Learning Forum

[AaronHuang]

Hi,

I am currently working on optimizing an athermal waveguide in Lumerical FDE (2024R2) using Optislang (2024R1). To facilitate this, I’ve draft a custom script intended to extract several key parameters, such as the figure of merit (FoM), mode numbers, loss, group index (ng), and effective index (neff), which are then used as inputs/results in Optislang.

Here is the script:

#returns a list of results this script produces

function listScriptResults(){

    return {'neff_slope': 0, 'FoM': 0, 'mode_no': 0, 'loss': 0, 'ng_300K': 0, 'neff_300K': 0};  

}

 

#returns a list of parameters this script produces

function listScriptParameters(){

    return {'start_temp': 300, 'end_temp': 400, 'temp_spacing': 10};  

}

 

#returns results

function computeScriptResults(parameters){  #argument is a struct with fields giving values of different parameters

    # Temperature range and spacing

    start_temp = parameters.start_temp;

    end_temp = parameters.end_temp;

    temp_spacing = parameters.temp_spacing;

    temps = linspace(start_temp, end_temp, (end_temp - start_temp)/temp_spacing + 1);

 

    # Create a matrix to store temperature, effective index, loss, and group index values

    data_matrix = matrix(length(temps), 4);

 

    # Loop over the temperature range

    for (i = 1:length(temps)) {

        temp = temps(i);

 

        # Set the temperature

        switchtolayout;

        setnamed("::model", "temperature", temp);

 

        # Run the FDE simulation

        findmodes;

 

        # Get the effective index, loss, and group index of the fundamental mode

        neff = getdata('mode1', "neff");

        loss = getdata('mode1', "loss");

        ng = getdata('mode1', "ng");

        

        # Store the temperature, effective index, loss, and group index in the matrix

        data_matrix(i, 1) = temp;

        data_matrix(i, 2) = neff;

        data_matrix(i, 3) = loss;

        data_matrix(i, 4) = ng;

    }

 

    # Extract temperature, effective index, loss, and group index arrays from the matrix

    temps = data_matrix(:, 1);

    neff_values = data_matrix(:, 2);

    neff_300K = data_matrix(1, 2);

    loss_values = data_matrix(:, 3);

    ng_300K = data_matrix(1, 4);

 

    # Perform a polynomial fit (linear fit) to the real part of the neff values

    p = polyfit(temps, real(neff_values), 1); # Linear fit

 

    # The slope of the linear fit is the second coefficient

    neff_slope = p(2);

    FoM = abs(neff_slope);

    FoM_loss = mean(abs(loss_values));

    mode_no = nummodes;

 

    return {'neff_slope': neff_slope, 'FoM': FoM, 'mode_no': mode_no, 'loss': FoM_loss, 'ng_300K': ng_300K, 'neff_300K': neff_300K};

}

I’ve validated the execution of this code in Lumerical FDE, and it returns the correct values as expected. However, when I attempted to integrate this script within a parametric system in Optislang, I encountered an error message: "ERROR: 'XXXXX' could not be evaluated: Object 'script.XXXXX' not known."

Could you advise on how I might resolve this issue? It seems that the script may not be recognized correctly by Optislang, and I am unsure how to address this.

Thank you in advance for your assistance!

[Guilin Sun]

I have little experience using this optimization. But give the following suggestions:

The error message "ERROR: Object not known" could be indicative of various issues. One possible reason for this error could be a corrupted optislang.ini file. To resolve this, you can locate the optislang.ini file at C:\Users\yourusername\AppData\Roaming\Dynardo, delete it, and restart Optislang. If this does not resolve the issue, you may need to delete the file again and reinstall Optislang . 

 

You may also refer to this example https://optics.ansys.com/hc/en-us/articles/4404715023251-Optimizing-Wire-Grid-Polarizer-optiSLang-Interoperability

 

and the documentation https://www.ansys.com/products/connect/ansys-optislang