Metric Adapter for Diverse Performances.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2a8c9618",
   "metadata": {},
   "source": [
    "### 1. Load exemplary waveforms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "24002365",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The file to check is: 140V.csv\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "waveform_path = \"./Waveforms\"\n",
    "files = os.listdir(waveform_path)\n",
    "\n",
    "selected_idx = 0\n",
    "print(f\"The file to check is: {files[selected_idx]}\")\n",
    "all_results = np.loadtxt(f\"{waveform_path}/{files[selected_idx]}\",\n",
    "                        skiprows=1, delimiter=\",\") # load all data\n",
    "\n",
    "iL_PA_RNN, vp_ES, vs_ES = [all_results[:, i] \n",
    "                           for i in range(all_results.shape[1])] # assign each column into a var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c0a60db0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x1080 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the waveforms in three diverse scenarios\n",
    "fig, axs = plt.subplots(2, 1, figsize=(8, 15))\n",
    "\n",
    "for i, (waves, title) in enumerate(zip([(vp_ES, vs_ES), (iL_PA_RNN,)],\n",
    "                                       [\"vp and vs waveforms\", \"iL waveforms\"])):\n",
    "    for wave in waves:\n",
    "        axs[i].plot(wave)\n",
    "    axs[i].set_title(title)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2dd5eb58",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "c11ebd3d",
   "metadata": {},
   "source": [
    "### 2. Performance Metric 1 - Current Stress\n",
    "* Peak-to-peak current stress: ipp\n",
    "* Root-mean-square current stress: irms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3dacf374",
   "metadata": {},
   "outputs": [],
   "source": [
    "def current_stress(iL, metric=\"ipp\", delta_T=200e-9):\n",
    "    assert metric in [\"ipp\", \"irms\"]\n",
    "    if metric == \"ipp\":\n",
    "        return max(iL)-min(iL)\n",
    "    elif metric == \"irms\":\n",
    "        return np.sqrt((iL**2).mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "50020cb0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ipp of PA-RNN is 4.20 A.\n",
      "\n",
      "\n",
      "irms of PA-RNN is 1.08 A.\n"
     ]
    }
   ],
   "source": [
    "# get peak-to-peak current stress\n",
    "ipp_PA_RNN = current_stress(iL_PA_RNN, metric=\"ipp\")\n",
    "print(f\"ipp of PA-RNN is {ipp_PA_RNN:.2f} A.\")\n",
    "\n",
    "print(\"\\n\")\n",
    "\n",
    "# get root-mean-square current stress\n",
    "irms_PA_RNN = current_stress(iL_PA_RNN, metric=\"irms\")\n",
    "print(f\"irms of PA-RNN is {irms_PA_RNN:.2f} A.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4a0c438e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "808f76b5",
   "metadata": {},
   "source": [
    "### 3. Performance Metric 2 - Soft Switching Performance\n",
    "* Zero voltage switching: ZVS\n",
    "* Zero current switching: ZCS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "496f6526",
   "metadata": {},
   "outputs": [],
   "source": [
    "def locate(v, V):\n",
    "    \"\"\"\n",
    "        Locate ip1, ip2, ip3, ip4\n",
    "        and is1, is2, is3, is4\n",
    "    \"\"\"\n",
    "    idx = [None]*4\n",
    "    v0 = v[0]\n",
    "    for i in range(1, len(v)+1):\n",
    "        i = i % len(v)\n",
    "        dV = v[i]-v0\n",
    "        if (dV > V/2):\n",
    "            if v0 < -V/2: idx[0] = i\n",
    "            else: idx[1] = i\n",
    "            if (dV > V*1.5):\n",
    "                idx[1] = i\n",
    "        elif (dV < -V/2):\n",
    "            if v0 > V/2: idx[2] = i\n",
    "            else: idx[3] = i\n",
    "            if (dV < -V*1.5):\n",
    "                idx[3] = i\n",
    "        v0 = v[i]\n",
    "    return idx\n",
    "\n",
    "Vo_map = {0: 140, 1: 160}\n",
    "def segment(vp, iL):\n",
    "    V = Vo_map[selected_idx]\n",
    "    # segment the NPC bridge to obtain each drain/collection current\n",
    "    s_pri = np.array(vp>V/2).astype(np.float32)\n",
    "    s_pri2 = np.array(vp>-V/2).astype(np.float32)\n",
    "    iL = np.array(iL).astype(np.float32)\n",
    "    iS1 = s_pri*iL\n",
    "    iS4 = -(1-s_pri2)*iL\n",
    "    iS2 = np.clip(s_pri+np.clip(s_pri2-s_pri, 0., 1.)*(iL>=0.), 0., 1.)*iL\n",
    "    iS3 = -np.clip((1-s_pri2)+np.clip(s_pri2-s_pri, 0., 1.)*(iL<=0.), 0., 1.)*iL\n",
    "    return iS1, iS2, iS3, iS4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9dbd172c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def soft_switching(iL, vp, vs):\n",
    "    iS1, iS2, iS3, iS4 = segment(vp, iL)\n",
    "    indices = locate(vp, Vin)\n",
    "    indices2 = locate(vs, Vo_map[selected_idx])\n",
    "    ZVS = np.zeros(12, )\n",
    "    ZCS = np.zeros(12, )\n",
    "    \n",
    "    \n",
    "    ##### ZVS\n",
    "    ZVS[0] = (iS1[indices[1]]<=1e-1).astype(int) # ZVS: S1\n",
    "    ZVS[7] = (iS1[indices[1]]<=1e-1).astype(int) # ZVS: S8\n",
    "    ZVS[1] = (iS2[indices[0]]<=1e-1).astype(int) # ZVS: S2\n",
    "    ZVS[6] = (iS2[indices[0]]<=1e-1).astype(int) # ZVS: S7\n",
    "    ZVS[2] = (iS3[indices[2]]<=1e-1).astype(int) # ZVS: S3\n",
    "    ZVS[5] = (iS3[indices[2]]<=1e-1).astype(int) # ZVS: S6\n",
    "    ZVS[3] = (iS4[indices[3]]<=1e-1).astype(int) # ZVS: S4\n",
    "    ZVS[4] = (iS4[indices[3]]<=1e-1).astype(int) # ZVS: S5\n",
    "    ZVS[8] = (iL[indices2[0]]>=-1e-1).astype(int) # ZVS: Q1\n",
    "    ZVS[9] = (iL[indices2[2]]<=1e-1).astype(int) # ZVS: Q2\n",
    "    ZVS[10] = (iL[indices2[3]]<=1e-1).astype(int) # ZVS: Q3\n",
    "    ZVS[11] = (iL[indices2[1]]>=-1e-1).astype(int) # ZVS: Q4\n",
    "\n",
    "    #### ZCS\n",
    "    ZCS[0] = (np.abs(iS1[indices[2]])<=threshold).astype(int) # ZCS: S1\n",
    "    ZCS[7] = (np.abs(iS1[indices[2]])<=threshold).astype(int) # ZCS: S8\n",
    "    ZCS[1] = (np.abs(iS2[indices[3]])<=threshold).astype(int) # ZCS: S2\n",
    "    ZCS[6] = (np.abs(iS2[indices[3]])<=threshold).astype(int) # ZCS: S7\n",
    "    ZCS[2] = (np.abs(iS3[indices[1]])<=threshold).astype(int) # ZCS: S3\n",
    "    ZCS[5] = (np.abs(iS3[indices[1]])<=threshold).astype(int) # ZCS: S6\n",
    "    ZCS[3] = (np.abs(iS4[indices[0]])<=threshold).astype(int) # ZCS: S4\n",
    "    ZCS[4] = (np.abs(iS4[indices[0]])<=threshold).astype(int) # ZCS: S5\n",
    "    ZCS[8] = (np.abs(iL[indices2[2]])<=threshold).astype(int) # ZCS: Q1\n",
    "    ZCS[9] = (np.abs(iL[indices2[0]])<=threshold).astype(int) # ZCS: Q2\n",
    "    ZCS[10] = (np.abs(iL[indices2[1]])<=threshold).astype(int) # ZCS: Q3\n",
    "    ZCS[11] = (np.abs(iL[indices2[3]])<=threshold).astype(int) # ZCS: Q4\n",
    "    return ZVS, ZCS, indices, indices2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5440e808",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ZVS and ZCS performance of experimental data:\n",
      "(array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.]), array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 0., 0.]))\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Vin = 300\n",
    "threshold = 2e-2\n",
    "\n",
    "\n",
    "ZVZCS_pred = soft_switching(iL_PA_RNN, vp_ES, vs_ES)\n",
    "print(\"ZVS and ZCS performance of experimental data:\")\n",
    "print(ZVZCS_pred[:2])\n",
    "\n",
    "indices, indices2 = ZVZCS_pred[2:]\n",
    "indices[2] -= 1\n",
    "indices[2] -= 1\n",
    "plt.figure()\n",
    "plt.plot(vp_ES/300)\n",
    "plt.plot(vs_ES/300)\n",
    "plt.scatter(indices, vp_ES[indices]/300, c=\"b\")\n",
    "plt.scatter(indices, iL_PA_RNN[indices]/iL_PA_RNN.max(), c=\"b\")\n",
    "plt.scatter(indices2, vs_ES[indices2]/300, c=\"r\")\n",
    "plt.scatter(indices2, iL_PA_RNN[indices2]/iL_PA_RNN.max(), c=\"r\")\n",
    "plt.plot(iL_PA_RNN/iL_PA_RNN.max())\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4fdf5a4c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b4657669",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "a9dfb2e3",
   "metadata": {},
   "source": [
    "### 4. Performance Metric 3 - Power\n",
    "* Transferred power: PL\n",
    "* Reactive power: Q"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c7a034ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "def power(vp, vs, iL, metric=\"PL\", delta_T=200e-9):\n",
    "    assert metric in [\"PL\", \"Q\"]\n",
    "    if metric == \"PL\":\n",
    "        return (vp*iL).mean()\n",
    "    elif metric == \"Q\":\n",
    "        vL = vp-n*vs\n",
    "        return np.sqrt((vL**2).mean())*np.sqrt((iL**2).mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c0d7f152",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PL of PA-RNN is 236.21 W.\n",
      "\n",
      "\n",
      "Q of PA-RNN is 68.83 W.\n"
     ]
    }
   ],
   "source": [
    "# get power transferred\n",
    "PL_PA_RNN = power(vp_ES, vs_ES, iL_PA_RNN, metric=\"PL\")\n",
    "print(f\"PL of PA-RNN is {PL_PA_RNN:.2f} W.\")\n",
    "\n",
    "print(\"\\n\")\n",
    "\n",
    "# get reactive power\n",
    "n = 2\n",
    "Q_PA_RNN = power(vp_ES, vs_ES, iL_PA_RNN, metric=\"Q\")\n",
    "print(f\"Q of PA-RNN is {Q_PA_RNN:.2f} W.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "008fedbf",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "389f6de7",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}