Issue 450 api option to customize which price and power sensors to take into account for scheduling

flexmeasures/api/v3_0/sensors.py

@@ -1,6 +1,6 @@
 from datetime import datetime, timedelta
 import json
-from typing import Optional
+from typing import List, Optional
 
 from flask import current_app
 from flask_classful import FlaskView, route
@@ -220,6 +220,15 @@ def get_data(self, response: dict):
             ),  # todo: allow unit to be set per field, using QuantityField("%", validate=validate.Range(min=0, max=1))
             "targets": fields.List(fields.Nested(TargetSchema), data_key="soc-targets"),
             # todo: add a duration parameter, instead of falling back to FLEXMEASURES_PLANNING_HORIZON
+            "up_deviation_price_sensor": SensorIdField(
+                data_key="consumption-price-sensor", required=False
+            ),
+            "down_deviation_price_sensor": SensorIdField(
+                data_key="feed-in-price-sensor", required=False
+            ),
+            "inflexible_device_sensors": fields.List(
+                SensorIdField, data_key="inflexible-device-sensors", required=False
+            ),
         },
         location="json",
     )
@@ -231,6 +240,9 @@ def trigger_schedule(  # noqa: C901
         unit: str,
         prior: datetime,
         roundtrip_efficiency: Optional[ur.Quantity] = None,
+        up_deviation_price_sensor: Optional[Sensor] = None,
+        down_deviation_price_sensor: Optional[Sensor] = None,
+        inflexible_device_sensors: Optional[List[Sensor]] = None,
         **kwargs,
     ):
         """
@@ -258,6 +270,10 @@ def trigger_schedule(  # noqa: C901
         with a target state of charge of 25 kWh at 4.00pm.
         The minimum and maximum soc are set to 10 and 25 kWh, respectively.
         Roundtrip efficiency for use in scheduling is set to 98%.
+        Consumption should be priced by sensor 9,
+        and feed-in (i.e. production) should be priced by sensor 10,
+        where the aggregate power flow includes the sum over sensors 13, 14 and 15.
+        Note that, if forecasts for sensors 13, 14 and 15 are not available, a schedule cannot be computed.
 
         .. code-block:: json
 
@@ -273,7 +289,10 @@ def trigger_schedule(  # noqa: C901
                 ],
                 "soc-min": 10,
                 "soc-max": 25,
-                "roundtrip-efficiency": 0.98
+                "roundtrip-efficiency": 0.98,
+                "consumption-price-sensor": 9,
+                "feed-in-price-sensor": 10,
+                "inflexible-device-sensors": [13, 14, 15]
             }
 
         **Example response**
@@ -395,6 +414,9 @@ def trigger_schedule(  # noqa: C901
             soc_min=soc_min,
             soc_max=soc_max,
             roundtrip_efficiency=roundtrip_efficiency,
+            up_deviation_price_sensor=up_deviation_price_sensor,
+            down_deviation_price_sensor=down_deviation_price_sensor,
+            inflexible_device_sensors=inflexible_device_sensors,
             enqueue=True,
         )
 

flexmeasures/cli/data_add.py

@@ -922,7 +922,8 @@ def create_schedule(
             soc_min=soc_min,
             soc_max=soc_max,
             roundtrip_efficiency=roundtrip_efficiency,
-            price_sensor=optimization_context_sensor,
+            up_deviation_price_sensor=optimization_context_sensor,
+            down_deviation_price_sensor=optimization_context_sensor,
         )
         if job:
             print(f"New scheduling job {job.id} has been added to the queue.")
@@ -938,7 +939,8 @@ def create_schedule(
             soc_min=soc_min,
             soc_max=soc_max,
             roundtrip_efficiency=roundtrip_efficiency,
-            price_sensor=optimization_context_sensor,
+            up_deviation_price_sensor=optimization_context_sensor,
+            down_deviation_price_sensor=optimization_context_sensor,
         )
         if success:
             print("New schedule is stored.")

flexmeasures/data/models/planning/battery.py

@@ -1,4 +1,4 @@
-from typing import Optional, Union
+from typing import List, Optional, Union
 from datetime import datetime, timedelta
 
 import pandas as pd
@@ -10,6 +10,7 @@
     initialize_series,
     add_tiny_price_slope,
     get_prices,
+    inflexible_device_forecasts,
     fallback_charging_policy,
 )
 
@@ -25,7 +26,9 @@ def schedule_battery(
     soc_max: Optional[float] = None,
     roundtrip_efficiency: Optional[float] = None,
     prefer_charging_sooner: bool = True,
-    price_sensor: Optional[Sensor] = None,
+    up_deviation_price_sensor: Optional[Sensor] = None,
+    down_deviation_price_sensor: Optional[Sensor] = None,
+    inflexible_device_sensors: Optional[List[Sensor]] = None,
     round_to_decimals: Optional[int] = 6,
 ) -> Union[pd.Series, None]:
     """Schedule a battery asset based directly on the latest beliefs regarding market prices within the specified time
@@ -56,13 +59,21 @@ def schedule_battery(
         soc_max = sensor.get_attribute("max_soc_in_mwh")
 
     # Check for known prices or price forecasts, trimming planning window accordingly
-    prices, (start, end) = get_prices(
+    up_deviation_prices, (start, end) = get_prices(
         (start, end),
         resolution,
-        price_sensor=price_sensor,
+        price_sensor=up_deviation_price_sensor,
         sensor=sensor,
         allow_trimmed_query_window=True,
     )
+    down_deviation_prices, (start, end) = get_prices(
+        (start, end),
+        resolution,
+        price_sensor=down_deviation_price_sensor,
+        sensor=sensor,
+        allow_trimmed_query_window=True,
+    )
+
     start = pd.Timestamp(start).tz_convert("UTC")
     end = pd.Timestamp(end).tz_convert("UTC")
     if soc_targets is not None:
@@ -73,20 +84,23 @@ def schedule_battery(
     # Add tiny price slope to prefer charging now rather than later, and discharging later rather than now.
     # We penalise the future with at most 1 per thousand times the price spread.
     if prefer_charging_sooner:
-        prices = add_tiny_price_slope(prices, "event_value")
+        up_deviation_prices = add_tiny_price_slope(up_deviation_prices, "event_value")
+        down_deviation_prices = add_tiny_price_slope(
+            down_deviation_prices, "event_value"
+        )
 
     # Set up commitments to optimise for
     commitment_quantities = [initialize_series(0, start, end, resolution)]
 
     # Todo: convert to EUR/(deviation of commitment, which is in MW)
     commitment_upwards_deviation_price = [
-        prices.loc[start : end - resolution]["event_value"]
+        up_deviation_prices.loc[start : end - resolution]["event_value"]
     ]
     commitment_downwards_deviation_price = [
-        prices.loc[start : end - resolution]["event_value"]
+        down_deviation_prices.loc[start : end - resolution]["event_value"]
     ]
 
-    # Set up device constraints (only one device for this EMS)
+    # Set up device constraints (only one flexible device for this EMS, plus inflexible devices)
     columns = [
         "equals",
         "max",
@@ -97,7 +111,18 @@ def schedule_battery(
         "derivative down efficiency",
         "derivative up efficiency",
     ]
-    device_constraints = [initialize_df(columns, start, end, resolution)]
+    if inflexible_device_sensors is None:
+        inflexible_device_sensors = []
+    device_constraints = [
+        initialize_df(columns, start, end, resolution)
+        for i in range(1 + len(inflexible_device_sensors))
+    ]
+    for i, inflexible_sensor in enumerate(inflexible_device_sensors):
+        device_constraints[i + 1]["derivative equals"] = inflexible_device_forecasts(
+            (start, end),
+            resolution,
+            inflexible_sensor,
+        )
     if soc_targets is not None:
         device_constraints[0]["equals"] = soc_targets.shift(
             -1, freq=resolution
@@ -121,9 +146,13 @@ def schedule_battery(
     device_constraints[0]["derivative down efficiency"] = roundtrip_efficiency**0.5
     device_constraints[0]["derivative up efficiency"] = roundtrip_efficiency**0.5
 
-    # Set up EMS constraints (no additional constraints)
+    # Set up EMS constraints
     columns = ["derivative max", "derivative min"]
     ems_constraints = initialize_df(columns, start, end, resolution)
+    ems_capacity = sensor.generic_asset.get_attribute("capacity_in_mw")
+    if ems_capacity is not None:
+        ems_constraints["derivative min"] = ems_capacity * -1
+        ems_constraints["derivative max"] = ems_capacity
 
     ems_schedule, expected_costs, scheduler_results = device_scheduler(
         device_constraints,

flexmeasures/data/models/planning/charging_station.py

@@ -1,4 +1,4 @@
-from typing import Optional, Union
+from typing import List, Optional, Union
 from datetime import datetime, timedelta
 
 import pandas as pd
@@ -10,6 +10,7 @@
     initialize_series,
     add_tiny_price_slope,
     get_prices,
+    inflexible_device_forecasts,
     fallback_charging_policy,
 )
 
@@ -25,7 +26,9 @@ def schedule_charging_station(
     soc_max: Optional[float] = None,
     roundtrip_efficiency: Optional[float] = None,
     prefer_charging_sooner: bool = True,
-    price_sensor: Optional[Sensor] = None,
+    up_deviation_price_sensor: Optional[Sensor] = None,
+    down_deviation_price_sensor: Optional[Sensor] = None,
+    inflexible_device_sensors: Optional[List[Sensor]] = None,
     round_to_decimals: Optional[int] = 6,
 ) -> Union[pd.Series, None]:
     """Schedule a charging station asset based directly on the latest beliefs regarding market prices within the specified time
@@ -53,13 +56,21 @@ def schedule_charging_station(
         soc_max = sensor.get_attribute("max_soc_in_mwh", max(soc_targets.values))
 
     # Check for known prices or price forecasts, trimming planning window accordingly
-    prices, (start, end) = get_prices(
+    up_deviation_prices, (start, end) = get_prices(
         (start, end),
         resolution,
-        price_sensor=price_sensor,
+        price_sensor=up_deviation_price_sensor,
         sensor=sensor,
         allow_trimmed_query_window=True,
     )
+    down_deviation_prices, (start, end) = get_prices(
+        (start, end),
+        resolution,
+        price_sensor=down_deviation_price_sensor,
+        sensor=sensor,
+        allow_trimmed_query_window=True,
+    )
+
     # soc targets are at the end of each time slot, while prices are indexed by the start of each time slot
     soc_targets = soc_targets.tz_convert("UTC")
     start = pd.Timestamp(start).tz_convert("UTC")
@@ -69,20 +80,23 @@ def schedule_charging_station(
     # Add tiny price slope to prefer charging now rather than later, and discharging later rather than now.
     # We penalise the future with at most 1 per thousand times the price spread.
     if prefer_charging_sooner:
-        prices = add_tiny_price_slope(prices, "event_value")
+        up_deviation_prices = add_tiny_price_slope(up_deviation_prices, "event_value")
+        down_deviation_prices = add_tiny_price_slope(
+            down_deviation_prices, "event_value"
+        )
 
     # Set up commitments to optimise for
     commitment_quantities = [initialize_series(0, start, end, resolution)]
 
     # Todo: convert to EUR/(deviation of commitment, which is in MW)
     commitment_upwards_deviation_price = [
-        prices.loc[start : end - resolution]["event_value"]
+        up_deviation_prices.loc[start : end - resolution]["event_value"]
     ]
     commitment_downwards_deviation_price = [
-        prices.loc[start : end - resolution]["event_value"]
+        down_deviation_prices.loc[start : end - resolution]["event_value"]
     ]
 
-    # Set up device constraints (only one device for this EMS)
+    # Set up device constraints (only one flexible device for this EMS, plus inflexible devices)
     columns = [
         "equals",
         "max",
@@ -91,7 +105,17 @@ def schedule_charging_station(
         "derivative max",
         "derivative min",
     ]
-    device_constraints = [initialize_df(columns, start, end, resolution)]
+    if inflexible_device_sensors is None:
+        inflexible_device_sensors = []
+    device_constraints = [initialize_df(columns, start, end, resolution)] * (
+        1 + len(inflexible_device_sensors)
+    )
+    for i, inflexible_sensor in enumerate(inflexible_device_sensors):
+        device_constraints[i + 1]["derivative equals"] = inflexible_device_forecasts(
+            (start, end),
+            resolution,
+            inflexible_sensor,
+        )
     device_constraints[0]["equals"] = soc_targets.shift(-1, freq=resolution).values * (
         timedelta(hours=1) / resolution
     ) - soc_at_start * (
@@ -121,9 +145,13 @@ def schedule_charging_station(
     device_constraints[0]["derivative down efficiency"] = roundtrip_efficiency**0.5
     device_constraints[0]["derivative up efficiency"] = roundtrip_efficiency**0.5
 
-    # Set up EMS constraints (no additional constraints)
+    # Set up EMS constraints
     columns = ["derivative max", "derivative min"]
     ems_constraints = initialize_df(columns, start, end, resolution)
+    ems_capacity = sensor.generic_asset.get_attribute("capacity_in_mw")
+    if ems_capacity is not None:
+        ems_constraints["derivative min"] = ems_capacity * -1
+        ems_constraints["derivative max"] = ems_capacity
 
     ems_schedule, expected_costs, scheduler_results = device_scheduler(
         device_constraints,

flexmeasures/data/models/planning/exceptions.py

@@ -6,6 +6,10 @@ class UnknownMarketException(Exception):
     pass
 
 
+class UnknownForecastException(Exception):
+    pass
+
+
 class UnknownPricesException(Exception):
     pass
 

flexmeasures/data/models/planning/solver.py

@@ -173,15 +173,21 @@ def ems_derivative_min_select(m, j):
 
     def device_derivative_down_efficiency(m, d, j):
         try:
-            return device_constraints[d]["derivative down efficiency"].iloc[j]
+            eff = device_constraints[d]["derivative down efficiency"].iloc[j]
         except KeyError:
             return 1
+        if np.isnan(eff):
+            return 1
+        return eff
 
     def device_derivative_up_efficiency(m, d, j):
         try:
-            return device_constraints[d]["derivative up efficiency"].iloc[j]
+            eff = device_constraints[d]["derivative up efficiency"].iloc[j]
         except KeyError:
             return 1
+        if np.isnan(eff):
+            return 1
+        return eff
 
     model.up_price = Param(model.c, model.j, initialize=price_up_select)
     model.down_price = Param(model.c, model.j, initialize=price_down_select)
@@ -237,17 +243,19 @@ def device_derivative_bounds(m, d, j):
         )
 
     def device_down_derivative_bounds(m, d, j):
+        """Strictly non-positive."""
         return (
-            m.device_derivative_min[d, j],
+            min(m.device_derivative_min[d, j], 0),
             m.device_power_down[d, j],
             0,
         )
 
     def device_up_derivative_bounds(m, d, j):
+        """Strictly non-negative."""
         return (
             0,
             m.device_power_up[d, j],
-            m.device_derivative_max[d, j],
+            max(0, m.device_derivative_max[d, j]),
         )
 
     def ems_derivative_bounds(m, j):