"Vyu\u017Eit\u00ED neuronov\u00FDch s\u00EDt\u00ED v materi\u00E1lov\u00E9m v\u00FDzkumu"@cs . "CZ - \u010Cesk\u00E1 republika" . "LII" . "[B70B8A83A64E]" . "27360" . . "P(GA106/05/2596)" . "65-69" . . . . . . "Exploitation of Neural Networks In Material Research" . . . . "Theoretical knowledge of physical metallurgy doesn't express comprehensively all physical variables, which influence resultant product manufacture quality. Therefore data files contained in IRA and ARA steel diagrams of different chemical composition are the basic tool for heat treatment. It is possible to treat these data statistically and thus acquire empiric relations, which serve for partial processes course prediction proceeding at heat treatment. These relations were obtained so far on the basis of regression analysis of measured data. Real possibility of prediction of different steel parameter with exploitation of artificial intelligence elements offers at present. Model of prediction of steel mechanical properties after heat treatment using neural networks methods was created. By suitable connection of these results with knowledge of physical metallurgy and with practical data about heat treatment it's possible to obtain groundwork for creation of semi-empiric model of heat treatment" . . . "neural networks; prediction; model; metallurgy; steel"@en . . "RIV/61989100:27360/06:00014224" . "Exploitation of Neural Networks In Material Research"@en . "2" . "2"^^ . "5"^^ . "475290" . "Theoretical knowledge of physical metallurgy doesn't express comprehensively all physical variables, which influence resultant product manufacture quality. Therefore data files contained in IRA and ARA steel diagrams of different chemical composition are the basic tool for heat treatment. It is possible to treat these data statistically and thus acquire empiric relations, which serve for partial processes course prediction proceeding at heat treatment. These relations were obtained so far on the basis of regression analysis of measured data. Real possibility of prediction of different steel parameter with exploitation of artificial intelligence elements offers at present. Model of prediction of steel mechanical properties after heat treatment using neural networks methods was created. By suitable connection of these results with knowledge of physical metallurgy and with practical data about heat treatment it's possible to obtain groundwork for creation of semi-empiric model of heat treatment"@en . . "Vyu\u017Eit\u00ED neuronov\u00FDch s\u00EDt\u00ED v materi\u00E1lov\u00E9m v\u00FDzkumu"@cs . "Exploitation of Neural Networks In Material Research"@en . "Teoretick\u00E9 poznatky fyzik\u00E1ln\u00ED metalurgie nepostihuj\u00ED dosud komplexn\u011B v\u0161echny fyzik\u00E1ln\u00ED prom\u011Bnn\u00E9, kter\u00E9 ovliv\u0148uj\u00ED v\u00FDsledn\u00E9 mechanick\u00E9 vlastnosti v\u00FDrobku. Proto jsou z\u00E1kladn\u00ED pom\u016Fckou pro tepeln\u00E9 zpracov\u00E1n\u00ED IRA a ARA diagramy oceli o dan\u00E9m chemick\u00E9m slo\u017Een\u00ED. \u00DAdaje obsa\u017Een\u00E9 v IRA a ARA diagramech je mo\u017Eno statisticky zpracovat a takto doj\u00EDt k empirick\u00FDm vztah\u016Fm, kter\u00E9 slou\u017E\u00ED pro predikci pr\u016Fb\u011Bhu d\u00EDl\u010D\u00EDch proces\u016F prob\u00EDhaj\u00EDc\u00EDch p\u0159i tepeln\u00E9m zpracov\u00E1n\u00ED. Doposud byly tyto vztahy z\u00EDsk\u00E1v\u00E1ny na z\u00E1klad\u011B regresn\u00ED anal\u00FDzy nam\u011B\u0159en\u00FDch dat. V sou\u010Dasn\u00E9 dob\u011B se nask\u00FDt\u00E1 re\u00E1ln\u00E1 mo\u017Enost predikce r\u016Fzn\u00FDch parametr\u016F oceli p\u0159i tepeln\u00E9m zpracov\u00E1n\u00ED s vyu\u017Eit\u00EDm prvk\u016F um\u011Bl\u00E9 inteligence. Byl vytvo\u0159en neuronov\u00FD model pro predikci mechanick\u00FDch vlastnost\u00ED ocel\u00ED po tepeln\u00E9m zpracov\u00E1n\u00ED. Dan\u00E1 problematika, spolu s dal\u0161\u00EDmi technick\u00FDmi aplikacemi um\u011Bl\u00E9 inteligence, byla \u0159e\u0161ena v r\u00E1mci grantov\u00E9ho projektu GA\u010CR 106/05/2596."@cs . "2"^^ . . . "Jan\u010D\u00EDkov\u00E1, Zora" . "RIV/61989100:27360/06:00014224!RIV07-GA0-27360___" . "Exploitation of Neural Networks In Material Research" . "1210-0471" . . "Heger, Milan" . "Sborn\u00EDk v\u011Bdeck\u00FDch prac\u00ED vysok\u00E9 \u0161koly b\u00E1\u0148sk\u00E9-Technick\u00E9 univerzity Ostrava" . . .