Merge pull request #210 from agoenergy/137-improve-market-scanning-tab

137 improve market scanning tab

app/tab_market_scanning.py

@@ -4,11 +4,15 @@
 import plotly.express as px
 import streamlit as st
 
-from app.ptxboa_functions import config_number_columns, remove_subregions
+from app.ptxboa_functions import (
+    config_number_columns,
+    prepare_and_download_df_as_excel,
+    remove_subregions,
+)
 from ptxboa.api import PtxboaAPI
 
 
-def content_market_scanning(api: PtxboaAPI, res_costs: pd.DataFrame) -> None:
+def content_market_scanning(api: PtxboaAPI, res_costs: pd.DataFrame, cd: dict) -> None:
     """Create content for the "market scanning" sheet.
 
     Parameters
@@ -17,6 +21,8 @@ def content_market_scanning(api: PtxboaAPI, res_costs: pd.DataFrame) -> None:
         an instance of the api class
     res_costs : pd.DataFrame
         Results.
+    cd: dict
+        context data.
     """
     with st.expander("What is this?"):
         st.markdown(
@@ -47,39 +53,195 @@ def content_market_scanning(api: PtxboaAPI, res_costs: pd.DataFrame) -> None:
     )
 
     # merge costs and distances:
-    df_plot = pd.DataFrame()
-    df_plot["total costs"] = res_costs["Total"]
-    df_plot = df_plot.merge(distances, left_index=True, right_index=True)
+    df = pd.DataFrame()
+    df["total costs"] = res_costs["Total"]
+    df = df.merge(distances, left_index=True, right_index=True)
+
+    # merge RE supply potential from context data:
+    df = df.merge(
+        cd["supply"].set_index("country_name")[
+            ["re_tech_pot_EWI", "re_tech_pot_PTXAtlas"]
+        ],
+        left_index=True,
+        right_index=True,
+        how="left",
+    )
+
+    # prettify column names:
+    df.rename(
+        {
+            "total costs": f"Total costs ({st.session_state['output_unit']})",
+            "shipping distance": "Shipping distance (km)",
+            "pipeline distance": "Pipeline distance (km)",
+            "re_tech_pot_EWI": "RE technical potential (EWI) (TWh/a)",
+            "re_tech_pot_PTXAtlas": "RE technical potential (PTX Atlas) (TWh/a)",
+        },
+        inplace=True,
+        axis=1,
+    )
+
+    # replace nan entries:
+    df = df.replace({"no potential": None, "no analysis ": None})
+    df = df.astype(float)
 
     # do not show subregions:
-    df_plot = remove_subregions(api, df_plot, st.session_state["country"])
+    df = remove_subregions(api, df, st.session_state["country"])
+
+    with st.container(border=True):
+        st.markdown(
+            "### Costs and transportation distances from different supply regions"
+            f" to {st.session_state['country']}"
+        )
+        c1, c2 = st.columns(2)
+        with c1:
+            # select which distance to show:
+            selected_distance = st.radio(
+                "Select parameter to show on horizontal axis:",
+                ["Shipping distance (km)", "Pipeline distance (km)"],
+                key="selected_distance_supply_regions",
+            )
 
-    # create plot:st.session_state
-    [c1, c2] = st.columns([1, 5])
-    with c1:
-        # select which distance to show:
+        with c2:
+            # select parameter for marker size:
+            parameter_for_marker_size = st.radio(
+                "Select parameter to scale marker size:",
+                [
+                    "RE technical potential (EWI) (TWh/a)",
+                    "RE technical potential (PTX Atlas) (TWh/a)",
+                    "None",
+                ],
+            )
+
+        # create plot:
+        df_plot = df.copy().round(0)
+
+        # distinguish between selected region and others:
+        df_plot["category"] = "other regions"
+        df_plot.at[st.session_state["region"], "category"] = "selected supply region"
+
+        if parameter_for_marker_size == "None":
+            fig = px.scatter(
+                df_plot,
+                x=selected_distance,
+                y=f"Total costs ({st.session_state['output_unit']})",
+                color="category",
+                color_discrete_sequence=["#1A667B", "#D05094"],
+                text=df_plot.index,
+            )
+            fig.update_traces(textposition="top center")
+        else:
+            df_plot = df_plot.loc[df_plot[parameter_for_marker_size] > 0]
+            fig = px.scatter(
+                df_plot,
+                x=selected_distance,
+                y=f"Total costs ({st.session_state['output_unit']})",
+                size=parameter_for_marker_size,
+                size_max=50,
+                color="category",
+                color_discrete_sequence=["#1A667B", "#D05094"],
+                text=df_plot.index,
+            )
+            fig.update_traces(textposition="top center")
+
+        st.plotly_chart(fig, use_container_width=True)
+
+        # show data in tabular form:
+        with st.expander("**Data**"):
+            column_config = config_number_columns(df, format="%.0f")
+            st.dataframe(
+                df,
+                use_container_width=True,
+                column_config=column_config,
+            )
+
+            fn = "market_scanning_supply_regions"
+            prepare_and_download_df_as_excel(df, filename=fn)
+
+    with st.container(border=True):
+        st.markdown(
+            f"### Transportation distances from {st.session_state['region']}"
+            " to different target countries"
+        )
+
+        # filter shipping and pipeline distances:
+        df = input_data.loc[
+            (
+                input_data["parameter_code"].isin(
+                    ["shipping distance", "pipeline distance"]
+                )
+            )
+            & (input_data["source_region_code"] == st.session_state["region"]),
+            ["target_country_code", "parameter_code", "value"],
+        ]
+        df = df.pivot_table(
+            index="target_country_code",
+            columns="parameter_code",
+            values="value",
+            aggfunc="sum",
+        )
+
+        # merge H2 demand from context data:
+
+        # H2 demand data hard coded from excel tool.
+        # TODO: improve data representation in context_data.xlsx
+        # and take data from there.
+        df["h2_demand_2030"] = None
+
+        df.at["China", "h2_demand_2030"] = 35
+        df.at["France", "h2_demand_2030"] = None
+        df.at["Germany", "h2_demand_2030"] = 3
+        df.at["India", "h2_demand_2030"] = 10
+        df.at["Japan", "h2_demand_2030"] = 3
+        df.at["Netherlands", "h2_demand_2030"] = None
+        df.at["South Korea", "h2_demand_2030"] = 2
+        df.at["Spain", "h2_demand_2030"] = None
+        df.at["USA", "h2_demand_2030"] = 10
+
+        # EU data is missing completely in context_data.xlsx:
+        df.at["EU", "shipping distance"] = 1104
+        df.at["EU", "pipeline distance"] = 2000
+        df.at["EU", "h2_demand_2030"] = 20
+
+        df["h2_demand_2030"] = df["h2_demand_2030"].astype(float)
+
+        # prettify column names:
+        df.rename(
+            {
+                "shipping distance": "Shipping distance (km)",
+                "pipeline distance": "Pipeline distance (km)",
+                "h2_demand_2030": "Projected H2 demand in 2030 (Mt/a)",
+            },
+            inplace=True,
+            axis=1,
+        )
+
+        # select parameter to display on x axis:
         selected_distance = st.radio(
-            "Select parameter:",
-            ["shipping distance", "pipeline distance"],
+            "Select parameter to show on horizontal axis:",
+            ["Shipping distance (km)", "Pipeline distance (km)"],
+            key="selected_distance_target_countries",
         )
-    with c2:
+
+        # create plot:
+        df_plot = df.copy().round(0)
+
+        # distinguish between selected region and others:
+        df_plot["category"] = "other countries"
+        df_plot.at[st.session_state["country"], "category"] = "selected target country"
+
         fig = px.scatter(
             df_plot,
             x=selected_distance,
-            y="total costs",
-            title="Costs and transportation distances",
-            height=600,
-        )
-        # Add text above markers
-        fig.update_traces(
+            y="Projected H2 demand in 2030 (Mt/a)",
+            color="category",
+            color_discrete_sequence=["#1A667B", "#D05094"],
             text=df_plot.index,
-            textposition="top center",
-            mode="markers+text",
         )
+        fig.update_traces(textposition="top center")
 
-        st.plotly_chart(fig)
+        st.plotly_chart(fig, use_container_width=True)
 
-    # show data in tabular form:
-    st.markdown("**Data:**")
-    column_config = config_number_columns(df_plot, format="%.1f")
-    st.dataframe(df_plot, use_container_width=True, column_config=column_config)
+        with st.expander("**Data**"):
+            st.dataframe(df, use_container_width=True)
+            fn = "market_scanning_target_countries"
+            prepare_and_download_df_as_excel(df, filename=fn)

ptxboa_streamlit.py

@@ -1,5 +1,10 @@
 # -*- coding: utf-8 -*-
-"""Mockup streamlit app."""
+"""
+PtX BOA streamlit app, main file.
+
+Execution:
+>>> streamlit run  ptxboa_streamlit.py
+"""
 # TODO how do I use the treamlit logger?
 import logging
 
@@ -39,7 +44,7 @@
 logger.addHandler(log_handler)
 
 
-logger.info("Updating app...")
+logger.debug("Updating app...")
 
 # app layout:
 
@@ -191,7 +196,7 @@
     )
 
 if st.session_state[st.session_state["tab_key"]] == "Market scanning":
-    content_market_scanning(api, costs_per_region)
+    content_market_scanning(api, costs_per_region, cd)
 
 if st.session_state[st.session_state["tab_key"]] == "Input data":
     content_input_data(api)