examples/01_microgenerator.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2b04f6ae-c630-45b5-9143-3398de977134",
   "metadata": {},
   "source": [
    "# Example: microgenerator simulation\n",
    "In this tutorial we'll create and simulate a microgenerator using `FFS`."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c631a323-148a-4908-81fe-067b81826a95",
   "metadata": {},
   "source": [
    "## Background\n",
    "<img src='./_img/mcrg.png' width=600px></img>\n",
    "\n",
    "In `FFS`, we can simulate and optimize energy harvesters that have the same architecture and basic properties as shown above.\n",
    "Namely, the energy harvester must be a single-axis, cylindrical microgenerators that consist of one or more evenly-spaced coils wound around the outside of the body.\n",
    "Inside the body is a moving magnet assembly, suspended by a magnetic spring, that consists of one or more evenly-spaced magnets.\n",
    "\n",
    "We simulate these devices by composing a number of \"submodels\" together, each representing parts of the mechanical system, the electrical system or the interaction between these two systems. This means we'll need to specify a number of different components:\n",
    "\n",
    "- The magnet assembly\n",
    "- The coil(s)\n",
    "- The flux curve model\n",
    "- The mechanical friction\n",
    "- The electro-mechanical damping (coupling) force\n",
    "- The magnetic spring behaviour\n",
    "- The target load\n",
    "- The rectification voltage drop\n",
    "- The impact dynamics\n",
    "- The input excitation\n",
    "- The device height\n",
    "- The behavioural system model\n",
    "\n",
    "This may seem a bit daunting at first-glance, but fortunately `FFS` has been designed to make defining and simulating these devices as painless as possible. Let's dive in."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0a539e61-5a1e-4418-b8aa-231f5f2bc752",
   "metadata": {},
   "source": [
    "## Creating a base model\n",
    "Let's start by first creating a base microgenerator model, without any components or submodels attached to it.\n",
    "First, we'll have to import `FFS`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c2f9ab63-1443-46c9-bb55-f5b8f803ed79",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "import ffs  # FFS import"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f27e2be5-c4bb-4daf-a536-fcbe33e0c469",
   "metadata": {
    "tags": []
   },
   "source": [
    "We define all models using the `UnifiedModel` class from the `unified` module:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e7f3efb9-1f4d-416a-b90b-5ee899ddb902",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = ffs.unified.UnifiedModel()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "906f770a-9fde-42d1-87c0-befb639a1556",
   "metadata": {},
   "source": [
    "Due to the iterative and exploratory nature of energy harvester design, `FFS` makes it easy to inspect its various components.\n",
    "We can `print` most components to get an overview of their properties:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0b40757a-8710-4c18-be23-8a883a3a318c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unified Model: \n",
      "  height: None\n",
      "  governing_equations: None\n",
      "  time: None\n",
      "  raw_solution: None\n",
      "  post_processing_pipeline: {}\n",
      "  magnetic_spring: None\n",
      "  magnet_assembly: None\n",
      "  mechanical_spring: None\n",
      "  mechanical_damper: None\n",
      "  input_excitation: None\n",
      "  flux_model: None\n",
      "  coil_configuration: None\n",
      "  rectification_drop: None\n",
      "  load_model: None\n",
      "  coupling_model: None\n",
      "  extra_components: {}\n"
     ]
    }
   ],
   "source": [
    "print(model)  # Print our base model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aaae3696-cf5d-4f59-a8a7-c906bdc1c085",
   "metadata": {},
   "source": [
    "## Creating components"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21c3b4d7-590e-41f6-9d86-b8306a42b8bd",
   "metadata": {},
   "source": [
    "Currently, we can see that our `model` is completely empty -- no components are currently attached to it.\n",
    "\n",
    "Let's define some components that we can attach.\n",
    "\n",
    "### Device height\n",
    "Let's start by declaring the height that we want our microgenerator to have:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "eca999a0-a066-405b-a8d1-5db7b54171be",
   "metadata": {},
   "outputs": [],
   "source": [
    "device_height = 120 / 1000   # must be in metres!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b4e25c3-e404-4daf-9468-637a1536f069",
   "metadata": {},
   "source": [
    "### Magnet assembly\n",
    "Let's create a magnet assembly, which lives in the `mechanical_componets` module of `FFS`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8e447dee-87c0-401a-bf4b-7d3b1d5f7cbe",
   "metadata": {},
   "outputs": [],
   "source": [
    "magnet_assembly = ffs.mechanical_components.magnet_assembly.MagnetAssembly(\n",
    "    m=1,  # Number of magnets\n",
    "    l_m_mm=10,  # Magnet length\n",
    "    l_mcd_mm=0,  # Spacing between magnets (not relevant to this example)\n",
    "    dia_magnet_mm=10,  # Magnet diameter\n",
    "    dia_spacer_mm=0  # Spacer diamer (also not relevant here)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2cb1b507-c213-452d-bd3e-8c9168783eda",
   "metadata": {},
   "source": [
    "To look-up the meaning of each parameter, we can use the `?` operator to lookup any component's documentation (`FFS` is fully documented): "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "49f0ae4d-963b-4f7b-aca6-741d5aba3a38",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mInit signature:\u001b[0m\n",
       "\u001b[0mffs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmechanical_components\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmagnet_assembly\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mMagnetAssembly\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mm\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mint\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0ml_m_mm\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0ml_mcd_mm\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mdia_magnet_mm\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mdia_spacer_mm\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mmat_magnet\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'NdFeB'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mmat_spacer\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'iron'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mDocstring:\u001b[0m     \n",
       "Represent a magnet assembly.\n",
       "\n",
       "Parameters\n",
       "----------\n",
       "m : int\n",
       "    Number of magnets.\n",
       "l_m_mm : float\n",
       "    Height of the magnets in mm.\n",
       "l_mcd_mm : float\n",
       "    Distance between the centers of each magnet, in mm.\n",
       "dia_magnet_mm: float\n",
       "    Diameter of magnets in mm.\n",
       "dia_spacer_mm : float\n",
       "    Diameter of spacer in mm\n",
       "mat_magnet : str\n",
       "    Magnet material key. Optional.\n",
       "mat_spacer: str\n",
       "    Spacer material key. Optional.\n",
       "\u001b[0;31mInit docstring:\u001b[0m Constructor\n",
       "\u001b[0;31mFile:\u001b[0m           ~/Nextcloud/nextcloud/PhD/Python/unified_model/ffs/mechanical_components/magnet_assembly.py\n",
       "\u001b[0;31mType:\u001b[0m           type\n",
       "\u001b[0;31mSubclasses:\u001b[0m     \n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ffs.mechanical_components.magnet_assembly.MagnetAssembly?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6bd48640-0d3a-45d9-b7cd-3895d9663f29",
   "metadata": {},
   "source": [
    "We can also inspect most `FFS` objects using `print`, like we saw earlier:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "a5fc5056-c4fd-4207-b935-86021f43d2ab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MagnetAssembly(n_magnet=1, l_m_mm=10, l_mcd_mm=0, dia_magnet_mm=10, dia_spacer_mm=0)\n"
     ]
    }
   ],
   "source": [
    "print(magnet_assembly)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0e3c7839-7554-4635-887d-3b521e2c60a4",
   "metadata": {},
   "source": [
    "### Mechanical damper\n",
    "Let's create our mechanical damper, which allows us to model mechanical losses due to friction:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d058012d-b170-4ffc-9356-1376ee7a5cb4",
   "metadata": {},
   "outputs": [],
   "source": [
    "mechanical_damper = ffs.mechanical_components.damper.MassProportionalDamper(\n",
    "    damping_coefficient=4.272,\n",
    "    magnet_assembly=magnet_assembly  # This depends on our `magnet_assembly`!\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0017769-89fb-435d-bf7e-62ea4b5ba6a5",
   "metadata": {},
   "source": [
    "Note that in order to define our mechanical damper, we needed to pass in our `magnet_assembly` we created earlier! Now, you might be wondering,\n",
    "\n",
    "\"if I change my magnet assembly configuration later, will I also have to respecify my mechanical damper?\" \n",
    "\n",
    "The answer is no! To make iteration and experimentaiton easier, **`FFS` automatically keeps track of dependencies, and updates components whenever their dependencies are changed.**\n",
    "We'll demonstrate this in a later tutorial, but for now, you can trust that `FFS` will make sure components are updated as their dependencies change.\n",
    "\n",
    "### Coil configuration\n",
    "Let's define a coil configuration, which specifies all the properties of our coils:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a0c81d5e-4eae-442b-99b6-a3f8e127a4cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "coil_configuration = ffs.electrical_components.coil.CoilConfiguration(\n",
    "    c=1,  # Number of coils\n",
    "    n_z=15,  # Vertical windings\n",
    "    n_w=17,  # Horizontal windings\n",
    "    l_ccd_mm=0,  # Coil spacing\n",
    "    ohm_per_mm=1079 / 1000 / 1000,  # Ohm per mm of wire\n",
    "    coil_wire_radius_mm=0.143 / 2,  # Wire radius\n",
    "    coil_center_mm=72-13,  # Coil center location\n",
    "    inner_tube_radius_mm=5.5,  # Inner tube radius of microgenerator body\n",
    "    tube_wall_thickness_mm=2  # Wall thickness of the microgenerator body\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bff9d69-298c-4ef5-a80f-5cf12e1058fa",
   "metadata": {},
   "source": [
    "### Magnetic spring\n",
    "Let's define our magnetic spring, that models the magnetic-levitation spring dynamics of the magnet assembly:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "39aadd08-da0d-461a-91d3-35b9410242bb",
   "metadata": {},
   "outputs": [],
   "source": [
    "magnetic_spring = ffs \\\n",
    ".mechanical_components.magnetic_spring.MagneticSpringInterp(\n",
    "    fea_data_file='../data/magnetic-spring/10x10alt.csv',\n",
    "    magnet_assembly=magnet_assembly\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "65b381ca-ecaf-4f48-b9eb-14f3445ecbcb",
   "metadata": {},
   "source": [
    "In this case, we use simulated FEA data to model our spring.\n",
    "The FEA data file specifies the force experienced between two pole-matched 10mm x 10mm cylindrical magnets (which are the same as the magnets defined in our magnet assembly)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88d5ae13-3011-48cf-9613-8ce14ef55256",
   "metadata": {},
   "source": [
    "### Flux model\n",
    "Next, let's define our flux model. Our flux model allows us to predict a the magnetic flux captured by the coil(s) given the coil properties, magnet assembly and their relative position.\n",
    "In our case, we'll be using a pre-trained flux curve model. In this particular case, our pre-trained model is specific to our 10mm x 10mm cylindrical magnets and choice of copper wire:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "7701a5e5-f6a1-4c3e-94b5-092ec4bad31e",
   "metadata": {},
   "outputs": [],
   "source": [
    "flux_model = ffs.electrical_components.flux.FluxModelPretrained(\n",
    "    coil_configuration=coil_configuration,\n",
    "    magnet_assembly=magnet_assembly,\n",
    "    curve_model_path='../data/flux_curve_model/flux_curve_model_2022_08_09.model'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f94233a0-5dea-457e-ad68-cb59add0bddd",
   "metadata": {},
   "source": [
    "### Coupling\n",
    "Here we define out coupling model, that models the interaction between the electrical and mechanical systems:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "82a89874-24ce-47f5-abdc-ead5364170cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "coupling = ffs.coupling.CouplingModel(coupling_constant=5.096)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "283658f5-c489-4bd4-87ec-87cf88abad50",
   "metadata": {},
   "source": [
    "### Impact dynamics\n",
    "Let's also define the impact dynamics that occur when the magnet assembly impacts the top of the microgenerator device.\n",
    "In this case, we model it as a lossy mechanical spring  / collision:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "acc15361-1824-4f9c-8520-3c8171d6ea2a",
   "metadata": {},
   "outputs": [],
   "source": [
    "mechanical_spring = ffs.mechanical_components.mechanical_spring.MechanicalSpring(\n",
    "    magnet_assembly=magnet_assembly,\n",
    "    damping_coefficient=3.108\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f832804-dff3-4c7f-9d9b-9574c988a990",
   "metadata": {},
   "source": [
    "### Target load\n",
    "Let's define our target load as a 30 Ohm resistor:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "44d96616-141f-4304-b0ab-08d43402604d",
   "metadata": {},
   "outputs": [],
   "source": [
    "load_model = ffs.electrical_components.load.SimpleLoad(R=30)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "977c25b7-980b-45b4-8371-908f5de3acdd",
   "metadata": {},
   "source": [
    "### Rectification drop\n",
    "`FFS` allows us to model the rectification drop from a full-wave bridge rectifier. In this case, like with the device height, we only need to specify a float value:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "3f1222f5-9a4c-478a-b4f5-20f4802af073",
   "metadata": {},
   "outputs": [],
   "source": [
    "v_rect_drop = 0.05  # 50 millivolts"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b84ac1e-5854-4772-9fcc-80d232f44770",
   "metadata": {},
   "source": [
    "### Input excitation\n",
    "We also need to add an input excitation, which will drive the microgenerator. We'll use some footstep-like accelerometer measurements:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f2db2c09-8494-4b3a-977a-5f1b56c67027",
   "metadata": {},
   "outputs": [],
   "source": [
    "input_excitation = ffs.mechanical_components.input_excitation.accelerometer.AccelerometerInput(\n",
    "    raw_accelerometer_data_path='../data/2019-05-23/A/log_01_acc.csv',\n",
    "    accel_column='z_G',\n",
    "    time_column='time(ms)',\n",
    "    accel_unit='g',\n",
    "    time_unit='ms'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf6f9bcb-4f50-496f-b73d-bef4713995b5",
   "metadata": {},
   "source": [
    "## Defining the system behaviour model\n",
    "The last thing we need to specify is the the system behavioural model, or what's known as the `governing_equations` function.\n",
    "The function defines all behaviour and interaction of the various components within the microgenerator. \n",
    "\n",
    "If you wanted to simulate your own electromechanical energy harvester model, you'd need to define your own `governing_equations` function.\n",
    "\n",
    "Fortunately, `FFS` ships with a `unified_ode` governing equations function (which all of this research is built upon), which is what we'll also here:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "58705657-9500-44fd-84f0-1fd7ef617364",
   "metadata": {},
   "outputs": [],
   "source": [
    "governing_equations = ffs.governing_equations.unified_ode"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90718482-8f19-4f66-94ad-0667d0c2a0ba",
   "metadata": {},
   "source": [
    "## Assembling the model\n",
    "Finally, let's add all our components, input excitation and governing equations to our model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "3d4941bb-aca2-4cc5-aba0-4d831fedd515",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = model \\\n",
    "    .with_height(device_height) \\\n",
    "    .with_magnet_assembly(magnet_assembly) \\\n",
    "    .with_coil_configuration(coil_configuration) \\\n",
    "    .with_magnetic_spring(magnetic_spring) \\\n",
    "    .with_mechanical_damper(mechanical_damper) \\\n",
    "    .with_mechanical_spring(mechanical_spring) \\\n",
    "    .with_coupling_model(coupling) \\\n",
    "    .with_flux_model(flux_model) \\\n",
    "    .with_rectification_drop(v_rect_drop) \\\n",
    "    .with_load_model(load_model) \\\n",
    "    .with_input_excitation(input_excitation) \\\n",
    "    .with_governing_equations(governing_equations)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "784ff578-fdf8-4e92-95a0-a7d43d4d3c79",
   "metadata": {},
   "source": [
    "In addition to being able `print` our various models and components, the `UnifiedModel` class also provides a `.summarize` method, which summarizes the microgenerator configuration for us:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "18d7876f-4b22-4ec5-a212-f85e7247f72f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unified Model: \n",
      "  height: 0.12\n",
      "  governing_equations: <function unified_ode at 0x284f685e0>\n",
      "  time: None\n",
      "  raw_solution: None\n",
      "  post_processing_pipeline: {}\n",
      "  magnetic_spring: MagneticSpringInterp(../data/magnetic-spring/10x10alt.csv, auto)\n",
      "  magnet_assembly: MagnetAssembly(n_magnet=1, l_m_mm=10, l_mcd_mm=0, dia_magnet_mm=10, dia_spacer_mm=0)\n",
      "  mechanical_spring: MechanicalSpring(position=0.12, magnet_length=0.01, magnet_assembly_length=0.01, strength=10000000.0, damping_coefficient=3.108)\n",
      "  mechanical_damper: MassProportionalDamper(damping_coefficient=4.272, magnet_assembly_mass=0.005890486225480863)\n",
      "  input_excitation: AccelerometerInput('../data/2019-05-23/A/log_01_acc.csv')\n",
      "  flux_model: FluxModelPretrained(curve_model_path=../data/flux_curve_model/flux_curve_model_2022_08_09.model, curve_model=CurveModel(regressor=LinearRegression(), x_arr.shape=(93,), kernel_space.shape=(93, 53), feature_engineering_pipeline=PolynomialFeatures(degree=5)), flux_model_interp=FluxModelInterp(c=1, c_c=0.059, l_ccd=0.0, m=1, l_mcd=0.0, flux_model=<ffs.utils.utils.FastInterpolator object at 0x2859c0f10>, dflux_model=<ffs.utils.utils.FastInterpolator object at 0x2859c0fa0>), flux_model=<ffs.utils.utils.FastInterpolator object at 0x2859c0f10>, dflux_model=<ffs.utils.utils.FastInterpolator object at 0x2859c0fa0>)\n",
      "  coil_configuration: CoilModel(c=1, n_z=15, n_w=17, l_ccd_mm=0, ohm_per_mm=0.0010789999999999999, tube_wall_thickness_mm=2, coil_wire_radius_mm=0.0715, coil_center_mm=59, inner_tube_radius_mm=5.5, custom_coil_resistance=None)\n",
      "  rectification_drop: 0.05\n",
      "  load_model: SimpleLoad: 30 Ohms\n",
      "  coupling_model: CouplingModel(coupling_constant=5.096)\n",
      "  extra_components: {}\n"
     ]
    }
   ],
   "source": [
    "print(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f91f07d9-9a86-4b7f-80e7-fac34df0ceac",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "  ┌────┐         120\n",
      "  │    │         115\n",
      "  │    │         110\n",
      "  │    │         105\n",
      "  │    │         100\n",
      "  │    │         95\n",
      "  │    │         90\n",
      "  │    │         85\n",
      "  │    │         80\n",
      "┌─│────│─┐       75\n",
      "└─│████│─┘       70\n",
      "  │████│         65\n",
      "  │    │         60\n",
      "  │    │         55\n",
      "  │    │         50\n",
      "  │    │         45\n",
      "  │    │         40\n",
      "  │    │         35\n",
      "  │    │         30\n",
      "  │    │         25\n",
      "  │    │         20\n",
      "  │    │         15\n",
      "  │████│         10\n",
      "  │████│         5\n",
      "  └────┘         0\n",
      "====================\n",
      "🧲 There are 1 magnet(s) in the magnet assembly, each 10mm long with a diameter of 10mm.\n",
      "🧲 The magnets' centers are 0mm apart.\n",
      "🧲 The magnet assembly has a weight of 0.0578N.\n",
      "🧲 The magnet assembly hovers 46.015mm above the fixed magnet.\n",
      "-----------\n",
      "⚡ There are 1 coils, each with 255 windings (15 vertical X 17 horizontal).\n",
      "⚡ This gives each a coil an estimated height of ~2.14mm and width of ~2.43mm.\n",
      "⚡ The coils' centers are 0mm apart.\n",
      "⚡ The first coil's center is 59mm above the fixed magnet.\n",
      "⚡ The estimated total coil resistance is 15.314459828575176Ω.\n",
      "-----------\n",
      "📏 The device has a height of 120.0mm.\n",
      "-----------\n",
      "🎯 The device is powering a 30Ω load.\n",
      "====================\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model.summarize()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae9f5ad4-d190-445f-9fce-68f52dc24131",
   "metadata": {},
   "source": [
    "## Running a simulation\n",
    "We can now predict the time-domain behaviour of our model using the `.solve` method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "ea30f222-db4b-4346-9832-93624813051d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Run the simulation!\n",
    "model.solve( \n",
    "    t_start=0,\n",
    "    t_end=8,\n",
    "    y0=[0., 0., 0.04, 0., 0.],  # initial conditions\n",
    "    t_eval=np.linspace(0, 8, 1000),\n",
    "    t_max_step=1e-3\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "66a15b78-6c09-41b2-9a78-7a4d9cec73d9",
   "metadata": {},
   "source": [
    "A reminder that, if we're ever unsure, we can lookup the documentation using `?`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "5e46485c-bf79-4c13-9586-da169ff3764e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m\n",
       "\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msolve\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mt_start\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'float'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mt_end\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'float'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0my0\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'List[float]'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mt_eval\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'Union[List, np.ndarray]'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mt_max_step\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'float'\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0.0001\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mmethod\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'str'\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'RK45'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;34m'None'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mDocstring:\u001b[0m\n",
       "Solve the unified model.\n",
       "\n",
       "Parameters\n",
       "----------\n",
       "t_start : float\n",
       "    The start time of the simulation.\n",
       "t_end : float\n",
       "    The end time of the simulation\n",
       "y0 : ndarray\n",
       "    The initial values of `y`, or the result vector that is passed\n",
       "    to the governing equations.\n",
       "t_eval : ndarray[float]\n",
       "    Times at which to store the computed solution.\n",
       "t_max_step : float, optional\n",
       "    The maximum time step (in seconds) to be used when solving the\n",
       "    unified model. Default value is 1e-4.\n",
       "method : str, optional\n",
       "    The solver to use. Default value is 'RK45'. See the documentation\n",
       "    for `scipy.integrate.solve_ivp` for possible solution methods.\n",
       "\n",
       "See Also\n",
       "--------\n",
       "scipy.integrate.solve_ivp : Function used to solve the governing\n",
       "    equations of the unified model.\n",
       "\u001b[0;31mFile:\u001b[0m      ~/Nextcloud/nextcloud/PhD/Python/unified_model/ffs/unified.py\n",
       "\u001b[0;31mType:\u001b[0m      method\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.solve?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b678cde1-f479-4f92-b0ff-577379f94aa5",
   "metadata": {},
   "source": [
    "## Retrieving solutions\n",
    "\n",
    "We can retrieve our simulation results using the `.get_result` method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "8b0f470d-40a6-4852-b59d-dfbd34a6f245",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_result\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;34m'pd.DataFrame'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mDocstring:\u001b[0m\n",
       "Get a dataframe of the results using expressions.\n",
       "\n",
       "*Any* reasonable expression is possible. You can refer to each of the\n",
       "differential equations that is represented by the governing equations\n",
       "using the letter 'x' with the number appended. For example `x1` refers\n",
       "to the first differential equation, `x2` to the second, etc.\n",
       "\n",
       "Each expression is available as a column in the returned pandas\n",
       "dataframe, with the column name being the key of the passed expression.\n",
       "\n",
       "Parameters\n",
       "----------\n",
       "**kwargs\n",
       "    Each key is the name of the column of the returned dataframe.\n",
       "    Each value is the expression to be evaluated.\n",
       "\n",
       "Returns\n",
       "-------\n",
       "pandas dataframe\n",
       "    Output dataframe containing the evaluated expressions.\n",
       "\n",
       "See Also\n",
       "--------\n",
       "ffs.utils.utils.parse_output_expression : helper function\n",
       "    that contains the parsing logic.\n",
       "\n",
       "Example\n",
       "--------\n",
       ">>> # Here we use previously-built and solved unified model\n",
       ">>> ffs\n",
       "<ffs.unified.UnifiedModel at 0x7fa9e45a83c8>\n",
       ">>> print(ffs.raw_solution)\n",
       "[[1 2 3 4 5]\n",
       " [1 1 1 1 1]]\n",
       ">>> ffs.get_result(an_expr='x1',\n",
       "                             another_expr='x2-x1',\n",
       "                             third_expr='x1*x2')\n",
       "   an_expr  another_expr  third_expr\n",
       "0        1             0           1\n",
       "1        2            -1           2\n",
       "2        3            -2           3\n",
       "3        4            -3           4\n",
       "4        5            -4           5\n",
       "\u001b[0;31mFile:\u001b[0m      ~/Nextcloud/nextcloud/PhD/Python/unified_model/ffs/unified.py\n",
       "\u001b[0;31mType:\u001b[0m      method\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.get_result?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4c8f10a-3752-49fd-b55d-c1ef8f77d29c",
   "metadata": {},
   "source": [
    "As mentioned in the documentation above, fetching results utilizes an expression syntax. This allows us to easily specify exactly which solution expressions should be returned.\n",
    "In the case of the `unified_ode` governing equations function that we selected to use, each differential equation's solution is accessible as follows:\n",
    "\n",
    "- `t` -- time\n",
    "- `x1` -- tube position\n",
    "- `x2` -- tube velocity\n",
    "- `x3` -- magnet assembly position\n",
    "- `x4` -- magnet assembly velocity\n",
    "- `x5` -- effective flux linkage\n",
    "\n",
    "We also have access to a gradient function `g`, which allows us to compute the gradient of an expression."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ba046d6-ef7b-4c3c-a968-9464fc0d51e0",
   "metadata": {},
   "source": [
    "Let's retrieve the time, relative magnet assembly position and the load voltage using `.get_result`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "1fb496e2-338f-4dfd-a398-9d31edad6970",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>time</th>\n",
       "      <th>y_diff</th>\n",
       "      <th>y_diff_prime</th>\n",
       "      <th>v_load</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.040000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.008008</td>\n",
       "      <td>0.040166</td>\n",
       "      <td>0.040909</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.016016</td>\n",
       "      <td>0.040645</td>\n",
       "      <td>0.077915</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.024024</td>\n",
       "      <td>0.041396</td>\n",
       "      <td>0.108568</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.032032</td>\n",
       "      <td>0.042361</td>\n",
       "      <td>0.131073</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>7.967968</td>\n",
       "      <td>0.049186</td>\n",
       "      <td>0.993367</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>7.975976</td>\n",
       "      <td>0.049186</td>\n",
       "      <td>0.994131</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>7.983984</td>\n",
       "      <td>0.049186</td>\n",
       "      <td>0.994894</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>7.991992</td>\n",
       "      <td>0.049187</td>\n",
       "      <td>0.995652</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.049188</td>\n",
       "      <td>0.996406</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         time    y_diff  y_diff_prime  v_load\n",
       "0    0.000000  0.040000      0.000000     0.0\n",
       "1    0.008008  0.040166      0.040909     0.0\n",
       "2    0.016016  0.040645      0.077915     0.0\n",
       "3    0.024024  0.041396      0.108568     0.0\n",
       "4    0.032032  0.042361      0.131073     0.0\n",
       "..        ...       ...           ...     ...\n",
       "995  7.967968  0.049186      0.993367     0.0\n",
       "996  7.975976  0.049186      0.994131     0.0\n",
       "997  7.983984  0.049186      0.994894     0.0\n",
       "998  7.991992  0.049187      0.995652     0.0\n",
       "999  8.000000  0.049188      0.996406     0.0\n",
       "\n",
       "[1000 rows x 4 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = model.get_result(\n",
    "    time='t',  # Time\n",
    "    y_diff='x3-x1',  # Relative magnet assembly position\n",
    "    y_diff_prime='x4-x2',  # Relative magnet assembly velocity\n",
    "    v_load='g(t, x5)'  # Compute the time-derivative of the effective flux linkage\n",
    ")\n",
    "\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8b6ab67d-c870-46ff-af5e-0458d9c5b1bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "         time    y_diff  y_diff_prime  v_load\n",
      "0    0.000000  0.040000      0.000000     0.0\n",
      "1    0.008008  0.040166      0.040909     0.0\n",
      "2    0.016016  0.040645      0.077915     0.0\n",
      "3    0.024024  0.041396      0.108568     0.0\n",
      "4    0.032032  0.042361      0.131073     0.0\n",
      "..        ...       ...           ...     ...\n",
      "995  7.967968  0.049186      0.993367     0.0\n",
      "996  7.975976  0.049186      0.994131     0.0\n",
      "997  7.983984  0.049186      0.994894     0.0\n",
      "998  7.991992  0.049187      0.995652     0.0\n",
      "999  8.000000  0.049188      0.996406     0.0\n",
      "\n",
      "[1000 rows x 4 columns]\n"
     ]
    }
   ],
   "source": [
    "print(results)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c19890f-f2fb-4c62-a49d-96be23e78e33",
   "metadata": {},
   "source": [
    "Let's plot the results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "2a8d9ca1-afe3-4432-8d6f-fe3a9395ab14",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='time'>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "results.plot(x='time', y='y_diff')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "624e25ac-3f45-43b6-a5c2-fdf6b203e35c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='time'>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "results.plot(x='time', y='v_load')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "7bd53ba6-5beb-4b8e-aca9-fe6a94bdb186",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mInit signature:\u001b[0m \u001b[0mffs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mevaluate\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mSample\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0macc_path\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madc_path\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvideo_labels_path\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mDocstring:\u001b[0m      A class for holding groundtruth sample data\n",
       "\u001b[0;31mFile:\u001b[0m           ~/Nextcloud/nextcloud/PhD/Python/unified_model/ffs/utils/utils.py\n",
       "\u001b[0;31mType:\u001b[0m           type\n",
       "\u001b[0;31mSubclasses:\u001b[0m     \n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ffs.evaluate.Sample?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "1dd8bd55-1d44-41e9-a5fe-6f612d18f843",
   "metadata": {},
   "outputs": [],
   "source": [
    "sample = ffs.evaluate.Sample(\n",
    "    acc_path='../data/2019-05-23/A/log_04_acc.csv',\n",
    "    adc_path='../data/2019-05-23/A/log_04_adc.csv',\n",
    "    video_labels_path='../data/2019-05-23/A/a004_transcoded_subsampled_labels_2019-10-31-10:24:04.csv'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "2f3790c0-8029-4a93-92c5-55d4f8c6541f",
   "metadata": {},
   "outputs": [],
   "source": [
    "solve_kwargs = dict(\n",
    "    t_start=0,\n",
    "    t_end=8,\n",
    "    y0=[0., 0., 0.04, 0., 0.],  # initial conditions\n",
    "    t_eval=np.linspace(0, 8, 1000),\n",
    "    t_max_step=1e-3\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "d1f8e45c-a15a-4f8c-8cc0-4006c265263b",
   "metadata": {},
   "outputs": [],
   "source": [
    "results, _ = model.score_sample(\n",
    "    sample=sample,\n",
    "    solve_kwargs=solve_kwargs,\n",
    "    y_diff_expr='x3-x1',\n",
    "    mech_metrics_dict={'y_diff_dtw_distance_norm': ffs.metrics.dtw_euclid_norm_by_length},\n",
    "    v_load_expr='g(t, x5)',\n",
    "    elec_metrics_dict={'rms_perc_diff': ffs.metrics.root_mean_square_percentage_diff}\n",
    "                       \n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "fd6fe6c8-8247-4fbd-97b8-0fac67a7d4e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "evaluators = _"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "565c1786-4c9c-4150-b32d-d2d307429bce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAABL3ElEQVR4nO29eZhcVbX3/1k1Tz13ZyCdkAQIEKYAgQSQQUEBFZFJhJ8IKgJ68RWnK3i9iPJevfeKyntVBBQEESUIKogREDBGL2FIIEASSEiAJB0y9Nxd87R/f+xT1VXV1fNQ1dX78zz9VPU5+5xa1af6e1atvfZaopTCYDAYDJWLrdQGGAwGg2FiMUJvMBgMFY4ReoPBYKhwjNAbDAZDhWOE3mAwGCocR6kNKKSxsVHNnz+/1GYYDAbDlGLdunVtSqmmYvvKTujnz5/P2rVrS22GwWAwTClEZPtA+0zoxmAwGCocI/QGg8FQ4RihNxgMhgqn7GL0BoOhckgkErS0tBCNRkttSsXg8Xhobm7G6XQO+xgj9AaDYcJoaWmhqqqK+fPnIyKlNmfKo5Sivb2dlpYWFixYMOzjTOjGYDBMGNFolIaGBiPy44SI0NDQMOJvSEboDQbDhGJEfnwZzd/TCP1UpXUzvL261FYYDIYpgBH6qcpPj4d7zym1FQZDWdPe3s6SJUtYsmQJs2bNYs6cOdnf4/H4uL5WV1cXt91227iec7wwk7EGg6FiaWhoYP369QDcdNNNBAIBvvrVrw55XDKZxOEYmTxmhP7zn//8aEydUIxHP9VJp0ttgcEwpfj5z3/Occcdx1FHHcUFF1xAOBwG4IorruCaa65h2bJl/Ou//ivbtm1j+fLlHHHEEXzzm98kEAhkz/H973+f4447jiOPPJJvfetbAFx//fVs27aNJUuW8LWvfa0k720gjEc/1UlGwOUvtRUGw5B8+08b2fRuz7iec/F+1XzrnMNGdMz555/PZz/7WQC++c1vctddd/GFL3wB0Omgzz77LHa7nQ9/+MN88Ytf5JJLLuH222/PHv/kk0/y5ptv8sILL6CU4iMf+QirV6/mP//zP9mwYUP2G0Q5YTz6qU48XGoLDIYpxYYNGzj55JM54ogjuP/++9m4cWN230UXXYTdbgdgzZo1XHTRRQBceuml2TFPPvkkTz75JEcffTTHHHMMb7zxBm+++ebkvokRYjz6qU4yUmoLDIZhMVLPe6K44oor+OMf/8hRRx3FPffcw6pVq7L7/P6hvx0rpbjhhhu4+uqr87a/884742zp+GE8+oF4ezW0lfddGoBkrNQWGAxTit7eXmbPnk0ikeD+++8fcNzy5ct5+OGHAXjggQey288880zuvvtugsEgALt27WLfvn1UVVXR29s7scaPEiP0xWjfplMX7zu/1JYMjRF6g2FE3HzzzSxbtoyTTjqJQw45ZMBxt956Kz/84Q858sgj2bp1KzU1NQB84AMf4NJLL+WEE07giCOO4MILL6S3t5eGhgZOOukkDj/88LKbjBWlVKltyGPp0qWq5I1HXvg5rLRSsL60EWqaS2tPMW7SHzqufAaajy2tLQbDALz++usceuihpTZjVITDYbxeLyLCAw88wG9/+1seeeSRUpsFFP+7isg6pdTSYuNNjL4Y+17ve763TIU+Q9JUBTQYJoJ169Zx7bXXopSitraWu+++u9QmjRoj9MVo3QxNh0Lr69D6Biw6s9QW5ZNO9T1PmdCNwTARnHzyybzyyiulNmNcMDH6YvTsgpmLwVMD3S2ltqY/uXF5E6M3GAxDYIS+EKWgdw9UzYbq5jIV+pxwjRF6g8EwBEboC4n16Nz0qlk6Nt+9s9QW9cd49AaDYQQYoS+kd49+rJoNNXOge1dp7SlGrkdvYvQGg2EIKkrod+7cyc6dY/TAe3frx6pZEJgJkQ5IJcdu3DgSi4aLPp8KjMs1MhhGgN1uZ8mSJRx++OFcdNFF2SJmo+GKK67goYceAuDKK69k06ZNA45dtWoVzz77bPb322+/nV/96lejfu2xUFFCP3fuXObOnTu2k/Tu1Y+BmeBv1M/D7WM75zjT2tVXGKqjqzxX4g3EuFwjg2EEeL1e1q9fz4YNG3C5XHkFykCXJB4Nv/jFL1i8ePGA+wuF/pprruGTn/zkqF5rrFSU0K9YsYIVK1aM7STRLv3orQd/k34eah3bOceZnt5g9nk4EiqhJSNnXK6RwTBKTj75ZLZu3cqqVas4+eST+chHPsLixYtJpVJ87Wtfy5YevuOOOwBd1+baa6/l4IMP5owzzmDfvn3Zc5122mlkFnc+/vjjHHPMMRx11FGcfvrpvPPOO9x+++386Ec/YsmSJfzjH//gpptu4pZbbgFg/fr1LF++nCOPPJLzzjuPzs7O7Dm//vWvc/zxx7No0SL+8Y9/jMv7rqg8+p/97GcAXHzxxaM/SbRbP3qqy1boe4N9Qh+NTK2iZuNyjQxTk79cD3teG99zzjoCzv7PYQ1NJpP85S9/4ayzzgLgpZdeYsOGDSxYsIA777yTmpoaXnzxRWKxGCeddBIf+MAHePnll9m8eTObNm1i7969LF68mE9/+tN5521tbeWzn/0sq1evZsGCBXR0dFBfX88111yT1+jk6aefzh7zyU9+kh//+Meceuqp3HjjjXz729/m1ltvzdr5wgsvsHLlSr797W/z1FNPjfnPVFFCPy5EusDpB7szR+jbSmpSIcFQX7gmHptaQm8wTDaRSIQlS5YA2qP/zGc+w7PPPsvxxx/PggULAF16+NVXX83G37u7u3nzzTdZvXo1l1xyCXa7nf3224/3ve99/c7/3HPPccopp2TPVV9fP6g93d3ddHV1ceqppwJw+eWXZ8shg66XD3DssceOW0VMI/SFRLvBW6uf+xr0Y5l59JGcyaR0wgi9YYowTM97vMnE6AvJLUmslOLHP/4xZ56Zvwp+5cqVE21eP9xuN6AnkUc7f1DIsGL0InKWiGwWka0icn2R/aeIyEsikhSRCwv2XS4ib1o/l4+L1RNJtEuviAXw1ILNUXZCnyvuyuTRGwxj5swzz+RnP/sZiUQCgC1bthAKhTjllFNYsWIFqVSK3bt387e//a3fscuXL2f16tW8/fbbAHR0dAAMWLa4pqaGurq6bPz9vvvuy3r3E8WQHr2I2IGfAu8HWoAXReRRpVRuXtEO4ArgqwXH1gPfApYCClhnHds5PuZPANHuPqG32bRXX2ZCrxJmZazBMJ5ceeWVvPPOOxxzzDEopWhqauKPf/wj5513Hs888wyLFy9m3rx5nHDCCf2ObWpq4s477+T8888nnU4zY8YM/vrXv3LOOedw4YUX8sgjj/DjH/8475h7772Xa665hnA4zMKFC/nlL385oe9vyDLFInICcJNS6kzr9xsAlFLfKzL2HuAxpdRD1u+XAKcppa62fr8DWKWU+u1ArzeWMsVtbTqW3tjYOKrjAbj9ZKjeDy61MkN+ugwaD4KLfz36c44zf7jjW5y3+1Z68bM5cDxLv/rHUps0bMblGhmmDFO5THE5MxFliucAuStcWoBlw7Sn2LFzCgeJyFXAVQDz5s0b5qn7My7iEe2GGTl/QG+dnqAtJywvPmwLQCpeYmNGhhF4g2HyKYs8eqXUnUqppUqppU1NTaM+zz333MM999wzNmNyQzeg4/SZ3PpywRL6iD2AbYqVQBiXa2QwGEbEcIR+F5C7lLHZ2jYcxnLsiBmziKTTuqhZrtB7a8vOo5dklBQ2EnYv9rQRekN5U25d7KY6o/l7DkfoXwQOEpEFIuICPg48OszzPwF8QETqRKQO+IC1rTyJB0GlC4S+DEM3qRgJcZKyuXGkp1boxjC98Hg8tLe3G7EfJ5RStLe34/F4RnTckDF6pVRSRK5FC7QduFsptVFEvgOsVUo9KiLHAX8A6oBzROTbSqnDlFIdInIz+mYB8B2lVMeILJxMsqtiC0I38V5IJfQiqjLAloqRFJcW+jJOYDIYmpubaWlpobW1vDLXpjIej4fm5pG1Nx3Wgiml1EpgZcG2G3Oev4gOyxQ79m5gajRbLCb0mcVT0e6+ImclxpaKkbS5UXYXTmU8ekP54nQ6sytGDaWjLCZjy4aiQl+nH8sofGNPx0mJC2VzYCM19AEGg2FaU1ElEMa8XHmg0A2UVeaNPR0j5XIjNgc2NbWEvhRLyg2G6U5FCb3P5xvbCQYL3UTKJxbuSMdJ21xgd2AjjVIKESm1WcNizNfIYDCMmIoK3dx2223cdtttoz9BVuhr+7aVWegmlVa4VIy0XXv0DlLEU+lSmzVsxnyNDAbDiKkooX/wwQd58MEHR3+CTHjGXd23rcxCN9FECrckUA43NrsDOyliyakj9GO+RgaDYcRUlNCPmWg3uAJgz4lolVnoJpJI4SaBsnsQuxMHaWKJqSP0BoNh8jFCn0th+QPQufNOf9mEbiJxLfTkefRTa0LWYDBMLkbocykm9KDbCsZ6+m8vAVHLoxenB7E7tEc/hUI3BoNh8jFCn8tAQu+uhlj/BgKlIGLF6MXhwZ7x6E3oxmAwDEJFpVeuWrVqbCeIdkF1kQW+7qqy8eh16CZOzOm1YvRTK+tmzNfIYDCMGOPR55AMd/HEWxEeXteSv8NdVV4ePQlsLo+O0Ysat76SBoOhMqkoob/lllu45ZZbRn18KtzFu1E3t63amr/DUw3R8vDoMzF6u9OD2PQXskQyUWKrhs9Yr5HBYBg5FSX0jz32GI899tjoDk6ncSaD9ODj3a4o6XROWdUy8uijsRgOSWN3ebFZaaCpxNTx6Md0jQwGw6ioKKEfE/FebCh6lI9IIsWenpwG3O6asonRx6PaLofbi82hhT6ZNBUsDQbDwBihz2CVP3D6dcmDHR3hvn3uKt2UJF36fPV4TNvlcHkRe0boS2+XwWAoX4zQW4S62wGYNWMmAPt6c1r0eaySCPHgZJvVj4Ql9E63F7vVCCU1hWL0BoNh8qkoofd6vXi93lEd293ZBsCsmVroW3OF3l2lH8tgQjZpCb3d5cHmsIQ+NXVCN2O5RgaDYXRUVB79X/7yl1EfG+7RHn1j4wxc9ij7enNj9JZHXwYTsulYBABx+rDZtT2DTsYqBWVUwngs18hgMIyOivLox0KkVxctq6ptoDHgoq03x0vOePRlMCGbjltzB04f9qxHP0Do5k9fhO8fAHs3TpJ1BoOhHKkoob/55pu5+eabR3VsIqh7llfXNVLtddITzRHPMvLoVSIj9H0x+nSqiEff+Q6suwfC7fDPWyfLvCEZyzUyGAyjo6KE/umnn+bpp58e1bHJcBcAdXUNVHuc9OYKfWYyNtOYpISoeJ/QZ9Iri07GbnlSP847EbY+BenyKJMwlmtkMBhGR0UJ/VhIR7oIKi8et5sqj4PeaI6XnA3dlN6jJ6Fj9Dh9OKzQTbpYCYTdr9Bjr+P24EkQ6YC2zZNopMFgKCeM0FvYot2EbH4AqjyOsg3dSDIj9DnplUVi9Ondr7I+3swfdjfoDXs2TJaJBoOhzDBCb2FP9BCxBQCo9jrzPXqXH8RWFpOxfULvy4ZuVKHQp9PQ9gZvqHlsU/uRtjlhrxF6g2G6UlHplQ0NDaM+1pXoJerQIZpM6EYphYjo9MQyqXdjS1ppn04vYsssmCoI3QT3YkvF2aFmkMRBMLCA6tY3JtnS4ozlGhkMhtFRUUL/8MMPj/pYT6qXmEfXoq/yOEmlFaF4ioDb+hO5y6OCpT3VF7rBZgdAFWbddO8EYDeNAHS45lDduX3SbByMsVwjg8EwOoYVuhGRs0Rks4hsFZHri+x3i8gKa//zIjLf2u4UkXtF5DUReV1Ebhhn+8cFpRT+dJC01V3Kb4l7OJY7IdvXTvDnq9/ic79el1/hcpJwpKIkxalF3ipTnC4M3VhCXzf7AAJuB7ttM6Fru148ZTAYph1DCr2I2IGfAmcDi4FLRGRxwbDPAJ1KqQOBHwH/ZW2/CHArpY4AjgWuztwEJoIbbriBG24Y+b0kHE9RRRg8tQD4nPbs9ixW6EYpxX+sfJ2/bNjDhncnP93SkY6SsHn0LxmhTxd49F1a6F0N+zOrxsP2dBMkwhBqnUxTizLaa2QwGEbPcEI3xwNblVJvAYjIA8C5wKacMecCN1nPHwJ+IiICKMAvIg7AC8SBCYt/rFmzZlTHdfSGmSsRbL5aAHwuLfSRRIHQh9vZ29NXA+fVlm6ObK4drbkjRimFMx0lac8IvRW6KYjRp7tb6FV+6urrmR3uZku3DuHQ+Q4EZkyavcUY7TUyGAyjZzihmznAzpzfW6xtRccopZJAN9CAFv0QsBvYAdyilOoofAERuUpE1orI2tbWyfc6u6yCZk5/PQAe18Ae/Za9fROyb7eFJs9IIJ5K4yFOqlDoCzz6eNdu9qlaZlZ7aKpyszmqSy/TtWMyzTUYDGXCRKdXHg+kgP2ABcBXRGRh4SCl1J1KqaVKqaVNTU0TbFJ/gl1a6D0BLYiZ0E2kiNBnyhe7HDZaOsNMJtF4Gi8x0nar+uMAoZtUsJUOqmgKuKnzudgatRZ89e6eTHMNBkOZMByh3wXMzfm92dpWdIwVpqkB2oFLgceVUgml1D7gf4GlYzV6vAl1a6H31egQh8+lBbRf6CbWmy1ffPTcWnZ2RCbVzkgihYc4aUd+jL4w68YWbqNNVVPtdVLrdbI37kI5vNC7Z1LtNRgM5cFwhP5F4CARWSAiLuDjwKMFYx4FLreeXwg8o5RS6HDN+wBExA8sByYsobu5uZnm5uYRHxe1Klf6a7XQe136zxKOF2TdJEK094bxuewcOCPA7u7JFfpwPIlXYlq0oU/oCzx6e6SdDlVNlcdBrc8JCKnAzLIQ+tFeI4PBMHqGnIxVSiVF5FrgCcAO3K2U2igi3wHWKqUeBe4C7hORrUAH+mYAOlvnlyKyERDgl0qpVyfijQD8+te/HtVxmcqV3ioduvFmPPq80I1eNdvb3UlTlZuGgJuuSIJUWmG3TU6994xHr5w+vcESenI9+lQSV7yLDqoJuB3U+lwAxL0zcQT3ToqdgzHaa2QwGEbPsBZMKaVWAisLtt2Y8zyKTqUsPC5YbHu5kalcKd6CGH1h6AYI9XTRFGikwe9CKegMx2kMuCfFzmgiRQ1xvVgKik/GRvRNq11VUeVxWh49RFyN+HpNYTODYTpSUbVurrvuOq677rqRHxjt0o/WginvQFk3QCLSTZ3fRb1fe8odoclr4xeJp/FJFCn06HOblof0fEO7qqHK46Dao4U+5G4qi9DNqK+RwWAYNRVVAmH9+vWjOk6i3SRx4LAE1O2wIaI96CyW0KtYL1UeR2mEPpEiQIR4pj5+VuhzPHprUVS3VON22PC79U2r19mgm5vHevvKLpeA0V4jg8EweirKox8tzkQPEXsg21tVRPA57QUevRZXezxIlbtEQh9PECCKzWsJdbEYfVj3vo25avX7sOYbehxWMbHe0sfpDQbD5GKEHvAnOok46/K2eV2OoqEbeyKYF/vujgzQr3UCSEaC2ERhs0JMmRg9KkforS5YKXd+3Z4usd5faN+k2GowGMqHaS/0sWSKGtVF3J1fPtfrshHJS6/UQu8jTMDjyFa1DEYL6sxMIKmIFnGHtzB0k3NDyrQ7tG4GmXIOHTZL6ING6A2G6UZFxegXLVo04mM6QnEa6CHpOyBvu8dhJ5bM6bNqCX0VEQJuB36XAxHye8tOMCmrTLLTl/Ho9eWT3Bh9rIc0Nhweba/TbsPlsNGmrJtDiYV+NNfIYDCMjYoS+jvvvHPEx7QH48yTbjr8jXnb3U5bvtC7dB59gAhVHgc2mxBwOeiNTZ5Hr6K6zk6h0OeHbnoIiY8qK9sGIOB20Ja2umSVOHQzmmtkMBjGxrQP3XT29FItERxV+VUd3Q47sWROSMRmJ+XwExAt9AABj2NSQzeZevi2TNaN6LCM5IZuYj304s/G5kGHb0JxwN8EZbBoymAwTC4VJfRXXXUVV1111YiOCXfq3HJXzcy87W6HjXiuRw8knQHLo9fecqbl4GQhcatyprVKF5uNNDakwKMP4sPj7Lu0fpeDUDwJ/hkQLG1N+tFcI4PBMDYqKnSzZcuWER8T7dJC76ubnbfd7bARLAjLJOw+qiSSnYgNuB39xkwkEgtaxvXlwafFjqTyJ2N7lBe3w57d5HNbqaKBGSUP3YzmGhkMhrFRUR79aEj06FCGr25W3na3w04ske/Rx+x+AuQIvcc5qZOxtkRG6Kuz25TYEJUfuulWPtyOvkvrc9kJxZJa6E3WjcEw7Zj2Qp8pGWAL5NfBdzls+TF6IG7XMXqPVQunyjO5k7H2uCX01sQwQFoc2FSqr39ttIfutBd3TujG47ATTaT7hN70jjUYphXTXugdYS30hS32isXoo3YfASLZWjhV7smN0duTQeK4wOHKblNix06KRFrbqmLdlkffF7rxOO1Ekykdo0/F+nLtDQbDtKCiYvRLliwZ8TGuWBtR3Hhc/rzt/dIrgYj4qZYIHiss4nc7dEhkknAlQ0RsPlw527TQp0mkFG67gmgPvXjzQjdup02HoTI3s1AreGsnze5cRnONDAbD2Kgoob/11ltHfIwn3kmvow5PwXZ34YIpIGzzsR8RHHYtoj6XnUgihVIKkYmvSe9KhYjZ829IShzao0+mwZZAUISVhxpngUefSPUJfXAvNB404fYWYzTXyGAwjI1pH7oJJDsIO+r6bXcXidGH8RIgko1xe112lELHvycBTzpEvFDobXYcpEmk0pDQHa8iuPM8eh2jt0I3kDch++y2Nm55YjOptInbGwyVSkV59J/4xCeAkXUxqk51EXXP7bfd7bCTSCnSaYXN6iAVwotDLEF1+bINSsLxZDZuP5F40mESjiIevaSIp9KgdLPyCK78yVinjWgyDQFrrUCoL5f+yyteYU9PlGP3r+O9h+TPU0wEo7lGBoNhbFSUR9/S0kJLS8uwxydTaeroJuFp6LfPZXnE8VSftx7EavgR0wuXMiWA86pcTiA+FSbpCORvzHr0KuvRR5W732RsKq1IuGv0alprdWx3OMGenigA/9zaNinvYaTXyGAwjJ2KEvqR0hOJU08vaW9/oc+EPnJz6XvSVgs/S+gzXnxey8EJIpFK41dhUq58oVc2HaNPptKQyPHoHfkePUA0qawyCDp081ZbMDtmy97eiX4LBoOhRExvoe/ch1NSSKB/yCIT+siN0/coa8rWqjmTKQEcKeLR72gP840/vEZXeHwak4TjKQISIe0s6A5lefTxVBoS2juP4urn0ev3ks5bNLWnW48/aEaAd9pD42KnwWAoP6a10Ie7tOAVLpYCskKZm3nTnc4Ifb5HXyx08/0nN/Ob53ew4sWd42NrLEGACKqwDaDl0evQjeXRK1e/BVNAX+aNVQZhtyX0yxbWs6szYiZkDYYKpaImY0844YQRjY92a8Fz1RQT+v4efVcqX+gzMfpIon8u/aZ39aKkl3Z0jsimgYhEwrgkhbgLY/QOHKT6Zd14cjz6jOjr1bEzYd/rAOzpieJ22Fg0s4q00rX5m6rc42LvQIz0GhkMhrFTUUL/ve99b0TjEz06+8RbUyR0kxX6HI8+ZYlgVuiLe/TJVJodHdq7fnNfkPEgGtQ3DvFU5++wObCT0Hn0lkcf7Zd1k+PR+5t01o1S7O2JMrPawwxL3Ft7YxMu9CO9RgaDYexM69BNKqgbaftrZ/bb53b2D920Fwi911lc6N/tipJIKRoDLna0h/vq0IyBeKgLALu3v9A7sNIrMx69Ksijz76XlPboU3GIdtEbTVLtddAYsIQ+GBuznQaDofyoKKG/4IILuOCCC4Y9XlkFzarq+3v0Lnv/rJuOREboB5+M3durY99L5taRTCvaQ2OfkE2EtUdv99bkbRebA7ukSeamVxZOxjpyQzd9i6aC0SQBtyPrxbf2TrzQj/QaGQyGsVNRQt/e3k57e/uwx9si7YSVG4cn0G9f0aybhI2kOIfMo2+3POPF+2nve6+Vqz4WEiEd63cF6vO2i92eE6MfKL2yIHQDWuhjWugbLI++fRI8+pFeI4PBMHaGJfQicpaIbBaRrSJyfZH9bhFZYe1/XkTm5+w7UkTWiMhGEXlNRArLypQMR6yTbqkuuq9YjD6aTOsSBFa5YI/ThghE4vmTsRkPfvFsfe5MGuNYSFlC763KF3psTuyZ0E0yJ72yaIw+Z3VscG9W6P0uO3ab0B2ZvNr6BoNh8hhS6EXEDvwUOBtYDFwiIosLhn0G6FRKHQj8CPgv61gH8GvgGqXUYcBpQNmoiTveSdBeU3yfFfrIlCpOpxXxZJq4w5/16EUEr9Pez6PvCGqhPyzj0feOXejTEUvoa/KbmNvyVsaGSdrcKGzZ0BPkLJjKLWwWaiUUS+J3OxARarxOI/QGQ4UyHI/+eGCrUuotpVQceAA4t2DMucC91vOHgNNFl3P8APCqUuoVAKVUu1JqcuoFDANvoouwYyChz/foo1YIJ+kIZIUedJw+XLAytj0Up8rtYEa1Dol0hcdBQK0a8r6q/AJsYs/k0evJ2ITNjcMm2QqbkOPRJ1PgqQWbE4L76LU8esAIvcFQwQwnvXIOkLvqpwVYNtAYpVRSRLqBBmARoETkCaAJeEAp9d+FLyAiVwFXAcybN2+k7yHL6aefPqLxgVQ3Hd7ir1eYR5+pUJl05gu912UnWhijD8WpD+gJUZ/LTsc4TMbaot0kseEoWDAldkdO9cowCXFnhT1D34KpNNhs4G8i1buXeDI96UI/0mtkMBjGzkTn0TuA9wDHAWHgaRFZp5R6OneQUupO4E6ApUuXjjoX8d///d9HNL5adZNw9S9RDDkrYy2Bz9SzSTsDEOvKjvM5Hf1CN52hOHU+3R6kzueicxzKINjiPQTxU1tQ917sDmyZ0E0qQVKceROxkLtgyrIzMIN0ry5s5s8R+vEq1zAYI71GBoNh7AwndLMLyK3j22xtKzrGisvXAO1o73+1UqpNKRUGVgLHjNXocSERxU+UZJGCZpCbdWOFbiyRTLn6e/SFoZtgLEmVRwtond9J5zh49M5ENyFb/+wgm93Zl3WTjJHA0V/oswXa+oReWUIf8OQIvQndGAwVyXCE/kXgIBFZICIu4OPAowVjHgUut55fCDyjlFLAE8ARIuKzbgCnApvGx/T+nH322Zx99tnDGhvr1atilbe+6P5sHr0VusnmyrurIdKVHed12vtl3QRzYt/aoy8uoPc/v51rf/OSFukhcCV6Cduq+m232XUefSKZ7vPoC0I3IoLbYdWkB/DPQML6/U926GYk18hgMIwPQ4ZurJj7tWjRtgN3K6U2ish3gLVKqUeBu4D7RGQr0IG+GaCU6hSRH6JvFgpYqZT68wS9FyKRyLDHhrv24gbEX9yjt9kEl72vb2xG8NOeOoh2QVrHu30uO3t68gUyk80CWuh3WuUQckmlFf/2hw0AXHhsM6cdPHjTD2+ql1hhLXr0gqmsR5+KEy/i0UNOO0GAwAzs4TaEdF7opieSyGu0MhGM5BoZDIbxYVgxeqXUSnTYJXfbjTnPo8BFAxz7a3SKZVkRtSpX2otUrszgdtiy6ZXZFbLeOlBpiHWDtw6v1Tc2l3yP3ll0Mvbdrj7BW7e9cxhCHyTomd1ve2YyNp5SkIoXDd2A/uaRK/Q2laSGUJ5Hn1YQjCep9jgHtcVgMEwtKmpl7EiI9ejyB86qxgHHuHL6xmbSK8VnfQMIdwBWg/CcyVilFKEcoa/1ueiJJnVjkBzeauur/77x3Z4h7fWrIElXkcVdhR69cuSVP8jgcdr6ettaufSN0p21s9qrH3uj/StxdoTihGL9txsMhqnBtBX6RHalafHQDVgNwhP5Hr34rZh+ZgFTwYKpSCJFWpETutHecU+BgL7dqlfXHj+/fuimH0pRpYKk3EVy/m26Z2wyN3TjLB66yX7zsJqEN0l3djI248X3FMTpg7Ek7//h37no9jXjUpzNYDBMPhVVpvjDH/7wsMcmI3oBkr+6eHol6AqWhQumHP58j97rcuR59EHL880IaK2VZtkVjlPvd2XH7eiI4HPZOWpuDb9asx2lFCLFY+OJSA8uSen5gUKyHaZ06CamfEU9eq8rN3SjyyA00U3AlfHoiwv982+10x6K0x6K80pLF0fPG/jvNRxGco0MBsP4UFFC/9WvfnXYY1ORXlJKqKoqXusGLI/eEviMR28PWEIf6QvdxFNpkqk0DruNUEyPD7i12NZYHn1h6mJbMMaMKjdz633EkmlagzFmVBUvAxTuaqMGoFiGkC1nZWwqQVzZi3r0+TF6PS/RJF34LTsz6aCFoZvcximrt7SNWehHco0MBsP4MG1DNyrWSwgv1V7XgGO00Ofn0bsCVkw/J0YPfQuqgpZQ+i1PudbylLsLUizbQzHq/S7m1vkA2NkxcDZKKNPysFiGkM2BPbMyNhkjlh446yYbuvHUkhQnM+092VIJ2dBNNN/Ot1pDLGzyc+CMAK/t6h7QRoPBUL5UlNCfdtppnHbaacMbHOshiDdb8KsYrtysG+vRVVUHSNajz5QbyIRvCkM3mRWyhatj24NxGgJuZtdqL36wUsbhbp3z7i6sXAlZoU8mk5BKEFX24qEbZ86ksQhBRx2z7X2TwAN59G+3hVjY6Gfx7Gpe3z30pPFQjOgaGQyGcaGihH4k2OJBwuIbMC4OugxCn0evH91OJ3hrB/ToQ9EEZ9peoKlX92WtzYRu+nn0cRr8LpoCQzf9yGQIFWt5iE2/fiqVglTcEvqB8uhz2iLa65hhyxX6/jH6dFrxdluIBY1+Dp5Vxa6uCL1Rs3rWYKEU/PHz8P0DYecLpbbGMAjTVujtiSBRm2/QMXkx+mQKp12w20THyiP5Qp/JvGnc8hvucN3KgY+eB53vUOVxIpIfo0+nFZ2hOA0BF3U+F3abDCr0iaAWen9dMaHXnngqmdBCny6edeN12fpi9ECn1NFAXyjG5bDhcdrozaRR7nkNbjmIX9tuZP8qmN/gBwYPMRmmGS1rYf39ugfx4zeU2hrDIExboXcmQyTs/kHHuJ196ZXRRDpbBRJfA4R1lyRPQd/Y+VvvY4+yJixf+Dl2m1DtcdKdE7rpiSZIphUNfjc2m9Dgdw0q9JmmI7WDCH06lUCl4kTT9j47c/A48hd2tVNDverMG1PtcfZ59I/fgITbOM62heP2rGBevb4p7iiyytcwTXntd+Dwwnu/CbvWwr7XS22RYQCmrdC7UiESjiGEPid0E0um+jzlwAywioJl2glG4inoeJva0FvckfwwqYWnw4bfQzpNrS+/YFh7KE6ztHLe2svg56ez2NdN22Bt/CId9CovVX5v/32iRT2dTForYwfIurFW8OoSRNCqaqlJd+lSDhZVHoeejO3cDu/8g+1HfYm/p45k/jsrmFerQ0zFyjnw4i/gR4fDY1+ClFlYNW3Y+RzMPQ6OuUz/vnnl4OMH44Wf68/Q49/QISHDuFJRQv+xj32Mj33sY8Ma60mHdW35QcgN3UQT6b5Jzuo50LsbKIjR73wegGfV4diPOB9634WWF6n1OvNi9O3BON933EF1cBvs28T10R/ROkgXKnu0kx6pKl6DJuvRx5F0krhyDrAy1o5SfZPKe9NV2ElnQ1Cgc+l7o0nY9gwAbzaewe9Sp+IO76Zm7xpqvE62dxQs7tr5Ivz5K+Bww9q7YdX3ir+JPa/BvefwsVk7+djyeQP/MyfjsGcDBFsH/HsYyoB4SF+n5uOhahbsdwxs/svozrXjeVj5VUDguZ/C83cUH9e+DVZcBvd8GLY8Mfg5wx3G6cihcvLoE1E+f/GZIAKd7+ivlL56sBev2+JVYdKuwYU+P+smx6Ovng2xHogF8WZj9EnYt4GkuNjrnIcc/B6wu+D1R6nxnZvn0Sd3vMCJ9k3sWXojsxpqOeTPX2ZB+gXg5KJ2OGKdBItUrgSyk7E2q1/sYLVuQK8H8DjtvJu01g/07gG/ThmtyoSYdr4A/ia2q1n8NX0syl2NvPJb5tZfRktnQYx+1XfB1whXr4aV/wr//CEsOhPmHt83pn0b/OqjIDY+/7658O4jehLvnFv1DQK08G94GP7ydQi3AQIHng7LPwcHnK6vK+h/4Dceg02P6OczFsPC0+CA92bfB5Eu2PMqvLseunaAuwqq99M36Jo5UN2sr1/nO/qGHe7Qdnhq9WfG1wCeai1mkU6IBXWfYKX0uTzV4PDov73YIJ3SPyoF6aT1PG3tt8aA3qbSgLKe59zsskkBxZIDlDVW5RwzEq93mEXqBklMyFLdDE2L9I1bpaB5qd5+8Nnwt+/qb7pVM0dgG/DMzXq19r88B7/7FDx1Eyz6ANQv7BsT3Af3nqOvha8OfvMx+NAP4bjP5J/rrb/D49fDvk3groGln4JTvqqvW4Z0Gt5ZDRv/oM830/oMzT5aN+bJ0N0CLS9CqA08NdB0CDQuAqe13kUpva97p/6c2F36s+Nv0p8jm12/VrxXd4iL9gDWZ8htfYYcHv2ZiVufMZtT68s4UzlCv3cj4Z+9FwCfM+cD667RXy3P/I++bek0PhUF1wDimTk0L48+x6Ov2k8/9u7G62wGrNDNng3scc/Hk3ZrMVj4Xtj0KLUzL2BHTpmD5k130q182JdeDrVV9D75PS6OPUw6/eWiXrsr0U3UWVvcSMujd6S00A9WvRL0N48anLwZb9RXv+MtmHU4ANUeBy0dYf3NZO4yWkNxlMMDh50Prz3IAc2XsbE9R+i3r4Ftz9B7yk3ctXoXSw/6Cu95++/wh2vgmn+Cy6f/Ee6/UAvbp58g7JkF//whvud+oIX2Y7+CVEzfJDb/GeYshbP+E9rfhHX3wK8v0P9gi87UHuTbqyGdgLoFUDdfH7P+14Do35Mx/U0q+8ZrtGCnjXc3Ljg8cMMuaNuif286RD8e8mH423/om3Ch+A5Gy1p45x9w5nfB5YcP/whuWw6P/h+4/E/65pOMw4Of1DfkTz8OMw7Vnv2fv6yv9wmf1zfXVf8Jq/9b3yDOuAl2vwr/eyu88lvtMMw8XAv3Kw9A13Yttt5a2PAQPP0dnWQx/z2Agl0vQ09Lf3vFBrXzwO6G4J5si8+i41wBy0EYugx5lsMvgAvvHv74YVI5Ql83nw8+oUV31c//DRJhfZfd8Rys+QksPjfrZcYiPbhFIZ6hhD4/Ru/J9egBet7FN2t/wJqM3buBna5jCWT+rIs/Am8+waGz3mJ1xPKg27Yyd+8z/CR1LtfU1YPdxhv7X8qJW/8fXS0bqZ13eD87vMkeglVzihtpCb09pQU4gaNfPXrQWTeghT6RSrMlOcMS+m3ZMVUeJ7ZIGwS3wbGX07orRlPAjSy5FF66hzPUGp7oPKKvXMOq74J/BtdtO4ant70JwIr338yyf1wBf/13OO0bWqh73tX/tI0H8kErh37Vj+/SXv0PDtb/CA4PvP9mOOFfst9SOPmrsPH38Nxt8OxPoH4BLLsajrgQZi/RIpBOwe718OZT0LZZe0RNi2D2UXqMv1F7VaFW/Y/bvQt6dul/wrr52tP3NUAqrr8JRDr0RHu0W4/x1mpBcAX060W79beBVEL/qLS+BjZ7nwdvc/TZptLa80X0P79kHm30edqWdz5YbFrEOof0HTccD3zY8e5hjFv/G3jpXv2e2rZosau1WnHOOBQaDtTftEYi9Gt+qv++x3xS/14zB97/HXjsOnjpV3r7yq/AjjVwwV2w3xI97mO/gt9fCU/cAK8/qq9Z2xZY8v/Bh34ATms+a/nntIg/dZP+XWww/2Q4/UZ9c3J6tDOy7W+w7Wn9OjannnuYe63WjJq5+vz7XofWN6DtTX1N578HGg/SfwNvvf4Mhdv1T3Cf/py4q/Q3RU+Ndv6QbDSAZAQSUf15cQf0Z6zhwOH/7UZA5Qi9vyFblZGj/7++7fEQ/PdC/TXNEvpQTxduwO4ZuPwBaI8+lVYkU2liiXSfp5zr0c/ToiShvRBqZVvtAvwZz//gD4LYOTa4mu7ImbrW+//+iKQ4+b3jQ3zBWpXasegi4m/+BLX2lzDvB3k2pNKKqnQPrQM0SOkT+sE9em/Owq5QLEkQHxFXPd72PqGv9jo4MP667jowdxltm+M0Blz671a/kOO7VhJNLKYtGKepdQ28vZrE+/+Dv/85xKdOms8Lb3fwlbUJVh9/NbYX7tCTtDYnfPz+/FAOaLFuOkSHa5w+OPJjULd//hiHC476uP4ZCJsd5hyrfwYcY9PhhKqZg4/LfH4MxXnnn9YTBa1btChlbsoi2pn6560Qatf/j0PRtVPfGJZ/Lj+0cszl+nPx+A36//atv8HJX9GfmQxOD1x0r75RvPqgDh9e9G/ahtwb4Nzj4YrH9Gv1vKu9/cLS5P5GOPIi/TMQgRn6ZjZFqZjJ2GgixTvtIWLJNHt7omzZ26tXqbr8+mJvfzY7NtSj0wodviLVIHPIbSeoPfrMZGzGo9+Fy27DbhMCXZsBeFPmZytX4quHg97Pkfv+SJUKEdr5GrzyAM9WfxCp6hOV6ob9eCJ9HIE3fgeJ/Bh4e0+YasJ9NXYKsf7RHOm+GH1hc3DoC91Ek6ns6t2gf38dusnY4XFylNqMsjlh9hLaemM0Btz6H+f4q5jZ9TJn2V5k957dsPJrUN3Ma7MuIJlWLF/YwHVnLKKlM8Kjs74AH/0ZvOdLcPXfddilGLMOhzO+Bad+rb/IG8qbti36m1Muh52vPd2X7xveOVZ9T39+l12dv91m05+fOcdA62Z43zd1CmchNjuc9H/gc/+ET/8FDvvowN9yaufCvGX9RX6aUDEe/Ss7u9jXE2NPd5Rl3+3rPf6hI2bzPzMOx77ubv1V2mYnEuwCwDmE0Pe1E0xbMXrrvujyaw+iczsigtdpp6Z3KwBvpJupduf8Wd/7DVxvvpffu76F74EYeGq533MJDdJXY6epys3/pE7nnPhzsPGPsOSS7L49+/YwQxSeqgE+oJZH77SEXtejHzhGH433CX20aj60r8mOqfY4WGR7k+TMI3E6PXSE4hw+x/rWs/QzRNfez22t/4/0A3eCSsAnHualXfrGdPTcWhoDbhbNDPCTVW9xznWX6MVlhsohI6LJqI5xH1mQ4TbrcD0v9ez/wFGXDDwpq5TO0Fp/P5z0xb7wTy61c7UnbhgXKsajX7awgaOaa9m/wc/NHz2c/7nkaK58zwL+/NpuHttT2/fhBKIhPYHi8Q8RuslkqiRT+R49QMMBWW/Y67JTF9oG/iZ2J/z4XTnjZh/Fayf9hCAewnWL4fI/8U7US4PfnR3SVOVmTXox3d558OoDeTbs3avTOGsaBggrZIQ+Nbysm0gilW0ikqg7QE8oWT1wa9yKo2QboRnHoJSiIxSnPmOnw0XyE3/kx6nz2Dzzg/CZJ2Dhqby8s4s5tV5mVHuw2YTrzljE1n1BHllf2D/eUDG0b9Nx+sZF/fed9T2Ih+G3F2fLhOQR7oDfXa4nUg94X3FP3TDuVIxHD3D1Zz8NwGXLdRjgI0ftRyyZ5sEXNnCuCz0RV7+QeKgLAE9g8JK7GcGMF3r0oGN9b/0d0Ln0TZG3YdYhhFv6+sVmSB50Fhc8VcO9px3PqTObaA/+laXz+zz6ao8Dl93OGzXvYdn2h/S8gksv5mrbp4W+oWlWcSMtoXepKEgmRl+8Hj1ooXfE9PtIzThM79y7Aea/h9mRN/FIgt31R+OIp4in0tT7+9JTA7WN3OO6hH1Ns/mPOUcAsH5HF0fPq82OOeuwWRw+p5rvPLaJUCyJ22GnLRSjMxRHKWg65kyWLRhgvsFQ5lgefZueeKfxoP5DZhwKF92js2R+cQacd4ee2EzGdLbL09/WE9pn3AQn/p++GL9hQqkoob/iiiv6bbvy5AV8+nlL0Ht02l0irIt5+aprBz1fRjAzMfo8Aa0/QKdtxcN4HTZmhrdD00kEt/UX+hpvX/ORVFrRGY7TmNOERERoqnKz3nU0y1K/0XnsB+hU0Z52LfTO6iE8+nQMbJmsm4E9+mgijWClGs7SYs0eLfQzu18DYG/NUdiDumRDfc43D4C59T52Wrn0+3qi7OqK8KmT5veZYxN+eukxXHnvWv79kY3Z7R6nDZsIYdeRLFs0MZkFhkmiTc9H0VBE6AEOPgs++Ud4+Eq46ww9aRtq1QI/dzl86Ja+z55hUqgooW9r08W/Ghv7+sDOq/ehAvtBgmx+dSqqhT5QrOxvDhkPPpbQHn1eSeP6Bfqx822aHV14VYhU48FEE+lsLfoMmQqW3ZEEXeE4aQUNgXwBbaxysz5lLRDZ/UpW6EOW0OMfKEavBdytdAmFgUI3GfGPJFLZloDeuv30efdoga9te4kW1UibrR5POCP0+QvOmuu8vLGnF4CXd3YB9GtGsn+Dnye/dEp2cVVjwJ39RrHwS7+lt6vIV3pD+ZOJ0bduhpp5ep3EQOx/Inx+Dbx0ny6V4D1RZ8TkLn4zTBoVJfQXXqjTr1atWpXdJiIcOHcWwbd8BHq0aKajWqg8gcFj9C5LMKNWjD7Po284QD+2b+NA2anH1R0EhLJdmzLUePtKFXeEMgKa3/CkKeDina60npjavR7QXahUcB84GUTo9SX0Sd9kbLGsG2/OZGzcptcGBDxOvYhkz6ugFP69L/J0+mCi0SSdlp2ZevoZmut8PP36PpRSvLyjC6ddOGy//n9HEWFufX8h2P3773Hf0x6+fbEpaztladtSPGxTiKcGTrwWuHbCTTIMTsVMxg7GgTMCdKgAaaviJLEeYjgRh3vQ4zKecTCWJK3I9+gzCxta32BBWgt9b7XeVhi6cdptBNwOusIJ2qyQSEOh0Fe5dQXLWUfqFX3Aw+taaJRuUq6avlIBhVhC70Wfd6j0ytzJWL/bob9Ct74BrZuxh/ayNn0wPZEE7aGMnfmv21znzbY+fHlHJ4tnVxd9vYEwvtxUJhOj3wJNB5fWFMOImBZCf0CTFvqo1cBDYrrpyFBksm4ypXvzBM1dpcX+3fUckNhKqzQQtNUC/YUetFffFYlnq1Q2VhWEbgJuOkIx0k2H6NIAyRiPvvIuB/kj2KsGWchjCb0Hfd44TjxFQjdOuw2nXYgkdHql22HDabfB/ifpFX1PfweA59Wh9OZ69AWhm0zrw+3tYV7b1T3mHrKGKcpwPHpD2VBRoZuBOGBGgC5VRaJXC70t3jtk0xHo8+gzoud1FXiu+x0N259lfgxe40DqM55y4Th0nL47nOgT+oIYfVOVm7SCYNUBVKsUrds3sfHdHubPDA++YtOK0XtF25gWR7YPbCEeh24QHkumCWRuRgtO0QXgNv8Zmg6htW0+PdEEKaVw2qVvnEVznV5a/uTGPYTjKZbMrR3YNkPl0mg8+qnEsDx6ETlLRDaLyFYRub7IfreIrLD2Py8i8wv2zxORoIh8dZzsHhEHNPnpJIBErBWxySAJ+wiE3ioxXDjJytxl0LOLxvguXlQHE4rnhEQKyNSkbwvGsNsk2zQ8Q6alYLtHp4Zuem0dAI3SPXB8Hvpi9JZHb3MOHI7yuLTQh2LJbE9bXD698tBVBaffSLXXQW80SUcwTr3f1a/V4oJGvU7g5/94G4Cl841HP23I/SyY0M2UYkihFxE78FPgbGAxcImILC4Y9hmgUyl1IPAj4L8K9v8QGGWx6uHzuc99js997nP9tld5nMScNTjjXQA4k2ESjsFLFENf6CbT2LufR7/4o7qWC/CnxPGEYrp2fb8bAlDrddEVjtPWq3vFFlapzIRydjl0Ybb2t19jXr0PZ6R1eB59RugdrgGHep12wvEUwWgy38YTr4UbdsIhH8p2meoIx/tNxAI47DaO2V+L+5xaL811Q98wc6k59oMs++AgtWsM5Y+vsa8ktGFKMJzQzfHAVqXUWwAi8gBwLrApZ8y5wE3W84eAn4iIKKWUiHwUeBso6Fgx/lx88cUD7rP5G/D2hiCVwJ0OkXYOUA0yh4Alhpk2f75CoQ80wWee5MG1O9i5xkm3FcsvzLoBqPHp/W3BWL+wDfR59HsjdtLVzdg7t/LBYwLI+h5dR33AN2bF6K3QzWATzH63g1AsSW80SZWn4NJb3lqmy5RS/TODMlx58kKef6sjL39+uFQvPpUjR3GcoQxIW60o5y0vrR2GETOc0M0cYGfO7y3WtqJjlFJJoBtoEJEA8HXg24O9gIhcJSJrRWRta+voOwvt3LmTnTt3Ft3nqdYeSLi7lYAKo9yDp1ZCn2Dvywp9kfvinGMINx4FwJ7uiHVcMY/eaWXdxGgI9BfQJsuj39cbo8M7n4W0cFaz1We2du7ARmazbrSNdufAHr0W8SS9sSRVnuINWao9ustURyhO3QBCf+qiJjZ+50yuPHlh0f2Dkehppat194iPM5QBGS9+6adKa4dhxEx01s1NwI+UUsHBBiml7lRKLVVKLW1qGn11ucsuu4zLLrus6L6qOl1gaeeuFgISQYYoUQw6TOFx2thntfnr59Fb1Fohjp0dWuj7ecvoGH0yrdjWGmJmtafffr/bQWPAxdttQd5IzuQA224O91tNDWoGE/pM6EbfFOyDxOirPTr+3htNFLVR265DN3t7osyoGvhczgEmfIdizyO38Ltb+k3zGKYCR10K170GB55RaksMI2Q4oZtdQK7SNFvbio1pEREHUAO0A8uAC0Xkv4FaIC0iUaXUT8Zq+Eipb9RCv2nbdj5EmJB/8MqVGQJuJ/t6tLdcLPYOOiwDsL0jhNMu2cVJea9v5aMHY0nmFVlIBHDQjCo27w3yj84G3kMMWp63XmBojz4TunEMIvRVHie9UV2+eSChr/M5ebdb39hmFbkhGaYxdkfxSpOGsmc4Qv8icJCILEAL+seBSwvGPApcDqwBLgSeUUopcpqgishNQLAUIg8wc5aONu3YvhWXpPBVD6+wVsBtz6ZE9puMtcisfN3RHqbG6+yXqQIwv6FP3OfWe4ueZ9HMAPeu2Y5N5oAb3V3J4YXAID04c0I3SWXD5SoekoE+j34woZ9T12fbrBoj9AZDJTDk928r5n4t8ATwOvCgUmqjiHxHRD5iDbsLHZPfCnwZKLvv5jNm6mYhsX26tHBVzfDSAgM5gjhg6MYS+ne7o1R7iwvt/EZ/9vmCxuIZP+85SIetNqn9UWLXja1nHJrfsLiQnJWxCRxFv01kqPLoCeFUWhFwF7czN4vGePQGQ2UwrAVTSqmVwMqCbTfmPI8Cg/ThAqXUTaOwb9ywWa3N5sk+ANz+YQq9NbHqsMmQMXqgX358hga/i1nVHvb0RDl0dvFete89uIlLjp/HUc01yLpDdfngTI/MgbBi9D6J0aN8A06yQv7cwUAefe63jUUzB++pazAYpgYVtTL2K1/5ysA7nT6S4mKuWFk9w5iMBbKeb62v/+KhDNU5olkzgNCLCL/57DJSaVW0Xjzoyd/vnW+Vb/V+Cf50nW52PBi2vteO4xhQwIG8m8BA4w6aUUW930WVxzFg1s1YqF9+PicdPD3buRkMpaKihP6cc84ZeKcI4q/n0Fi7LlnsHp63mhHETKnhYjjsNmq8OizSNEimysKmoRdpZTniQjj8gqFLuuYIfQJHv5IFueSKe22RxVAAdpvw6LUnDTjxPFaqD17OIcvMhJ7BMJlUVFGzzZs3s3nz5gH3230N1Cf26F+GkUcPZFMMBwrJZJhtTVzOrik+0ToqhlO3O1folX3Q0E1uxczGIrn8GZrrfBPizQPE2lrYt/PtCTm3wWAoTkUJ/dVXX83VV1898ABfTqaNZ3jplfvVauEeqtF1xlvev2FkJQHGjPSFgRI4qBvkm0fmvUDfStzJZs/K/+HRnw66fs5gMIwzFSX0Q+LNmYAdrFBYDpmc9wNnDB52+cL7DuLwOdWcfNAkx59tNhCrty2OQVMic/cNVN7AYDBUHhUVox8Sn868wekfvA1aDu85qJEbzj6E844evDbOKYuaOGVRiSYZbQ5IxbE73Rw9d+BsIo/TTlOVG7vIgKWMDQZD5TE9hd43vMVSoJf6X33qARNk0DhhCf2i2fUwSOgG4Kkvn8oQUSiDwVBhTC+hr7eKcCVjpbVjvMlMyNqHDscMlP5pMBgql4oS+m9+85uDD5hl5agvPG3CbZlUrEVT2MtfxJtOvoRTFg1SX99gMIw7FSX0Z5wxRFW9WUfA+b+ABScPPm6qkcm8GYZHX2qqFh7DAUsGKdJmMBjGnYoS+vXr1wOwZMmS4gNE4MhBKzVMTTKe/BTw6CN7tvHuW0F0szKDwTAZVJTQX3fddQCsWrWqpHZMOpmuUlPAo9/9xO08/qwbvnhBqU0xGKYNJseuEnBY+fFTQOgNBsPkY4S+Esh69OUfujEYDJOPEfpKYAp59CaF32CYfIzQVwIZgZ8CQm8wGCafipqM/e53v1tqE0pD1qMv/9DNrNM/xamlKhVhMExTKkroTzzxxFKbUBqmUNaNf95hNB+yX6nNMBimFRUVunn22Wd59tlnS23G5GNVr8Q5jrXwJ4jQjo3sfGN9qc0wGKYVFeXRf+Mb3wCmYR69SuvHKSD0e565h2eed8HXP1FqUwyGaUNFefTTFqX04xQQeoPBMPkYoa8IMkI/yd2tDAbDlMAIfSWQTunHKeLRq1IbYDBMM4zQVwKZGL2toqZcDAbDOFFRynDrrbeW2oTScPj5sPWvUNNcakuGpPnsz3HyQY2lNsNgmFZUlNAPWJ640jny47D/iVA3v9SWDIlv9gHMXDCr1GYYDNOKYYVuROQsEdksIltF5Poi+90issLa/7yIzLe2v19E1onIa9bj+8bZ/jyeeuopnnrqqYl8ifLEZpsSIg/Qu+0l3n7luVKbYTBMK0SpwafGRMQObAHeD7QALwKXKKU25Yz5PHCkUuoaEfk4cJ5S6mIRORrYq5R6V0QOB55QSs0Z7PWWLl2q1q5dO6o3c9pppwHTMI9+ClG14Cjq/S62b3ix1KYYDBWFiKxTSi0ttm84Hv3xwFal1FtKqTjwAHBuwZhzgXut5w8Bp4uIKKVeVkq9a23fCHhFxD3yt2AwGAyG0TIcoZ8D7Mz5vcXaVnSMUioJdAMNBWMuAF5SSsUKX0BErhKRtSKytrW1dbi2GwwGg2EYTEp6pYgcBvwXcHWx/UqpO5VSS5VSS5uaTGVDg8FgGE+GI/S7gLk5vzdb24qOEREHUAO0W783A38APqmU2jZWgw0Gg8EwMoaTXvkicJCILEAL+seBSwvGPApcDqwBLgSeUUopEakF/gxcr5T633GzegDuuOOOiX4JwxjZ/yPXceKBhVE9g8EwkQwp9EqppIhcCzwB2IG7lVIbReQ7wFql1KPAXcB9IrIV6EDfDACuBQ4EbhSRG61tH1BK7RvvNwJw8MEHT8RpDeOIp2kuDXNmltoMg2FaMawFU0qplcDKgm035jyPAhcVOe7/Av93jDYOmz/96U8AnHPOOZP1koYR0r15DW+GauH8I0ptisEwbaiolbE/+MEPACP05cyefz5EzO+C732h1KYYDNMGU9TMMKlIqQ0wGKYhRugNBoOhwjFCb5h8TEF6g2FSMUJvmFxM7MZgmHQqajL2vvvuK7UJhiFYeOHXOWGhqUdvMEwmFSX0c+fOHXqQoaS4a2dQ3Tij1GYYDNOKigrdrFixghUrVpTaDMMgdLy6itf/9/FSm2EwTCsqyqP/2c9+BsDFF19cYksMA7H3hT8R8zqBfy21KQbDtKGiPHqDwWAw9McIvWHSMdmVBsPkYoTeYDAYKhwj9IZJxaTRGwyTT0VNxj700EOlNsEwBIsu/RbHLagvtRkGw7SiooS+sdEsxCl3XIFafNV1pTbDYJhWVFTo5p577uGee+4ptRmGQdi37nE2/O2RUpthMEwrKsqjz4j8FVdcUVI7DAOzb92TRL0O4MYhxxoMhvGhojx6g8FgMPTHCL3BYDBUOEboDQaDocIxQm+YfMzSWINhUqmoydiVK1eW2gTDECz+1HdZur9JrzQYJpOKEnqfz1dqEwxD4HB7cLq9pTbDYJhWVFTo5rbbbuO2224rtRmGQdi95hHWP/lgqc0wGKYVFSX0Dz74IA8+aESknGl79e9sWfNkqc0wGKYVFSX0BoPBYOjPsIReRM4Skc0islVEri+y3y0iK6z9z4vI/Jx9N1jbN4vImeNou8FgMBiGwZBCLyJ24KfA2cBi4BIRWVww7DNAp1LqQOBHwH9Zxy4GPg4cBpwF3Gadz2AwGAyTxHCybo4Htiql3gIQkQeAc4FNOWPOBW6ynj8E/ERExNr+gFIqBrwtIlut860ZH/MNU5HOcJz3//DvpTbDYCg7Tju4iX/7UKEfPXaGI/RzgJ05v7cAywYao5RKikg30GBtf67g2DmFLyAiVwFXAcybN2+4tvdj1apVoz7WMDn8+P5HWLOtrdRmGAxlycxqz4Sctyzy6JVSdwJ3AixdutSsm6xgLlu+P5ct37/UZhgM04rhTMbuAubm/N5sbSs6RkQcQA3QPsxjDQaDwTCBDEfoXwQOEpEFIuJCT64+WjDmUeBy6/mFwDNKKWVt/7iVlbMAOAh4YXxMNxgMBsNwGDJ0Y8XcrwWeAOzA3UqpjSLyHWCtUupR4C7gPmuytQN9M8Aa9yB64jYJ/ItSKjVB78VgMBgMRRDteJcPS5cuVWvXri21GQaDwTClEJF1SqmlxfaZlbEGg8FQ4RihNxgMhgrHCL3BYDBUOEboDQaDocIpu8lYEWkFto/hFI1AOS69NHaNDGPXyDB2jYxKtGt/pVRTsR1lJ/RjRUTWDjTzXEqMXSPD2DUyjF0jY7rZZUI3BoPBUOEYoTcYDIYKpxKF/s5SGzAAxq6RYewaGcaukTGt7Kq4GL3BYDAY8qlEj95gMBgMORihNxgMhgqnYoR+qAbmpUJE7haRfSKyodS2ZBCRuSLyNxHZJCIbReSLpbYJQEQ8IvKCiLxi2fXtUtuUi4jYReRlEXms1LZkEJF3ROQ1EVkvImVTDVBEakXkIRF5Q0ReF5ETysCmg62/U+anR0SuK7VdACLyJeszv0FEfisi49pqqiJi9FbD8S3A+9HtCl8ELlFKbRr0wElARE4BgsCvlFKHl9oeABGZDcxWSr0kIlXAOuCjpf57WX2G/UqpoIg4gX8CX1RKPTfEoZOCiHwZWApUK6U+XGp7QAs9sFQpVVaLf0TkXuAfSqlfWH0sfEqprhKblcXSjF3AMqXUWBZojoctc9Cf9cVKqYhV2n2lUuqe8XqNSvHosw3MlVJxINPAvOQopVaja/SXDUqp3Uqpl6znvcDrFOnlO9koTdD61Wn9lIUnIiLNwIeAX5TalnJHRGqAU9B9KlBKxctJ5C1OB7aVWuRzcABeq0OfD3h3PE9eKUJfrIF5yYVrKiAi84GjgedLbAqQDY+sB/YBf1VKlYVdwK3AvwLpEttRiAKeFJF1InJVqY2xWAC0Ar+0Ql2/EBF/qY0q4OPAb0ttBIBSahdwC7AD2A10K6WeHM/XqBShN4wCEQkADwPXKaV6Sm0PgFIqpZRagu4vfLyIlDzcJSIfBvYppdaV2pYivEcpdQxwNvAvVqiw1DiAY4CfKaWOBkJAOc2buYCPAL8rtS0AIlKHjkAsAPYD/CLyifF8jUoRetOEfIRYMfCHgfuVUr8vtT2FWF/1/wacVWJTAE4CPmLFwx8A3icivy6tSRrLG0QptQ/4AzqMWWpagJacb2MPoYW/XDgbeEkptbfUhlicAbytlGpVSiWA3wMnjucLVIrQD6eBucHCmvS8C3hdKfXDUtuTQUSaRKTWeu5FT66/UVKjAKXUDUqpZqXUfPRn6xml1Lh6XKNBRPzWZDpWaOQDQMmzu5RSe4CdInKwtel0dN/ocuESyiRsY7EDWC4iPut/83T0vNm4MWRz8KnAQA3MS2wWACLyW+A0oFFEWoBvKaXuKq1VnARcBrxmxcMBvqGUWlk6kwCYDdxrZUTYgAeVUmWTyliGzAT+oLUBB/AbpdTjpTUpyxeA+y3H6y3gUyW2B8jeEN8PXF1qWzIopZ4XkYeAl4Ak8DLjXAqhItIrDQaDwTAwlRK6MRgMBsMAGKE3GAyGCscIvcFgMFQ4RugNBoOhwjFCbzAYDBWOEXqDwWCocIzQGwwGQ4Xz/wMyO3oV3AhYtgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mech_evaluator = evaluators['mech']\n",
    "mech_evaluator.poof()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "37201409-f210-4d5e-b6cf-88a0b93ae972",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'y_diff_dtw_distance_norm': 0.004593695478977788,\n",
       " 'rms_perc_diff': 0.029389835098693005}"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "6704eee0-cf8a-4dcb-9a8a-d66d8bc8333c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'height': 0.12,\n",
       " 'magnet_assembly': {'m': 1,\n",
       "  'l_m_mm': 10,\n",
       "  'l_mcd_mm': 0,\n",
       "  'dia_magnet_mm': 10,\n",
       "  'dia_spacer_mm': 0},\n",
       " 'magnetic_spring': {'fea_data_file': '/Users/michael/Nextcloud/nextcloud/PhD/Python/unified_model/data/magnetic-spring/10x10alt.csv',\n",
       "  'filter_callable': 'auto',\n",
       "  'magnet_assembly': 'dep:magnet_assembly'},\n",
       " 'mechanical_spring': {'magnet_assembly': 'dep:magnet_assembly',\n",
       "  'strength': 10000000.0,\n",
       "  'damping_coefficient': 3.108},\n",
       " 'mechanical_damper': {'damping_coefficient': 4.272,\n",
       "  'magnet_assembly': 'dep:magnet_assembly'},\n",
       " 'input_excitation': {'raw_accelerometer_data_path': '/Users/michael/Nextcloud/nextcloud/PhD/Python/unified_model/data/2019-05-23/A/log_04_acc.csv',\n",
       "  'accel_column': 'z_G',\n",
       "  'time_column': 'time(ms)',\n",
       "  'accel_unit': 'g',\n",
       "  'time_unit': 'ms',\n",
       "  'smooth': True,\n",
       "  'interpolate': True},\n",
       " 'coil_configuration': {'c': 1,\n",
       "  'n_z': 15,\n",
       "  'n_w': 17,\n",
       "  'l_ccd_mm': 0,\n",
       "  'ohm_per_mm': 0.0010789999999999999,\n",
       "  'tube_wall_thickness_mm': 2,\n",
       "  'coil_wire_radius_mm': 0.0715,\n",
       "  'coil_center_mm': 59,\n",
       "  'inner_tube_radius_mm': 5.5,\n",
       "  'custom_coil_resistance': None},\n",
       " 'flux_model': {'coil_configuration': 'dep:coil_configuration',\n",
       "  'magnet_assembly': 'dep:magnet_assembly',\n",
       "  'curve_model_path': '../data/flux_curve_model/flux_curve_model_2022_08_09.model'},\n",
       " 'rectification_drop': 0.05,\n",
       " 'load_model': {'R': 30},\n",
       " 'coupling_model': {'coupling_constant': 5.096},\n",
       " 'extra_components': None,\n",
       " 'governing_equations': {'module_path': 'ffs.governing_equations',\n",
       "  'func_name': 'unified_ode'}}"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.get_config()"
   ]
  }
 ],
 "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}