Enhancement/update rated power in tie (#268)

* changing turbine capacity to rated power from asset table in TIE * rerunning example 3 notebook * update plantdata TIE test for asset table * remove unused prepare_scada method in TIE * update TIE test results * updating TIE example notebook 3 * updating changelog for TIE power curve filtering change * rerun schema generation --------- Co-authored-by: RHammond2 <[email protected]>
NREL · Feb 9, 2024 · 4fbdfd1 · 4fbdfd1
1 parent a117550
commit 4fbdfd1
Show file tree

Hide file tree

Showing 25 changed files with 258 additions and 399 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -30,6 +30,9 @@ All notable changes to this project will be documented in this file. If you make
   results for the TIE and wake loss regression tests.
 - `openoa.utils.timeseries.gap_fill_data_frame()` now returns the original data if there is no data
   to fill in, avoiding a Pandas `concat` deprecation warning about pending behavioral changes.
+- The turbine capacity value used for power curve filtering in `TurbineLongTermGrossEnergy` is
+  changed to the rated power from the asset table instead of the maximum power from SCADA. This
+  makes the power curve filtering more robust to turbine power outliers above rated power.
 
 ## [3.0.1 - 2023-12-22]
 

diff --git a/examples/03_turbine_ideal_energy.ipynb b/examples/03_turbine_ideal_energy.ipynb
diff --git a/openoa/analysis/turbine_long_term_gross_energy.py b/openoa/analysis/turbine_long_term_gross_energy.py
@@ -310,45 +310,6 @@ def setup_inputs(self) -> None:
 
         self._inputs = pd.DataFrame(inputs)
 
-    @logged_method_call
-    def prepare_scada(self) -> None:
-        """
-        Performs the following manipulations:
-         1. Creates a copy of the SCADA data
-         2. Sorts it by turbine asset_id, then timestamp (the two index columns)
-         3. Drops any rows that don't have any windspeed or energy data
-         4. Flags windspeed values outside the range [0, 40]
-         5. Flags windspeed values that have stayed the same for at least 3 straight readings
-         6. Flags power values outside the range [0, turbine capacity]
-         7. Flags windspeed and power values that don't mutually coincide within a reasonable range
-         8. Combine the flags using an "or" combination to be a new column in scada: "valid"
-        """
-        self.scada = (
-            self.plant.scada.swaplevel().sort_index().dropna(subset=["WMET_HorWdSpd", "WTUR_SupWh"])
-        )
-        turbine_capacity = self.scada.groupby(level="asset_id").max()["WTUR_W"]
-        flag_range = filters.range_flag(self.scada.loc[:, "WMET_HorWdSpd"], below=0, above=40)
-        flag_frozen = filters.unresponsive_flag(self.scada.loc[:, "WMET_HorWdSpd"], threshold=3)
-        flag_neg = pd.Series(index=self.scada.index, dtype=bool)
-        flag_window = pd.Series(index=self.scada.index, dtype=bool)
-        for t in self.turbine_ids:
-            ix_turb = self.scada.index.get_level_values("asset_id") == t
-            flag_neg.loc[ix_turb] = filters.range_flag(
-                self.scada.loc[ix_turb, "power"], below=0, above=turbine_capacity.loc[t]
-            )
-            flag_window.loc[ix_turb] = filters.window_range_flag(
-                window_col=self.scada.loc[ix_turb, "WMET_HorWdSpd"],
-                window_start=5.0,
-                window_end=40,
-                value_col=self.scada.loc[ix_turb, "WTUR_W"],
-                value_min=0.02 * turbine_capacity.loc[t],
-                value_max=1.2 * turbine_capacity.loc[t],
-            )
-
-        flag_final = ~(flag_range | flag_frozen | flag_neg | flag_window).values
-        self.scada.assign(valid=flag_final)
-        self.scada.assign(valid_run=flag_final)
-
     def sort_scada_by_turbine(self) -> None:
         """
         Sorts the SCADA DataFrame by the asset_id and timestamp index columns, respectively.
@@ -370,42 +331,50 @@ def filter_turbine_data(self) -> None:
         """
         Apply a set of filtering algorithms to the turbine wind speed vs power curve to flag
         data not representative of normal turbine operation
+
+        Performs the following manipulations:
+         1. Drops any scada rows that don't have any windspeed or energy data
+         2. Flags windspeed values outside the range [0, 40]
+         3. Flags windspeed values that have stayed the same for at least 3 straight readings
+         4. Flags power values less than 2% of turbine capacity when wind speed above cut-in
+         5. Flags windspeed and power values that don't mutually coincide within a reasonable range
+         6. Combine the flags using an "or" combination to be a new column in scada: "flag_final"
         """
 
         dic = self.scada_dict
 
         # Loop through turbines
         for t in self.turbine_ids:
             scada_df = self.scada_dict[t]
-            turbine_capacity_real = scada_df.WTUR_W.max()
+            turbine_capacity = self.plant.asset.loc[t, "rated_power"]
 
             max_bin = (
-                self._run.max_power_filter * turbine_capacity_real
+                self._run.max_power_filter * turbine_capacity
             )  # Set maximum range for using bin-filter
 
             scada_df.dropna(
                 subset=["WMET_HorWdSpd", "WTUR_SupWh"], inplace=True
             )  # Drop any data where scada wind speed or energy is NaN
 
             scada_df = scada_df.assign(
-                flag_neg=filters.range_flag(scada_df.WTUR_W, lower=0, upper=turbine_capacity_real),
+                flag_neg=filters.range_flag(scada_df.WTUR_W, lower=0, upper=scada_df.WTUR_W.max()),
                 flag_range=filters.range_flag(scada_df.WMET_HorWdSpd, lower=0, upper=40),
                 flag_frozen=filters.unresponsive_flag(scada_df.WMET_HorWdSpd, threshold=3),
                 flag_window=filters.window_range_flag(
                     window_col=dic[t].loc[:, "WMET_HorWdSpd"],
                     window_start=5.0,
                     window_end=40,
                     value_col=dic[t].loc[:, "WTUR_W"],
-                    value_min=0.02 * turbine_capacity_real,
-                    value_max=1.2 * turbine_capacity_real,
+                    value_min=0.02 * turbine_capacity,
+                    value_max=1.2 * turbine_capacity,
                 ),
                 flag_bin=filters.bin_filter(
                     bin_col=dic[t].loc[:, "WTUR_W"],
                     value_col=dic[t].loc[:, "WMET_HorWdSpd"],
-                    bin_width=0.06 * turbine_capacity_real,
+                    bin_width=0.06 * turbine_capacity,
                     threshold=self._run.wind_bin_thresh,
                     center_type="median",
-                    bin_min=np.round(0.01 * turbine_capacity_real),
+                    bin_min=np.round(0.01 * turbine_capacity),
                     bin_max=np.round(max_bin),
                     threshold_type="std",
                     direction="all",

diff --git a/openoa/schema/base_electrical_losses_schema.json b/openoa/schema/base_electrical_losses_schema.json
@@ -6,16 +6,16 @@
       "units": "kWh"
     },
     "frequency": [
-      "us",
-      "ns",
       "ME",
-      "W",
+      "ns",
       "D",
-      "s",
-      "MS",
       "h",
       "ms",
-      "min"
+      "MS",
+      "min",
+      "us",
+      "s",
+      "W"
     ]
   },
   "scada": {
@@ -30,13 +30,13 @@
       "units": null
     },
     "frequency": [
-      "us",
       "ns",
       "D",
-      "s",
       "h",
       "ms",
-      "min"
+      "min",
+      "us",
+      "s"
     ]
   }
 }
diff --git a/openoa/schema/base_electrical_losses_schema.yml b/openoa/schema/base_electrical_losses_schema.yml
@@ -8,26 +8,26 @@ scada:
     dtype: float
     units: kW
   frequency:
-  - us
   - ns
   - D
-  - s
   - h
   - ms
   - min
+  - us
+  - s
 meter:
   MMTR_SupWh:
     name: MMTR_SupWh
     dtype: float
     units: kWh
   frequency:
-  - us
-  - ns
   - ME
-  - W
+  - ns
   - D
-  - s
-  - MS
   - h
   - ms
+  - MS
   - min
+  - us
+  - s
+  - W
diff --git a/openoa/schema/base_monte_carlo_aep_schema.json b/openoa/schema/base_monte_carlo_aep_schema.json
@@ -11,16 +11,16 @@
       "units": "kWh"
     },
     "frequency": [
-      "us",
-      "ns",
       "ME",
-      "W",
+      "ns",
       "D",
-      "s",
-      "MS",
       "h",
       "ms",
-      "min"
+      "MS",
+      "min",
+      "us",
+      "s",
+      "W"
     ]
   },
   "meter": {
@@ -30,16 +30,16 @@
       "units": "kWh"
     },
     "frequency": [
-      "us",
-      "ns",
       "ME",
-      "W",
+      "ns",
       "D",
-      "s",
-      "MS",
       "h",
       "ms",
-      "min"
+      "MS",
+      "min",
+      "us",
+      "s",
+      "W"
     ]
   },
   "reanalysis": {
@@ -54,16 +54,16 @@
       "units": "m/s"
     },
     "frequency": [
-      "us",
-      "ns",
       "ME",
-      "W",
+      "ns",
       "D",
-      "s",
-      "MS",
       "h",
       "ms",
-      "min"
+      "MS",
+      "min",
+      "us",
+      "s",
+      "W"
     ]
   }
 }
diff --git a/openoa/schema/base_monte_carlo_aep_schema.yml b/openoa/schema/base_monte_carlo_aep_schema.yml
@@ -4,16 +4,16 @@ meter:
     dtype: float
     units: kWh
   frequency:
-  - us
-  - ns
   - ME
-  - W
+  - ns
   - D
-  - s
-  - MS
   - h
   - ms
+  - MS
   - min
+  - us
+  - s
+  - W
 curtail:
   IAVL_ExtPwrDnWh:
     name: IAVL_ExtPwrDnWh
@@ -24,16 +24,16 @@ curtail:
     dtype: float
     units: kWh
   frequency:
-  - us
-  - ns
   - ME
-  - W
+  - ns
   - D
-  - s
-  - MS
   - h
   - ms
+  - MS
   - min
+  - us
+  - s
+  - W
 reanalysis:
   WMETR_HorWdSpd:
     name: WMETR_HorWdSpd
@@ -44,13 +44,13 @@ reanalysis:
     dtype: float
     units: kg/m^3
   frequency:
-  - us
-  - ns
   - ME
-  - W
+  - ns
   - D
-  - s
-  - MS
   - h
   - ms
+  - MS
   - min
+  - us
+  - s
+  - W
diff --git a/openoa/schema/base_tie_schema.json b/openoa/schema/base_tie_schema.json
@@ -1,4 +1,11 @@
 {
+  "asset": {
+    "rated_power": {
+      "dtype": "float",
+      "name": "rated_power",
+      "units": "kW"
+    }
+  },
   "reanalysis": {
     "WMETR_AirDen": {
       "dtype": "float",
@@ -16,13 +23,13 @@
       "units": "m/s"
     },
     "frequency": [
-      "us",
       "ns",
       "D",
-      "s",
       "h",
       "ms",
-      "min"
+      "min",
+      "us",
+      "s"
     ]
   },
   "scada": {
@@ -42,13 +49,13 @@
       "units": null
     },
     "frequency": [
-      "us",
       "ns",
       "D",
-      "s",
       "h",
       "ms",
-      "min"
+      "min",
+      "us",
+      "s"
     ]
   }
 }
diff --git a/openoa/schema/base_tie_schema.yml b/openoa/schema/base_tie_schema.yml
@@ -12,13 +12,18 @@ scada:
     dtype: float
     units: m/s
   frequency:
-  - us
   - ns
   - D
-  - s
   - h
   - ms
   - min
+  - us
+  - s
+asset:
+  rated_power:
+    name: rated_power
+    dtype: float
+    units: kW
 reanalysis:
   WMETR_HorWdSpd:
     name: WMETR_HorWdSpd
@@ -33,10 +38,10 @@ reanalysis:
     dtype: float
     units: kg/m^3
   frequency:
-  - us
   - ns
   - D
-  - s
   - h
   - ms
   - min
+  - us
+  - s