[Diy_efi] engine modeling

[Alexis PAVLOV]

Hi Tim,

That's a good idea :-)
BUT, I'm afraid once you'll have designed your control algos
you'll see that they work poorly on real engines. Because there
is a big difference between models and real badly adjusted,
badly cooled and unbalanced engines. Especially if you want
to modelize base functions like ignition or injection in order
to control the engine torque.
Of course this approach is very used in research papers and
PhD works. But, the science is nothing without the experience.
And tests often show some difference between research algos
and real observations.

I'd rather suggest you doing exhaustive measures on real engines
and then applying your control algorithms to these logs.

Alexei




Tim Drury wrote:
> 
> I skimmed through the archives and haven't seen any mention
> of anyone doing any sort of engine modeling via matlab/octave
> or something else.
> 
> Has anyone done any work on spark-ignition 4-stroke engine
> modeling they'd like to share?
> 
> I'm considering building some kind of modular, reconfigurable
> engine model to act as a stub against engine control and data
> acquisition algorithms.
> 
> Is there sufficient interest on this list to attempt this
> publicly via sourceforge or should I just keep to myself :)
> 
> -tim drury
> 
> _______________________________________________
> diy_efi mailing list
> diy_efi at diy-efi.org
> http://lists.diy-efi.org/mailman/listinfo/diy_efi
_______________________________________________
diy_efi mailing list
diy_efi at diy-efi.org
http://lists.diy-efi.org/mailman/listinfo/diy_efi