Data_Simulation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Data Simulation.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyNogXyAoVGO9cJ1N72pJgy0",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Yustira/ANN-MT-1D/blob/main/Data_Simulation.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YLofVu5_HTCB"
      },
      "source": [
        "!pip install SimPEG==0.11.6\n",
        "!pip install git+https://github.com/simpeg/tle-magnetotelluric_inversion\n",
        "!pip install pymatsolver==0.1.3"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "csjwVel1Hm7C"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from scipy.constants import mu_0\n",
        "from SimPEG import (Mesh, Maps, SolverLU, DataMisfit, Regularization,\n",
        "                    Optimization, InvProblem, Inversion, Directives, Utils)\n",
        "from tleMT.MT1D import MT1DProblem, MT1DSurvey, MT1DSrc, ZxyRx, Survey, AppResPhaRx\n",
        "try: \n",
        "    from pymatsolver import PardisoSolver as Solver\n",
        "except importError:\n",
        "    from SimPEG import SolverLU as SolverLU\n",
        "%matplotlib inline"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_XDtAs52rT8p"
      },
      "source": [
        "def build_mesh(frequency, max_depth_core, ncell_per_skind = 1e2, n_skind = 2, max_hz_core = 5e1):\n",
        "    rxloc = np.r_[0]\n",
        "    rx = ZxyRx(rxloc, component='both', frequency=frequency)\n",
        "    src = MT1DSrc([rx])\n",
        "    survey = MT1DSurvey([src])\n",
        "    max_depth_core = max_depth_core\n",
        "    mesh = survey.setMesh(sigma = 1e-2, max_depth_core = max_depth_core, ncell_per_skind = ncell_per_skind, \n",
        "                          n_skind = n_skind, core_meshType = 'log', max_hz_core = max_hz_core)\n",
        "    return mesh, survey"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uhCGKGzvtigf"
      },
      "source": [
        "def plot_mesh(mesh):\n",
        "    fig, ax = plt.subplots(1, 1, figsize=(8, 4))\n",
        "    mesh.plotGrid(centers=True, faces=True, ax=ax)\n",
        "    ax.legend(['centers', 'faces'])\n",
        "    ax.invert_xaxis()\n",
        "    ax.set_xlabel('Z (m)')\n",
        "    ax.grid(which='both', lw=0.5)"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-U3AowoOuTLc"
      },
      "source": [
        "def model_ref(mesh, survey, layers_tops, rho_layers, loglog=True):\n",
        "    prob = MT1DProblem(mesh, sigmaMap = Maps.ExpMap(mesh), verbose = True, Solver = Solver)\n",
        "    prob.pair(survey)\n",
        "    rho = np.ones(mesh.nC)*np.nan\n",
        "\n",
        "    for layers_top, rho_layer in zip(layers_tops, rho_layers):\n",
        "        inds = mesh.vectorCCx < layers_top\n",
        "        rho[inds] = rho_layer\n",
        "\n",
        "    sigma = 1/rho\n",
        "    z = np.repeat(mesh.vectorCCx[1:], 2, axis=0)\n",
        "    z = np.r_[mesh.vectorCCx[0], z, mesh.vectorCCx[-1]]\n",
        "    rho_plt = np.repeat(rho, 2, axis=0)\n",
        "    fig, ax = plt.subplots(1, 1, figsize=(15, 5))\n",
        "\n",
        "    if loglog == True:\n",
        "        ax.loglog(-z, rho_plt, lw=2)\n",
        "    else:\n",
        "        ax.semilogy(-z, rho_plt, lw=2)\n",
        "\n",
        "    ax.set_ylabel(\"$\\\\rho$ $(\\\\Omega m)$\", fontsize=14)\n",
        "    ax.set_xlabel(\"z (m)\", fontsize=14)\n",
        "    ax.grid(True, which='both', linewidth=0.4)\n",
        "    ax.set_xlim(-z.max(), max_depth_core)\n",
        "    ax.set_ylim(5e0, 2e3)\n",
        "\n",
        "    return z, sigma, rho"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2KrxFL2OyHpw"
      },
      "source": [
        "def survey_dobs(survey, sigma, std):\n",
        "    mtrue = np.log(sigma) \n",
        "    dtrue = survey.dpred(mtrue) \n",
        "    np.random.seed(1)\n",
        "    uncert = std*np.abs(dtrue)\n",
        "    noise = uncert*np.random.randn(survey.nFreq*2)\n",
        "    survey.dobs = dtrue+noise\n",
        "    return survey.dobs, dtrue\n",
        "\n",
        "def omega(frequency):\n",
        "    return 2*np.pi*frequency\n",
        "\n",
        "def appres_phase_from_data(survey, data, frequency):\n",
        "    Zxy_real = data.reshape((survey.nFreq, 2))[:, 0]\n",
        "    Zxy_imag = data.reshape((survey.nFreq, 2))[:, 1]\n",
        "    Zxy = Zxy_real + 1j*Zxy_imag\n",
        "    app_res = abs(Zxy)**2/(mu_0*omega(frequency))\n",
        "    phase = np.rad2deg(np.arctan(Zxy_imag/Zxy_real))\n",
        "    return app_res, phase"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YSq4tCjZ4mAa"
      },
      "source": [
        "def fitting(periods, app_res_true, app_res_obs, phase_true, phase_obs, tp1, tp2, title):\n",
        "    fig, ax = plt.subplots(2, 1, figsize=(9, 4*2))\n",
        "    ax[0].loglog(periods, app_res_obs, tp1, lw=1.5, label='$d_{noise}$')\n",
        "    ax[0].loglog(periods, app_res_true, tp2, color='orange', lw=2.5, label='$d_{true}$')\n",
        "    ax[0].set_title(title, fontsize=14)\n",
        "    ax[0].set_ylabel('$\\\\rho_a$ $(\\\\Omega m)$', fontsize=14)\n",
        "    ax[0].set_ylim(5e0, 2e3)\n",
        "\n",
        "    ax[1].semilogx(periods, phase_obs, tp1, lw=1.5, label='$d_{noise}$')\n",
        "    ax[1].semilogx(periods, phase_true, tp2, color='orange', lw=2.5, label='$d_{true}$')\n",
        "    ax[1].set_ylabel('$\\\\phi$ $(^\\circ)$', fontsize=14)\n",
        "    ax[1].set_ylim([0, 90])\n",
        "\n",
        "    for a in ax:\n",
        "        a.set_xlabel('T (s)', fontsize=14)\n",
        "        a.legend(fontsize=10)\n",
        "        a.grid(True, which='both', lw=0.5)\n",
        "        a.set_xlim([periods.min(), periods.max()])\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.show() \n",
        "    return fig"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fAp_CFLGXiq7"
      },
      "source": [
        "def save_data(name, periods, app_res, phase):    \n",
        "    d_obs = np.array([periods, app_res, phase])\n",
        "    np.savetxt(str(name+\".csv\"), d_obs.transpose(), delimiter=',', fmt='%f', header='periods, rho, phi')"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "id": "JqeugtffexAz",
        "outputId": "1ccb3453-5a1c-423b-b290-397312bdb6d5"
      },
      "source": [
        "layers = np.r_[0, -750, -1000, -2500, -3100]\n",
        "rho_layers = np.r_[70, 8, 60, 250, 850]\n",
        "frequency = np.logspace(-4, 4, 50)\n",
        "periods = 1/frequency\n",
        "max_depth_core = 4000\n",
        "mesh, survey = build_mesh(frequency, max_depth_core)\n",
        "#plot_mesh(mesh)\n",
        "z, sigma, rho = model_ref(mesh, survey, layers, rho_layers, False)"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            ">> Smallest cell size = 0 m\n",
            ">> Padding distance = 1000000 m\n",
            ">> # of padding cells 33\n",
            ">> # of core cells cells 371\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 692
        },
        "id": "i3O9PzKamKO1",
        "outputId": "7c0790bf-3dc3-415f-ce8d-9b4dae4f8b37"
      },
      "source": [
        "std = 0.05\n",
        "d_obs, d_true = survey_dobs(survey, sigma, std)\n",
        "app_res_true, phase_true = appres_phase_from_data(survey, d_true, frequency)\n",
        "app_res_obs, phase_obs = appres_phase_from_data(survey, d_obs, frequency)\n",
        "fig = fitting(periods, app_res_true, app_res_obs, phase_true, phase_obs, '.-', '-', \"Site 10 (Noise 5%)\")"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            ">> Compute fields\n",
            "Delete Matrices\n",
            "Factorize A matrix\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/discretize/operators/differential_operators.py:2334: UserWarning: cell_gradient_BC is deprecated and is not longer used. See cell_gradient\n",
            "  warnings.warn(\"cell_gradient_BC is deprecated and is not longer used. See cell_gradient\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 648x576 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MLEldk3am_Z4"
      },
      "source": [
        "fig.savefig('site10_5.jpg')\n",
        "save_data('site10_5', periods, app_res_obs, phase_obs)"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T9hZkMOmvF83"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}