ABRP, the Taycan and the magic number

[JimBob]

(First sorry, I can't figure out how to get the columns to align)


In the Hitch Hiker's Guide to the Galaxy, the answer to the ultimate question of the meaning of everything is 42.

I have determined for the Taycan the answer is 170.


I was looking at doing a trip from Toronto to Victoria, British Columbia this summer, conditions permitting. I can either fly or drive.

So, I run this through ABRP and get a route, but it has a hell of a lot stops to charge at.

Next question is, how accurate is all of this?

I have several tests over a known distance which I know to a high degree of accuracy.

ABRP allows inputs of all kinds of information such as speed, temperature, route etc. which can be known with quite good accuracy but there is one number that the model is based on. That is The Reference Consumption. So what the hell is this? There are hints as to how the model is calculated but its not explicitly stated.

No problem, I can run my known data through the ABRP planner.

Well it doesn't work very well. Refer to the first line of the table. The Actual Wh/km was observed at 258.6.

So now its trial and error to get the ABRP model to agree with my test.

I want to start out at 100% SOC and end at 55% SOC. The mean temperature is -8.7C and the speed is 115 kph or 115% or the reference speed. And there are to be no charging stops along the route.

So I start plugging in numbers and get this. The first result where there are no charging stops is the ultimate answer.

Reference Consumption WH/KM @ 110 KMH From ABRP​	No of Charging Stops​	Total Energy Consumed kWh​	Per Wh/km​	Actual Wh/km​
255	1	49.1	323	258.6
250	1	48.3	318
245	1	47.5	313
240	1	46.8	308
235	1	46	303
230	1	45.2	297
225	1	44.4	292
220	1	43.6	287
215	1	42.8	282

210	1	42	277
205	1	41.3	272
200	1	40.5	266
195	1	39.7	261
190	1	38.9	256
185	1	38.1	251
180	1	37.3	246
175	1	36.5	240
170	0	35.7	235	258.6

To interpret.

I measured the energy consumption on the trip to be 258.6 Wh/km. When I ran it through ABRP, the first time I hit zero charging stops the energy consumption was reported as 235 Wh/km. This looks like the underlying error in the ABRP model. The rest, I will be diplomatic and call factor of safety.

Is this fully settled? No. More tests need to be run. But the underlying premise of ABRP as a planner is wonderful.

Sponsored

 

[andrewket]

(First sorry, I can't figure out how to get the columns to align)


In the Hitch Hiker's Guide to the Galaxy, the answer to the ultimate question of the meaning of everything is 42.

I have determined for the Taycan the answer is 170.


I was looking at doing a trip from Toronto to Victoria, British Columbia this summer, conditions permitting. I can either fly or drive.

So, I run this through ABRP and get a route, but it has a hell of a lot stops to charge at.

Next question is, how accurate is all of this?

I have several tests over a known distance which I know to a high degree of accuracy.

ABRP allows inputs of all kinds of information such as speed, temperature, route etc. which can be known with quite good accuracy but there is one number that the model is based on. That is The Reference Consumption. So what the hell is this? There are hints as to how the model is calculated but its not explicitly stated.

No problem, I can run my known data through the ABRP planner.

Well it doesn't work very well. Refer to the first line of the table. The Actual Wh/km was observed at 258.6.

So now its trial and error to get the ABRP model to agree with my test.

I want to start out at 100% SOC and end at 55% SOC. The mean temperature is -8.7C and the speed is 115 kph or 115% or the reference speed. And there are to be no charging stops along the route.

So I start plugging in numbers and get this. The first result where there are no charging stops is the ultimate answer.

Reference Consumption WH/KM @ 110 KMH From ABRP​	No of Charging Stops​	Total Energy Consumed kWh​	Per Wh/km​	Actual Wh/km​
255	1	49.1	323	258.6
250	1	48.3	318
245	1	47.5	313
240	1	46.8	308
235	1	46	303
230	1	45.2	297
225	1	44.4	292
220	1	43.6	287
215	1	42.8	282

210	1	42	277
205	1	41.3	272
200	1	40.5	266
195	1	39.7	261
190	1	38.9	256
185	1	38.1	251
180	1	37.3	246
175	1	36.5	240
170	0	35.7	235	258.6

To interpret.

I measured the energy consumption on the trip to be 258.6 Wh/km. When I ran it through ABRP, the first time I hit zero charging stops the energy consumption was reported as 235 Wh/km. This looks like the underlying error in the ABRP model. The rest, I will be diplomatic and call factor of safety.

Is this fully settled? No. More tests need to be run. But the underlying premise of ABRP as a planner is wonderful.

Nice work. You might consider sharing your model, wheels, and tires. You found the “magic” number for your car, but all the variants will be different.

 

[JimBob]

The car is a 4S with 20 inch PZero snow tires.

I don't see that changing these is going to affect the outcomes much. The default consumption provided by the ABRP platform just appears so far from reality that it swamps everything else. And you don't need to be exact, just close like in horse shoes.

ABRP appears to use one number, default consumption, which is used to determine range across all kinds of different conditions. Those conditions should be able to be accurately determined by the user and can be input into their model. As I understand it, something like Tronity can be used to update the default consumption, but as I don't use it, I don't know what its accuracy is.

If use Tronity, let it update the ABRP consumption, then run that number against a real world test and see how it matches up.

Anyone can run the same test, if they have accurate data for a distance run. Just make the ABRP match up with their data and see what happens. That's all that I have done.

Sponsored