Mismatch Component Name in COSMO

[swingg278]

1. The example does not work OTB, in the complist we have "full name" while in the "table_params" it is abbreviation.
2. Code suggested is not very robust, please find below an updated version to ease step up.
3. Cosmosac does not compile on Mac Book (Apple Silicon) - need to adapt Catch
   Find and replace:
   #define CATCH_TRAP() __asm__("int $3\n" : : ) to #define CATCH_TRAP() __builtin_trap()

import matplotlib.pyplot as plt
import cCOSMO
import os
from cosmopharm import SLE, COSMOSAC
from cosmopharm.utils import create_components, read_params

# Define all paths in one place
params_file = "data/sle/table_params.xlsx"
comp_list_path = "../../COSMOSAC/profiles/UD/complist.txt"
sigma3_path = "../../COSMOSAC/profiles/UD/sigma3/"

# Define components with mapping between abbreviated names (params) and full names (COSMO)
component_mapping = {
    'SIM': 'SIMVASTATINE',           # Mapping from params file to COSMO database
    'PLGA50': 'LACTIC_ACID',     # Example of mapping
}

# Abbreviated names used in the parameter file
params_names = ['SIM', 'PLGA50']

# Corresponding full names from the COSMO database
cosmo_names = [component_mapping[name] for name in params_names]

print(f"📋 Component mapping:")
for short, full in zip(params_names, cosmo_names):
    print(f"   {short} → {full}")

def verify_paths_exist(paths_dict):
    """Test function to verify that all required paths exist."""
    missing_paths = []
    for path_name, path in paths_dict.items():
        if not os.path.exists(path):
            missing_paths.append(f"{path_name}: {path}")
    
    if missing_paths:
        print("⚠️  Error: The following paths do not exist:")
        for missing_path in missing_paths:
            print(f"   - {missing_path}")
        return False
    else:
        print("✅ All required paths exist.")
        return True

# Verify paths before proceeding
print("Checking file paths...")
paths_to_check = {
    'params_file': params_file,
    'comp_list_path': comp_list_path,
    'sigma3_path': sigma3_path
}

if not verify_paths_exist(paths_to_check):
    print("Program stopped due to missing paths.")
    exit(1)

# Load parameters and create components
print("📋 Loading parameters...")
try:
    parameters = read_params(params_file)
    print(f"✅ Parameters successfully loaded from {params_file}")
    
    # Debug: show available components in parameters
    if hasattr(parameters, 'columns'):
        print("📊 Columns available in parameters:")
        for col in parameters.columns:
            print(f"   - {col}")
    
    if hasattr(parameters, 'index'):
        print("🧪 Components available in parameters:")
        for idx, comp in enumerate(parameters.index[:10]):  # Show the first 10
            print(f"   {idx+1}. {comp}")
        if len(parameters.index) > 10:
            print(f"   ... and {len(parameters.index) - 10} more")
    
    print(f"\n🔍 Attempting to create components: {params_names}")
    mixture = create_components(params_names, parameters)
    print("✅ Components successfully created")
    
except ValueError as e:
    print(f"❌ Error while creating components: {e}")
    print("\n💡 Suggestions:")
    print("1. Make sure the component names exactly match those in the parameter file")
    print("2. Check for case sensitivity (uppercase/lowercase)")
    print("3. Use component names that exist in the database")
    print("\nStopping the program.")
    exit(1)
except Exception as e:
    print(f"❌ Unexpected error while loading parameters: {e}")
    print("Stopping the program.")
    exit(1)

# Initialize COSMO-SAC model - replace paths with your local paths to COSMO profiles
db = cCOSMO.DelawareProfileDatabase(
    comp_list_path,
    sigma3_path)

# Debug: show a few available identifiers for diagnostics
print("🔍 Checking available identifiers in the database...")

# Try to normalize and add each component individually
for name in cosmo_names:
    try:
        print(f"Attempting to normalize identifier: '{name}'")
        iden = db.normalize_identifier(name)
        print(f"✅ Normalized identifier for '{name}': '{iden}'")
        db.add_profile(iden)
        print(f"✅ Profile added for '{name}'")
    except ValueError as e:
        print(f"❌ Error for '{name}': {e}")
        print(f"💡 Suggestion: Check that '{name}' exists in {comp_list_path}")
        # Show the first available identifiers for reference
        print("📋 Here are some available identifiers in the database:")
        with open(comp_list_path, 'r') as f:
            lines = f.readlines()[1:11]  # Skip header, show first 10
            for line in lines:
                if line.strip():
                    parts = line.strip().split()
                    if len(parts) >= 4:
                        print(f"   ID: {parts[0]}, Name: {parts[3]}")
        print("Stopping the program due to missing identifiers.")
        exit(1)

COSMO = cCOSMO.COSMO3(cosmo_names, db)

# Setup the COSMO-SAC model with components
actmodel = COSMOSAC(COSMO, mixture=mixture)

# Calculate solubility (SLE)
sle = SLE(actmodel=actmodel)
solubility = sle.solubility(mix='real')

# Output the solubility
print(solubility[['T', 'w', 'x']].to_string(index=False))

# Plot results
plt.plot(*solubility[['w','T']].values.T,'.-', label='Solubility (w)')

# Settings
plt.xlim(0,1)
plt.ylim(300,500)
# Adding title and labels
plt.title('Solubility vs. Temperature')
plt.ylabel("T / K")
xlabel = {'w':'Weight', 'x':'Mole'}
plt.xlabel(f"Weight fraction {mixture[0].name}")
plt.legend()
# Save the figure to a PNG or PDF file
plt.savefig('solubility_plot.png')  # Saves the plot as a PNG file
# plt.savefig('solubility_plot.pdf')  # Saves the plot as a PDF file
plt.show()