Today, I’m going to add in a little real world complexity. My last post discussed two different methods for implementing fleet EV charge control to limit the kW load from charging multiple Fleet Electric Vehicles – either simultaneous or sequential charging. I hope that after reading that post you have a high-level understanding of the how the 2 most common peak load management schemes could be implemented. However, that post contained significant omissions that have a tremendous impact on the effectiveness of any EV energy cost savings available. In today’s post, I’ll correct one of those omissions. Perhaps a real-world example will help to explain:
Several years ago, I received a call from a CA utility about the ability of one of our systems to participate in an innovative demand response (“DR”) program. (Look for an upcoming post about DR and EVs.) EvAuto, our EV control system, includes a connection to that utility’s DRAS server so I confidently answered that we could participate. I “listened” to him and prepared how I’d answer his question about how much load we could shed during an afternoon peak period. Instead, he threw a curve ball by asking if we could increase total load from 10 am until 1 pm if notified to do so. I was surprised and thought that I’d misunderstood his request, but he then explained how they anticipated an excess of solar and wind energy available at those times of day.
Sadly, I had to tell him that we couldn’t reliably participate. Why? Well, the fleet was normally on the road during those times.
That’s when I developed my 3rd maxim: “You can’t charge what’s not there.” Like all maxims it’s obvious and simple which is what allows it to be easily applied.
Let’s apply this maxim to our hypothetical two vehicle fleet. As you may recall, our focus was to reduce the peak load from charging, and we examined the impacts of PLM on vehicle availability and the people that operate them. Now let’s introduce the absence of the vehicles to our analysis by blocking out time for the vehicle operations.
I’ve blocked out approximately 9.5 hours for the vehicles to complete their route, return to the facility and be connected. Looking at the above figure, you can see that due to the time overlap, our PLM strategy must change, or we’ll leave our vehicles partially charged. As a start we can move charging so that we don’t start charging until the vehicles are on site.
Unfortunately, for both PLM methods there’s either one vehicle or both vehicles that aren’t fully charged. There’s a couple of options available to make sure this doesn’t happen, but unfortunately both result in setting the higher peak that we had hoped to avoid.
While simple, the above example brings forth real world questions because in both cases, we can see that the charge is nearly completed, and (as described in a previous post) it’s likely that the vehicle received enough charge to complete its route. In a real-world, complex environment other factors may actually help to solve this issue. Here we assume that all vehicles require a full charge and arrive / depart at the same time. Most fleets include many more vehicles, and in my experience some vehicles return early and need less charge. The trick is how to take advantage of these variances to successfully lower charging cost.
This post is part of a series explaining how fleet electric vehicle operators can save money charging their EVs. I’m breaking the concepts down into smaller pieces to introduce them to both new and experienced energy managers. For some, these concepts will seem overly simplistic, but I hope to offer easy to understand pieces to help all readers. In other posts I discuss different methods of EV Peak Load Management (“PLM”), operator specific needs and wants, impact of battery chemistry, Demand Response and how Control Dynamix EvAuto can control the cost of EV charging.