[time-nuts] Neural net to control oven temperature ?

[Ben Bradley]

The article link in my post doesn't have valid links to the figures
(Pease hand-drawn schematics), but these links work:
https://web.archive.org/web/20121113202641/https://www.electronicdesign.com/files/29/6131/figure_01.gif
https://web.archive.org/web/20121113202709/https://www.electronicdesign.com/files/29/6131/figure_02.gif
https://web.archive.org/web/20121113202731/https://www.electronicdesign.com/files/29/6131/figure_03.gif

On Wed, Jul 10, 2019 at 1:06 AM Ben Bradley <ben.pi.bradley at gmail.com> wrote:
>
> I recall Bob Pease in one of his many "What's all this ...stuff"
> columns made a small oven and PID temperature controller that he
> claimed kept the temperature within 0.001 degrees or something like
> that. This would make machine learning severe overkill. Temp control
> is slow enough (and generates/uses a small enough data set) that for
> ML or any other method, an ARM or even AVR microcontroller might be
> enough to do it.
>
> I did a quick Google search, this column makes no such claim (it's
> about temp controllers in general), but he surely wrote several times
> about PID and/or temperature control. He has a lot of hints and ideas
> in this column, like having different sensors for the P and I, placed
> strategically for better operation:
>
> https://www.electronicdesign.com/analog/whats-all-p-i-d-stuff-anyhow
>
> On Wed, Jul 10, 2019 at 12:05 AM paul swed <paulswedb at gmail.com> wrote:
> >
> > I will mention that TI has a neural net chip/eval board now for as I recall
> > $99.
> > Like so many things maybe it makes sense.
> > Regards
> > Paul
> > WB8TSL
> >
> > On Tue, Jul 9, 2019 at 11:02 PM Chase Turner <seapeatea at gmail.com> wrote:
> >
> > > Hi Glen,
> > >
> > > This is actually something I know a little about.
> > >
> > > Neural nets are most useful for feature selection, that is, finding the
> > > important x that is a function of y, in a very large sea of x variables. In
> > > this case, we already know what's important, which is temperature
> > > stability. So, a neural net would be a bit much when we already know what
> > > feature is important for function. Additionally, unless I'm mistaken, oven
> > > control is probably a linear relationship of some sort or another, and
> > > neural nets are much better suited for examining and revealing insights
> > > about non-linear data.
> > >
> > > If you have a method by which you can collect the necessary data that has a
> > > bearing on the oven functionality, you'd probably be better off training a
> > > logistic classifier, and using it instead. That said, both methods would be
> > > overkill, imo- I'd use a PID instead.
> > >
> > > Best,
> > > Chase
> > >
> > > On Tue, Jul 9, 2019 at 10:00 PM Glen English VK1XX <
> > > glenlist at pacificmedia.com.au> wrote:
> > >
> > > > Has anyone tried to use a Neural net to control oven tmep, rather than
> > > > the ye olde PID ?
> > > >
> > > > IE the algorithm learns from previous beheviour and successfully
> > > > predicts behaviour (or not).
> > > >
> > > > I'm sure there are a few out there proficient with machine learning
> > > > algorithms.
> > > >
> > > > Might make a good masters thesis I bet.
> > > >
> > > > Given that oven control based on inputs and whatever is not random,
> > > > unlike say flicker etc.
> > > >
> > > > glen
> > > >
> > > >
> > > >
> > > > _______________________________________________
> > > > time-nuts mailing list -- time-nuts at lists.febo.com
> > > > To unsubscribe, go to
> > > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> > > > and follow the instructions there.
> > > >
> > > _______________________________________________
> > > time-nuts mailing list -- time-nuts at lists.febo.com
> > > To unsubscribe, go to
> > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> > > and follow the instructions there.
> > >
> > _______________________________________________
> > time-nuts mailing list -- time-nuts at lists.febo.com
> > To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> > and follow the instructions there.