Author Topic: How Can I solve floating point division by zero error?  (Read 891 times)

dipensahajony

  • Posts: 2
    • View Profile
    • Institute of Water Modelling
How Can I solve floating point division by zero error?
« on: March 05, 2018, 06:03:16 am »
Hello there,
I am operating a 1D drainage model for south west region of Bangladesh. I am getting "Floating point division by zero" error during simulation. If anybody knows the solution please help me out! thank you in advance.
:-) :-) :-)

Aurelian

  • Posts: 40
    • View Profile
Re: How Can I solve floating point division by zero error?
« Reply #1 on: March 08, 2018, 04:27:51 am »
Hello,

Can you give us more details about the model and the context of the error ? What modules of Mike 11 are activated in the model ? When do the errors occur? Can you attach a print-screen of the error ?

Kind regards,
Aurelian

dipensahajony

  • Posts: 2
    • View Profile
    • Institute of Water Modelling
Re: How Can I solve floating point division by zero error?
« Reply #2 on: March 11, 2018, 11:17:38 am »
Dear Aurelian,
Thanks for your reply. I am using hydrodynamic module of MIKE11. After a certain time during simulation it shows that error. screenshot is attached for you. please feel free for further quarries if you need.
Thank you and BR
Dipen
:-) :-) :-)

Aurelian

  • Posts: 40
    • View Profile
Re: How Can I solve floating point division by zero error?
« Reply #3 on: March 15, 2018, 04:47:18 pm »
Dear Dipen,

Thank you for the print screen. Apparently, the problem is generated by the instability of the model. I observed that you have many warnings which indicate strong oscillation in the model. My opinion is that the eror "Floating point division by zero" is just a consequence of the instabilities.

Recommendations:
1. Analyse the results file (it should cover 67% of simulation period) and try to identify the problems here;
1. Check out the geometry of the cross sections and the junctions between branches;
2. Try smaller time steps for simulation: 1 min, 30 sec, even 5 sec;

Best regrds,
Aurelian


khanhtriet

  • Posts: 2
    • View Profile
Re: How Can I solve floating point division by zero error?
« Reply #4 on: June 28, 2018, 01:58:46 pm »
Dear Dipen,

have you solved your issue? If yes, I would love hear about your solution.

best regards,
Triet


khanhtriet

  • Posts: 2
    • View Profile
Re: How Can I solve floating point division by zero error?
« Reply #5 on: June 28, 2018, 02:04:10 pm »
Dear Aurelian,

could you elaborate more on your second recommendation: "Check out the geometry of the cross sections and the junctions between branches"? Does it mean to equal the "lowest point", i.e No. 2 in makers of cross-sections at the junctions? Thank you!

best regards,
Triet