{"id":"W4224209662","doi":"10.2118/209245-ms","title":"Towards a Machine Learning Based Dynamic Surrogate Modeling and Optimization of Steam Injection Policy in SAGD","year":2022,"lang":"en","type":"article","venue":"SPE Western Regional Meeting","topic":"Reservoir Engineering and Simulation Methods","field":"Engineering","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Robustness (evolution); Randomness; Computer science; Surrogate model; Time horizon; Mathematical optimization; Reinforcement learning; Engineering; Machine learning; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005487648,0.0001177053,0.0001692418,0.0003930017,0.00007500263,0.00001584279,0.00007309847,0.00003895147,0.000008045276],"category_scores_gemma":[0.00007589578,0.0001399642,0.00003597833,0.0003800127,0.00001060505,0.00008675599,0.00004845066,0.000269845,2.202993e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001658422,"about_ca_system_score_gemma":0.00003049339,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003752111,"about_ca_topic_score_gemma":0.00004132178,"domain_scores_codex":[0.9990779,0.0001136911,0.0002781641,0.0001493895,0.0002174687,0.0001634398],"domain_scores_gemma":[0.9997245,0.00007371307,0.00005157255,0.0000868409,0.00002518041,0.00003818457],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001974027,0.000007948295,0.006126122,0.00008825547,0.000007360997,0.000001725647,0.0002400345,0.9920641,0.000408879,0.00001198527,2.916076e-7,0.001023576],"study_design_scores_gemma":[0.0005040007,0.00004011373,0.0005447472,0.00006898462,0.000004332581,0.000009175822,0.0001331407,0.998427,0.00004781634,0.00005156385,0.00004890141,0.0001202087],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6934741,0.0003587417,0.3057003,0.00008752901,0.00006264602,0.00007996685,0.000003221403,0.0001108131,0.0001227438],"genre_scores_gemma":[0.9790648,0.00006599835,0.02071073,0.0000125677,0.00003031207,0.00001379947,0.00003216073,0.00003724856,0.00003234166],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2855908,"threshold_uncertainty_score":0.5707573,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01899077654364116,"score_gpt":0.2701831385919762,"score_spread":0.2511923620483351,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}