{"id":"W7124142497","doi":"10.1109/codit66093.2025.11321361","title":"Machine Learning and Derivative-Free Optimization for PID Tuning: Case Study of Improved Black Liquor Concentration Control","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Control Systems Optimization","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canetique (Canada); University of Waterloo; Natural Resources Canada","funders":"","keywords":"PID controller; Black liquor; Boosting (machine learning); Bayesian optimization; Paper machine; Kraft paper; Kraft process; Black box","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":[0.001842993,0.0005783309,0.0009849534,0.000457084,0.00047837,0.0008369743,0.0006749877,0.001240899,0.0008897214],"category_scores_gemma":[0.002430157,0.0002666744,0.0004958542,0.000448152,0.0006727631,0.0004623371,0.0005788995,0.0009410572,0.0001208984],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007866836,"about_ca_system_score_gemma":0.0006310729,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009774521,"about_ca_topic_score_gemma":0.00770745,"domain_scores_codex":[0.9994316,0.0002267166,0.00003005947,0.00008071819,0.0001518877,0.00007906326],"domain_scores_gemma":[0.9987225,0.0008836248,0.0001014533,0.00008484635,0.0001660612,0.00004145829],"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":[0.0001531759,0.0001130868,0.001327094,0.0001377878,0.00002808035,0.0003185759,0.0001064743,0.9659533,0.004061224,0.002522076,0.0003098112,0.02496933],"study_design_scores_gemma":[0.0000155402,0.0001021314,0.0004998426,0.000005094817,0.000007508619,0.00002077343,0.00001522801,0.9950177,0.003398445,0.0004508984,0.0004597432,0.000007000365],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.712258,0.001442244,0.273729,0.0006848822,0.00004324836,0.0001253694,0.0001063746,0.0008085723,0.01080219],"genre_scores_gemma":[0.982128,0.0001035738,0.01659491,0.00001808511,0.000006200772,0.00002209694,0.00001691912,0.00001522659,0.001094948],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009774521,"threshold_uncertainty_score":0.01943523,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007381979492292272,"score_gpt":0.2454626712530163,"score_spread":0.2380806917607241,"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."}}