Re: [foxboro] 24 hr rolling average
- From: "LS Kong" <lskong@xxxxxxxxxxxxx>
- To: <foxboro@xxxxxxxxxxxxx>
- Date: Fri, 30 Nov 2001 11:17:11 +0800
Hmmm.... no attachment allowed.
Let do it this way:
=================
INDEPENDENT_SEQUENCE{*******************************
RUNNING AVERAGE CALCULATION **
06 DEC 2000: LS Kong - Made Generic **
***************************************************}
CONSTANTS
WINDOW = 1440; { 1440 Minutes / 24 Hours Window }
SAMPLE = 48; { 48 Sample in Window Period }
VARIABLES
QUEUE : R[SAMPLE]; { Queue Length=No of Samples }
I : I;
USER_LABELS
NEW_READING : RI0001; { Current reading }
AVERAGE : RO0001; { Calculated average }
SUM : RO0002; { Sum of All Readings }
HEAD : IO0001; { Current head of queue }
STATEMENTS
{ Initialisation }
WAIT 10; { Wait for things to settle }
QUEUE := SET_ARRAY
(NEW_READING); { Set all values in }
{ array to current value }
HEAD := 1; { Head of queue }
<<loop>>
{ Push current reading to the head of the queue }
QUEUE[HEAD] := NEW_READING;
{ Calculate Sum of All Readings }
SUM := 0;
FOR I := 1 TO SAMPLE DO
SUM := SUM + QUEUE[I];
ENDFOR;
{ Calculate Average over [SAMPLE] number of readings }
AVERAGE := SUM / SAMPLE;
{ Move the head of queue to the next reading }
IF HEAD < SAMPLE THEN
HEAD := HEAD + 1;
ELSE
HEAD := 1;
ENDIF;
RO0002 := 0;
RO0003 := QUEUE [SAMPLE - 12];
RO0004 := QUEUE [SAMPLE - 11];
RO0005 := QUEUE [SAMPLE - 10];
RO0006 := QUEUE [SAMPLE - 9];
RO0007 := QUEUE [SAMPLE - 8];
RO0008 := QUEUE [SAMPLE - 7];
RO0009 := QUEUE [SAMPLE - 6];
RO0010 := QUEUE [SAMPLE - 5];
RO0011 := QUEUE [SAMPLE - 4];
RO0012 := QUEUE [SAMPLE - 3];
RO0013 := QUEUE [SAMPLE - 2];
RO0014 := QUEUE [SAMPLE - 1];
RO0015 := QUEUE [SAMPLE];
{ Endless Loop }
WAIT WINDOW / SAMPLE * 60; { wait til next calculation cycle }
GOTO loop;
ENDSEQUENCE
----- Original Message -----
From: LS Kong <lskong@xxxxxxxxxxxxx>
To: <foxboro@xxxxxxxxxxxxx>
Sent: Friday, November 30, 2001 11:06 AM
Subject: Re: [foxboro] 24 hr rolling average
>
> Not the best but this should work.
>
>
> ----- Original Message -----
> From: Matic.Hank <Hank.Matic@xxxxxxxxx>
> To: <foxboro@xxxxxxxxxxxxx>
> Sent: Friday, November 30, 2001 2:37 AM
> Subject: Re: [foxboro] 24 hr rolling average
>
>
> > We do a rolling three hour average with a sequence block , taking a
sample
> > once per minute. To trend the average , you need only to put the
sequence
> > block real output into a historian and assign that value to a trend
point.
> > If you like , I can send you a copy of the code.
> >
> > -----Original Message-----
> > From: Jaudon, Michael [mailto:MJAUDON@xxxxxxx]
> > Sent: Thursday, November 29, 2001 7:27 AM
> > To: 'foxboro@xxxxxxxxxxxxx'
> > Subject: [foxboro] 24 hr rolling average
> >
> >
> >
> > I know we have already ventured down this path once but at the time
> rolling
> > average didn't interest me and I didn't give it a passing thought...now
it
> > does.
> >
> > My question is: How would you program for a 24 hour rolling average (my
> case
> > is a flow) say sampled every 5 minutes (288 samples per day)? I also
need
> > to trend the result.
> >
> > Regards,
> >
> > Michael L. Jaudon
> > Kerr-McGee Chemical LLC
> > (662)343-8710
> > mjaudon@xxxxxxx
> >
> >
> >
>
___________________________________________________________________
This list is neither sponsored nor endorsed by The Foxboro Company.
Use the information obtained here at your own risk. For disclaimer,
see http://www.thecassandraproject.org/disclaimer.html#maillist
list info: //www.freelists.org/list/foxboro
subscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
unsubscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave
Other related posts:
