LLM vs EOS Thermodynamics

🧪🧠📊🌡️This project compares classical thermodynamic models with AI-driven predictions.

🧪 – Thermodynamic systems and EOS modeling 🧠 – AI/LLM intelligence 📊 – Data analysis, predictions, visualization 🌡️ – Key thermodynamic variables (T, P, etc.)

📌 Features

• Vapor–liquid equilibrium (VLE) calculations
• Cubic Equations of State (EOS): Peng–Robinson (PR), Soave–Redlich–Kwong (SRK)
• Activity coefficient models: NRTL, UNIQUAC, Wilson, UNIFAC
• Antoine equation Psat tools
• AI/ML predictions via NeuraGPT-mini (Mistral-7B with LoRA adapter)
• Comparisons across:
  • K-values and K-value ranges
  • Vapor fraction (ϕV)
  • Bubble/dew temperatures and pressures
  • T-x-y / P-x-y diagrams
  • Psat(T) vs experimental

🚀 Run

python -m venv .venv && . .venv/Scripts/activate
pip install -r requirements.txt
streamlit run app/streamlit_app.py