet is the foundation necessary to complete an EVI needs assessment and all other analysis.

Data collection—particularly related to vehicle mileage and fuel usage on vehicles without telematics—was identified by the consultant team as the largest challenge faced during the City of Fremont’s project. The effort required to collect, clean and organize data into a usable format caused significant project delays and risked budget overrun. It is recommended that other agencies considering a similar project spend time cleaning and organizing data prior to engaging a consultant or consider increasing their budget for the consultant to take on this task more fully. The Fleet Electrification Planning Tool contains a data template sheet to assist other organizations in identifying which data fields on most important to help them efficiently complete a fleet electrification analysis. Additionally, if organizations are considering a fleet electrification project in the future, investing in a robust vehicle telematics system now will improve the speed and robustness of a future analysis.

Finally, continuing to track robust data on new EVs can enable accurate fuel cost tracking. As the penetration of electric vehicles in a fleet increases, it may become difficult to track fuel costs across departments if electricity bills are paid from different budgets by different departments. It may also prove difficult if EVI is attached directly to a building electrical meter rather than on a separate meter. Being able to track charging at the vehicle level enables organizations to validate the fuel savings expected to result from electrification.

Public Fleets Are Not Likely to be Charging Constrained:
With the exception of mission critical vehicles like police or fire department vehicles, most public agency fleets have ample time to charge in comparison to their daily energy needs. On average, standard vehicles in Fremont’s fleet were estimated to need 4-10 kWh (approximately 10 – 40 miles per day) of electricity each day. When compared to a standard dwell-time of 12+ hours and the size of many EV batteries (often in excess of 50 kWh or 200 – 350 miles), it is evident that Level 2 (6.6 kW) charging is sufficient in most cases and that vehicles will not have to charge every night. The ability to charge vehicles every couple of days enables organizations to increase their vehicle to charging port ration, reduce their infrastructure needs, saving costs.

Graphic illustrates the percent of vehicles parked and corresponding dwell times over an average week at the City of Fremont’s Maintenance Center

Managed Charging Results in the Most Significant Cost Savings:
Across all charging scenarios that were modeled, managed charging reduced the cost of charging. In context of this study, managed charging referred to the practice of adjusting an EV charging profile (i.e. moving the time when a vehicle is charging) to optimize for cost and charge when electricity is cheaper, or to reduce coincident peak load. This is usually achieved in real-time with software that controls the amount of electricity being dispensed by a charging station. Unmanaged charging is the practice of allowing EV charging profiles to match natural driver/vehicle behavior, such that charging begins at the time of plug in and ends at the driver plug out time, or when the vehicle battery is fully charged (whichever occurs first). Managed charging provides the most benefit in a charging scenario that does not include solar or storage.

Graphic compares the expected cost of charging across managed and unmanaged charging scenarios for two different utility PG&E rates (B-10 & B-EV)at the City of Fremont’s Development Center. The analysis assumes an annual utility rate escalation of 3%.

Due to the long dwell times of vehicles in public fleets, there is ample opportunity to move charging into the night and early morning and avoid the most expensive electricity. This is particularly important with the recent shift in TOU peak periods to 4-9pm in California. Based on the analysis of vehicle duty cycles, much of the charging occurring in an unmanaged scenario would occur during these peak periods, increasing the cost of charging and reducing the potential total cost of ownership savings from EVs.

Graphic compares the projected vehicle charging demand on an average day under an umanaged charging strategy and a managed one.

Managed Charging is At Odds with Real-Time Carbon Emissions:
Real-time carbon emissions refer to the carbon emissions directly associated with the exact electrons being used to charge an electric vehicle. The carbon intensity of electricity used by the City of Fremont on a minute-by-minute basis differs from the overall carbon intensity of the electricity product that the City receives from EBCE. For the purposes of carbon accounting on an annual basis, the electricity product provided by EBCE to the City has a carbon intensity of zero. While managed charging reduces electricity costs, shifting charging to the overnight hours can increase real-time carbon emissions because the carbon intensity of the electrical grid in California is higher at night For example, at the City of Fremont’s Development Center managed charging resulted in a 20% increase in real-time carbon emissions compared to an unmanaged scenario.

PG&E’s B-EV Rate Can Be a Cost-Effective Option:
Without the presence of existing solar, the electric vehicle specific rate (B-EV) from PG&E may be the most cost-effective rate on which to charge electric vehicles. Based on a utility escalation rate of 3%, the EV specific rate resulted in a lower levelized cost of charging (LCOC) for the City of Fremont than adding new solar and putting EV charging on the same meter as the new solar.

To take advantage of the B-EV rate, an organization will need to install EV charging stations wired to their own dedicated utility meter. While this scenario offers the lowest possible cost for EV charging for facilities that do not already have existing solar on favorable NEM tariffs (as discussed above), it is important to note that it would likely result in more significant real-time carbon emissions. The actual cost savings realized by PG&E’s B-EV rate over time will also be highly dependent on the escalation of PG&E’s electricity rates in the long-term, with a larger utility rate escalation resulting in diminished savings potential.

This study focused specifically on the SF Bay Area and PG&E’s rates. However, utilities around the country are exploring EV-specific rates to provide similar savings to charging costs and manage stress on the grid.

The Added Financial Value of Energy Storage is Situational:
Use of stationary battery storage as a charging cost mitigation strategy is very much site dependent. When considering the cost of EV charging, the primary goal (beyond shifting charging to off-peak periods) is minimizing peak demand charges. If an organization cannot do this by selecting a site with no demand charges (such as a site with existing solar on an A-6 rate), then battery storage could prove to be a useful strategy to mitigate excessive demand charges. However, on PG&E’s B-10 rate, storage was not demonstrated to be effective at reducing overall EV charging costs. An analysis of solar + storage systems on Fremont’s facilities showed that the vast majority of savings was from the solar itself, and that any marginal savings associated with storage were not enough to justify the additional upfront cost.

That said, there are potential co-benefits of storage such as resilience (of both the facility and the fleet) or reduction of coincident carbon emissions that an organization should consider. The analysis done in this project primarily addressed storage through lens of charging cost optimization. As Fremont considers various backup power options to support both building operations and EV fleet charging in the event of a long-term grid outage, then solar + storage may become a significant strategy. With this in mind, storage for resiliency benefits should be evaluated in any EV Fleet CIP project planning process.

Public Fleets Are Great Uses Cases for Vehicle Grid Integration (VGI):
VGI is a broad term that refers to a variety of ways that electric vehicles can be used to support the electrical grid by providing managed charging or grid services. VGI can be further broken down into four categories:

• V1G – unidirectional, managed charging like the simple managed charging scenario modeled for Fremont’s vehicles in this study
• V2B/V2H – vehicle to building or vehicle to home; when a vehicle is used to power a building or home and manage electricity costs or provide resilience
• V2L – vehicle to load; when a vehicle is used to power other loads, or recharge other vehicles
• V2G – bidirectional charging/discharging for grid services

V1G is commonly deployed in fleet charging scenarios to minimize charging costs and balance loads. V2B and V2L are also possible today and are currently being deployed (See Working Group Meeting 4). Widespread application of V2G still faces challenges related to utility interconnection standards and vehicle OEM warranties that discourage discharging a vehicle’s battery.

Regardless of the exact VGI type, the fact that public fleets tend to drive a small number of miles each day and spend ~12 (or more) hours parked each day, (see “Public Fleets Are Not Likely to be Charging Constrained”) make these vehicles great use cases for VGI. This analysis showed that, by 2030, the City of Fremont will have an estimated 15 MWh of battery storage located in vehicles across its four priority sites. For reference, the average daily electrical load of all four priority facilities combined is ~3.3 MWh. As VGI applications mature, electric vehicles in public fleets may have significant resilience value by providing flexible back-up power to public facilities during multi-hour, or even multi-day, outages.