4. MAGNets tutorial.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e83378e1",
   "metadata": {},
   "source": [
    "# MAGNets Tutorial"
   ]
  },
  {
   "attachments": {
    "image-2.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "d4a89321",
   "metadata": {},
   "source": [
    "In this Jupyter Notebook, our goal is to explore **model skeletonization** options provided by **MAGNets**.\n",
    "\n",
    "For our analyses, we will use BWFLNet, a model of a real water distribution network. The Bristol Water Field Lab (BWFL) network is part of the water distribution network of the city of Bristol, UK, and it serves approximately 8,000 customer connections. BWFLnet is an operational network in Bristol, UK, operated by Bristol Water. The data provided is a the product of a long term research partnership between Bristol Water and Infrasense Labs at Imperial College London. The BWFLnet model and data set can be found [here](https://data.mendeley.com/datasets/srt4vr5k38/1).\n",
    "\n",
    "![image-2.png](attachment:image-2.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85aeac14",
   "metadata": {},
   "source": [
    "# 1. Getting Started"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86f8076a",
   "metadata": {},
   "source": [
    "## 1.1. Imports"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "202f6eea",
   "metadata": {},
   "source": [
    "Our first step is to install and import the Python packages we need into this file:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1b04f5c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install wntr matplotlib numpy magnets viswaternet "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f98adcc1-4d05-4a75-83f6-93c39b7fb476",
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "#import packages\n",
    "import wntr\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import time\n",
    "import viswaternet as vis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63275c67",
   "metadata": {},
   "source": [
    "## 1.2. Create water network model "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5165299a",
   "metadata": {},
   "source": [
    "Let us begin by defining the .INP file of the model and initializing a water network model object using WNTR. Note that this model has been slightly modified from the original model in the following ways:\n",
    "1. some demand patterns are exactly the same as others so we have merged them\n",
    "2. all throttle control valves (TCVs) in the model have been replaced with equivalent pipes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4cdb2ad3-4fad-4020-a9a7-b57e13640ae4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# import bwfl\n",
    "inp = 'networks/BWFLnet_modified.inp'\n",
    "\n",
    "# initialize water network model object\n",
    "wn = wntr.network.WaterNetworkModel(inp)\n",
    "\n",
    "# initialize viswaternet model object\n",
    "model=vis.VisWNModel(inp)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b5dd128",
   "metadata": {},
   "source": [
    "## 1.3. Run hydraulic simulation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5fcf9f6",
   "metadata": {},
   "source": [
    "We will now run a hydraulic simulation with WNTR (using the EPANET 2.2 engine), store the simulation time, and examine the number of components in the network."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e49bcbb0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# run wntr simulation and record simulation time\n",
    "sim = wntr.sim.EpanetSimulator(wn)\n",
    "\n",
    "t1 = time.time()\n",
    "results = sim.run_sim()\n",
    "t2 = time.time()\n",
    "\n",
    "simtime = t2-t1\n",
    "print('The simulation run time was ', simtime, 's')\n",
    "\n",
    "# describe the number of components in the system\n",
    "wn.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5b2712b1",
   "metadata": {},
   "source": [
    "## 1.4. Define critical nodes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2e9a311",
   "metadata": {},
   "source": [
    "Remember that this real network has several important junctions that we want to keep track of. Here, we define (1) a list of junction names representing locations of pressure sensors, (2) a list of junctions that are assigned different demand patterns, and (3) their neighboring nodes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5bc73d9d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# list of junctions where pressure sensors are installed\n",
    "pressure_sensors_list = ['node_1925', 'node_1938', 'node_1961', 'node_1971', \n",
    "                         'node_1983', 'node_0354', 'node_2012', 'node_2046', \n",
    "                         'node_2113', 'node_0785', 'node_2188', 'node_2197', \n",
    "                         'node_2210', 'node_2243', 'node_2265', 'node_2285', \n",
    "                         'node_1194', 'node_1202', 'node_2354', 'node_2428', \n",
    "                         'node_2440', 'node_2441', 'node_1738', 'node_2485', \n",
    "                         'node_1781', 'node_1876', 'node_2535']\n",
    "\n",
    "# list of junctions with different demand patterns ...\n",
    "diff_demand_pattern_list = ['node_0640', 'node_0870', 'node_0973', 'node_1122', 'node_1311', \n",
    "                            'node_1363', 'node_1388', 'node_1685', 'node_1813', 'node_2125', \n",
    "                            'node_2130', 'node_2167', 'node_2184', 'node_2220', 'node_2307']\n",
    "\n",
    "# ... and their neighbors\n",
    "diff_demand_neighbor_list = ['node_0595', 'node_0647', 'node_0648', 'node_2168', 'node_2152', \n",
    "                             'node_0907', 'node_0952', 'node_2199', 'node_2206', 'node_1121',\n",
    "                             'node_2260', 'node_1315', 'node_1340', 'node_2350', 'node_1385',\n",
    "                             'node_1386', 'node_1402', 'node_2439', 'node_2446', 'node_2456',\n",
    "                             'node_1807', 'node_0718', 'node_2140', 'node_2112', 'node_0755',\n",
    "                             'node_0890', 'node_0868', 'node_0913', 'node_0984', 'node_1037',\n",
    "                             'node_1223', 'node_1262']\n",
    "\n",
    "# list of all critical nodes\n",
    "critical_nodes = pressure_sensors_list + diff_demand_pattern_list + diff_demand_neighbor_list"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e3662ea",
   "metadata": {},
   "source": [
    "# 2. Model Reduction Using MAGNets"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ecaa7d28",
   "metadata": {},
   "source": [
    "We see that the BWFLNet model has over 2,500 nodes and links and 5 valves. Hydraulic models of complex water networks, such as this one, can be expensive to run in terms of computational effort and time. Therefore, we can turn to **reduced** or **skeletonized** models to speed up model running times and facilitate real-time decision making. \n",
    "\n",
    "**Model reduction** or **skeletonization** refers to the method of reducing the size of a network model to an equivalent model with fewer components while preserving the pressure at nodes."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca2e0954",
   "metadata": {},
   "source": [
    "Let us use the Python package **MAGNets (Model AGgregation and reduction of water distribution Networks)** to perform model reduction. MAGNets is a Python package designed to perform the reduction and aggregation of water distribution network models based on the variable elimination method proposed by [Ulanicki et al (1996)](https://www.researchgate.net/publication/273796660_Simplification_of_Water_Distribution_Network_Models). The software is capable of reducing a network around an optional operating point and allows the user to customize which junctions to keep in the reduced model as well as the level of reduction. \n",
    "\n",
    "**Note**: By default, all reservoirs, tanks, pumps, valves, and their adjacent junctions will remain in the reduced model."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2773e5b",
   "metadata": {},
   "source": [
    "## 2.1. Using MAGNets"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d662e5c",
   "metadata": {},
   "source": [
    "We begin by **importing** the MAGNets Python package into the file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0280fc85-e80c-4795-a8cc-e0bae8da24ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "import magnets as mg"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7f96147",
   "metadata": {},
   "source": [
    "The user can then call on the **```reduce_model```** function to reduce a hydraulic model of a water distribution network. \n",
    "\n",
    "```wn2 = mg.reduction.reduce_model(inp_file, op_pt, nodes_to_keep, max_nodal_degree, save_filename)```\n",
    "\n",
    "\n",
    "The parameters of the `reduce_model` function are described as follows:\n",
    "\n",
    "1. `inp_file`: the EPANET-compatible .INP file of the water distribution network model.\n",
    "\n",
    "2. `op_pt`: (optional, default = 0) the operating point, or the reporting time step of the hydraulic simulation at which the non-linear headloss equations are linearized.\n",
    "\n",
    "3. `nodes_to_keep`: (optional, default = []) a list of nodes the user wishes to keep in the reduced model.\n",
    "\n",
    "4. `max_nodal_degree`: (optional, default = None) the maximum nodal degree of nodes being removed from the model. The nodal degree of a node is equal to the number of pipes incident to the node.\n",
    "\n",
    "5. `save_filename`: (optional, default = None) a user-specified name assigned to the newly generated .INP file of the reduced model.\n",
    "\n",
    "`wn2` contains the water network model object of the reduced model. A .INP file of the reduced model is also written into the directory that contains the .INP file of the original network."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24ece0cb",
   "metadata": {},
   "source": [
    "## 2.2. Reduce model with ```max_nodal_degree``` = 1 (branch trimming)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c517592e",
   "metadata": {},
   "source": [
    "In the first example, we will remove all branches. We first specify:\n",
    "1.\tMaximum nodal degree = 1\n",
    "2.\tOperating point = 0\n",
    "3.\tThe list of nodes we want to keep in the reduced model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a4d8e924-5ae7-473d-9f5a-c3ee585d9b39",
   "metadata": {},
   "outputs": [],
   "source": [
    "#define inp_file\n",
    "inp = 'networks/BWFLnet_modified.inp'\n",
    "\n",
    "#specify operating point\n",
    "op_pt = 0\n",
    "\n",
    "#specify max nodal degree\n",
    "max_nodal_degree1 = 1\n",
    "\n",
    "#call model reduction\n",
    "t1 = time.time()\n",
    "wn2 = mg.reduction.reduce_model(inp_file = inp,\n",
    "                                op_pt = op_pt,\n",
    "                                nodes_to_keep = critical_nodes,\n",
    "                                max_nodal_degree = max_nodal_degree1,\n",
    "                                save_filename = 'networks/BWFLnet_modified_level1')  # wn2 is a wntr network model\n",
    "t2 = time.time()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68262f64",
   "metadata": {},
   "source": [
    "### Plot the full and reduced models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e1c045f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#define style object\n",
    "style8 = vis.NetworkStyle(base_node_size=1, draw_base_legend = False,\n",
    "                          reservoir_size=100, reservoir_border_width = 2, \n",
    "                          valve_size =100, valve_color ='xkcd:periwinkle')\n",
    "sensor_color = 'xkcd:orange'\n",
    "unique_dp_color = 'xkcd:teal'\n",
    "\n",
    "#plot full network\n",
    "fig, ax = plt.subplots(1, 2, figsize=(12,5))\n",
    "ax[0].set_frame_on(False)\n",
    "model=vis.VisWNModel(inp)\n",
    "model.plot_basic_elements(ax=ax[0], style = style8)\n",
    "# highlight pressure sensors\n",
    "model.draw_nodes(ax=ax[0], node_list=pressure_sensors_list, node_color=sensor_color, node_size=50,\n",
    "                 node_shape ='^', node_border_width = 1, node_border_color = 'k')  \n",
    "# highlight nodes with different demand patterns\n",
    "model.draw_nodes(ax=ax[0], node_list=diff_demand_pattern_list, node_color=unique_dp_color,  node_size=20,\n",
    "                 node_shape ='o', node_border_width = 1, node_border_color = 'k')\n",
    "\n",
    "#plot reduced network\n",
    "model1=vis.VisWNModel('networks/BWFLnet_modified_level1.inp')\n",
    "ax[1].set_frame_on(False)\n",
    "model1.plot_basic_elements(ax=ax[1], style = style8)\n",
    "# highlight pressure sensors\n",
    "model.draw_nodes(ax=ax[1], node_list=pressure_sensors_list, node_color=sensor_color, node_size=50,\n",
    "                 node_shape ='^', node_border_width = 1, node_border_color = 'k')  \n",
    "# highlight nodes with different demand patterns\n",
    "model.draw_nodes(ax=ax[1], node_list=diff_demand_pattern_list, node_color=unique_dp_color,  node_size=20,\n",
    "                 node_shape ='o', node_border_width = 1, node_border_color = 'k')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f640235c",
   "metadata": {},
   "source": [
    "### Display reduction time and number of elements"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f96fc985",
   "metadata": {},
   "outputs": [],
   "source": [
    "print('Total reduction time:', str(round(t2-t1, 2)), 's')\n",
    "print('Number of junctions in the original network:', \"\\033[1m\", wn.num_junctions, \"\\033[0;0m\", 'vs. number of junctions in the reduced network:',\"\\033[1m\", wn2.num_junctions, \"\\033[0;0m\")\n",
    "print('Number of pipes in the original network:', \"\\033[1m\",wn.num_pipes,\"\\033[0;0m\", 'vs. number of pipes in the reduced network:',\"\\033[1m\", wn2.num_pipes, \"\\033[0;0m\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "281fc719",
   "metadata": {},
   "source": [
    "We generated a much smaller model keeping all critical nodes. The process only took a few seconds."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58d718bd",
   "metadata": {},
   "source": [
    "## 2.3. Reduce nodes with ```max_nodal_degree``` = 2 (branch trimming + merging pipes in series and parallel)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95a4932c-ca48-49b0-b6d1-727400b067b6",
   "metadata": {},
   "source": [
    "Next, we will reduce the same model defining a different  ```max_nodal_degree ``` each time to see how this impacts the results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e998fd42-8a4a-4670-a8b3-afc3a1751ebc",
   "metadata": {},
   "outputs": [],
   "source": [
    "max_nodal_degree2 = 2\n",
    "\n",
    "#call model reduction\n",
    "t1 = time.time()\n",
    "wn3 = mg.reduction.reduce_model(inp, op_pt, critical_nodes, max_nodal_degree2, 'networks/BWFLnet_modified_level2')\n",
    "t2 = time.time()\n",
    "\n",
    "#plot full network\n",
    "fig, ax = plt.subplots(1, 2, figsize=(12,5))\n",
    "ax[0].set_frame_on(False)\n",
    "model=vis.VisWNModel(inp)\n",
    "model.plot_basic_elements(ax=ax[0], style = style8)\n",
    "# highlight pressure sensors\n",
    "model.draw_nodes(ax=ax[0], node_list=pressure_sensors_list, node_color=sensor_color, node_size=50,\n",
    "                 node_shape ='^', node_border_width = 1, node_border_color = 'k')  \n",
    "# highlight nodes with different demand patterns\n",
    "model.draw_nodes(ax=ax[0], node_list=diff_demand_pattern_list, node_color=unique_dp_color,  node_size=20,\n",
    "                 node_shape ='o', node_border_width = 1, node_border_color = 'k')\n",
    "\n",
    "#plot reduced network\n",
    "model1=vis.VisWNModel('networks/BWFLnet_modified_level2.inp')\n",
    "ax[1].set_frame_on(False)\n",
    "model1.plot_basic_elements(ax=ax[1], style = style8)\n",
    "# highlight pressure sensors\n",
    "model.draw_nodes(ax=ax[1], node_list=pressure_sensors_list, node_color=sensor_color, node_size=50,\n",
    "                 node_shape ='^', node_border_width = 1, node_border_color = 'k')  \n",
    "# highlight nodes with different demand patterns\n",
    "model.draw_nodes(ax=ax[1], node_list=diff_demand_pattern_list, node_color=unique_dp_color,  node_size=20,\n",
    "                 node_shape ='o', node_border_width = 1, node_border_color = 'k')\n",
    "\n",
    "# Display reduction time and number of elements\n",
    "print('Total reduction time:', str(round(t2-t1, 2)), 's')\n",
    "print('Number of junctions in the original network:', \"\\033[1m\", wn.num_junctions, \"\\033[0;0m\", 'vs. number of junctions in the reduced network:',\"\\033[1m\", wn3.num_junctions, \"\\033[0;0m\")\n",
    "print('Number of pipes in the original network:', \"\\033[1m\",wn.num_pipes,\"\\033[0;0m\", 'vs. number of pipes in the reduced network:',\"\\033[1m\", wn3.num_pipes, \"\\033[0;0m\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c1d0350",
   "metadata": {},
   "source": [
    "## 2.4. Reduce nodes with ```max_nodal_degree``` = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fcf547c6-3ea4-4775-aad1-3ef7d14bb40f",
   "metadata": {},
   "outputs": [],
   "source": [
    "max_nodal_degree3 = 3\n",
    "\n",
    "#call model reduction\n",
    "t1 = time.time()\n",
    "wn4 = mg.reduction.reduce_model(inp, op_pt, critical_nodes, max_nodal_degree3, 'networks/BWFLnet_modified_level3')\n",
    "t2 = time.time()\n",
    "\n",
    "#plot full network\n",
    "fig, ax = plt.subplots(1, 2, figsize=(12,5))\n",
    "ax[0].set_frame_on(False)\n",
    "model=vis.VisWNModel(inp)\n",
    "model.plot_basic_elements(ax=ax[0], style = style8)\n",
    "# highlight pressure sensors\n",
    "model.draw_nodes(ax=ax[0], node_list=pressure_sensors_list, node_color=sensor_color, node_size=50,\n",
    "                 node_shape ='^', node_border_width = 1, node_border_color = 'k')  \n",
    "# highlight nodes with different demand patterns\n",
    "model.draw_nodes(ax=ax[0], node_list=diff_demand_pattern_list, node_color=unique_dp_color,  node_size=20,\n",
    "                 node_shape ='o', node_border_width = 1, node_border_color = 'k')\n",
    "\n",
    "#plot reduced network\n",
    "model1=vis.VisWNModel('networks/BWFLnet_modified_level3.inp')\n",
    "ax[1].set_frame_on(False)\n",
    "model1.plot_basic_elements(ax=ax[1], style = style8)\n",
    "# highlight pressure sensors\n",
    "model.draw_nodes(ax=ax[1], node_list=pressure_sensors_list, node_color=sensor_color, node_size=50,\n",
    "                 node_shape ='^', node_border_width = 1, node_border_color = 'k')  \n",
    "# highlight nodes with different demand patterns\n",
    "model.draw_nodes(ax=ax[1], node_list=diff_demand_pattern_list, node_color=unique_dp_color,  node_size=20,\n",
    "                 node_shape ='o', node_border_width = 1, node_border_color = 'k')\n",
    "# Display reduction time and number of elements\n",
    "print('Total reduction time:', str(round(t2-t1, 2)), 's')\n",
    "print('Number of junctions in the original network:', \"\\033[1m\", wn.num_junctions, \"\\033[0;0m\", 'vs. number of junctions in the reduced network:',\"\\033[1m\", wn4.num_junctions, \"\\033[0;0m\")\n",
    "print('Number of pipes in the original network:', \"\\033[1m\",wn.num_pipes,\"\\033[0;0m\", 'vs. number of pipes in the reduced network:',\"\\033[1m\", wn4.num_pipes, \"\\033[0;0m\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b7c662b",
   "metadata": {},
   "source": [
    "## 2.5. Reduce nodes with ```max_nodal_degree``` = None "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "168d9ce5",
   "metadata": {},
   "source": [
    "All junctions that are not in the critical_nodes list or connected to reservoirs, tanks, pumps, or valves will be removed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "33d5d5b0-b7eb-431b-915a-cc770150e01f",
   "metadata": {},
   "outputs": [],
   "source": [
    "max_nodal_degree = None\n",
    "\n",
    "#call model reduction\n",
    "t1 = time.time()\n",
    "wn5 = mg.reduction.reduce_model(inp, op_pt, critical_nodes, max_nodal_degree,'networks/BWFLnet_modified_full')\n",
    "t2 = time.time()\n",
    "\n",
    "#plot full network\n",
    "fig, ax = plt.subplots(1, 2, figsize=(12,5))\n",
    "ax[0].set_frame_on(False)\n",
    "model=vis.VisWNModel(inp)\n",
    "model.plot_basic_elements(ax=ax[0], style = style8)\n",
    "# highlight pressure sensors\n",
    "model.draw_nodes(ax=ax[0], node_list=pressure_sensors_list, node_color=sensor_color, node_size=50,\n",
    "                 node_shape ='^', node_border_width = 1, node_border_color = 'k')  \n",
    "# highlight nodes with different demand patterns\n",
    "model.draw_nodes(ax=ax[0], node_list=diff_demand_pattern_list, node_color=unique_dp_color,  node_size=20,\n",
    "                 node_shape ='o', node_border_width = 1, node_border_color = 'k')\n",
    "\n",
    "#plot reduced network\n",
    "model1=vis.VisWNModel('networks/BWFLnet_modified_full.inp')\n",
    "ax[1].set_frame_on(False)\n",
    "model1.plot_basic_elements(ax=ax[1], style = style8)\n",
    "# highlight pressure sensors\n",
    "model.draw_nodes(ax=ax[1], node_list=pressure_sensors_list, node_color=sensor_color, node_size=50,\n",
    "                 node_shape ='^', node_border_width = 1, node_border_color = 'k')  \n",
    "# highlight nodes with different demand patterns\n",
    "model.draw_nodes(ax=ax[1], node_list=diff_demand_pattern_list, node_color=unique_dp_color,  node_size=20,\n",
    "                 node_shape ='o', node_border_width = 1, node_border_color = 'k')\n",
    "plt.setp(ax, ylim=ax[0].get_ylim())\n",
    "plt.setp(ax, xlim=ax[0].get_xlim())\n",
    "\n",
    "# Display reduction time and number of elements\n",
    "print('Total reduction time:', str(round(t2-t1, 2)), 's')\n",
    "print('Number of junctions in the original network:', \"\\033[1m\", wn.num_junctions, \"\\033[0;0m\", 'vs. number of junctions in the reduced network:',\"\\033[1m\", wn5.num_junctions, \"\\033[0;0m\")\n",
    "print('Number of pipes in the original network:', \"\\033[1m\",wn.num_pipes,\"\\033[0;0m\", 'vs. number of pipes in the reduced network:',\"\\033[1m\", wn5.num_pipes, \"\\033[0;0m\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f9b7bba-71a3-420e-a0bd-d2997983f4fc",
   "metadata": {},
   "source": [
    "As can be seen from the graphs above, the reduced networks vary greatly and the reduction time increases as the maximum nodal degree increases. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff74ada0",
   "metadata": {},
   "source": [
    "## 2.6. Comparing simulated pressures\n",
    "\n",
    "We will now compare the pressures at two critical nodes to evaluate how the reduced models compare to the full model."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b386f4ee",
   "metadata": {},
   "source": [
    "### Run simulations and track running times"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b10fe70d-b89c-4e17-92c8-d4746ddedd53",
   "metadata": {},
   "outputs": [],
   "source": [
    "sim = wntr.sim.EpanetSimulator(wn)\n",
    "t1 = time.time()\n",
    "results = sim.run_sim()\n",
    "t2 = time.time()\n",
    "sim_time_orig = t2-t1\n",
    "\n",
    "sim2 = wntr.sim.EpanetSimulator(wn2)\n",
    "t1 = time.time()\n",
    "results2 = sim2.run_sim()\n",
    "t2 = time.time()\n",
    "sim_time_1 = t2-t1\n",
    "\n",
    "sim3 = wntr.sim.EpanetSimulator(wn3)\n",
    "t1 = time.time()\n",
    "results3 = sim3.run_sim()\n",
    "t2 = time.time()\n",
    "sim_time_2 = t2-t1\n",
    "\n",
    "sim4 = wntr.sim.EpanetSimulator(wn4)\n",
    "t1 = time.time()\n",
    "results4 = sim4.run_sim()\n",
    "t2 = time.time()\n",
    "sim_time_3 = t2-t1\n",
    "\n",
    "sim5 = wntr.sim.EpanetSimulator(wn5)\n",
    "t1 = time.time()\n",
    "results5 = sim5.run_sim()\n",
    "t2 = time.time()\n",
    "sim_time_none = t2-t1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b64d63dc",
   "metadata": {},
   "source": [
    "### Plot time series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c283313f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#comparing the head at two different nodes in the regular and reduced networks: \n",
    "node1, node2 = 'node_0469', 'node_1194'\n",
    "\n",
    "#obtain head data for each network\n",
    "head=results.node['head']\n",
    "head2 = results2.node['head']\n",
    "head3=results3.node['head']\n",
    "head4=results4.node['head']\n",
    "head5=results5.node['head']\n",
    "\n",
    "#obtain data for critical nodes\n",
    "head_at_node1, head_at_node2 = head.loc[:,node1], head.loc[:,node2]\n",
    "reduced_head_at_node1_2, reduced_head_at_node2_2 = head2.loc[:,node1], head2.loc[:,node2]\n",
    "reduced_head_at_node1_3, reduced_head_at_node2_3 = head3.loc[:,node1], head3.loc[:,node2]\n",
    "reduced_head_at_node1_4, reduced_head_at_node2_4 = head4.loc[:,node1], head4.loc[:,node2]\n",
    "reduced_head_at_node1_5, reduced_head_at_node2_5 = head5.loc[:,node1], head5.loc[:,node2]\n",
    "\n",
    "#plot reduced and original model pressure head time series at two critical nodes \n",
    "fig, ax = plt.subplots(1,2,figsize=(10,4))\n",
    "ax[0].plot(head_at_node1.index/3600, head_at_node1, lw = 2, color = 'k')\n",
    "ax[0].plot(reduced_head_at_node1_2.index/3600, reduced_head_at_node1_2, linestyle = '--', lw = 2)\n",
    "ax[0].plot(reduced_head_at_node1_3.index/3600, reduced_head_at_node1_3, linestyle = '--', lw = 2)\n",
    "ax[0].plot(reduced_head_at_node1_4.index/3600, reduced_head_at_node1_4, linestyle = '-.', lw = 2)\n",
    "ax[0].plot(reduced_head_at_node1_5.index/3600, reduced_head_at_node1_5, linestyle = ':', lw = 2)\n",
    "ax[0].set_title(node1)\n",
    "ax[0].set_xlabel('Time [hr]')\n",
    "ax[0].set_ylabel('Pressure head [m]')\n",
    "\n",
    "\n",
    "ax[1].plot(head_at_node2.index/3600, head_at_node2, lw = 2, color = 'k')\n",
    "ax[1].plot(reduced_head_at_node2_2.index/3600, reduced_head_at_node2_2, linestyle = '--', lw = 2)\n",
    "ax[1].plot(reduced_head_at_node2_3.index/3600, reduced_head_at_node2_3, linestyle = '--', lw = 2)\n",
    "ax[1].plot(reduced_head_at_node2_4.index/3600, reduced_head_at_node2_4, linestyle = '-.', lw = 2)\n",
    "ax[1].plot(reduced_head_at_node2_5.index/3600, reduced_head_at_node2_5, linestyle = ':', lw = 2)\n",
    "ax[1].set_title(node2)\n",
    "ax[1].set_xlabel('Time [hr]')\n",
    "ax[1].legend(['Original model', 'Reduced model: max_nodal_degree = 1','Reduced model: max_nodal_degree = 2', 'Reduced model: max_nodal_degree = 3','Reduced model: max_nodal_degree = None'],bbox_to_anchor=(1.05, 1), loc='upper left', borderaxespad=0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bf5800e",
   "metadata": {},
   "source": [
    "## 2.7. Compare running times"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4617901e",
   "metadata": {},
   "source": [
    "Let us examine the simulation running times (in seconds) and speed up (decrease in running time compared to the original model)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dbb506f0",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "data = np.array([[sim_time_orig, sim_time_orig/sim_time_orig], \n",
    "                 [sim_time_1, sim_time_orig/sim_time_1], \n",
    "                 [sim_time_2, sim_time_orig/sim_time_2], \n",
    "                 [sim_time_3, sim_time_orig/sim_time_3], \n",
    "                 [sim_time_none, sim_time_orig/sim_time_none]]).T\n",
    "pd.DataFrame(data, \n",
    "             columns=['Original model', 'Reduced model: degree 1','Reduced model: degree 2', 'Reduced model:degree 3','Reduced model: degree None'],\n",
    "            index=['Running time [s]', 'Speedup']).round(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b610bfb5",
   "metadata": {},
   "source": [
    "We can see that the reduced models run much faster than the original model. All of these results demonstrate that reduced models can be used in a range of studies and applications to **speed up simulation running times** while **preserving pressures throughout the system**. \n",
    "\n",
    "**Note**: Users should carefully choose the: (1) nodes to keep, (2) degree of reduction, and (3) operating point, when creating the reduced models. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f08e4841",
   "metadata": {},
   "source": [
    "# 3. Exercise"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9298ad3",
   "metadata": {},
   "source": [
    "Use MAGNets to reduce network **ky2**. Try changing the operating point (```op_pt```) and (```max_nodal_degree```) arguments in **Cell 3.1.** to see how the pressure head results and reduced model topology change in **Cell 3.2.**:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12dcc4d9",
   "metadata": {},
   "source": [
    "## 3.1.  Perform model reduction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "917edda2",
   "metadata": {},
   "outputs": [],
   "source": [
    "#define inp_file\n",
    "inp_ky4 = 'networks/ky2.inp'\n",
    "wn_ky4 = wntr.network.WaterNetworkModel(inp_ky4)\n",
    "\n",
    "op_pt = 0                         # change the operating point: e.g., 5, 10, 15, 20  \n",
    "max_nodal_degree_ky4 = 1          # change the max nodal degree, e.g., 1, 2, 3, None\n",
    "\n",
    "#call model reduction\n",
    "t1 = time.time()\n",
    "wn_ky4_red = mg.reduction.reduce_model(inp_file = inp_ky4,\n",
    "                                op_pt = op_pt,\n",
    "                                nodes_to_keep = [],\n",
    "                                max_nodal_degree = max_nodal_degree_ky4,\n",
    "                                save_filename = 'networks/reduced ky2')  \n",
    "t2 = time.time()\n",
    "\n",
    "#display reduction time and number of elements\n",
    "print('Total reduction time:', str(round(t2-t1, 2)), 's')\n",
    "print('Number of junctions in the original network:', \"\\033[1m\", wn_ky4.num_junctions, \"\\033[0;0m\", 'vs. number of junctions in the reduced network:',\"\\033[1m\", wn_ky4_red.num_junctions, \"\\033[0;0m\")\n",
    "print('Number of pipes in the original network:', \"\\033[1m\",wn_ky4.num_pipes,\"\\033[0;0m\", 'vs. number of pipes in the reduced network:',\"\\033[1m\", wn_ky4_red.num_pipes, \"\\033[0;0m\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0939e382",
   "metadata": {},
   "source": [
    "## 3.2. Plot network layout and time series comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "724c0182",
   "metadata": {},
   "outputs": [],
   "source": [
    "#plot full network\n",
    "fig, ax = plt.subplots(1, 2, figsize=(14,5))\n",
    "ax[0].set_frame_on(False)\n",
    "model=vis.VisWNModel(network_model=wn_ky4)\n",
    "model.plot_basic_elements(ax=ax[0], style = style8)\n",
    "\n",
    "#plot reduced network\n",
    "model1=vis.VisWNModel(network_model=wn_ky4_red)\n",
    "ax[1].set_frame_on(False)\n",
    "model1.plot_basic_elements(ax=ax[1], style = style8)\n",
    "plt.setp(ax, ylim=ax[0].get_ylim())\n",
    "plt.setp(ax, xlim=ax[0].get_xlim())\n",
    "\n",
    "#comparing the head at two different nodes in the regular and reduced networks: \n",
    "node1, node2 = 'T-1', 'T-2'\n",
    "sim_ky4 = wntr.sim.EpanetSimulator(wn_ky4)\n",
    "results_ky4 = sim_ky4.run_sim()\n",
    "\n",
    "sim_ky4_red = wntr.sim.EpanetSimulator(wn_ky4_red)\n",
    "results_ky4_red = sim_ky4_red.run_sim()\n",
    "\n",
    "#obtain head data for each network\n",
    "head, head2 = results_ky4.node['head'], results_ky4_red.node['head']\n",
    "\n",
    "#obtain data for critical node 1\n",
    "head_at_node1, reduced_head_at_node1 = head.loc[:,node1], head2.loc[:,node1]\n",
    "\n",
    "#obtain data for critical node 2\n",
    "head_at_node2, reduced_head_at_node2 = head.loc[:,node2], head2.loc[:,node2]\n",
    "\n",
    "#plot reduced and original model pressure head time series at two critical nodes \n",
    "fig, ax = plt.subplots(1,2,figsize=(14,4))\n",
    "ax[0].plot(head_at_node1.index/3600, head_at_node1, color='k', lw = 2)\n",
    "ax[0].plot(reduced_head_at_node1.index/3600, reduced_head_at_node1, \n",
    "           linestyle = '--', color='orange', lw = 2)\n",
    "ax[0].set_title(node1)\n",
    "ax[0].set_xlabel('Time [hr]')\n",
    "ax[0].set_ylabel('Pressure head [m]')\n",
    "ax[0].legend(['Original model', 'Reduced model'])\n",
    "\n",
    "ax[1].plot(head_at_node2.index/3600, head_at_node2, color='k', lw = 2)\n",
    "ax[1].plot(reduced_head_at_node2.index/3600, reduced_head_at_node2, \n",
    "           linestyle = '--', color = 'orange', lw = 2)\n",
    "ax[1].set_title(node2)\n",
    "ax[1].set_xlabel('Time [hr]')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4943be40",
   "metadata": {},
   "source": [
    "We hope you have found this tutorial useful. For more information:\n",
    "\n",
    "[MAGNets Github repository](https://github.com/meghnathomas/MAGNets)\n",
    "\n",
    "[MAGNets paper in the Journal of Water Resources Planning and Management](https://ascelibrary.org/doi/full/10.1061/JWRMD5.WRENG-5486)"
   ]
  }
 ],
 "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
}