Author |
Message |
FuulLCTS (fuullcts)
New member Username: fuullcts
Post Number: 1 Registered: 05-2009
| Posted on Monday, May 04, 2009 - 09:37 am: | |
Hello, I am new to use FlexPDE (FlexPDE6)and I have already a problem! I want to use a script written in FlexPDE5 (it runs succefully on it) but when I try to run it on FlexPDE6, it indicates "No Evaluation Location Specified" in a popup with an highlighted term. I find no explanation for this popup and so I don't know how to correct my script. Could you help me? |
Robert G. Nelson (rgnelson)
Moderator Username: rgnelson
Post Number: 1251 Registered: 06-2003
| Posted on Monday, May 04, 2009 - 01:38 pm: | |
I assume this diagnostic occurs in a HISTORY plot? If you plot a HISTORY of a spatially varying quantity and do not specify a location for the evaluation, version 6 will issue the diagnostic you report. Version 5 did not diagnose this error, so the plot is meaningless, as you have no idea where the evaluation was made. You can correct this error by specifying the location at which you want to compute the plotted function, using an "AT(...)" qualifier.
|
Fuul (fuullcts)
New member Username: fuullcts
Post Number: 2 Registered: 05-2009
| Posted on Tuesday, May 05, 2009 - 03:51 am: | |
It do not occur in a history plot but in a DEFINITIONS entry. The highlighted term is exactely "(temp" in : vabla = IF (temp > 927) THEN 1E-3*(1.941674124E-7*temp^2 - 3.1115595256E-4*temp + 1.2123168158E-1) ELSE 0 But temp is not spatially dependent.
|
Robert G. Nelson (rgnelson)
Moderator Username: rgnelson
Post Number: 1252 Registered: 06-2003
| Posted on Tuesday, May 05, 2009 - 02:36 pm: | |
1. TEMP is by definition spatially-dependent. It is a VARIABLE, defined by a diffusion equation. 2. The diagnostic has been issued at the point of encountering the offending reference, instead of the outer evaluation that creates the conflict. Most diagnostics of this kind trace the reference to the initiating statement, but TIME_INTEGRAL apparently does not. We will correct this diagnostic in a future release. 3. The offense arises from the statement "xabla = time_integral(vabla)" (the line following the diagnostic). TIME_INTEGRAL() must operate on a scalar value, it cannot create a field of values. You can correct this by localizing the TEMP value using VAL() or EVAL(), as in: "xabla = time_integral(VAL(vabla,epai))"
|
Fuul (fuullcts)
Junior Member Username: fuullcts
Post Number: 3 Registered: 05-2009
| Posted on Wednesday, May 06, 2009 - 10:41 am: | |
Thanks! Now it runs perfectly. |