The abatement potential of EVs over ICEVs and HEVs, when modeled using high resolution grid emissions rates, is highly nuanced and time dependent. Investigating this nuance and the barriers and opportunities it creates pertaining to commercial EVs is a main contribution of this research effort. Grid emissions attributable to EV charging generated by our simulations are compared to tailpipe emissions from conventional and hybridized gasoline and diesel vehicles. Fuel consumption was calculated using the combined fuel economy of each vehicle type and vehicle miles traveled. CO2 emissions were calculated using the CO2 emitted from burning one gallon of gasoline and diesel .For this Phase II effort, the team leveraged the proven power of the system-of-systems model developed and described in detail in the Phase I report to explore emissions resulting from a series of distinct scenarios for charging events of electrified commercial vehicles. The model was produced in MATLAB/Simulink and enables the integration of three sub-system models enabling in turn comprehensive, quantitative simulations of EV deployment for multiple driving cases under varying charging schedules and grid emissions assumptions. The architecture of the Simulink model is displayed in Figure 2-6. The initialization code, written in MATLAB, may be found in the Appendix. The current version of the model does not automatically account for charging losses occurring during the transfer of energy from the charging infrastructure to the electric vehicle’s battery system. These losses were corrected in a post-processing step wherein each total cumulative emissions output from the simulated charging events was multiplied by 1/η. Following the findings by Channegowda et al and assumptions employed by our Phase I report, cannabis grow setup we selected η=0.88 for Level 2 and η=0.90 for Level 3 charging systems. Simulation results are presented in the following sections.
These, as with the team’s Phase I results, represent what is believed to be an innovative method leveraging new data that yields a distribution of projections for CO2 emissions across a variety of assumptions and plausible scenarios.Table 3-3 depicts the cumulative energy consumption in kWh for each vehicle type included in the study at different levels of daily VMT. These values are a primary input for the simulations, informing the model as to what quantity of energy must be replaced during a charging event to return the battery to its full state of charge. To ensure consistency of comparisons, simulation outputs were only collected for scenarios where the battery was fully recharged during the charging event. Figure 3-1, Figure 3-2, and Figure 3-3 serve as comparative visualizations of simulation results for light-duty truck emissions rates in kilograms of CO2 per kilometer. Each plot depicts relative emissions rates for a battery-electric truck under different emissions profiles and level 2 charging schedule assumptions for each VMT level at different times of the year. Emissions rates are plotted along with emissions from an ICEV baseline as well as an HEV to enable comparisons. The plots represent additional corroborating evidence in favor of several pertinent takeaways from our Phase I study. The daily variance of electrical power grid emissions rates resulting from the switching on and off of dispatchable marginal generation resources by the grid operator in anticipation of or in response to evolving demand for power is high, especially during summer and winter months when temperature fluctuations are expected to be more extreme. Using monthly or annual average emissions rates fails to account for the nuanced fluctuations of real-world grid emissions rates , implying the true environmental benefits of EVs compared to ICEVs or HEVs are often misrepresented.
Are environmental benefits then over- or understated? The directionality is not uniform. Vehicle use, charging patterns, and seasonal fluctuation of grid characteristics play important roles in determining actual environmental outcomes, positive or negative.The above figures reveal that in these cases at least, employing hourly grid emissions rate data to guide inquiry into environmental benefits of EVs, while still at a higher resolution than annual or monthly average rates, minimizes the variation of marginal resources. As was concluded in the Phase I report, EVs are likely to require marginal resources en masse because they act together to force demand projections out of the expected regime, especially at growing rates of adoption. This Phase I conclusion is presented with the caveat, however, as Phase I dealt with personal EVs as opposed to the commercial EVs we are concerned with here. Still, this conclusion may well hold for commercial EVs, especially concerning the commercial applications explored in this phase of our research. The nature of many residential services, moving companies, and refuse operations is to operate during daytime hours , leaving the largest window of logical charging times for the evening or overnight hours with the occasional exception. If most commercial vehicle activity occurs during the day, then most commercial EV charging will occur overnight. During a transition period of EV adoption growth, utilities will be able to accommodate the associated generation demand to support commercial EVs, but these needs will most likely be met by marginal resources. This has implications when considering the environmental benefits of commercial EVs because marginal resource emissions rates possess the highest temporal variations. When using marginal grid emissions rate assumptions, the per-kilometer emissions of a battery-electric light truck traveling 20 miles per day were found to vary by as much as 131% depending on the charging schedule used .
Table 3-4 reports the percentage improvement of CO2 emissions for an EV relative to an ICEV in the light truck category, assuming 20 miles per day of business use, under various charging profiles. Note that studies that assume an average CO2 emissions rate may predict a very different environmental impact than studies that take certain marginal dispatch factors into account. Simple shifts in charging times that may have little impact on existing business operations may have sizeable impacts on emissions attributable to charging events, indicating substantial value for managed charging implementation.While the Phase I study explored mostly passenger cars, as presented above, this study built upon it to explore light trucks in service fleet operations. Next, we turn to Medium Duty EV use cases, which appear to hold much near-term promise, and may be financially attractive without significant subsidies. Presented in Figures 3-4, 3-5 and 3-6 are simulations of CO2 emissions for a Moving Truck application of various vehicle miles traveled during three principal seasons . While in some cases the variation in emissions rates for the different scenarios can be quite high, it is important to understand the cause and implications. For the scenarios with more VMT, there is a proportionally greater quantity of energy that needs to be replaced during the charging event. This can have the result of moderating the variance in emissions compared to shorter VMT scenarios. We infer this is because the emissions intensity of the electricity being used to charge the EV’s battery pack is diluted the longer it is being actively recharged. At smaller levels of energy displacement, the vehicle’s battery can be recharged entirely within one or several hours of charging. If this charging occurs within peak hours or whenever the carbon intensity of the grid and its marginal resources are highest, then the subsequent carbon intensity of the EV is similarly high. During longer charging events replenishing greater amounts of energy to the vehicle’s battery, the charging event is likely to last beyond periods of peaking grid carbon intensity resulting in the deposit of less carbon-intensive electrical power to the battery for at least some of the charging event. This mechanism holds implications pertinent to commercial EVs, whose larger battery capacities, greater rates of energy consumption, and generally higher VMT necessitate longer charging events. Commercial EVs may have more opportunities than private light-duty EVs to reduce the carbon intensity of their operations by charging across periods of high and low grid emissions intensities alike. Still, pipp mobile systems the demands of business may predicate faster charging times to keep vehicles on the road and minimize range and scheduling anxieties.
Growing penetrations of Level 3 systems would reduce charging times, raising carbon intensities absent charging management programs. The MD moving truck example for the various VMT cases, EV charging profiles and marginal emissions assumptions demonstrates that more research may be valuable to quantify with greater certainty how each of these factors can influence fleet emissions. Now is a particularly important time to evaluate these factors, as many MD EV cases are in a growth stage based on market and policy considerations. The foregoing is also contingent on trends in emissions for grid generation. While it is beyond the scope of this study to forecast how grid generation will evolve to meet new demands, it is important to co-develop tools that can simulate the associated CO2 impacts of different EV and grid growth scenarios. As was the case with the personal vehicles in Phase I, there exists a future threshold of commercial EV penetration that will trigger a realignment of the existing trends observed in grid emissions. Commercial EVs demand more power on average than private vehicles and therefore will have greater effects on marginal emissions trends. Increased power demand at common charging times will give rise to additional marginal resources, which are likely to be fossil in nature, at least in the near-term. The range of results depicted in the various scenarios across a range of vehicles suggests that all key stakeholders may ultimately benefit frombetter foresight to charging events on both short term and longer-term time scales. Thus, the ability to manage charging events becomes an essential part of the suite of regulatory levers and cyber-physical infrastructure that can facilitate effective growth of commercial EV deployments. Efforts to optimize the benefits of managed charging seem to depend upon the effective acquisition, analysis, and conveyance of high-resolution information concerning EV charging and its various externalities between system users, grid operators, and policymakers. This research, by way of integrated systems modeling and use case simulation, argues that using averaged emissions rates, especially at lower temporal resolutions, may obscure information. By refining the temporal resolution of the input data for both charging events and grid generation profiles, stakeholders can be better equipped to optimize environmental benefits from EV growth, as well as inform business decisions, and guide policymaking.In the near- and intermediate-terms, coordinated EV charging is a critical component of an effective transition to battery-electric vehicles for a growing range of use cases. The prevailing generation dispatch mechanisms and physical constraints of the electrical power grid are inherently sensitive to shifting demand trends and require consideration of hourly factors. These sensitivities are expected to heighten as electrification increases from a multitude of sources, including not only EVs, but other intermittent, seasonal, and continuous loads . Anticipating the timing and magnitude of demand spikes, such as those resulting from EVs, is necessary from multiple time domains. First, as we have stated, an hourly or even sub-hourly view of charging and grid dispatch may be important to ensure emissions benefits are understood. And second, from a resource planning perspective, grid operators seek to optimize dispatch and infrastructure funding decision-making processesover a much longer time scale. Thankfully, a number of support policies and Federal funding resources are stimulating new R&D, pilots, and demonstrations at the intersection of grid modernization, charging communication protocol, smart charge management, EVs, and the increasingly distributed grid in order to mitigate unintended demand peaks resulting from both higher current charging and higher penetration of EVs.In our simulations of CO2 output, hybrids performed on par with many EVs under marginal grid emissions scenarios. The recent EV policy environment included monetary incentives like tax credits to offset the relatively high purchase prices of EVs over their ICEV counterparts. This research may suggest that hybrids can provide some important benefits in parallel with EV growth, including reduced environmental impacts, and opportunities to reduce infrastructure investments by leveraging existing assets. It is also quite clear that a diverse fleet mix may provide certain strategic advantages, and that “winner take-all” policy architecture may be unwise. This research therefore emphasizes the need to conduct comparative studies as policy options are explored. Doing so can ensure that public dollars are more effective in reducin the imacts of transortation while maintain in the current expectations in performance. This can also help inform public and private investment decisions and accelerate paths to scale and impact.