Author |
Message |
Zheng Wei (weiz5022)
New member Username: weiz5022
Post Number: 1 Registered: 01-2010
| Posted on Friday, January 15, 2010 - 08:01 am: | |
hello, everyone. i want to model soil water content distribution over time in a 1D system used a richard's equation whice had a source/sink item,i had seen the example of richard's equation whiche had no source/sink item in the software. but if i knew the initial and final water content distribution, how can i acquire the source/sink item and soil water content distribution, is this difficult? |
Zheng Wei (weiz5022)
New member Username: weiz5022
Post Number: 2 Registered: 01-2010
| Posted on Friday, January 15, 2010 - 09:22 pm: | |
would you give me some advice,please? thank you very much |
Marek Nelson (mgnelson)
Moderator Username: mgnelson
Post Number: 173 Registered: 07-2007
| Posted on Saturday, January 16, 2010 - 08:00 pm: | |
This is likely an impossible task. If all you know are the initial and final water content, then there are many different possible source histories/distributions that can get you from the initial to the final state. There is not a unique solution. You would have to impose some additional condition in order to make the solution unique.
|
Zheng Wei (weiz5022)
Junior Member Username: weiz5022
Post Number: 3 Registered: 01-2010
| Posted on Saturday, January 16, 2010 - 11:13 pm: | |
thank you for your help firstly,I need to know the source/sink item ,then the solution become unique. the example of richard equation took too long time, is this another question?
|
Robert G. Nelson (rgnelson)
Moderator Username: rgnelson
Post Number: 1318 Registered: 06-2003
| Posted on Sunday, January 17, 2010 - 03:55 pm: | |
1) If you know the source/sink history, then the solution will be unique. If you know the initial and final distribution and want to find the source/sink history, the solution is NOT unique. 2) In our Richard's equation example, the conductivity changes by four orders of magnitude. This causes the wave front to be extremely steep. To accurately model such a steep front, we need very dense meshing at the front and very short timesteps. Even so, our test completes in about one minute.
|
Zheng Wei (weiz5022)
Member Username: weiz5022
Post Number: 4 Registered: 01-2010
| Posted on Monday, January 18, 2010 - 06:32 am: | |
in my computer,it took about 30 minutes,but had not end at all. I didn't know what had happened. thank you for your help.
|
Robert G. Nelson (rgnelson)
Moderator Username: rgnelson
Post Number: 1319 Registered: 06-2003
| Posted on Monday, January 18, 2010 - 11:49 pm: | |
Sounds like your copy of FlexPDE is corrupted. You should download a new copy.
|
Zheng Wei (weiz5022)
Member Username: weiz5022
Post Number: 5 Registered: 01-2010
| Posted on Tuesday, January 19, 2010 - 12:40 am: | |
I am a foreign student, the software is very good, but how can i acquire the student version? thank you |
Marek Nelson (mgnelson)
Moderator Username: mgnelson
Post Number: 175 Registered: 07-2007
| Posted on Tuesday, January 19, 2010 - 01:51 pm: | |
There is a link to the free version 5 student download at www.pdesolutions.com/student6.html . Also, the version 6 evaluation download at www.pdesolutions.com/sdmenu6.html will run in student mode. |