Prescribing values for ohmic heating

[yannxiaoyanhu]

Hi,

I hope this message finds you well.

I attempted to assign values for ohmic heating in the configuration. I tried the following two ways, and both failed.

Method 1: ohmic = {
'mode': 'PRESCRIBED',
'prescribed_values': ([0.0], [0.0], [0.0]), # time, rho, ohmic
}
Error message:
Value error, Input to TimeVaryingArray unsupported. Input was of type: <class 'list'> [type=value_error, input_value=[0.0], input_type=list]
For further information visit https://errors.pydantic.dev/2.11/v/value_error

Method 2: ohmic = {
'mode': 'PRESCRIBED',
'prescribed_values': {0.0: ([0.0], [0.0])}, # time: (rho, ohmic)
}
Error message:
ValidationError: 1 validation error for ToraxConfig
sources.ohmic.standard.prescribed_values
Input should be a valid tuple [type=tuple_type, input_value={0.0: ([0.0], [0.0])}, input_type=dict]

I wonder if it is possible to give an example.

Thank you.

[theo-brown]

Hi! Thanks for your question. I agree that the PRESCRIBED mode documentation is a bit weak at the moment, that's probably partly my fault!

Firstly, in the specific case where you want no ohmic heating, you can just omit the ohmic source from the sources list, which represents ohmic heating only. However, note that this will not completely eliminate ohmic current. To eliminate the ohmic current, the best thing at the moment is to use the use_v_loop_lcfs_boundary_condition in profile_conditions. #1611 will hopefully be an extension to this that allows you to eliminate ohmic current altogether.

For PRESCRIBED mode in general, a few things to note:

1. Sources output a tuple, containing the values for each equation they effect (psi, electron heat, ion heat). This means that your prescribed_values also have to be a tuple of time-varying arrays, even if it's only affecting one equation. To check what equations are affected, find the affected_core_profiles property of the source.
2. The values section of the time-varying array must be 2D, even if it's only got one element, as its dimensions are (n_time, n_rho).

The correct way of prescribing a fixed flat profile at value x for ohmic heating is therefore:

'ohmic': {
        'mode': 'PRESCRIBED',
        'prescribed_values': (([0.0], [0.0], [[x]]),),
}

[yannxiaoyanhu]

Thanks for your reply!

Is it correct that if I want to set the following ohmic heating scheme:
1. time=0, heating=1e5 W/m^3 at rho=0.2, and heating=1.5e W/m^3 at rho=0.4; 2. time=1.0, heating=3e5 W/m^3 at rho=0.2, and heating=3.5e5 W/m^3 at rho=0.4

Then the configuration should be
'ohmic': {
'mode': 'PRESCRIBED',
'prescribed_values': (([0.0, 1.0], [0.2, 0.4], [[1e5, 1.5e5], [3e5, 3.5e5]]),),
}

Also, if I can only estimate an average ohmic heating by vloop * ip in the dataset, how should I set the configuration accordingly?

[theo-brown]

Yes, I think your example is correct (by eye)! You can always quickly try it on the example case and see whether it works.

Not sure about your follow-up question, someone with a bit more scientific/experimental experience might be better suited to help.

[theo-brown]

As this is not a bug with the code but a question about usage, I'm going to transfer this to a discussion.

[yannxiaoyanhu]

Thanks a lot :)