notebooks/sankey.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a813a8a4-ee0f-4daf-abcb-d0315b1309fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pypsa\n",
    "import pandas as pd\n",
    "from pypsa.descriptors import get_switchable_as_dense as as_dense\n",
    "import plotly.graph_objects as go\n",
    "import numpy as np\n",
    "import yaml\n",
    "import os\n",
    "from matplotlib.colors import to_rgba"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b9b14597-75aa-4f95-95e3-76520d0762a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "PATH = \"../workflows/pypsa-eur-sec/\"\n",
    "SCENARIO = \"elec_s_181_lvopt__Co2L0-3H-T-H-B-I-A-solar+p3-linemaxext10-noH2network-onwind+p0_2050\"\n",
    "RUN = \"20221227-main\"\n",
    "OUTPUT = \"../results/graphics-20221227/\"\n",
    "\n",
    "# SCENARIO = \"elec_s_181_lv1.0__Co2L0-25H-T-H-B-I-A-solar+p3-linemaxext15-seq1000_2050\"\n",
    "# RUN = \"20230125-carbon-small\"\n",
    "# OUTPUT = \"../results/graphics-carbon/\"\n",
    "\n",
    "OUTPUT_SCENARIO = f\"{OUTPUT}/{RUN}/{SCENARIO}/\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "79a34bd8",
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.exists(OUTPUT_SCENARIO):\n",
    "    os.makedirs(OUTPUT_SCENARIO)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "81756437-b180-4c56-9bc4-dc7ed263377f",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(PATH + \"config.main.yaml\") as file:\n",
    "    config = yaml.safe_load(file)\n",
    "\n",
    "colors = config[\"plotting\"][\"tech_colors\"]\n",
    "\n",
    "colors[\"electricity grid\"] = colors[\"electricity\"]\n",
    "colors[\"ground-sourced ambient\"] = colors[\"ground heat pump\"]\n",
    "colors[\"air-sourced ambient\"] = colors[\"air heat pump\"]\n",
    "colors[\"co2 atmosphere\"] = colors[\"co2\"]\n",
    "colors[\"co2 stored\"] = colors[\"co2\"]\n",
    "colors[\"net co2 emissions\"] = colors[\"co2\"]\n",
    "colors[\"co2 sequestration\"] = colors[\"co2\"]\n",
    "colors[\"fossil oil\"] = colors[\"oil\"]\n",
    "colors[\"fossil gas\"] = colors[\"gas\"]\n",
    "colors[\"biogas to gas\"] = colors[\"biogas\"]\n",
    "colors[\"process emissions from feedstocks\"] = colors[\"process emissions\"]\n",
    "colors[\"methanol emissions\"] = colors[\"shipping methanol emissions\"]\n",
    "\n",
    "gas_boilers = [\n",
    "    \"residential rural gas boiler\",\n",
    "    \"services rural gas boiler\",\n",
    "    \"residential urban decentral gas boiler\",\n",
    "    \"services urban decentral gas boiler\",\n",
    "    \"urban central gas boiler\",\n",
    "]\n",
    "for gas_boiler in gas_boilers:\n",
    "    colors[gas_boiler] = colors[\"gas boiler\"]\n",
    "\n",
    "colors[\"urban central gas CHP\"] = colors[\"CHP\"]\n",
    "colors[\"urban central gas CHP CC\"] = colors[\"CHP\"]\n",
    "colors[\"urban central solid biomass CHP\"] = colors[\"CHP\"]\n",
    "colors[\"urban central solid biomass CHP CC\"] = colors[\"CHP\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a9799dfb-d6a4-4540-8758-4293378fe560",
   "metadata": {},
   "outputs": [],
   "source": [
    "n = pypsa.Network(f\"{PATH}results/{RUN}/postnetworks/{SCENARIO}.nc\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "44b51f0a-8819-42ee-81ef-dccdf0f755b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "fn = f\"{PATH}resources/industrial_energy_demand_elec_s_181_2050.csv\"\n",
    "feedstock_emissions = (\n",
    "    pd.read_csv(fn, index_col=0)[\"process emission from feedstock\"].sum() * 1e6\n",
    ")  # t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4fe330b0-5696-420e-8472-eb74cee1e60f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_sankey(n):\n",
    "    columns = [\"label\", \"source\", \"target\", \"value\"]\n",
    "\n",
    "    gen = (\n",
    "        (n.snapshot_weightings.generators @ n.generators_t.p)\n",
    "        .groupby(\n",
    "            [\n",
    "                n.generators.carrier,\n",
    "                n.generators.carrier,\n",
    "                n.generators.bus.map(n.buses.carrier),\n",
    "            ]\n",
    "        )\n",
    "        .sum()\n",
    "        .div(1e6)\n",
    "    )  # TWh\n",
    "\n",
    "    gen.index.set_names(columns[:-1], inplace=True)\n",
    "    gen = gen.reset_index(name=\"value\")\n",
    "    gen = gen.loc[gen.value > 0.1]\n",
    "\n",
    "    gen[\"source\"] = gen[\"source\"].replace({\"gas\": \"fossil gas\", \"oil\": \"fossil oil\"})\n",
    "\n",
    "    sto = (\n",
    "        (n.snapshot_weightings.generators @ n.stores_t.p)\n",
    "        .groupby(\n",
    "            [n.stores.carrier, n.stores.carrier, n.stores.bus.map(n.buses.carrier)]\n",
    "        )\n",
    "        .sum()\n",
    "        .div(1e6)\n",
    "    )\n",
    "    sto.index.set_names(columns[:-1], inplace=True)\n",
    "    sto = sto.reset_index(name=\"value\")\n",
    "    sto = sto.loc[sto.value > 0.1]\n",
    "\n",
    "    su = (\n",
    "        (n.snapshot_weightings.generators @ n.storage_units_t.p)\n",
    "        .groupby(\n",
    "            [\n",
    "                n.storage_units.carrier,\n",
    "                n.storage_units.carrier,\n",
    "                n.storage_units.bus.map(n.buses.carrier),\n",
    "            ]\n",
    "        )\n",
    "        .sum()\n",
    "        .div(1e6)\n",
    "    )\n",
    "    su.index.set_names(columns[:-1], inplace=True)\n",
    "    su = su.reset_index(name=\"value\")\n",
    "    su = su.loc[su.value > 0.1]\n",
    "\n",
    "    load = (\n",
    "        (n.snapshot_weightings.generators @ as_dense(n, \"Load\", \"p_set\"))\n",
    "        .groupby([n.loads.carrier, n.loads.carrier, n.loads.bus.map(n.buses.carrier)])\n",
    "        .sum()\n",
    "        .div(1e6)\n",
    "        .swaplevel()\n",
    "    )  # TWh\n",
    "    load.index.set_names(columns[:-1], inplace=True)\n",
    "    load = load.reset_index(name=\"value\")\n",
    "\n",
    "    load = load.loc[~load.label.str.contains(\"emissions\")]\n",
    "    load.target += \" demand\"\n",
    "\n",
    "    for i in range(5):\n",
    "        n.links[f\"total_e{i}\"] = (\n",
    "            n.snapshot_weightings.generators @ n.links_t[f\"p{i}\"]\n",
    "        ).div(\n",
    "            1e6\n",
    "        )  # TWh\n",
    "        n.links[f\"carrier_bus{i}\"] = n.links[f\"bus{i}\"].map(n.buses.carrier)\n",
    "\n",
    "    def calculate_losses(x):\n",
    "        energy_ports = x.loc[\n",
    "            x.index.str.contains(\"carrier_bus\") & ~x.str.contains(\"co2\", na=False)\n",
    "        ].index.str.replace(\"carrier_bus\", \"total_e\")\n",
    "        return -x.loc[energy_ports].sum()\n",
    "\n",
    "    n.links[\"total_e5\"] = n.links.apply(calculate_losses, axis=1)\n",
    "    n.links[\"carrier_bus5\"] = \"losses\"\n",
    "\n",
    "    df = pd.concat(\n",
    "        [\n",
    "            n.links.groupby([\"carrier\", \"carrier_bus0\", \"carrier_bus\" + str(i)]).sum()[\n",
    "                \"total_e\" + str(i)\n",
    "            ]\n",
    "            for i in range(1, 6)\n",
    "        ]\n",
    "    ).reset_index()\n",
    "    df.columns = columns\n",
    "\n",
    "    # fix heat pump energy balance\n",
    "\n",
    "    hp = n.links.loc[n.links.carrier.str.contains(\"heat pump\")]\n",
    "\n",
    "    hp_t_elec = n.links_t.p0.filter(like=\"heat pump\")\n",
    "\n",
    "    grouper = [hp[\"carrier\"], hp[\"carrier_bus0\"], hp[\"carrier_bus1\"]]\n",
    "    hp_elec = (\n",
    "        (-n.snapshot_weightings.generators @ hp_t_elec)\n",
    "        .groupby(grouper)\n",
    "        .sum()\n",
    "        .div(1e6)\n",
    "        .reset_index()\n",
    "    )\n",
    "    hp_elec.columns = columns\n",
    "\n",
    "    df = df.loc[~(df.label.str.contains(\"heat pump\") & (df.target == \"losses\"))]\n",
    "\n",
    "    df.loc[df.label.str.contains(\"heat pump\"), \"value\"] -= hp_elec[\"value\"].values\n",
    "\n",
    "    df.loc[df.label.str.contains(\"air heat pump\"), \"source\"] = \"air-sourced ambient\"\n",
    "    df.loc[\n",
    "        df.label.str.contains(\"ground heat pump\"), \"source\"\n",
    "    ] = \"ground-sourced ambient\"\n",
    "\n",
    "    df = pd.concat([df, hp_elec])\n",
    "    df = df.set_index([\"label\", \"source\", \"target\"]).squeeze()\n",
    "    df = pd.concat(\n",
    "        [\n",
    "            df.loc[df < 0].mul(-1),\n",
    "            df.loc[df > 0].swaplevel(1, 2),\n",
    "        ]\n",
    "    ).reset_index()\n",
    "    df.columns = columns\n",
    "\n",
    "    # make DAC demand\n",
    "    df.loc[df.label == \"DAC\", \"target\"] = \"DAC\"\n",
    "\n",
    "    to_concat = [df, gen, su, sto, load]\n",
    "    connections = pd.concat(to_concat).sort_index().reset_index(drop=True)\n",
    "\n",
    "    # correct methanolisation\n",
    "\n",
    "    correction = connections.loc[\n",
    "        (connections.label == \"methanolisation\") & (connections.source == \"AC\"), \"value\"\n",
    "    ].sum()\n",
    "    connections.loc[\n",
    "        (connections.label == \"methanolisation\") & (connections.target == \"losses\"),\n",
    "        \"value\",\n",
    "    ] -= correction\n",
    "\n",
    "    connections.loc[\n",
    "        (connections.label == \"methanolisation\") & (connections.source == \"AC\"),\n",
    "        \"target\",\n",
    "    ] = \"losses\"\n",
    "\n",
    "    # aggregation\n",
    "\n",
    "    src_contains = connections.source.str.contains\n",
    "    trg_contains = connections.target.str.contains\n",
    "\n",
    "    connections.loc[src_contains(\"low voltage\"), \"source\"] = \"AC\"\n",
    "    connections.loc[trg_contains(\"low voltage\"), \"target\"] = \"AC\"\n",
    "    connections.loc[src_contains(\"water tank\"), \"source\"] = \"water tank\"\n",
    "    connections.loc[trg_contains(\"water tank\"), \"target\"] = \"water tank\"\n",
    "    connections.loc[src_contains(\"solar thermal\"), \"source\"] = \"solar thermal\"\n",
    "    connections.loc[src_contains(\"battery\"), \"source\"] = \"battery\"\n",
    "    connections.loc[trg_contains(\"battery\"), \"target\"] = \"battery\"\n",
    "    connections.loc[src_contains(\"Li ion\"), \"source\"] = \"battery\"\n",
    "    connections.loc[trg_contains(\"Li ion\"), \"target\"] = \"battery\"\n",
    "\n",
    "    connections.loc[src_contains(\"heat\") & ~src_contains(\"demand\"), \"source\"] = \"heat\"\n",
    "    connections.loc[trg_contains(\"heat\") & ~trg_contains(\"demand\"), \"target\"] = \"heat\"\n",
    "\n",
    "    connections = connections.loc[\n",
    "        ~(connections.source == connections.target)\n",
    "        & ~connections.source.str.contains(\"co2\")\n",
    "        & ~connections.target.str.contains(\"co2\")\n",
    "        & ~connections.source.str.contains(\"emissions\")\n",
    "        & ~connections.source.isin([\"gas for industry\", \"solid biomass for industry\"])\n",
    "        & (connections.value >= 0.5)\n",
    "        & ~connections.source.str.contains(\"losses\")\n",
    "    ]\n",
    "\n",
    "    where = connections.label == \"urban central gas boiler\"\n",
    "    connections.loc[where] = connections.loc[where].replace(\"losses\", \"fossil gas\")\n",
    "\n",
    "    connections.replace(\"AC\", \"electricity grid\", inplace=True)\n",
    "\n",
    "    return connections"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c8c24326-1d02-49bb-985a-f10421c5d36f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_sankey(connections, fn=None):\n",
    "    labels = np.unique(connections[[\"source\", \"target\"]])\n",
    "\n",
    "    nodes = pd.Series({v: i for i, v in enumerate(labels)})\n",
    "\n",
    "    node_colors = pd.Series(nodes.index.map(colors).fillna(\"grey\"), index=nodes.index)\n",
    "\n",
    "    link_colors = [\n",
    "        \"rgba{}\".format(to_rgba(node_colors[src], alpha=0.5))\n",
    "        for src in connections.source\n",
    "    ]\n",
    "\n",
    "    fig = go.Figure(\n",
    "        go.Sankey(\n",
    "            arrangement=\"snap\",  # [snap, nodepad, perpendicular, fixed]\n",
    "            valuesuffix=\"TWh\",\n",
    "            valueformat=\".1f\",\n",
    "            node=dict(pad=20, thickness=20, label=nodes.index, color=node_colors),\n",
    "            link=dict(\n",
    "                source=connections.source.map(nodes),\n",
    "                target=connections.target.map(nodes),\n",
    "                value=connections.value,\n",
    "                label=connections.label,\n",
    "                color=link_colors,\n",
    "            ),\n",
    "        )\n",
    "    )\n",
    "\n",
    "    fig.update_layout(title=f\"Sankey Diagram: {SCENARIO}\", font_size=15)\n",
    "\n",
    "    if fn is not None:\n",
    "        fig.write_html(fn + \".html\")\n",
    "        fig.write_image(fn + \".pdf\", width=1000, height=1000)\n",
    "    else:\n",
    "        fig.update_layout(height=1000)\n",
    "        return fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "939c2ede-5e2d-4f3d-b842-7a7c2ad6d576",
   "metadata": {},
   "outputs": [],
   "source": [
    "connections = prepare_sankey(n)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "612b962b-4172-461d-b942-d0f6785d1da7",
   "metadata": {},
   "outputs": [],
   "source": [
    "connections.to_csv(f\"{OUTPUT_SCENARIO}sankey.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77f19428-3a4d-4c16-bacd-0c41510699a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_sankey(connections, fn=f\"{OUTPUT_SCENARIO}sankey\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13e72d0e-0490-4bde-afbb-4e26c43037d2",
   "metadata": {},
   "source": [
    "## Carbon Sankey"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ee2f4306-c18d-49c2-9c45-0b34dcade7db",
   "metadata": {},
   "outputs": [],
   "source": [
    "columns = [\"label\", \"source\", \"target\", \"value\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d63ad57a-015b-46b6-bb0e-83332d05ac41",
   "metadata": {},
   "outputs": [],
   "source": [
    "collection = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7ae026d-94e3-4210-9a81-668ae7b5e18a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# DAC\n",
    "value = -(n.snapshot_weightings.generators @ n.links_t.p1.filter(like=\"DAC\")).sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(label=\"DAC\", source=\"co2 atmosphere\", target=\"co2 stored\", value=value)\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b18d849d-e852-41da-be61-02b9a7a0435e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# process emissions\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators @ n.links_t.p1.filter(regex=\"process emissions$\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"process emissions\",\n",
    "            source=\"process emissions\",\n",
    "            target=\"co2 atmosphere\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "71a90650-e9ae-4b29-a4e2-3892e698b016",
   "metadata": {},
   "outputs": [],
   "source": [
    "# process emissions CC\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators @ n.links_t.p1.filter(regex=\"process emissions CC\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"process emissions CC\",\n",
    "            source=\"process emissions\",\n",
    "            target=\"co2 atmosphere\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")\n",
    "\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators @ n.links_t.p2.filter(regex=\"process emissions CC\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"process emissions CC\",\n",
    "            source=\"process emissions\",\n",
    "            target=\"co2 stored\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "92726aaa-916d-49c1-b66a-c279906fb25e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# OCGT\n",
    "value = -(n.snapshot_weightings.generators @ n.links_t.p2.filter(like=\"OCGT\")).sum()\n",
    "collection.append(\n",
    "    pd.Series(dict(label=\"OCGT\", source=\"gas\", target=\"co2 atmosphere\", value=value))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "be7ae16a-8448-4b95-b070-c6de183df771",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Sabatier\n",
    "value = (n.snapshot_weightings.generators @ n.links_t.p2.filter(like=\"Sabatier\")).sum()\n",
    "collection.append(\n",
    "    pd.Series(dict(label=\"Sabatier\", source=\"co2 stored\", target=\"gas\", value=value))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "109512fe-39c6-4b09-b72b-366e56ab5335",
   "metadata": {},
   "outputs": [],
   "source": [
    "# SMR\n",
    "value = -(n.snapshot_weightings.generators @ n.links_t.p2.filter(regex=\"SMR$\")).sum()\n",
    "collection.append(\n",
    "    pd.Series(dict(label=\"SMR\", source=\"gas\", target=\"co2 atmosphere\", value=value))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8968182a-40f9-490c-812e-c957e3f32ad9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# SMR CC\n",
    "value = -(n.snapshot_weightings.generators @ n.links_t.p2.filter(regex=\"SMR CC\")).sum()\n",
    "collection.append(\n",
    "    pd.Series(dict(label=\"SMR CC\", source=\"gas\", target=\"co2 atmosphere\", value=value))\n",
    ")\n",
    "\n",
    "value = -(n.snapshot_weightings.generators @ n.links_t.p3.filter(like=\"SMR CC\")).sum()\n",
    "collection.append(\n",
    "    pd.Series(dict(label=\"SMR CC\", source=\"gas\", target=\"co2 stored\", value=value))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eadcd058-9818-4d53-8860-976139a5edaa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# gas boiler\n",
    "gas_boilers = [\n",
    "    \"residential rural gas boiler\",\n",
    "    \"services rural gas boiler\",\n",
    "    \"residential urban decentral gas boiler\",\n",
    "    \"services urban decentral gas boiler\",\n",
    "    \"urban central gas boiler\",\n",
    "]\n",
    "for gas_boiler in gas_boilers:\n",
    "    value = -(\n",
    "        n.snapshot_weightings.generators @ n.links_t.p2.filter(like=gas_boiler)\n",
    "    ).sum()\n",
    "    collection.append(\n",
    "        pd.Series(\n",
    "            dict(label=gas_boiler, source=\"gas\", target=\"co2 atmosphere\", value=value)\n",
    "        )\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "44d533d7-88d8-4bf5-9fec-2bb135e99004",
   "metadata": {},
   "outputs": [],
   "source": [
    "# biogas to gas\n",
    "value = (\n",
    "    n.snapshot_weightings.generators @ n.links_t.p2.filter(like=\"biogas to gas\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"biogas to gas\", source=\"co2 atmosphere\", target=\"biogas\", value=value\n",
    "        )\n",
    "    )\n",
    ")\n",
    "collection.append(\n",
    "    pd.Series(dict(label=\"biogas to gas\", source=\"biogas\", target=\"gas\", value=value))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5165e029-7cf1-4b47-b4c1-220e818e44b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# solid biomass for industry CC\n",
    "value = (\n",
    "    n.snapshot_weightings.generators\n",
    "    @ n.links_t.p2.filter(like=\"solid biomass for industry CC\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"solid biomass for industry CC\",\n",
    "            source=\"co2 atmosphere\",\n",
    "            target=\"solid biomass\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")\n",
    "\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators\n",
    "    @ n.links_t.p3.filter(like=\"solid biomass for industry CC\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"solid biomass for industry CC\",\n",
    "            source=\"solid biomass\",\n",
    "            target=\"co2 stored\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "61cf749b-eb1b-4ad0-bc55-d12c58070959",
   "metadata": {},
   "outputs": [],
   "source": [
    "# gas for industry\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators @ n.links_t.p1.filter(regex=\"gas for industry$\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"gas for industry\", source=\"gas\", target=\"co2 atmosphere\", value=value\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dc5a0c8e-4978-4555-ae0c-a9b4b2bedbbe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# gas for industry CC\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators @ n.links_t.p2.filter(like=\"gas for industry CC\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"gas for industry CC\",\n",
    "            source=\"gas\",\n",
    "            target=\"co2 atmosphere\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")\n",
    "\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators @ n.links_t.p3.filter(like=\"gas for industry CC\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"gas for industry CC\", source=\"gas\", target=\"co2 stored\", value=value\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9b352b17-fb26-49c4-a40c-bfb21e8f40f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fischer-Tropsch\n",
    "value = (\n",
    "    n.snapshot_weightings.generators @ n.links_t.p2.filter(like=\"Fischer-Tropsch\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(label=\"Fischer-Tropsch\", source=\"co2 stored\", target=\"oil\", value=value)\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fc782acb-8bde-4c5a-ba1b-3b72509877ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "# methanolisation\n",
    "value = (\n",
    "    n.snapshot_weightings.generators @ n.links_t.p3.filter(like=\"methanolisation\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"methanolisation\", source=\"co2 stored\", target=\"methanol\", value=value\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2d666402-5085-4ec5-93dd-11ef98a88f0a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# urban central gas CHP\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators @ n.links_t.p3.filter(like=\"urban central gas CHP\")\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"urban central gas CHP\",\n",
    "            source=\"gas\",\n",
    "            target=\"co2 atmosphere\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f060458d-03c6-4f09-b42c-6aaf2b8bb30a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# urban central gas CHP CC\n",
    "tech = \"urban central gas CHP CC\"\n",
    "value = -(n.snapshot_weightings.generators @ n.links_t.p3.filter(like=tech)).sum()\n",
    "collection.append(\n",
    "    pd.Series(dict(label=tech, source=\"gas\", target=\"co2 atmosphere\", value=value))\n",
    ")\n",
    "\n",
    "value = -(n.snapshot_weightings.generators @ n.links_t.p4.filter(like=tech)).sum()\n",
    "collection.append(\n",
    "    pd.Series(dict(label=tech, source=\"gas\", target=\"co2 stored\", value=value))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d08c781c-2679-4937-b7fb-94e0208c2157",
   "metadata": {},
   "outputs": [],
   "source": [
    "# urban solid biomass CHP CC\n",
    "tech = \"urban central solid biomass CHP CC\"\n",
    "\n",
    "value = (n.snapshot_weightings.generators @ n.links_t.p3.filter(like=tech)).sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(label=tech, source=\"co2 atmosphere\", target=\"solid biomass\", value=value)\n",
    "    )\n",
    ")\n",
    "\n",
    "value = -(n.snapshot_weightings.generators @ n.links_t.p4.filter(like=tech)).sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(label=tech, source=\"solid biomass\", target=\"co2 stored\", value=value)\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5fba7c72-3675-417f-8134-8e62d81f78b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# oil emissions\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators\n",
    "    @ as_dense(n, \"Load\", \"p_set\").filter(regex=\"^oil emissions\", axis=1)\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(label=\"oil emissions\", source=\"oil\", target=\"co2 atmosphere\", value=value)\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6b22ef69-00c7-4f47-8aa1-66a5c172a116",
   "metadata": {},
   "outputs": [],
   "source": [
    "# methanol emissions\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators\n",
    "    @ as_dense(n, \"Load\", \"p_set\").filter(like=\"methanol emissions\", axis=1)\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"methanol emissions\",\n",
    "            source=\"methanol\",\n",
    "            target=\"co2 atmosphere\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a73599fc-742f-43b5-bcf3-1ce0c0b180b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# agriculture machinery oil emissions\n",
    "value = -(\n",
    "    n.snapshot_weightings.generators\n",
    "    @ as_dense(n, \"Load\", \"p_set\").filter(\n",
    "        like=\"agriculture machinery oil emissions\", axis=1\n",
    "    )\n",
    ").sum()\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"agriculture machinery oil emissions\",\n",
    "            source=\"oil\",\n",
    "            target=\"co2 atmosphere\",\n",
    "            value=value,\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7982a9a2-860d-4641-8496-50c9a856109a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# feedstock_emissions\n",
    "collection.append(\n",
    "    pd.Series(\n",
    "        dict(\n",
    "            label=\"process emissions from feedstocks\",\n",
    "            source=\"oil\",\n",
    "            target=\"process emissions\",\n",
    "            value=feedstock_emissions,\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0cfdf0b7-2cec-46c3-8e88-e51ec531ac8e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat(collection, axis=1).T\n",
    "df.value /= 1e6  # Mt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "38bb932c-daa8-443f-8cdc-f33a8eb73cc6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fossil gas\n",
    "co2_intensity = 0.2  # t/MWh\n",
    "value = (n.snapshot_weightings.generators @ n.generators_t.p.filter(like=\"gas\")).div(\n",
    "    1e6\n",
    ").sum() * co2_intensity\n",
    "row = pd.DataFrame(\n",
    "    [dict(label=\"fossil gas\", source=\"fossil gas\", target=\"gas\", value=value)]\n",
    ")\n",
    "df = pd.concat([df, row], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0ab3c3da-f13a-47b5-a167-7fafea895644",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fossil oil\n",
    "co2_intensity = 0.27  # t/MWh\n",
    "value = (n.snapshot_weightings.generators @ n.generators_t.p.filter(like=\"oil\")).div(\n",
    "    1e6\n",
    ").sum() * co2_intensity\n",
    "row = pd.DataFrame(\n",
    "    [dict(label=\"fossil oil\", source=\"fossil oil\", target=\"oil\", value=value)]\n",
    ")\n",
    "df = pd.concat([df, row], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "45f607f5-0edd-45fd-a260-43fbaefdfb94",
   "metadata": {},
   "outputs": [],
   "source": [
    "# sequestration\n",
    "value = (\n",
    "    df.loc[df.target == \"co2 stored\", \"value\"].sum()\n",
    "    - df.loc[df.source == \"co2 stored\", \"value\"].sum()\n",
    ")\n",
    "row = pd.DataFrame(\n",
    "    [\n",
    "        dict(\n",
    "            label=\"co2 sequestration\",\n",
    "            source=\"co2 stored\",\n",
    "            target=\"co2 sequestration\",\n",
    "            value=value,\n",
    "        )\n",
    "    ]\n",
    ")\n",
    "df = pd.concat([df, row], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b0fdafe7-d1d3-4ed4-9ec4-609ad71a3547",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# net co2 emissions\n",
    "value = (\n",
    "    df.loc[df.target == \"co2 atmosphere\", \"value\"].sum()\n",
    "    - df.loc[df.source == \"co2 atmosphere\", \"value\"].sum()\n",
    ")\n",
    "row = pd.DataFrame(\n",
    "    [\n",
    "        dict(\n",
    "            label=\"net co2 emissions\",\n",
    "            source=\"co2 atmosphere\",\n",
    "            target=\"net co2 emissions\",\n",
    "            value=value,\n",
    "        )\n",
    "    ]\n",
    ")\n",
    "df = pd.concat([df, row], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "948a92fd-b51e-4ad0-b918-a0a86c31cc62",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_carbon_sankey(co2, fn=None):\n",
    "    labels = np.unique(co2[[\"source\", \"target\"]])\n",
    "\n",
    "    nodes = pd.Series({v: i for i, v in enumerate(labels)})\n",
    "\n",
    "    node_colors = pd.Series(nodes.index.map(colors).fillna(\"grey\"), index=nodes.index)\n",
    "\n",
    "    link_colors = [\n",
    "        \"rgba{}\".format(to_rgba(colors[src], alpha=0.5)) for src in co2.label\n",
    "    ]\n",
    "\n",
    "    fig = go.Figure(\n",
    "        go.Sankey(\n",
    "            arrangement=\"freeform\",  # [snap, nodepad, perpendicular, fixed]\n",
    "            valuesuffix=\" MtCO2\",\n",
    "            valueformat=\".1f\",\n",
    "            node=dict(pad=5, thickness=5, label=nodes.index, color=node_colors),\n",
    "            link=dict(\n",
    "                source=co2.source.map(nodes),\n",
    "                target=co2.target.map(nodes),\n",
    "                value=co2.value,\n",
    "                label=co2.label,\n",
    "                color=link_colors,\n",
    "            ),\n",
    "        )\n",
    "    )\n",
    "\n",
    "    fig.update_layout(\n",
    "        title=f\"Carbon Flow Sankey:<br>{SCENARIO}\",\n",
    "    )\n",
    "\n",
    "    if fn is not None:\n",
    "        fig.write_html(fn + \".html\")\n",
    "        fig.write_image(fn + \".pdf\", width=1200, height=1000)\n",
    "    else:\n",
    "        fig.update_layout(height=1400, margin=dict(l=100, r=0, t=0, b=150))\n",
    "        return fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e0c87987-4f02-4d80-95bb-de0890066d74",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(f\"{OUTPUT_SCENARIO}sankey-carbon.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f8dc14f6-af29-43de-a65a-e91abb3df0ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_carbon_sankey(df, fn=f\"{OUTPUT_SCENARIO}sankey-carbon\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fbf2b05d-2e0f-4ea3-b4d8-7356d8d2adbb",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "897ec0fe-b08b-4c40-aeef-672ac67f035f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d4c55b0c-e185-4d7e-ba4d-15fce7443234",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "05c60219-090b-4daf-a2a1-26741c186b50",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e0de7418-96f2-4bb9-b624-9699d72ac962",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f1059e7b-bed0-4bb7-b66d-d1becdea8fb7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "49d4f332-f695-4b85-b9d9-8e31d88229a4",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "db23460c-ad20-46b5-8126-645ac5aaddf2",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2cf58afe-fd3b-4684-ba2a-3038546ded65",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f70dad5e-4975-43e4-b525-45ba64b4cb59",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "494d35a45045e3fb9015f1d2cd094f8e1540ef3ea8b346f6f51a6ceb6be203c2"
  },
  "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": "3.10.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}