{"id":456,"date":"2025-11-21T16:51:48","date_gmt":"2025-11-21T16:51:48","guid":{"rendered":"https:\/\/sites.wp.odu.edu\/smartlab\/?page_id=456"},"modified":"2025-11-21T17:27:11","modified_gmt":"2025-11-21T17:27:11","slug":"gmm-lstm","status":"publish","type":"page","link":"https:\/\/sites.wp.odu.edu\/smartlab\/gmm-lstm\/","title":{"rendered":"GMM-LSTM"},"content":{"rendered":"\n<p style=\"font-size:15px\"><strong>A New Predictive Maintenance Approach: Novel Integration of GMM-LSTM for Prediction of Latent State and Failure Location of Rotating Machinery.<\/strong><\/p>\n\n\n\n<p style=\"font-size:15px\"><em><strong>Summary<\/strong><\/em>:<br>We propose a hybrid predictive-maintenance framework that couples Gaussian Mixture Models (for latent bearing-state classification) with LSTMs (for temporal prediction and failure-location inference). Using FFT and WPD features, the model achieves robust early-fault detection and precise localization on multiple bearing datasets.<\/p>\n\n\n\n<p style=\"font-size:15px\"><strong><em>Highlights:<\/em><\/strong><\/p>\n\n\n\n<ul>\n<li style=\"font-size:15px\">GMM \u2192 latent state classification; LSTM \u2192 time-aware failure-location prediction.<\/li>\n<\/ul>\n\n\n\n<ul>\n<li style=\"font-size:15px\">FFT + WPD feature stack for vibration-signal representation.<\/li>\n<\/ul>\n\n\n\n<ul>\n<li style=\"font-size:15px\">Demonstrated generalization with strong precision\/recall while mitigating overfitting.<\/li>\n<\/ul>\n\n\n\n<ul>\n<li style=\"font-size:15px\">Real-time-oriented pipeline for rotating-machinery health monitoring.<\/li>\n<\/ul>\n\n\n\n<p style=\"font-size:15px\"><strong><em>Duration<\/em><\/strong>: January 2023 \u2013 June 2024.<\/p>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\" \/>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p style=\"font-size:15px\"><strong>Associated Members:<\/strong><\/p>\n\n\n\n<div class=\"wp-block-columns is-layout-flex wp-container-core-columns-layout-1 wp-block-columns-is-layout-flex\">\n<div class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\"><div class=\"wp-block-image is-style-rounded\">\n<figure class=\"alignleft size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"793\" height=\"440\" src=\"http:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/john-park.jpg\" alt=\"\" class=\"wp-image-324\" style=\"object-fit:cover;width:250px;height:250px\" srcset=\"https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/john-park.jpg 793w, https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/john-park-300x166.jpg 300w, https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/john-park-768x426.jpg 768w\" sizes=\"(max-width: 793px) 100vw, 793px\" \/><figcaption class=\"wp-element-caption\"><a href=\"https:\/\/sites.wp.odu.edu\/smartlab\/faculty-student-body\/\">Dr. H. John Park<\/a><\/figcaption><\/figure><\/div><\/div>\n\n\n\n<div class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\"><div class=\"wp-block-image is-style-rounded\">\n<figure class=\"alignleft size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"480\" height=\"582\" src=\"http:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2023\/02\/504687aa-1.jpg\" alt=\"\" class=\"wp-image-161\" style=\"object-fit:cover;width:250px;height:250px\" \/><figcaption class=\"wp-element-caption\"><a href=\"https:\/\/sites.wp.odu.edu\/smartlab\/faculty-student-body\/\">Dr. Samuel F. Kovacic<\/a><\/figcaption><\/figure><\/div><\/div>\n\n\n\n<div class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\"><div class=\"wp-block-image is-style-rounded\">\n<figure class=\"alignleft size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"480\" height=\"582\" src=\"http:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/andres.jpg\" alt=\"\" class=\"wp-image-336\" style=\"object-fit:cover;width:250px;height:250px\" srcset=\"https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/andres.jpg 480w, https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/andres-247x300.jpg 247w\" sizes=\"(max-width: 480px) 100vw, 480px\" \/><figcaption class=\"wp-element-caption\"><a href=\"https:\/\/sites.wp.odu.edu\/smartlab\/faculty-student-body\/\">Dr. Andres Sousa-Poza<\/a><\/figcaption><\/figure><\/div><\/div>\n<\/div>\n\n\n\n<div class=\"wp-block-columns is-layout-flex wp-container-core-columns-layout-2 wp-block-columns-is-layout-flex\">\n<div class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\"><div class=\"wp-block-image is-style-rounded\">\n<figure class=\"alignleft size-medium is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"300\" height=\"300\" src=\"http:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/Arman-300x300.jpeg\" alt=\"\" class=\"wp-image-280\" style=\"object-fit:cover;width:250px;height:250px\" srcset=\"https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/Arman-300x300.jpeg 300w, https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/Arman-150x150.jpeg 150w, https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/Arman-768x768.jpeg 768w, https:\/\/sites.wp.odu.edu\/smartlab\/wp-content\/uploads\/sites\/32790\/2025\/11\/Arman.jpeg 800w\" sizes=\"(max-width: 300px) 100vw, 300px\" \/><figcaption class=\"wp-element-caption\">Arman Ghavidel, Ph.D. Candidate<\/figcaption><\/figure><\/div><\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>A New Predictive Maintenance Approach: Novel Integration of GMM-LSTM for Prediction of Latent State and Failure Location of Rotating Machinery. Summary:We propose a hybrid predictive-maintenance framework that couples Gaussian Mixture Models (for latent bearing-state classification) with LSTMs (for temporal prediction and failure-location inference). Using FFT and WPD features, the model achieves robust early-fault detection and [&hellip;]<\/p>\n","protected":false},"author":26318,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"_links":{"self":[{"href":"https:\/\/sites.wp.odu.edu\/smartlab\/wp-json\/wp\/v2\/pages\/456"}],"collection":[{"href":"https:\/\/sites.wp.odu.edu\/smartlab\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/sites.wp.odu.edu\/smartlab\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/sites.wp.odu.edu\/smartlab\/wp-json\/wp\/v2\/users\/26318"}],"replies":[{"embeddable":true,"href":"https:\/\/sites.wp.odu.edu\/smartlab\/wp-json\/wp\/v2\/comments?post=456"}],"version-history":[{"count":5,"href":"https:\/\/sites.wp.odu.edu\/smartlab\/wp-json\/wp\/v2\/pages\/456\/revisions"}],"predecessor-version":[{"id":487,"href":"https:\/\/sites.wp.odu.edu\/smartlab\/wp-json\/wp\/v2\/pages\/456\/revisions\/487"}],"wp:attachment":[{"href":"https:\/\/sites.wp.odu.edu\/smartlab\/wp-json\/wp\/v2\/media?parent=456"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}