How Predictive Energy Management Can Make PHEVs More Efficient


Leveraging outside of vehicle information could boost the energy efficiency of PHEVs on the order of about 20 to 30 percent

Almost every modern plug-in hybrid comes with a magnitude of sensors installed. They keep track of a variety of information relevant to the state the vehicle is in. This helps the onboard management system to decide when to draw energy from the batteries and power the electric motor or when to use the internal-combustion engine. In turn, this allows each hybrid to better utilize its dual powertrain in various driving conditions, helping it to save fuel and improve performance. But, what if your PHEV could use other, outside information to achieve even better efficiency?

In today’s highly connected world, there’s a tremendous and growing volume of information that can be gathered from outside the vehicle. This data includes information such as traffic, your vehicle’s trajectory, intended route, road grade, the status of traffic lights, upcoming elevation and other items. And we can utilize all that information, factoring in different data to allow your PHEV to be more efficient.

This is the core premise of predictive energy management strategies currently being developed and piloted in PHEVs. The findings from the research done until now are really interesting. If such outside information is leveraged, these predictive energy management strategies have shown that they could boost the energy efficiency of PHEVs in an order of about 20% to 30%.

The best laid out example is when a PHEV knows it will need to traverse a rather steep hill. If it could manage its power-split factor, it could factor in all its outside data and make its energy usage and fuel consumption a lot more optimized. The PHEV would utilize battery power to traverse the hill, then, knowing what the upwards and downwards gradients are, leave plenty of capacity in the batteries in order to recapture regenerative power on its descent. If the PHEV ascends over the crest of the hill with a 100% battery state, that leaves no space to store the energy it could recuperate with regenerative braking. In turn, regenerative power is wasted, and efficiency is not optimized.

This technology could even be better implemented in autonomous driving vehicles. With those, the human factor is taken out. In turn, this allows the vehicle to always use the most optimized route, road grade, and elevation, adjusting for current traffic and making for a much more calculated drive. With such implementation, predictive energy management strategies can lead to greatly enhanced decision making for the vehicle powertrain, allowing it to select the optimal energy source to combat the route & conditions ahead.

The premise is simple: utilize all the available data and make for a vehicle that uses predictive energy management strategies to save both money and emission rates. Commercial vehicles are most likely to present the first adopters and greater initial benefactors of such innovations. Fleet operators spend most of their funding on fuel and driver costs. The improvements in efficiency that would become available to them via predictive energy management strategies are especially compelling.

The addition of autonomous driving and an increased rollout of connected transportation makes for an even more compelling argument for this technology. With those circumstances in place, the predictive energy management strategies are set to play an increasingly prominent role in the future of transportation electrification. In the end, these implementations improve vehicle fuel consumption and power usage, but also, greatly help reduce Co2 emissions as well.

Source: Wards Auto

Categories: General

Tags: ,

Leave a Reply

8 Comments on "How Predictive Energy Management Can Make PHEVs More Efficient"

newest oldest most voted

Don’t the Mercedes PHEVs already do this? If you enter a destination in the navigation system, they will plan out and when do use the HVB.

Of course with 9 miles of AER it borders on a gimmick, but it’s a cool idea.

Actually, with less all-electric range, it’s even more important to use it as efficiently as possible…

As the batteries in PHEVs get larger with more range this need becomes a minority situation. It would still be useful and no reason not to implement it into the software though. Like in the current Volt for example, the avearage person pretty much does all commuting under EV power but the occasional trip or occasional longer daily drive will come up and the need for the range extender is necessary. Something like 10% of the time.

When the Volt range becomes the norm or even 100 EV miles is easily achieved with a battery that fits under the seats then the range extender will rarely be used. In our Volt we very rarely use the engine anymore now that we have a Model 3 and take it on trips.

” In our Volt we very rarely use the engine anymore now that we have a Model 3 and take it on trips.” This has always been the EV choice, lug a big battery or lug an extender. The extender has been a great option during the transistion to cheaper batteries which thanks to Tesla is rapidly coming. Which begs the question of what the masses will accept as their multi car options. A 300 AER + 100 AER? 150 AER? As to the article, managed charging + managed driving will improve range and time even more. I have spent some time with the Tesla supercharger route planner using various size batteries. The route planner has you stopping a lot more often than one would expect, though the charge times generally range from 15 -35 minutes which is shorter than you might expect. Tesla is maximizing the efficency of their battery chemistry for time spent charging + life of the battery. I think this will be more difficult to calculate for other brands who have to share a network of different chargers but still doable. Everybody is different, but I gladly take a 15-25 minute stop every 2-2.5 hours a… Read more »

I was looking at the EV Trip planner for my annual trip back to Wyoming and was surprised at how little time driving a Model 3 LR at average traffic speeds (faster than the limit), and as you say, more stops than expected, but shorter. It looks like I could do it in about 13.5 hours with 4 or 5 stops. Gas takes me as little as 12.5 with 3 stops, but usually 13-14 hours with 4 or 5 stops. I could make that trade-off. Most of the charges are from 40-180 miles range or so (optimal for rate of charge).

The Standard Model 3 would add another almost 2 hours beyond the Long Range model. Not a trade-off I would be willing to make (would mean an extra day and overnight in a hotel).

In other words: When the car is too old for map updates (like 10 years), it will become very inefficient on new streets, in the meaning of completely new build streets respectively when a street changed its altitude -> new bridge or street now underbridge etc. – additionally to the degradating battery…

And I’m quite sure that I already read something like that many years ago, IIRC concerning non-(PH)EV trucks…


Since most travel is over routes that the car has been on before, machine learning will probably give the most bang for the buck. A car can use it’s own GPS data and power use history to create an optimised energy plan for the current trip.

The example is poorly chosen IMHO. (It would make more sense for a plug-less hybrid…) For a PHEV, a much more relevant example would be when the car knows the remaining all-electric range won’t suffice to do the entire trip, and can decide to start using the combustion engine earlier when in favourable conditions, so as to keep some electricity for times where the combustion engine would be less efficient.