I think David wrote:
> 

> I wonder if the non-cyclic (and looks to me somewhat stochastic)
> behaviour is a result of the JVM implementation as you have replied
> earlier ?

I thought my previous post at www.manning.com/friedman-hill/ on this
topic was quite clear, but yes, thread scheduling on a multitaking OS
has an element of randomness to it, and so the HVAC simulation's
behavior does too.

> ... yielding a "perceived" set point of 71. This seems to be
> contrary to what your algorithm is trying to set .....

Some book quotes:

----------------------------------------------------------------------
Chapter 16 intro: "[the basic system] will keep the temperature stable
to within plus or minus one or two degrees.

Next, you'll improve the control program using fuzzy logic [to]
control the temperature within a small fraction of a degree."
----------------------------------------------------------------------
Section 16.1: "You can get an advanced degree in the engineering of
control systems. It's a vast and complex disclipline, and we can't
hope to cover it all here. Instead, I'll state some commonsense
principles to which your system should adhere, and then we'll come up
with a simple, ad hoc control algorithm to staisfy these
requirements." 
----------------------------------------------------------------------

Anyway, the first system does control the temperature within a few
degrees, and the second one does control it within a fraction of a
degree, and in both cases the range includes the setpoint. There is no
promise made that the mean temperature would evaluate to the setpoint
under the kind of analysis you're attempting to do, especially given
the oversimplified physics that are built into the simulator (and the
randomly selected values for Newton's cooling constant and the HOT and
COLD equilibrium values.)

Finally, note that physical thermostats always have a little screw you
can loosen and adjust to trim the constant offset between the
thermometer reading and the setpoint scale -- it would be a trivial
matter to add an "electronic screw" to the HVAC Controller.

So anyway -- you're reading way too much into this example.


---------------------------------------------------------
Ernest Friedman-Hill  
Science and Engineering PSEs        Phone: (925) 294-2154
Sandia National Labs                FAX:   (925) 294-2234
PO Box 969, MS 9012                 [EMAIL PROTECTED]
Livermore, CA 94550         http://herzberg.ca.sandia.gov

--------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to