System, method, and medium for end to end NMR based crude plant optimization

A method comprising:

• performing a nuclear magnetic resonance (NMR) scan of a sample of a hydrocarbon feedstock material to generate NMR data;
• and
• processing the NMR data, using a yield prediction model, to generate yield prediction data, the yield prediction data including a predicted yield for each of a plurality of products resulting from the processing of the hydrocarbon feedstock material by a hydrocarbon processing system;
• wherein:
• (i) the yield prediction model is a machine learning model trained using supervised learning based on training data comprising:
• NMR data for a plurality of hydrocarbon feedstock material batches;
• and
• yield data representative of, for each hydrocarbon feedstock material batch, a plurality of product yields generated by processing the hydrocarbon feedstock material batch via the hydrocarbon processing system;
• or
• (ii) the yield prediction model is a statistical model generated based on statistical correlations between:
• NMR data for a plurality of hydrocarbon feedstock material batches;
• and
• yield data representative of, for each hydrocarbon feedstock material batch, a plurality of product yields generated by processing the hydrocarbon feedstock material batch via the hydrocarbon processing system.

A method comprising:

• performing a nuclear magnetic resonance (NMR) scan of a sample of a hydrocarbon feedstock material to generate NMR data;
• and
• processing the NMR data, using a yield prediction model, to generate yield prediction data, the yield prediction data including a predicted yield for each of a plurality of products resulting from the processing of the hydrocarbon feedstock material by a hydrocarbon processing system,
• wherein the NMR data comprises a NMR fingerprint.