Jump to content
Jason (ABRP)

Alpha Feedback - BMW i3

Recommended Posts

Posted (edited)

Yes, IME, the i3 (sadly, just the i3 not the i3S here!) model is good, if quite conservative.

I think the reference ‘consumption’ for the plain i3 120Ah can safely be increased to 3.9 mi/kWh, at least.

IME, the only time the predicted SoC is significantly different to the actual SoC is when I enter a headwind in the settings.

I check windy.com to see the wind along the route I am planning, and use that figure (or a little less) in ABRP - but, IME the impact of the predicted SOC is always greater than it actually needs to be.

I don’t know whether this is because:

a) The ABRP model is over-playing the impact of a headwind on the energy consumption.

or

b) The actual wind affecting the car is not as significant as that stated by windy.com.

or, a combination of both.

As it is though, the wind setting is essentially not much use because the predicted impact of wind is not borne out by reality.

Edited by Bill N

Share this post


Link to post
Share on other sites
On 7/13/2019 at 1:45 PM, Bill N said:

Yes, IME, the i3 (sadly, just the i3 not the i3S here!) model is good, if quite conservative.

I think the reference ‘consumption’ for the plain i3 120Ah can safely be increased to 3.9 mi/kWh, at least.

IME, the only time the predicted SoC is significantly different to the actual SoC is when I enter a headwind in the settings.

I check windy.com to see the wind along the route I am planning, and use that figure (or a little less) in ABRP - but, IME the impact of the predicted SOC is always greater than it actually needs to be.

I don’t know whether this is because:

a) The ABRP model is over-playing the impact of a headwind on the energy consumption.

or

b) The actual wind affecting the car is not as significant as that stated by windy.com.

or, a combination of both.

As it is though, the wind setting is essentially not much use because the predicted impact of wind is not borne out by reality.

It's probably a combination of both.  It's possible we're overestimating the dragginess of the i3, and thus a high wind is treated as a bigger hit than it really is.  Have you checked direction of the wind and only taken the parallel component of the wind to your path? IE, wind coming out of the northwest when driving north would only have a 70% effect (trigonometry!).

I really wish we had a way to pull live data from the i3, then we could refine the model much more readily!

Share this post


Link to post
Share on other sites
11 minutes ago, Jason (ABRP) said:

It's probably a combination of both.  It's possible we're overestimating the dragginess of the i3, and thus a high wind is treated as a bigger hit than it really is.  Have you checked direction of the wind and only taken the parallel component of the wind to your path? IE, wind coming out of the northwest when driving north would only have a 70% effect (trigonometry!).

I really wish we had a way to pull live data from the i3, then we could refine the model much more readily!

Thanks Jason. Yeah, I do check the wind direction pre-departure via windy.com, and also use teslawinds.com whilst on route. If the wind direction would not result in either a direct headwind or tailwind, what I've done is just try to guesstimate a lower wind speed and entered that into ABRP - which still seems to result in a greater predicted impact than what actually materializes. It’s tricky though because (here in the UK especially) a route rarely follows a straight line - so you have to try and take into account the wind direction and then the variations in the direction of travel along the length of the route.

In theory, (though I appreciate it’s probably far from straightforward) it should be possible for ABRP to pull the forecast or current wind speed and direction data and apply that to the model whilst factoring in the car speed and direction at different points along the length of the route.

The other potential factor I’ve pondered is that whilst driving along a fairly busy 3 lane highway with other vehicles positioned to 3 (or sometimes all 4) sides of your own car, the actual airspeed of your car is possibly reduced and is not simply ground speed +/- wind speed because the air immediately surrounding your car is, to some extent, being ‘pulled along’ by the constant flow of moving objects.

Share this post


Link to post
Share on other sites

Your final point is very interesting, and would be hard to model.  Like a pseudo-drafting.

All these problems go away when we get a live data driving setup for the i3, since we account for all of the parameters and re-calibrate based on actual data from the car, accounting for the wind and other factors.  You could think of it like a "Smart GoM".

We have a thread up to work on this idea, but it hasn't seen a lot of traffic.  The main target has  been going after the PIDs, but if anyone wants to tinker with figuring out BMW's API, or any other means of getting live data from the i3, we'd be happy to work with you to figure it out!

 

Share this post


Link to post
Share on other sites

Yes, it's almost impossible to model that I would have thought - as it depends largely on the precise density and proximity of traffic at any moment in time. I was just really trying to think of other possible explanations for the discrepancy.

Live data driving would certainly help significantly once you are on the move, but I'm not sure if it would help so much when it comes to pre-planning a journey out of the car. Unless, maybe, ABRP used the journey data collected from cars, in conjunction with available wind data for the time and location of each journey, which you could then apply to the raw car data to refine the model - so the next time a user enters a forecast wind speed of X, when pre-planning a journey that involves driving at Y degrees relative to the forecast wind direction, the model knows to increase the energy consumption by Z (adjusted for ground speed).

I think the trouble is, there are so many localised factors that could potentially affect the actual wind that is affecting a car on a given road. So windy.com, or the nearest weather station, or whatever might say that the current wind is blowing SSW at 10mph, but the actual wind hitting the car might only be 7mph and might be blowing SW, due to local terrain, road side trees, binding, embankments, vehicles on the opposite etc etc. But, over time, and with more and more data points, it might be possible to define a good enough 'wind adjustment factor' for each car.

Share this post


Link to post
Share on other sites

Definitely!

And any wind data is better than no wind data, and we can adjust the effect of the wind based on real-world data and feedback as well.  It's definitely on our radar, but probably lower priority than building out some basic App support, and making it more convenient for you to connect to your car (or connect to your car at all in the i3's case).

Share this post


Link to post
Share on other sites

Hereby I upload my latest trip information. I have a bmw i3 2017 model 94 ah

we live in a part of Norway where the max speed is 80km/h, smal towns 60km/h. I get a max range of ca 210km from the battery

E89D502E-D84C-4FFC-83E6-88406FA40B04.jpeg

  • Like 1
  • Thanks 1

Share this post


Link to post
Share on other sites

2018 BMW i3 120Ah.  100 mile motorway journey.  Started at 100%.  Planner predicted arriving with 25%.  Actual 28.5%.  Adaptive cruise control set at 75mph, air-con set to 22 degs and outside weather sunny and 24 degs.  

Very pleased with my first use of the app. 

 

  • Like 1

Share this post


Link to post
Share on other sites

The route planner gives me a strange route. I want to go from Groenlo (The Netherlands) to Ahaus (Germany) to Enschede (The Netherlands). I can make this trip if it goes like I stated but the routeplanner thinks I need to charge in Enschede (The Netherland) so suggest a route from Groenlo (The Netherlands) to Charger in Enschede (The Netherlands) then to Ahaus (Germany) and back to Enschede (The Netherlands) ?

 

Share this post


Link to post
Share on other sites

Hi. 

I'm using ABRP with my i3 60Ah Rex - it works great, but I've notices that it seems to over-compensate for wind. Even moderate wind modifies consumption a lot - and more than in real life.

Thanks for a great app!

 

Nikolaj

Share this post


Link to post
Share on other sites
4 hours ago, Nikolaj said:

Hi. 

I'm using ABRP with my i3 60Ah Rex - it works great, but I've notices that it seems to over-compensate for wind. Even moderate wind modifies consumption a lot - and more than in real life.

Thanks for a great app!

 

Nikolaj

Glad you like it!

This is something we've been tracking for a while, there's a number of factors that reduce the effect wind has on your drive, like the cars in front of you disrupting the wind flow and making it not as impactful.  We've made some attempts at improving it, so it's definitely not as bad as it used to be, but still something we're working on improving.

Share this post


Link to post
Share on other sites

Is there any plan to include a model that takes into account the REX range?  As much as I like to use it as little as possible, it's there so i will make use of it.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


Contact Us

Bo - Lead Developer and Tesla owner: bo@abetterrouteplanner.com

Jason - New Car Models, Developer and Bolt owner : jason@abetterrouteplanner.com

Idreams - Forums Administrator, Forums Developer and Tesla owner : idreams@abetterrouteplanner.com

×
×
  • Create New...