Mal Goris wrote:
>
> So how does fuzzy logic (aka fuzzy control) compare with
> that? Firstly, fuzzy logic can be multiple input and mutiple
> output. So it really has to be compared with the LQG class of
> controllers not just PID. The point of fuzzy logic is that you don't
> need a model of the system, which as I said is expensive or impossible
> to obtain. All you need is an expert on the system (often this is the
> guy whose job the controller will replace) who gives a heuristic set
> of rules to control the system. It is almost no work at all to turn
> those rules into a fuzzy controller which you then drop into the
> system and turn on. It can work well with nonlinear or time-varying
> systems so long as the expert is able to spell out the rules with
> enough detail. Just like the PID controller and generally the LQG
> controllers as well, a fuzzy controller won't be the best but it will
> work satisfactorily and it is cheap (relative to the best controller)
> and easy and it works well enough.

If you think about it, a PID controller is just a highly refined
version of a fuzzy controller.
very warm = big +P and +I (over time)
warm = small +Pand +I (over time)
cold = small -P and -I (over time)
very cold = big -P and -I (over time)

Any rate of changes would relate to the D term

--
Friendly Regards

Tjaart van der Walt
mailto:tjaart@wasp.co.za
 ________________________________________________________
|        WASP International   http://wasp.co.za          |
|   R&D Engineer : GSM peripheral services development   |
|Vehicle tracking | Telemetry systems | GSM data transfer|
|Voice : +27-(0)11-622-8686  |  Fax : +27-(0)11-622-8973 |
|             WGS-84 : 26010.52'S 28006.19'E             |
|________________________________________________________|