Direct numerical simulation of zero-pressure-gradient turbulent boundary layer with passive scalars up to $Pr = 6$

This repository contains the statistical database extracted from the direct numerical simulation of zero-pressure gradient turbulent boundary layer ranging up to $Re_\theta = 1070$ with four different passive scalars corresponding to $Pr=1,2,4,6$.

For a detailed comparison of the obtained statistics against the documented data available in the literature, please take a look at our article

You are free to use the database. For the use of DNS database in scientific communications, we request you to cite our article as:

@article{balasubramanian_guastoni_schlatter_vinuesa_2023,
    title={Direct numerical simulation of a zero-pressure-gradient turbulent boundary layer with passive scalars up to {P}randtl number $Pr = 6$},
    volume={974},
    DOI={10.1017/jfm.2023.803},
    journal={Journal of Fluid Mechanics},
    publisher={Cambridge University Press},
    author={Balasubramanian, Arivazhagan G. and Guastoni, Luca and Schlatter, Philipp and Vinuesa, Ricardo},
    year={2023},
    pages={A49}}

Statistics dictionary

1. How to load the statistics file?

    x = np.load('TBL_statistics.npz')
   

   General key naming convention:

    quantity : flow / scalar
    flow_type : fixed here (redundant) - tbl
    Momentum Re : 420 / 628 / 830 / 1070
    Data : mean / rms / skewness / flatness / mean_params / xx_budget
    Variable : u / v / w / p / s1 / s2 / s3 / s4 / us1 / ...
    Scaling : outer / inner
   

   Only data at 4 streamwise locations are provided. For other locations, a request shall be made to the correspondance email provided in the article.
   For mean_params :

    cf_wrt_x / cf_wrt_Rex / Redisp_wrt_x / Retheta_wrt_x / Retau_wrt_x / h12_wrt_x / h12_wrt_Retheta
   

   is possible.

   To load any variable, the format is:

    var = x['{quantity}_{flow_type}_{momentum_Re}_{data}_{variable}_{scaling}']
   

   So, to load u-RMS plot in inner-scale at $Re_\theta = 628$,

    urms = x['flow_tbl_628_rms_u_inner']
   

2. How to plot the required data?
   First, load the required data as above.
   Then, plot it as:

    plt.plot(urms[0],urms[1])
   

3. For list of keys available,

    for i in x.keys():
        print(i)
   

Data request

If you need some additional data that might be of interest to you, please don't hesitate to contact us at:
Arivazhagan G B
Luca Guastoni
Ricardo Vinuesa

Upcoming

Please take a look at our work on Non-intrusive sensing in turbulent boundary layers via deep fully-convolutional neural networks which has used the data sampled from the above-discussed DNS simulation.