{"id":"W4281763532","doi":"10.1016/j.ijheatmasstransfer.2022.123112","title":"Deep reinforcement learning for heat exchanger shape optimization","year":2022,"lang":"en","type":"article","venue":"International Journal of Heat and Mass Transfer","topic":"Topology Optimization in Engineering","field":"Engineering","cited_by":66,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Solver; Shape optimization; Pressure drop; Heat transfer; Computational fluid dynamics; Reinforcement learning; Degrees of freedom (physics and chemistry); Parametric statistics; Artificial neural network; Topology optimization; Curse of dimensionality; Artificial intelligence; Mathematical optimization; Finite element method; Mechanics; Mathematics; Physics; Thermodynamics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008956314,0.0009053577,0.001519384,0.0004580345,0.000334741,0.0007072446,0.001308153,0.001887312,0.003471781],"category_scores_gemma":[0.002644703,0.0006100598,0.0005649549,0.0004466644,0.001003746,0.0008947968,0.001173463,0.001834589,0.0004837289],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001120855,"about_ca_system_score_gemma":0.001025569,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00824788,"about_ca_topic_score_gemma":0.007646594,"domain_scores_codex":[0.9997564,0.00008576498,0.0000102382,0.00005415044,0.00004978151,0.0000436287],"domain_scores_gemma":[0.9986682,0.0008767492,0.00009665373,0.00007999698,0.000195495,0.00008293806],"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":[0.00004666689,0.00004180473,0.0003367971,0.00002808533,0.00001888216,0.00002005503,0.00001285016,0.969758,0.0005373071,0.003107571,0.001035264,0.02505673],"study_design_scores_gemma":[0.000002585743,0.0000053938,0.00001415117,0.000001373601,8.920236e-7,0.000001097123,7.653917e-7,0.9989926,0.0000463375,0.0008939782,0.00004006087,7.211018e-7],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06169834,0.001065642,0.9316031,0.0006672365,0.0001497374,0.0000470682,0.0001457194,0.0009336361,0.003689612],"genre_scores_gemma":[0.9158659,0.0002819147,0.07675452,0.0003651217,0.0001022742,0.0001322822,0.0002674637,0.0001409357,0.006089562],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00824788,"threshold_uncertainty_score":0.01639974,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008812605859114631,"score_gpt":0.2222847997075518,"score_spread":0.2134721938484372,"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."}}