{"id":"W4319347401","doi":"10.1029/2022wr032653","title":"Efficient Optimization of Energy Recovery From Geothermal Reservoirs With Recurrent Neural Network Predictive Models","year":2023,"lang":"en","type":"article","venue":"Water Resources Research","topic":"Reservoir Engineering and Simulation Methods","field":"Engineering","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Office of Energy Efficiency; Office of Energy Efficiency and Renewable Energy; Energi Simulation; Geothermal Technologies Office; U.S. Department of Energy","keywords":"Geothermal gradient; Computer science; Model predictive control; Workflow; Artificial neural network; Reservoir simulation; Geothermal energy; Field (mathematics); Electricity generation; Recurrent neural network; Mathematical optimization; Data mining; Power (physics); Artificial intelligence; Engineering; Petroleum engineering; Geology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001006995,0.0001756198,0.0002441909,0.0003921431,0.0001235271,0.0000673443,0.0003231366,0.0001163742,0.00006980267],"category_scores_gemma":[0.00003456158,0.0001272367,0.0000617032,0.0008657114,0.00008722483,0.0000903145,0.0001537253,0.0003382128,0.00001130138],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007419298,"about_ca_system_score_gemma":0.00001014399,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001627917,"about_ca_topic_score_gemma":0.000008630685,"domain_scores_codex":[0.9975375,0.0003541133,0.0003108474,0.0002997273,0.0008157443,0.0006820569],"domain_scores_gemma":[0.9989007,0.0003462558,0.0000225305,0.0004188747,0.0001811441,0.0001305389],"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":[0.0002407884,0.00001377031,0.000138758,0.00004555106,0.00007258492,0.000009117824,0.00132522,0.9967844,0.0003484408,0.00001065327,0.0003696203,0.0006410253],"study_design_scores_gemma":[0.0003959432,0.0001437809,0.0002884024,0.00009214961,0.000006796624,6.652523e-7,0.0001019416,0.9962034,0.001245382,0.0002233067,0.001149541,0.0001487524],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8597028,0.000262063,0.1379323,0.00004698346,0.0001675288,0.0001705799,0.00003190037,0.0003349739,0.001350916],"genre_scores_gemma":[0.9941226,0.0001185476,0.004740451,0.000002913508,0.0002559326,0.00007238452,0.0001362538,0.00008779498,0.0004631598],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1344198,"threshold_uncertainty_score":0.518856,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0388982769591298,"score_gpt":0.2813159203590017,"score_spread":0.2424176433998719,"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."}}