Update sdm getting started to use IVPerformanceMatrix

README.md

@@ -19,10 +19,11 @@ finalized on the `master` branch. Likewise, the documentation badly needs updati
 
 PVfit is currently focused on direct-current (DC) PV module performance measurement and
 modeling. Following the standardized technical approach of most accredited PV
-calibration laboratories for measuring I-V curves using PV reference devices, PVfit
-makes considerable use of the effective irradiance ratio
+calibration laboratories for measuring current-voltage (I-V) curves using PV reference
+devices, PVfit makes considerable use of the effective irradiance ratio
 (F = Isc / Isc0 = M * Isc,ref / Isc0,ref) to quantify the *effective* irradiance on a PV
-device, in contrast to the common use of MET-station data
+device, in contrast to the common use of MET-station data. PVfit also supports inference
+of effective-irradiance ratio and cell temperature directly from I-V data, see
 ([poster](https://pvpmc.sandia.gov/download/7302/)). See
 [this paper](https://doi.org/10.1002/ese3.190) for a more detailed introduction. Email
 [Mark Campanelli](mailto:[email protected]) to be added to the

pvfit/modeling/dc/single_diode/model/demos/data.py

@@ -8,19 +8,17 @@
 
 import pandas
 
+from pvfit.measurement.iv.types import IVPerformanceMatrix
 from pvfit.modeling.dc.common import Material
 
 
 # Performance-matrix data for a 72-cell HIT module measured over variable
 # irradiance and temperature. See https://www.nrel.gov/docs/fy14osti/61610.pdf and
 # [email protected] for the complete data set.
 
-HIT_MODULE = {
-    "N_s": 72,
-    "material": Material.xSi,
-    "matrix": pandas.read_csv(
-        StringIO(
-            """T_degC,G_W_per_m2,I_sc_A,V_oc_V,I_mp_A,V_mp_V,P_mp_W
+matrix_data = pandas.read_csv(
+    StringIO(
+        """T_degC,G_W_per_m2,I_sc_A,V_oc_V,I_mp_A,V_mp_V,P_mp_W
 15,100,0.558,46.74,0.51,40.1,20.47
 25,100,0.57,45.37,0.526,38.39,20.19
 15,200,1.111,48.23,1.013,41.7,42.25
@@ -39,6 +37,20 @@
 25,1100,6.079,50.39,5.632,41.52,233.84
 50,1100,6.101,47.11,5.639,37.87,213.58
 65,1100,6.129,45.16,5.604,35.99,201.73"""
-        )
-    ),
+    )
+)
+
+iv_performance_matrix = IVPerformanceMatrix(
+    I_sc_A=matrix_data["I_sc_A"].to_numpy(),
+    I_mp_A=matrix_data["I_mp_A"].to_numpy(),
+    V_mp_V=matrix_data["V_mp_V"].to_numpy(),
+    V_oc_V=matrix_data["V_oc_V"].to_numpy(),
+    G_W_per_m2=matrix_data["G_W_per_m2"].to_numpy(),
+    T_degC=matrix_data["T_degC"].to_numpy(),
+)
+
+HIT_MODULE = {
+    "N_s": 72,
+    "material": Material.xSi,
+    "matrix": iv_performance_matrix,
 }

pvfit/modeling/dc/single_diode/model/demos/getting_started.py

@@ -10,7 +10,6 @@
 import numpy
 
 from pvfit.measurement.iv.types import FTData
-from pvfit.modeling.dc.common import G_hemi_W_per_m2_stc, T_degC_stc
 import pvfit.modeling.dc.single_diode.equation.simulation as sde_sim
 import pvfit.modeling.dc.single_diode.model.demos.data as data
 import pvfit.modeling.dc.single_diode.model.simple.auxiliary_equations as ae
@@ -20,60 +19,23 @@
 # By convention, variable names for numeric values include the units.
 
 N_s = data.HIT_MODULE["N_s"]
-matrix = data.HIT_MODULE["matrix"]
+material = data.HIT_MODULE["material"]
+iv_performance_matrix = data.HIT_MODULE["matrix"]
+ivft_data = iv_performance_matrix.ivft_data
 
 print(f"Performance matrix for HIT module with {N_s} cells in series:")
-print(matrix)
-
-# We will flatten all channels from data table into vectors for model fitting.
-V_V_data = numpy.full(0, numpy.nan)
-I_A_data = numpy.copy(V_V_data)
-F_data = numpy.copy(V_V_data)
-T_degC_data = numpy.copy(V_V_data)
-G_data_minimal = numpy.copy(V_V_data)
-F_data_minimal = numpy.copy(V_V_data)
-T_degC_data_minimal = numpy.copy(V_V_data)
-
-# Pick out Isc at STC, which is needed to determine F for each I-V curve.
-I_sc_A_0_meas_row_idx = numpy.logical_and(
-    matrix.loc[:, "T_degC"] == T_degC_stc,
-    matrix.loc[:, "G_W_per_m2"] == G_hemi_W_per_m2_stc,
-)
-I_sc_A_0_meas = matrix.loc[I_sc_A_0_meas_row_idx, "I_sc_A"].iloc[0]
-
-print("I_sc_A_0_meas", I_sc_A_0_meas)
-
-# Ordering of the I-V-F-T points does not matter, as long as it's consistent
-# between vectors.
-for index, row in matrix.iterrows():
-    V_V_data = numpy.concatenate(
-        (V_V_data, numpy.array([0.0, row["V_mp_V"], row["V_oc_V"]]))
-    )
-    I_A_data = numpy.concatenate(
-        (I_A_data, numpy.array([row["I_sc_A"], row["I_mp_A"], 0.0]))
-    )
-    # Use effective irradiance ratio F, not the ratio of irradiances.
-    F = row["I_sc_A"] / I_sc_A_0_meas
-    F_data = numpy.concatenate((F_data, numpy.full(3, F)))
-    T_degC_data = numpy.concatenate((T_degC_data, numpy.full(3, row["T_degC"])))
-    F_data_minimal = numpy.concatenate((F_data_minimal, numpy.array([F])))
-    T_degC_data_minimal = numpy.concatenate(
-        (T_degC_data_minimal, numpy.array([row["T_degC"]]))
-    )
-    G_data_minimal = numpy.concatenate(
-        (G_data_minimal, numpy.array([row["G_W_per_m2"]]))
-    )
+print(iv_performance_matrix)
 
-print(f"I_sc_A_0_meas = {I_sc_A_0_meas}")
-print(f"I_A_data =\n{I_A_data}")
-print(f"V_V_data =\n{V_V_data}")
-print(f"F_data =\n{F_data}")
-print(f"T_degC_data =\n{T_degC_data}")
+print(f"I_sc_A_0_data = {iv_performance_matrix.I_sc_A_0}")
+print(f"I_A_data =\n{ivft_data.I_A}")
+print(f"V_V_data =\n{ivft_data.V_V}")
+print(f"F_data =\n{ivft_data.F}")
+print(f"T_degC_data =\n{ivft_data.T_degC}")
 
 # TODO Demonstrate how to fit these parameters to data.
 model_parameters_fit = ModelParameters(
     N_s=N_s,
-    T_degC_0=T_degC_stc,
+    T_degC_0=iv_performance_matrix.T_degC_0,
     I_sc_A_0=5.531975277850156,
     I_rs_A_0=7.739216483430394e-11,
     n_0=1.086322575846391,
@@ -85,7 +47,8 @@
 # Compute parameters for I-V curve at STC.
 iv_curve_parameters_0 = sde_sim.iv_curve_parameters(
     model_parameters=ae.compute_sde_model_parameters(
-        ft_data=FTData(F=1, T_degC=T_degC_stc), model_parameters=model_parameters_fit
+        ft_data=FTData(F=1, T_degC=iv_performance_matrix.T_degC_0),
+        model_parameters=model_parameters_fit,
     )
 )
 
@@ -127,16 +90,16 @@
 # Now make a nice plot.
 fig, ax = pyplot.subplots(figsize=(8, 6))
 # Plot the data fits.
-for idx, (F, T_degC) in enumerate(zip(F_data_minimal, T_degC_data_minimal)):
+for idx, (F, T_degC) in enumerate(zip(iv_performance_matrix.F, iv_performance_matrix.T_degC)):
     # Plot Isc, Pmp, and Voc with same colors as fit lines.
     color = next(ax._get_lines.prop_cycler)["color"]
     ax.plot(
-        V_V_data[3 * idx : 3 * idx + 3],
-        I_A_data[3 * idx : 3 * idx + 3],
+        ivft_data.V_V[3 * idx : 3 * idx + 3],
+        ivft_data.I_A[3 * idx : 3 * idx + 3],
         "o",
         color=color,
     )
-    V_V = numpy.linspace(0, V_V_data[3 * idx + 2], 101)
+    V_V = numpy.linspace(0, ivft_data.V_V[3 * idx + 2], 101)
     ax.plot(
         V_V,
         sde_sim.I_at_V(