{"id":12548,"date":"2024-11-08T12:56:40","date_gmt":"2024-11-08T03:56:40","guid":{"rendered":"https:\/\/blogs.mathworks.com\/japan-community\/?p=12548"},"modified":"2024-11-08T12:57:10","modified_gmt":"2024-11-08T03:57:10","slug":"reduced-order-modeler-app-how-to-interactively-create-ai-based-reduced-order-models-roms-jp","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/japan-community\/2024\/11\/08\/reduced-order-modeler-app-how-to-interactively-create-ai-based-reduced-order-models-roms-jp\/","title":{"rendered":"Reduced Order Modeler App: \u4f4e\u6b21\u5143\u5316\u30e2\u30c7\u30eb\uff08ROM\uff09\u3092\u30a4\u30f3\u30bf\u30e9\u30af\u30c6\u30a3\u30d6\u306b\u69cb\u7bc9"},"content":{"rendered":"<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u203b\u3053\u306e\u6295\u7a3f\u306f 2024 \u5e74 10 \u6708 21 \u65e5\u306b<a href=\"https:\/\/blogs.mathworks.com\/deep-learning\/2024\/10\/21\/introducing-the-reduced-order-modeler-app-how-to-interactively-create-ai-based-reduced-order-models-roms\/?from=jp\">Artificial Intelligence \u3078 \u6295\u7a3f<\/a>\u3055\u308c\u305f\u3082\u306e\u306e\u6284\u8a33\u3067\u3059\u3002<\/div>\n<hr \/>\n<p>&nbsp;<\/p>\n<p>\u4f4e\u6b21\u5143\u5316\u30e2\u30c7\u30eb\uff08Reduced Order Model\uff09\u306f\u3001\u30e2\u30c7\u30eb\u306e\u8907\u96d1\u3055\u3092\u524a\u6e1b\u3057\u3064\u3064\u3001\u671f\u5f85\u3055\u308c\u308b\u7cbe\u5ea6\u3092\u7dad\u6301\u3059\u308b\u305f\u3081\u306e\u624b\u6cd5\u3068\u3057\u3066\u6ce8\u76ee\u3092\u96c6\u3081\u3066\u3044\u307e\u3059\u3002MathWorks \u306f AI \u3092\u6d3b\u7528\u3057\u305f\u4f4e\u6b21\u5143\u5316\u30e2\u30c7\u30eb\u3092\u30a4\u30f3\u30bf\u30e9\u30af\u30c6\u30a3\u30d6\u306b\u4f5c\u6210\u3059\u308b\u305f\u3081\u306b\u3001<a href=\"https:\/\/www.mathworks.com\/products\/reduced-order-modeling.html\">Reduced Order Modeler<\/a> \u30a2\u30d7\u30ea\u3092\u30ea\u30ea\u30fc\u30b9\u3057\u307e\u3057\u305f\u3002\u3053\u306e\u30a2\u30d7\u30ea\u306f\u3001\u30e6\u30fc\u30b6\u30fc\u304c\u5b9f\u9a13\u3092\u8a2d\u8a08\u3057\u3001\u305d\u306e\u5b9f\u9a13\u3092\u7528\u3044\u3066\u8a73\u7d30\u30e2\u30c7\u30eb\u304b\u3089\u5165\u51fa\u529b\u30c7\u30fc\u30bf\u3092\u751f\u6210\u3057\u3001\u3053\u306e\u30c7\u30fc\u30bf\u3092\u4f7f\u3063\u3066\u4f4e\u6b21\u5143\u5316\u30e2\u30c7\u30eb\uff08ROM\uff09\u3092\u5b66\u7fd2\u3057\u3001\u305d\u306e\u7d50\u679c\u5f97\u3089\u308c\u305f ROM \u3092 Simulink \u306b\u30a8\u30af\u30b9\u30dd\u30fc\u30c8\u3057\u3066\u5229\u7528\u3059\u308b\u3053\u3068\u3092\u53ef\u80fd\u306b\u3057\u307e\u3059\u3002\u5177\u4f53\u7684\u306a\u6d41\u308c\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3067\u3059\u3002<\/p>\n<h6><\/h6>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-16406 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/ROM_workflow.png\" alt=\"The Reduced Order Modeling (ROM) workflow includes designing experiments, running experiments, training the ROM, and exporting.\" width=\"1920\" height=\"752\" \/><\/p>\n<p><strong>Figure 1:<\/strong> ROM \u958b\u767a\u306e\u30ef\u30fc\u30af\u30d5\u30ed\u30fc<\/p>\n<p><span data-teams=\"true\"><span class=\"ui-provider a b c d e f g h i j k l m n o p q r s t u v w x y z ab ac ae af ag ah ai aj ak\" dir=\"ltr\">Reduced Order Modeler \u30a2\u30d7\u30ea\u3092\u4f7f\u7528\u3057\u3066 AI \u30d9\u30fc\u30b9\u306e ROM \u3092\u4f5c\u6210\u3059\u308b\u5177\u4f53\u7684\u306a\u65b9\u6cd5\u306b\u5165\u308b\u524d\u306b\u3001\u305d\u3082\u305d\u3082\u3069\u3053\u306b ROM \u306e\u30cb\u30fc\u30ba\u304c\u3042\u308b\u306e\u304b\u898b\u3066\u307f\u307e\u3057\u3087\u3046\u3002<\/span><\/span><\/p>\n<p style=\"font-size: 20px; color: #c04c0b;\"><strong>ROM \u3092\u4f7f\u3046\u3079\u304d\u30b1\u30fc\u30b9<\/strong><\/p>\n<p>Simulink \u306e\u6d3b\u7528\u65b9\u6cd5\u3068\u3057\u3066\u5236\u5fa1\u30b7\u30b9\u30c6\u30e0\u306e\u8a2d\u8a08\u3084\u30b7\u30b9\u30c6\u30e0\u30ec\u30d9\u30eb\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u304c\u3042\u308a\u307e\u3059\u3002\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u5b9f\u884c\u901f\u5ea6\u306f\u901f\u3051\u308c\u3070\u901f\u3044\u307b\u3069\u3001\u30a8\u30f3\u30b8\u30cb\u30a2\u306f\u8fc5\u901f\u306b\u8a2d\u8a08\u306b\u304a\u3051\u308b\u8a66\u884c\u932f\u8aa4\u3092\u7e70\u308a\u8fd4\u3059\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002\u305f\u3060\u3001\u8907\u96d1\u306a\u30b7\u30b9\u30c6\u30e0\u306e\u30e2\u30c7\u30eb\u306f\u591a\u304f\u306e\u30b3\u30f3\u30dd\u30fc\u30cd\u30f3\u30c8\u304b\u3089\u69cb\u6210\u3055\u308c\u308b\u3053\u3068\u304c\u591a\u304f\u3001\u6642\u306b\u306f\u8a08\u7b97\u6d41\u4f53\u529b\u5b66\uff08CFD\uff09\u3084\u6709\u9650\u8981\u7d20\u89e3\u6790\uff08FEA\uff09\u3001\u30b3\u30f3\u30d4\u30e5\u30fc\u30bf\u652f\u63f4\u5de5\u5b66\uff08CAE\uff09\u30c4\u30fc\u30eb\u3092\u7528\u3044\u3066\u65e2\u306b\u4f5c\u6210\u3057\u305f\u30b3\u30f3\u30dd\u30fc\u30cd\u30f3\u30c8\u306e\u30e2\u30c7\u30eb\u304c\u542b\u307e\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u3089\u306e\u9ad8\u7cbe\u5ea6\u30e2\u30c7\u30eb\u306f\u3001\u8a73\u7d30\u306a\u30b3\u30f3\u30dd\u30fc\u30cd\u30f3\u30c8\u8a2d\u8a08\u306b\u306f\u6709\u7528\u3067\u3059\u304c\u3001\u30b7\u30b9\u30c6\u30e0\u30ec\u30d9\u30eb\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3084\u5236\u5fa1\u8a2d\u8a08\u306b\u306f\u9045\u3059\u304e\u3066\u5b9f\u7528\u7684\u3067\u306f\u3042\u308a\u307e\u305b\u3093\u3002<\/p>\n<p>\u9ad8\u7cbe\u5ea6\u306a\u30e2\u30c7\u30eb\u3060\u3068\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3059\u308b\u306e\u306b\u6570\u6642\u9593\u3001\u3042\u308b\u3044\u306f\u6570\u65e5\u304b\u304b\u308b\u3053\u3068\u3082\u3042\u308a\u307e\u3059\u30021 \u56de\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u5b9f\u884c\u3059\u308b\u3060\u3051\u3067\u305d\u308c\u3060\u3051\u306e\u6642\u9593\u3092\u5f85\u305f\u306a\u3051\u308c\u3070\u306a\u3089\u306a\u3044\u3068\u3057\u305f\u3089\u3001\u8a2d\u8a08\u30d7\u30ed\u30bb\u30b9\u5168\u4f53\u3082\u5f53\u7136\u6642\u9593\u304c\u304b\u304b\u308a\u307e\u3059\u3002<span data-teams=\"true\"><span class=\"ui-provider a b c d e f g h i j k l m n o p q r s t u v w x y z ab ac ae af ag ah ai aj ak\" dir=\"ltr\">\u3067\u306f\u3001\u65b0\u898f\u3067\u3001\u3088\u308a\u901f\u304f\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3067\u304d\u308b\u30e2\u30c7\u30eb\u3092\u4e00\u304b\u3089\u4f5c\u6210\u3057\u306a\u3051\u308c\u3070\u306a\u3089\u306a\u3044\u306e\u3067\u3057\u3087\u3046\u304b\uff1f\u305d\u3093\u306a\u3053\u3068\u306f\u3042\u308a\u307e\u305b\u3093\u3002\u3053\u3053\u3067\u4f4e\u6b21\u5143\u5316\u30e2\u30c7\u30eb\uff08ROM\uff09\u306e\u51fa\u756a\u3067\u3059\u3002\u3053\u308c\u306f\u3001\u9ad8\u7cbe\u5ea6\u30e2\u30c7\u30eb\u3092\u518d\u5229\u7528\u3057\u3066<\/span><\/span>\u30b7\u30b9\u30c6\u30e0\u30ec\u30d9\u30eb\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3084\u5236\u5fa1\u8a2d\u8a08\u306b\u9069\u3057\u305f\u3001\u3088\u308a\u901f\u304f\u52d5\u4f5c\u3059\u308b\u8fd1\u4f3c\u30e2\u30c7\u30eb\u3092\u4f5c\u6210\u3059\u308b\u65b9\u6cd5\u3067\u3059\u3002\u4ed6\u306b\u3082\u3001\u4eee\u60f3\u30bb\u30f3\u30b5\u30fc\u30e2\u30c7\u30eb\u3001\u30cf\u30fc\u30c9\u30a6\u30a7\u30a2\u30a4\u30f3\u30b6\u30eb\u30fc\u30d7\uff08HIL\uff09\u30c6\u30b9\u30c8\u3001\u30c7\u30b8\u30bf\u30eb\u30c4\u30a4\u30f3\u306e\u751f\u6210\u306b\u3082 ROM \u3092\u5229\u7528\u3057\u307e\u3059\u3002<\/p>\n<p>\u305d\u308c\u3067\u306f\u3001Reduced Order Modeler \u30a2\u30d7\u30ea\u3092\u4f7f\u7528\u3057\u3066 ROM \u3092\u4f5c\u6210\u3059\u308b\u65b9\u6cd5\u3092\u898b\u3066\u3044\u304d\u307e\u3057\u3087\u3046\u3002\u5177\u4f53\u7684\u306b\u306f\u3001\u30b8\u30a7\u30c3\u30c8\u30a8\u30f3\u30b8\u30f3\u306e\u30bf\u30fc\u30d3\u30f3\u30d6\u30ec\u30fc\u30c9\u306e\u4f8b\u3092\u7528\u3044\u3066\u3001ROM \u958b\u767a\u30ef\u30fc\u30af\u30d5\u30ed\u30fc\uff08Figure 1\uff09\u306e\u30b9\u30c6\u30c3\u30d7\u3092\u3054\u6848\u5185\u3057\u307e\u3059\u3002<\/p>\n<p style=\"font-size: 20px; color: #c04c0b;\"><strong>ROM \u306e\u305f\u3081\u306e\u5b9f\u9a13\u8a08\u753b<\/strong><\/p>\n<p>\u30b8\u30a7\u30c3\u30c8\u30a8\u30f3\u30b8\u30f3\u306e\u30bf\u30fc\u30d3\u30f3\u30d6\u30ec\u30fc\u30c9\u30e2\u30c7\u30eb\uff08Figure 2\uff09\u306f\u3001\u9ad8\u7cbe\u5ea6\u306a\u30e2\u30c7\u30eb\u3067\u3042\u308a\u3001\u5468\u56f2\u6e29\u5ea6\u3001\u51b7\u5374\u6e29\u5ea6\u3001\u5727\u529b\u306b\u3055\u3089\u3055\u308c\u305f\u3068\u304d\u306e\u30d6\u30ec\u30fc\u30c9\u306e\u6700\u5927\u5909\u4f4d\u3092\u8a08\u7b97\u3057\u307e\u3059\u3002\u3053\u308c\u306b\u306f\u504f\u5fae\u5206\u65b9\u7a0b\u5f0f\u3092\u89e3\u304f\u5fc5\u8981\u304c\u3042\u308a\u8a08\u7b97\u8ca0\u8377\u304c\u9ad8\u304f\u3001\u30b7\u30b9\u30c6\u30e0\u30ec\u30d9\u30eb\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3084\u5236\u5fa1\u8a2d\u8a08\u306b\u306f\u9069\u3057\u3066\u3044\u307e\u305b\u3093\u3002\u3053\u306e\u554f\u984c\u306b\u5bfe\u51e6\u3059\u308b\u305f\u3081\u306b\u3001\u307e\u305a\u5b9f\u9a13\u8a08\u753b\u304b\u3089\u59cb\u3081\u3066 ROM \u3092\u4f5c\u6210\u3057\u307e\u3059\u3002<\/p>\n<h6><\/h6>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-16376 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/Figure2-Model.png\" alt=\"Simulink model of the high-fidelity model of a jet blade turbine\" width=\"1920\" height=\"1080\" \/><\/p>\n<p><strong>Figure 2:<\/strong> \u30b8\u30a7\u30c3\u30c8\u30a8\u30f3\u30b8\u30f3\u306e\u30bf\u30fc\u30d3\u30f3\u30d6\u30ec\u30fc\u30c9\u306e\u8a73\u7d30\u30e2\u30c7\u30eb<\/p>\n<h6><\/h6>\n<p>\u3053\u306e\u30b9\u30c6\u30c3\u30d7\u3067\u306f\u3001Simulink \u30e2\u30c7\u30eb\u5185\u3067\u95a2\u9023\u3059\u308b\u4fe1\u53f7\u3092\u9078\u629e\u3057\u3001\u305d\u308c\u3089\u306e\u5165\u51fa\u529b\u30bf\u30a4\u30d7\u3092\u6307\u5b9a\u3059\u308b\u3053\u3068\u3067\u3001\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u5165\u529b\u3068 ROM \u306e\u5165\u51fa\u529b\u3092\u5b9a\u7fa9\u3057\u307e\u3059\uff08Figure 3\uff09\u3002\u9ad8\u7cbe\u5ea6\u30e2\u30c7\u30eb\u306e\u6307\u5b9a\u3055\u308c\u305f\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u5165\u529b\u306b\u6442\u52d5\u4fe1\u53f7\u3092\u6ce8\u5165\u3057\u3001\u5bfe\u5fdc\u3059\u308b ROM \u306e\u5165\u51fa\u529b\u30c7\u30fc\u30bf\u3092\u53ce\u96c6\u3059\u308b\u305f\u3081\u306e\u5b9f\u9a13\u3092\u8a2d\u5b9a\u3057\u307e\u3059\u3002\u3053\u306e\u4f8b\u3067\u306f\u3001\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u5165\u529b\u3068\u3057\u3066\u5468\u56f2\u6e29\u5ea6\u3001\u51b7\u5374\u6e29\u5ea6\u3001\u5727\u529b\u306e\u4fe1\u53f7\u304c\u542b\u307e\u308c\u3001\u30e9\u30f3\u30c0\u30e0\u306a\u30d1\u30eb\u30b9\u304c\u6ce8\u5165\u3055\u308c\u307e\u3059\u3002ROM \u306e\u5165\u529b\u306f\u3053\u308c\u3089\u306e\u30d5\u30a3\u30eb\u30bf\u30ea\u30f3\u30b0\u3055\u308c\u305f\u4fe1\u53f7\u3067\u3042\u308a\u3001ROM \u306e\u5165\u529b\u306f\u5468\u56f2\u6e29\u5ea6\u3001\u51b7\u5374\u6e29\u5ea6\u3001\u5727\u529b\u4fe1\u53f7\u306b\u5bfe\u3059\u308b\u7269\u7406\u7684\u306a\u5909\u5316\u3092\u8868\u3057\u3001ROM \u306e\u51fa\u529b\u306f\u6700\u5927\u5909\u4f4d\u3068\u306a\u308a\u307e\u3059\u3002<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-16379 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/Figure3-IO_specification.png\" alt=\"UI to select ROM inputs and ouputs\" width=\"1269\" height=\"814\" \/><\/p>\n<p><strong>Figure 3:<\/strong> \u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3068 ROM \u306e\u5165\u51fa\u529b\u8a2d\u5b9a<\/p>\n<p>&nbsp;<\/p>\n<p>\u4fe1\u53f7\u3092\u9078\u629e\u30fb\u6307\u5b9a\u3057\u305f\u5f8c\u3001\u5165\u529b\u8a2d\u8a08\u7a7a\u9593\u306e\u30ab\u30d0\u30ec\u30c3\u30b8\u3092\u78ba\u8a8d\u3057\u307e\u3059\uff08Figure 4\uff09\u3002\u305d\u3057\u3066\u3001\u4fe1\u53f7\u6ce8\u5165\u30e2\u30fc\u30c9\u3001\u4fe1\u53f7\u30bf\u30a4\u30d7\u3001\u30d1\u30eb\u30b9\u5e45\u3001\u30d1\u30eb\u30b9\u306e\u6570\u3001\u304a\u3088\u3073\u30d1\u30eb\u30b9\u632f\u5e45\u7bc4\u56f2\u306a\u3069\u306e\u5b9f\u9a13\u8a2d\u5b9a\u306e\u30c7\u30d5\u30a9\u30eb\u30c8\u5024\u3092\u8abf\u6574\u3057\u307e\u3059\u3002<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-16382 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/Figure4-DesignExperiments.png\" alt=\"Designing experiments for the collecting ROM input and output data\" width=\"1920\" height=\"1080\" \/><\/p>\n<p><strong>Figure 4:<\/strong> ROM \u306e\u5165\u51fa\u529b\u30c7\u30fc\u30bf\u53ce\u96c6\u306e\u305f\u3081\u306e\u5b9f\u9a13\u8a2d\u5b9a<\/p>\n<p>&nbsp;<\/p>\n<p style=\"font-size: 20px; color: #c04c0b;\"><b>\u30c7\u30fc\u30bf\u4f5c\u6210\u306e\u305f\u3081\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u5b9f\u884c<\/b><\/p>\n<p>\u5b9f\u9a13\u8a2d\u5b9a\u304c\u5b8c\u4e86\u3057\u305f\u3089\u3001\u30dc\u30bf\u30f3\u3092\u30af\u30ea\u30c3\u30af\u3067\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u5b9f\u884c\u3057\u307e\u3059\u3002\u307e\u305f\u3001\u51e6\u7406\u52b9\u7387\u3092\u4e0a\u3052\u308b\u305f\u3081\u306b\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u4e26\u5217\u5b9f\u884c\u3082\u5fc5\u8981\u306b\u5fdc\u3058\u3066\u5b9f\u65bd\u3057\u307e\u3059\u3002\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u304c\u5b8c\u4e86\u3059\u308b\u3068\u3001\u4ee5\u4e0b\u306b\u793a\u3059\u3088\u3046\u306b\u5b9f\u9a13\u7d50\u679c\u304c\u78ba\u8a8d\u3067\u304d\u307e\u3059\u3002<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-16385 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/Figure5-Generated_IOData_from_experiments.png\" alt=\"Waveforms of generated ROM input and output data from the experiments\" width=\"1920\" height=\"1080\" \/><\/p>\n<p><strong>Figure 5:<\/strong> ROM \u5165\u51fa\u529b\u30c7\u30fc\u30bf\u4f5c\u6210<\/p>\n<h6><\/h6>\n<p style=\"font-size: 20px; color: #c04c0b;\"><strong>ROM \u306e\u5b66\u7fd2<\/strong><\/p>\n<p>\u6b21\u306e\u30b9\u30c6\u30c3\u30d7\u306f\u4f5c\u6210\u3057\u305f\u5165\u51fa\u529b\u30c7\u30fc\u30bf\u3092\u4f7f\u7528\u3057\u305f ROM \u306e\u5b66\u7fd2\u3067\u3059\u3002\u4f5c\u6210\u3057\u305f\u51fa\u529b\u30c7\u30fc\u30bf\u306e\u6027\u8cea\u306b\u5fdc\u3058\u3066\u3001\u6642\u9593\u4f9d\u5b58\uff08\u52d5\u7684\uff09\u304b\u975e\u6642\u9593\u4f9d\u5b58\uff08\u9759\u7684\uff09\u304b\u306b\u57fa\u3065\u3044\u3066\u3001\u5b66\u7fd2\u3059\u308b\u30e2\u30c7\u30eb\u3092\u9078\u629e\u3057\u307e\u3059\u3002<\/p>\n<ul>\n<li><span data-teams=\"true\"><span class=\"ui-provider a b c d e f g h i j k l m n o p q r s t u v w x y z ab ac ae af ag ah ai aj ak\" dir=\"ltr\">\u8a2d\u8a08\u6700\u9069\u5316\u3084\u8a2d\u8a08\u306e\u30ef\u30fc\u30af\u30d5\u30ed\u30fc\u3092\u9ad8\u901f\u5316\u3059\u308b\u305f\u3081\u306e\u3001\u591a\u5c64\u30d1\u30fc\u30bb\u30d7\u30c8\u30ed\u30f3\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3084\u5185\u633f\u30e2\u30c7\u30eb\u306a\u3069\u306e\u9759\u7684 ROM<\/span><\/span><\/li>\n<li><span data-teams=\"true\"><span class=\"ui-provider a b c d e f g h i j k l m n o p q r s t u v w x y z ab ac ae af ag ah ai aj ak\" dir=\"ltr\">\u30b7\u30b9\u30c6\u30e0\u30ec\u30d9\u30eb\u306e\u89e3\u6790\u3001\u5236\u5fa1\u8a2d\u8a08\u3001HIL \u30c6\u30b9\u30c8\u3059\u308b\u305f\u3081\u306e\u3001LSTM\u3001\u975e\u7dda\u5f62 ARX\u3001\u30cb\u30e5\u30fc\u30e9\u30eb\u72b6\u614b\u7a7a\u9593\u30e2\u30c7\u30eb\u306a\u3069\u306e\u52d5\u7684 ROM<\/span><\/span><\/li>\n<\/ul>\n<p>\u3053\u306e\u4f8b\u3067\u306f\u3001\u30b8\u30a7\u30c3\u30c8\u30a8\u30f3\u30b8\u30f3\u306e\u30bf\u30fc\u30d3\u30f3\u30d6\u30ec\u30fc\u30c9\u30b7\u30b9\u30c6\u30e0\u306e\u52d5\u7684\u7279\u6027\u3092\u52b9\u679c\u7684\u306b\u6349\u3048\u308b\u305f\u3081\u306b\u3001\u30cb\u30e5\u30fc\u30e9\u30eb\u72b6\u614b\u7a7a\u9593\u30e2\u30c7\u30eb\u3092\u4f7f\u7528\u3057\u307e\u3059\u3002Reduced Order Modeler \u30a2\u30d7\u30ea\u3067\u5bfe\u5fdc\u3059\u308b\u30c6\u30f3\u30d7\u30ec\u30fc\u30c8\uff08\u3053\u306e\u5834\u5408\u306f\u30cb\u30e5\u30fc\u30e9\u30eb\u72b6\u614b\u7a7a\u9593\u30e2\u30c7\u30eb\uff09\u3092\u30af\u30ea\u30c3\u30af\u3059\u308b\u3068\u3001\u81ea\u52d5\u7684\u306b Experiment Manager \u30a2\u30d7\u30ea\u304c\u8d77\u52d5\u3057\u307e\u3059\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u3055\u307e\u3056\u307e\u306a\u30cf\u30a4\u30d1\u30fc\u30d1\u30e9\u30e1\u30fc\u30bf\u306e\u7bc4\u56f2\u3092\u30b9\u30a4\u30fc\u30d7\u3057\u3066\u7570\u306a\u308b\u30e2\u30c7\u30eb\u3092\u5b66\u7fd2\u3057\u3001\u5404\u30e2\u30c7\u30eb\u306e\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u3092\u8a55\u4fa1\u30fb\u6bd4\u8f03\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-16391 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/Figure6-Training_results.png\" alt=\"Hyperparameters of the neural state space models trained with ROM input and output data\" width=\"1920\" height=\"1080\" \/><\/p>\n<p><strong>Figure 6:<\/strong> \u30cb\u30e5\u30fc\u30e9\u30eb\u72b6\u614b\u7a7a\u9593\u30e2\u30c7\u30eb\u306e\u5b66\u7fd2<\/p>\n<p>&nbsp;<\/p>\n<p>\u5b66\u7fd2\u304c\u5b8c\u4e86\u3059\u308b\u3068\u3001Loss\u3001TrainingMSE\u3001TestMSE \u3068\u3044\u3063\u305f\u6307\u6a19\u3092\u7528\u3044\u3066\u30e2\u30c7\u30eb\u3092\u8a55\u4fa1\u30fb\u6bd4\u8f03\u3057\u307e\u3059\u3002TrainingMSE \u3068 TestMSE \u306f\u3001\u305d\u308c\u305e\u308c\u5143\u306e\u5b66\u7fd2\u30c7\u30fc\u30bf\u3068\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u3092\u7528\u3044\u305f\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u51fa\u529b\u304b\u3089\u8a08\u7b97\u3055\u308c\u305f\u5e73\u5747\u4e8c\u4e57\u8aa4\u5dee\u3067\u3059\u3002TestMSE \u306e\u5024\u304c\u4f4e\u3044\u9806\u306b\u5b66\u7fd2\u6e08\u307f\u30e2\u30c7\u30eb\u3092\u4e26\u3079\u66ff\u3048\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002TestMSE \u304c\u4f4e\u304f\u3001TrainingMSE \u3082\u540c\u69d8\u306b\u4f4e\u304f\u3001Loss \u304c\u6700\u5c0f\u9650\u3067\u3042\u308b\u30e2\u30c7\u30eb\u304c\u3088\u3044\u5019\u88dc\u3068\u306a\u308a\u307e\u3059\u3002\u540c\u69d8\u306b\u3001LSTM \u3084\u975e\u7dda\u5f62 ARX \u30e2\u30c7\u30eb\u306a\u3069\u306e\u4ed6\u306e\u52d5\u7684 ROM \u30e2\u30c7\u30eb\u3092\u5b66\u7fd2\u3057\u3001\u305d\u306e\u7d50\u679c\u3092\u6bd4\u8f03\u3059\u308b\u3053\u3068\u3082\u53ef\u80fd\u3067\u3059\u3002<\/p>\n<p>&nbsp;<\/p>\n<p style=\"font-size: 20px; color: #c04c0b;\"><strong>ROM \u306e\u30a8\u30af\u30b9\u30dd\u30fc\u30c8<\/strong><\/p>\n<p>\u6700\u5f8c\u306e\u30b9\u30c6\u30c3\u30d7\u306f\u3001\u5b66\u7fd2\u6e08\u307f\u306e ROM \u3092 MATLAB \u30ef\u30fc\u30af\u30b9\u30da\u30fc\u30b9\u306b\u30a8\u30af\u30b9\u30dd\u30fc\u30c8\u3057\u3001Neural State Space \u30d6\u30ed\u30c3\u30af\uff08Figure 7\uff09\u3092\u4f7f\u7528\u3057\u3066 Simulink \u30e2\u30c7\u30eb\u306b\u53d6\u308a\u8fbc\u307f\u3001<span data-teams=\"true\"><span class=\"ui-provider a b c d e f g h i j k l m n o p q r s t u v w x y z ab ac ae af ag ah ai aj ak\" dir=\"ltr\">\u5143\u306e\u9ad8\u7cbe\u5ea6\u30e2\u30c7\u30eb\u3068\u306e\u6bd4\u8f03\u3092\u3059\u308b\u3053\u3068\u3067\u3059<\/span><\/span><\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-16394 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/Figure7-Bring_ROM_into_Simulink.png\" alt=\"Simulink model with ROM (instead of high-fidelity model)\" width=\"1920\" height=\"1080\" \/><\/p>\n<p><strong>Figure 7:<\/strong> \u5b66\u7fd2\u3057\u305f ROM \u3092 Simulink \u30e2\u30c7\u30eb\u306b\u53d6\u308a\u3053\u307f\u5143\u306e\u9ad8\u7cbe\u5ea6\u30e2\u30c7\u30eb\u3068\u6bd4\u8f03<\/p>\n<h6><\/h6>\n<p>\u7d50\u679c\uff08Figure 8\uff09\u3067\u793a\u3055\u308c\u3066\u3044\u308b\u3088\u3046\u306b\u3001\u5b66\u7fd2\u6e08\u307f\u306e\u30cb\u30e5\u30fc\u30e9\u30eb\u72b6\u614b\u7a7a\u9593\u30e2\u30c7\u30eb\u306f\u3001\u5143\u306e\u9ad8\u7cbe\u5ea6\u30e2\u30c7\u30eb\u306e\u30b7\u30b9\u30c6\u30e0\u52d5\u614b\u3092\u5fe0\u5b9f\u306b\u6349\u3048\u3066\u3044\u307e\u3059\u3002\u5b66\u7fd2\u6e08\u307f\u306e ROM \u306f 0.03 \u79d2\u3067\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u884c\u3046\u305f\u3081\u3001\u30b7\u30b9\u30c6\u30e0\u30ec\u30d9\u30eb\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3084\u5236\u5fa1\u8a2d\u8a08\u306b\u975e\u5e38\u306b\u9069\u3057\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306b\u5bfe\u3057\u5143\u306e\u9ad8\u7cbe\u5ea6\u306a\u30e2\u30c7\u30eb\u306f\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306b 30,000 \u79d2\u304b\u304b\u3063\u3066\u3044\u307e\u3059\u3002<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-16397 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/Figure8-Comparison_ROM_highfidelitymodel.png\" alt=\"Comparison of simulation results between ROM and high-fidelity model\" width=\"1920\" height=\"1080\" \/><\/p>\n<p>&nbsp;<\/p>\n<p>ROM \u306e\u6e96\u5099\u304c\u6574\u3063\u305f\u3089\u3001Simulink \u3067\u306e\u5236\u5fa1\u8a2d\u8a08\u3084 HIL \u30c6\u30b9\u30c8\u306b\u5229\u7528\u3057\u305f\u308a\u3001<span data-teams=\"true\"><span class=\"ui-provider a b c d e f g h i j k l m n o p q r s t u v w x y z ab ac ae af ag ah ai aj ak\" dir=\"ltr\">FMU\uff08Functional Mock-Up Unit\uff09\u3092\u4ecb\u3057\u3066 Simulink \u306e\u5916\u90e8\u3067\u4f7f\u7528\u3059\u308b\u305f\u3081\u306b\u30a8\u30af\u30b9\u30dd\u30fc\u30c8\u3059\u308b\u3053\u3068\u3082\u3067\u304d\u307e\u3059\u3002<\/span><\/span><\/p>\n<h6><\/h6>\n<p style=\"font-size: 20px; color: #c04c0b;\"><strong>\u307e\u3068\u3081<\/strong><\/p>\n<p>Reduced Order Modeler \u30a2\u30d7\u30ea\u3092\u8a66\u3057\u3066\u307f\u307e\u305b\u3093\u304b\uff1f<a href=\"https:\/\/www.mathworks.com\/matlabcentral\/fileexchange\/156364-reduced-order-modeling-support-package-beta\">\u3053\u3061\u3089<\/a>\u304b\u3089\u30b5\u30dd\u30fc\u30c8\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3067\u304d\u307e\u3059\u3002<\/p>\n<p>\u4f4e\u6b21\u5143\u30e2\u30c7\u30eb\u5316\u306e\u4e16\u754c\u3092\u3055\u3089\u306b\u6df1\u304f\u77e5\u308a\u3001\u305d\u306e\u5b9f\u8df5\u7684\u306a\u5fdc\u7528\u5148\u3092\u77e5\u308b\u306b\u306f\u3001<a href=\"https:\/\/www.mathworks.com\/videos\/series\/reduced-order-modeling.html\">ROM \u306b\u95a2\u3059\u308b\u30d3\u30c7\u30aa\u30b7\u30ea\u30fc\u30ba<\/a>\u3084\u3001\u30a6\u30a7\u30d3\u30ca\u30fc\u300c<a href=\"https:\/\/www.mathworks.com\/videos\/ai-with-model-based-design-reduced-order-modeling-1670274257624.html\">AI with Model-Based Design: Reduced Order Modeling<\/a>\u300d\u306e\u9332\u753b\u3092\u3054\u89a7\u304f\u3060\u3055\u3044\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<div class=\"overview-image\"><img decoding=\"async\"  class=\"img-responsive\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2024\/10\/ROM_workflow.png\" onError=\"this.style.display ='none';\" \/><\/div>\n<p>\u203b\u3053\u306e\u6295\u7a3f\u306f 2024 \u5e74 10 \u6708 21 \u65e5\u306bArtificial Intelligence \u3078 \u6295\u7a3f\u3055\u308c\u305f\u3082\u306e\u306e\u6284\u8a33\u3067\u3059\u3002<\/p>\n<p>&nbsp;<br \/>\n\u4f4e\u6b21\u5143\u5316\u30e2\u30c7\u30eb\uff08Reduced Order Model\uff09\u306f\u3001\u30e2\u30c7\u30eb\u306e\u8907\u96d1\u3055\u3092\u524a\u6e1b\u3057\u3064\u3064\u3001\u671f\u5f85\u3055\u308c\u308b\u7cbe\u5ea6\u3092\u7dad\u6301\u3059\u308b\u305f\u3081\u306e\u624b\u6cd5\u3068\u3057\u3066\u6ce8\u76ee\u3092\u96c6\u3081\u3066\u3044\u307e\u3059\u3002MathWorks \u306f AI \u3092\u6d3b\u7528\u3057\u305f\u4f4e\u6b21\u5143\u5316\u30e2\u30c7\u30eb\u3092\u30a4\u30f3\u30bf\u30e9\u30af\u30c6\u30a3\u30d6\u306b\u4f5c\u6210\u3059\u308b\u305f\u3081\u306b\u3001Reduced&#8230; <a class=\"read-more\" href=\"https:\/\/blogs.mathworks.com\/japan-community\/2024\/11\/08\/reduced-order-modeler-app-how-to-interactively-create-ai-based-reduced-order-models-roms-jp\/\">read more >><\/a><\/p>\n","protected":false},"author":159,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[61],"tags":[],"_links":{"self":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/12548"}],"collection":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/users\/159"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/comments?post=12548"}],"version-history":[{"count":15,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/12548\/revisions"}],"predecessor-version":[{"id":12647,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/12548\/revisions\/12647"}],"wp:attachment":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/media?parent=12548"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/categories?post=12548"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/tags?post=12548"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}