Upload alloy_data_preprocessing.py

alloy_data_preprocessing.py

@@ -0,0 +1,177 @@
+import pymatgen as mg
+import pandas as pd
+import numpy as np
+from pymatgen.core.structure import Composition
+
+
+def calculate_density(comp):
+    """Calculates densisty based on Rule of Mixtures (ROM)."""
+
+    # comp = Composition(formula)
+
+    weights = [comp.get_atomic_fraction(e) for e in comp.elements]
+    vols = np.array([e.molar_volume for e in comp.elements])
+    atomic_masses = np.array([e.atomic_mass for e in comp.elements])
+
+    val = np.sum(weights * atomic_masses) / np.sum(weights * vols)
+
+    return round(val, 1)
+
+
+def calculate_young_modulus(comp):
+    """Calculates Young Modulus based on Rule of Mixtures (ROM)."""
+
+    # comp = Composition(formula)
+
+    weights = np.array([comp.get_atomic_fraction(e) for e in comp.elements])
+    vols = np.array([e.molar_volume for e in comp.elements])
+    ym_vals = []
+    for e in comp.elements:
+        if str(e) == "C":  # use diamond form for carbon
+            ym_vals.append(1050)
+        elif str(e) == "B":  # use minimum value for Boron Carbide
+            ym_vals.append(362)
+        elif str(e) == "Mo":
+            ym_vals.append(329)
+        elif str(e) == "Co":
+            ym_vals.append(209)
+        else:
+            ym_vals.append(e.youngs_modulus)
+
+    # ym_vals = np.array([e.youngs_modulus for e in comp.elements])
+    ym_vals = np.array(ym_vals)
+
+    if None in ym_vals:
+        print(comp, ym_vals)
+        return ""
+
+    val = np.sum(weights * vols * ym_vals) / np.sum(weights * vols)
+    if val is np.nan:
+        val = 0
+
+    return int(round(val, 0))
+
+
+def calculate_electronegativity(comp):
+    return comp.average_electroneg
+
+
+def create_composition(comp_df):
+    ls_comp = comp_df.to_dict("records")
+    res = []
+    for comp_dict in ls_comp:
+        elem_fill = np.sum([comp_dict[e] for e in comp_dict])
+        comp_dict["Fe"] = 100 - elem_fill
+
+        # print(comp_dict)
+        compo = Composition.from_weight_dict(comp_dict)
+        res.append(compo)
+
+    comp_df["composition"] = res
+    return comp_df
+
+
+def calculate_electronegativity(comp):
+    return comp.average_electroneg
+
+
+def calculate_valence_electron_concentration(comp):
+    """
+    Using the formuma from https://www.sciencedirect.com/science/article/pii/S0927025622000015#s0100
+    VEC = Sum(j=1 to N)C(j)VEC(j)
+    where N is the number of alloying elements, C(j) and VEC(j) are the atomic percentage and the valence electron number of element j
+    """
+
+    weights = np.array([comp.get_atomic_fraction(e) for e in comp.elements])
+
+    val_ls = []
+    for e in comp.elements:
+        if str(e) == "Cr":
+            val_ls.append(6)
+        elif str(e) == "Mo":
+            # For Mo valence electron can vary from 2 to 6
+            val_ls.append(4)
+        else:
+            val_ls.append(e.valence[1])
+    val_ls = np.array(val_ls)
+
+    vec = np.sum(weights * val_ls)
+
+    return vec
+
+
+def calculate_configuration_entropy(comp):
+    """
+    Using the formuma from https://www.sciencedirect.com/science/article/pii/S0927025622000015#s0100
+    VEC = -R*Sum(j=1 to N)C(j)ln(C(j))
+    where N is the number of alloying elements, C(j) is the atomic percentage element j and R is the gas constant
+    The gas constant is omitted for now
+    """
+
+    weights = np.array([comp.get_atomic_fraction(e) for e in comp.elements])
+
+    ent = np.sum(weights * np.log(weights))
+
+    return ent
+
+
+def add_physics_features(df):
+    """
+    Adds the density and young modulus as additional columns
+    elem_df: pd.DataFrame containing the proportion of each elements
+
+    """
+    mapping = {"%C": "C", "%Co": "Co", "%Cr": "Cr", "%V": "V", "%Mo": "Mo", "%W": "W"}
+    if type(df) != pd.DataFrame:
+        # Fix for the case where the input df is not a dataframe but an array
+        print(df.shape)
+        if df.shape[1] < 10:
+            cols = ["%C", "%Co", "%Cr", "%V", "%Mo", "%W", "Temperature_C"]
+        else:
+            cols = [
+                "%C",
+                "%Co",
+                "%Cr",
+                "%V",
+                "%Mo",
+                "%W",
+                "M6C",
+                "M23C6",
+                "FCCA1#2",
+                "M2C",
+                "MC - SHP",
+                "MC ETA",
+                "%C matrice",
+                "%Co matrice",
+                "%Cr matrice",
+                "%V matrice",
+                "%Mo matrice",
+                "%W matrice",
+                "Temperature_C",
+            ]
+        df = pd.DataFrame(df, columns=cols)
+        print(df.shape)
+    elem_df = df[mapping.keys()]
+    elem_df.rename(columns=mapping, inplace=True)
+    elem_df["Fe"] = 100 - elem_df.sum(axis=1)
+    df_w_compo = create_composition(elem_df)
+
+    df["density"] = np.vectorize(calculate_density)(df_w_compo["composition"])
+    df["young_modulus"] = np.vectorize(calculate_young_modulus)(df_w_compo["composition"])
+    df["electronegativity"] = np.vectorize(calculate_electronegativity)(df_w_compo["composition"])
+    df["valence_electron_concentration"] = np.vectorize(calculate_valence_electron_concentration)(
+        df_w_compo["composition"]
+    )
+    df["configuration_entropy"] = np.vectorize(calculate_configuration_entropy)(df_w_compo["composition"])
+    return df
+
+
+if __name__ == "__main__":
+    df = pd.DataFrame([[0.3, 5, 3.9, 2.1, 5, 1.2]], columns=["%C", "%Co", "%Cr", "%V", "%Mo", "%W"])
+    df = pd.DataFrame([[0.3, 5, 3.9, 2.1, 5, 1.2]], columns=["C", "Co", "Cr", "V", "Mo", "W"])
+    df = pd.DataFrame([[7, 38]], columns=["Al", "Ni"])  # Debug density issue on gradio demo
+
+    # add_physics_features(df)
+    df = create_composition(df)
+    val = calculate_density(df["composition"].iloc[0])
+    print(val)