{"componentChunkName":"component---src-templates-research-details-js","path":"/research/molecular-composition-reconstruction-of-naphtha-fractions-through-data-driven-modeling-and-interpretable-optimization","result":{"data":{"researchItem":{"date":"09/2025","introduction":"Refineries often need molecular-level information about petroleum fractions, but directly measuring every molecular detail is expensive and time-consuming. This work reconstructs the molecular composition of naphtha from a limited set of bulk properties, such as density and boiling point information. A feedforward neural network first learns how molecular composition maps to measurable properties. The trained model is then embedded inside an inverse optimization problem, which searches for a realistic composition that would produce the observed properties. To keep the answer chemically plausible, the method uses database-derived reference compositions, carbon number patterns, compound-type distributions, and entropy-based regularization. Interpretable analysis with SHAP helps explain why selecting key bulk properties improves the reconstruction.","summary":"From a few routine fuel measurements, rebuild the molecular ingredient list of naphtha.","title":"Molecular composition reconstruction of naphtha fractions through data-driven modeling and interpretable optimization","url":"https://www.sciencedirect.com/science/article/pii/S0009250925014769"},"file":{"childImageSharp":{"fluid":{"base64":"data:image/jpeg;base64,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","aspectRatio":1.5625,"src":"/static/9f8298e51bf2647d176180a787947ce9/9842e/naphtha-composition-reconstruction.jpg","srcSet":"/static/9f8298e51bf2647d176180a787947ce9/07ab6/naphtha-composition-reconstruction.jpg 225w,\n/static/9f8298e51bf2647d176180a787947ce9/32fd5/naphtha-composition-reconstruction.jpg 450w,\n/static/9f8298e51bf2647d176180a787947ce9/9842e/naphtha-composition-reconstruction.jpg 900w,\n/static/9f8298e51bf2647d176180a787947ce9/724e2/naphtha-composition-reconstruction.jpg 1350w,\n/static/9f8298e51bf2647d176180a787947ce9/d1c6b/naphtha-composition-reconstruction.jpg 1383w","sizes":"(max-width: 900px) 100vw, 900px"}}}},"pageContext":{"slug":"molecular-composition-reconstruction-of-naphtha-fractions-through-data-driven-modeling-and-interpretable-optimization","image":"research_images/naphtha-composition-reconstruction.jpg"}},"staticQueryHashes":["362580002","426988268"]}