We’re still very early-on in the EV adoption process. Early adopters can get a jump on the advantages of a new technology, but it often comes at the cost of needing to climb a learning curve not as many have climbed. Early EVs and earlier EV conversions were a real guessing game for travel, and most people only used them for local jaunts due to low ranges. Over time, not only have ranges increased, but things like trip planning software have made the job of planning for charging a lot easier.
But, as charging opportunities proliferate, trip planning software will likely shift from being about getting there at all to getting there for the lowest price. Just look at gasoline apps like GasBuddy if you’re not sure about this. Gas is available everywhere, and people don’t usually have to carefully plan trips, but getting the lowest price is still very useful.
Another big challenge for EV drivers is taking good care of the vehicle’s expensive and temperature sensitive battery pack. Early conversions and early EVs like the Nissan LEAF required a lot of attention to minimize degradation, with DIY projects sometimes even requiring care to avoid fire.
But, as EVs have gotten better and more of the work was shifted to automated battery management systems, EV drivers don’t have to think much about this. Liquid cooling and improved battery chemistries have even led to degradation becoming a shrinking concern.
But, it still pays to take a few precautions, even if you don’t strictly have to. For example, it’s generally a good idea to try to keep an EV’s state of charge between 20% and 80%, unless you have a LiFePO4 (LFP) battery, in which case a 100% charge is OK. But, the average new EV buyer doesn’t know anything about battery chemistries, so they might miss out on an opportunity to avoid some degradation. Or, with an e-bike, it’s a good idea to store batteries in a temperature-controlled environment instead of leaving them out in the sun or cold to bake or freeze.
But, what if there was a way to avoid all of the extra special care and knowledge of batteries, but still get the benefits? Or, what if there are even better ways to care for batteries that enthusiast early adopters don’t know about? Or even better, what if management done so well that humans literally couldn’t do better manually was a thing? A company that recently popped up on CleanTechnica’s radar is working to do just that, and for EVs of all sizes.
The history of EVs is instructive here. For example, the earliest Chevy Volt PHEV models reserved as much as half of the battery’s energy storage capacity for a buffer, because knowledge of battery care was limited at the time. They didn’t really know how much you could really use the battery cells to get better long-term longevity, so they just cut the vehicle’s electric range to almost half.
But, at the same time, the company says that relying on emerging research to take advantage of more capacity through smaller buffers is a bit of a dice roll. It could be that we don’t know everything about how the batteries will really perform in the long run under the broad variety of conditions we expect them to work in, and a nasty surprise could result for some drivers.
Instead of making battery management systems (BMS) a cheap thing that performs only basic functions, Elysium wants to take battery management to the next level, with better hardware and far better software that manages things a lot more closely. The idea is to get companies to spend more on BMS, but be able to reap rewards due to better battery efficiency. They’re saying that with better software that adapts and takes advantage of the latest research (possibly with updates), manufacturers can use smaller batteries to deliver the same range and usability. This would more than make up for the small amount extra a superior BMS solution would cost.
Another thing the company has made clear on the website is that you can’t just send AI in to do the job. Machine learning is very useful, but it’s still often a “black box” where we don’t understand the underlying processes happening in a complex multi-layer set of statistical operations. As long as the results are good, who cares what happens in the AI’s neural network, right?
This approach alone isn’t great for battery management, though. While AI can be a quick way to get software to fit the problem it’s solving, the AI’s solution to real-world problems is only as good as its training data. There’s no guarantee of future performance, only a repeat of the best past performance.
Instead, the company wants to continually work on finding the best balance between AI techniques and more hard-coded approaches to force software to integrate scientific knowledge of batteries. The idea is to get the best of both worlds, with the excesses of each approach tamed by the other. They call this the “grey box” approach.
Another important challenge Elysia aims to tackle is determining the value of used EVs. It’s possible for a private buyer or a dealer at auction to make a big mistake buying a seemingly good used EV only to find that the battery is in much worse shape than they thought. This could lead to paying more than a vehicle is really worth, or not seeing the “creampuff” used cars that can be had cheap despite great remaining value.
With ICE vehicles, you can determine this by looking at the miles on the drivetrain and the history of the vehicle (crashes, maintenance, etc). But, knowing what useful life a battery has isn’t nearly as easy.
Once again, the “grey box” approach, utilizing the best AI and the best industry and academic battery knowledge available can save the day and avoid big headaches. Elysia has a very lengthy article about its approaches to this challenge here.
We’ll be reaching out to the company to see what solutions it is bringing to the table in 2023, but be sure to check out its website to learn more if this sounds like an exciting company!
Featured image from Elysia’s website, showing the broad range of vehicles it aims to better support.
Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.
EV Obsession Daily!
Tesla Sales in 2023, 2024, and 2030
CleanTechnica uses affiliate links. See our policy here.