{"id":"W4388691832","doi":"10.1109/tase.2023.3331347","title":"Unsupervised Fault Detection for Building Air Handling Unit Systems Using Deep Variational Mixture of Principal Component Analyzers","year":2023,"lang":"en","type":"article","venue":"IEEE Transactions on Automation Science and Engineering","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Autoencoder; Principal component analysis; Component (thermodynamics); Fault detection and isolation; Missing data; Computer science; Robust principal component analysis; Artificial intelligence; Deep learning; Artificial neural network; Function (biology); Data mining; Pattern recognition (psychology); Machine learning","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000623757,0.0001146729,0.0001320205,0.0007773323,0.0005118907,0.0001151465,0.0002473329,0.00006195065,0.000001045185],"category_scores_gemma":[0.000018342,0.0001198032,0.00005220434,0.002154438,0.00004892739,0.0006206603,0.000005531102,0.00009340258,0.000001390332],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001083794,"about_ca_system_score_gemma":0.00006884395,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002559781,"about_ca_topic_score_gemma":0.00000207331,"domain_scores_codex":[0.9988223,0.00001260429,0.0002877254,0.0002985115,0.0003688872,0.0002099791],"domain_scores_gemma":[0.9992479,0.0001066913,0.00009646907,0.0002026831,0.0002676363,0.00007861861],"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":[0.000001801705,0.00001067644,0.000002352419,0.00003227174,0.000007593403,1.269601e-7,0.0001461499,0.7543249,0.2385853,0.00176073,4.464881e-7,0.005127635],"study_design_scores_gemma":[0.0001401198,0.00003300919,0.0005678642,0.00003675538,0.00001041719,0.000009557947,0.00005108595,0.8812261,0.1176862,0.00004549146,0.00008509042,0.0001083844],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2141406,0.00000627292,0.7848225,0.00005110887,0.0002692807,0.0002601998,0.000006257875,0.000439187,0.000004523931],"genre_scores_gemma":[0.9455542,0.000007060735,0.05429441,0.00001131916,0.00002131233,0.00009520288,0.000001144604,0.000008672664,0.000006692799],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7314135,"threshold_uncertainty_score":0.4885434,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02259878845908984,"score_gpt":0.2654866127062673,"score_spread":0.2428878242471775,"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."}}