Project: 3D Driving Model - Page 2 - Fuel Economy, Hypermiling, EcoModding News and Forum

noeryan · 02-21-2009, 01:07 AM

Here is a very very simple version of the differential equations to be used. No Loops, IF/ELSE, etc. just good ole fashion Excel. I do not have access to 2007 at my current location (came home for weekend). The file is 2003. Check it out and edit it as much as you'd like and repost it.

I WANT FEEDBACK!!!

ALSO WANT A BETTER NAME! Something Stealthy.

dcb · 02-21-2009, 08:53 AM

Can you measure your cars weight and compute the cda and rr with some coastdown tests, then find a smooth hill on a non windy day and test your calculations? If you can predict your terminal speed reasonably well then you are probably in the ballpark.

Otherwise I fear we are doing your homework for you.

noeryan · 02-21-2009, 11:44 AM

Your doing my homework for me?

The Cd and Crr for my car IS 0.32 and 0.01. I've already calculated it. As for finding a half mile constant slope hill... I'll work on it. I was just trying to set up something that could do the calculations for me and still be accurate. I was hoping you guys would change the time steps, maybe the weight of your car and info and see if it sounds correct.. or even if the calculations are correct. I wanted to start this project as an open source, anyone who wants to get involved can help, type of project since the variables can be changed and applied to all of us.

I'm sorry guys I had a different intention of starting this thread... to get us involved in a project together and work off of eachother to get ideas flowing and a program written. If yall are just wanting me to do everything and post my results, I will but if it has potential to be commercialized I'll be trying that out.

jamesqf · 02-21-2009, 12:57 PM

Quote:

Originally Posted by noeryan

I've got some MATLAB experience, but it may be hard to get your hands on it if anyone wants to help out so I'm going to try to just use Excel and for now try to keep it open source so people can use it.

But if you want it open-source, why would you try to do it in Excel, which is proprietary and only runs on Windows? Octave Octave is an open-source program that's mostly Matlab compatible, which will run on any system.

dcb · 02-21-2009, 01:14 PM

FYI, it did open in openoffice, though I should look at Octave too.

I guess my hope would be in most projects is "less talk, more action", and am hoping to see some signs of action from the "idea guys", as implementation and testing is the real hard part. This sort of thing should pretty much write itself though. Right now it is a 1 dimensional driving model though

noeryan · 02-21-2009, 08:32 PM

You may be right about the Octave... I'm looking at it now. I figure most people run PCs with Windows and have access to Excel.

I don't have much knowledge in the software department. As for test runs and gathering data... not that hard if I have the right conditions and since the program I'm trying to write is customized to the location, I can write it for a common trip of mine and test it every day.

I'm trying to take this one step at a time. Give me the benefit of a doubt. If you aren't interested or have negative comments, find a different thread to hate on.

In much better light, I have created a sort of differential of my route to school. I used Terraserver-usa.com and spent about an hour logging the data which I have uploaded for your viewing pleasure.

Taking a quick snip off the end of Road 3 (between the last RL and the end of my route), you can gather several ideas for how to go about calculating the information. If using the trial and error, all of the acceleration equations would change the EP +1 each loop through until a final velocity satisfies the final and initial conditions. Remember, each cell in the excel sheet will be used to solve its own set of differential equations so like I said earlier, MATLAB would be a lot friendlier on the programmer.

A lot of hard thinking has been put into this today. There are various ways one could apply power while on route. I think I'll try tackling them as they come. Its very difficult to decide on my own a set way of doing the math. I believe I will initially start with the last set of differentials first and work forward and then reversing and working from front to back to make sure the system works as it should.

I believe accelerating at a constant rate to 75% of the Speed Limit is a necessity to keep with traffic. Afterwards, the section of the route is calculated maintaining a velocity of 50% to 110% of the Speed Limit.

I know the post is scattered... I've had it open adding things all day long... Any suggestions?

noeryan · 02-21-2009, 08:34 PM

Forgot the Upload!!!

jamesqf · 02-21-2009, 10:32 PM

Quote:

Originally Posted by noeryan

You may be right about the Octave... I'm looking at it now. I figure most people run PCs with Windows and have access to Excel.

Ah, but how about the people who do physics & programming? I think you'd find a lot of Linux users in that group.

noeryan · 02-21-2009, 11:03 PM

Ok, I'll give it a thought. I'm struggling with the software that I know of now. It'd be rough to have to learn new languages too.

dcb · 02-22-2009, 01:46 AM

What would you think about logging the altitude readings in 1 second resolution and adding a column to your first spreadsheet? Then just compute the slope row by row instead of having a constant?