The Development of the Machine-Tool Dynamic Passport Based on Neural Network Modeling of its Working Space Using nVidia CUDA Technology and Deep Learning Approaches

The authors propose a method of developing an individual dynamic passport of a machine-tool that allows them to determine optimal modes of operation and setup of the machine in an automated mode. The method is based on identifying the relationships between the input parameters of the cutting process using experimental data (mode of cutting, machined material, workpiece fixing method, tool geometry, geometric parameters of the workpiece) and the output parameters (amplitude, frequency, and fractal dimension of the attractor) using a neural network adapted to the regression analysis. After training, the regression neural network model of the cutting process is created, that is able to simulate any combination of the input parameters of the cutting process and analyze the output values, therefore determining the margin and area of the system stability in a wide setting range.  The resulting neural network model reflects the dynamics of the specific equipment that leads to high output rates of machining without compromising quality. The use of nVidia CUDA parallel computing algorithms significantly accelerates the learning process of the neural network, and therefore makes it possible to use them in operational diagnostics in production systems.

References

[1] Rezhimy rezaniia metallov: spravochnik [Metal cutting modes: a guide]. Ed. Baranovskii Iu.V. Moscow, NIITavtoprom publ., 1995. 456 p.

[2] Kudinov V.A. Dinamika stankov [Machine dynamics]. Moscow, Mashinostroenie publ., 1967. 359 p.

[3] Kabaldin Iu.G., Laptev I.L., Shatagin D.A., Seryi S.V. Diagnostika dinamicheskoi ustoichivosti i iznosa instrumenta v tekhnologicheskikh sistemakh na osnove iskusstvennogo intellekta s ispol’zovaniem vychislenii nVidia CUDA pri udalennom dostupe [Diagnosis of dynamic stability and tool wear in technological systems based on artificial intelligence using nVidia CUDA computing for remote access]. N. Novgorod, Nizhny Novgorod State Technical University publ., 2014. 112 p.

[4] Kabaldin Iu.G., Bilenko S.V., Seryi S.V. Upravlenie dinamicheskim kachestvom metallorezhushchikh sistem na osnove iskusstvennogo intellekta [Management dynamic quality of cutting systems based on artificial intelligence]. Komsomol’sk-na-Amure, Komsomolsk-na-Amure State Technical University publ., 2004. 240 p.

[5] Ezhov A.A., Shumskii S.A. Neirokomp’iuting i ego primeneniia v ekonomike i biznese [Neurocomputing and its application in economics and business]. Moscow, MIFI publ., 1998. 224 p.

[6] Gorban’ A.N., Dunin-Barkovskii V.L., Kirdin A.N., Mirkes E.M., Novokhod’ko A.Iu., Rossiev D.A., Terekhov S.A., Senashova M.Iu., Tsaregorodtsev V.G. Neiroinformatika [Neuroinformatics]. Novosibirsk, Nauka publ., 1998. 296 p.

[7] Uossermen F. Neirokomp’iuternaia tekhnika: teoriia i praktika [Neurocomputing equipment: theory and practice]. Moscow, Mir publ., 1992. 127 p.

[8] Zamiatin N.V., Medintsev D.V. Metodika neirosetevogo modelirovaniia slozhnykh system [Technique of neural modelling of complex systems]. Izvestiia Tomskogo politekhnicheskogo universiteta [Bulletin of the Tomsk Polytechnic University]. 2006, vol. 309, no. 8, pp. 100–106.

[9] Eremin D.M., Gartsev I.B. Iskusstvennye neironnye seti v intellektual’nykh sistemakh upravleniia [Artificial neural network in intelligent control systems]. Moscow, MIREA publ., 2004. 75 p.

[10] Sigeru Omatu, Marzuki Khalid, Rubiia Iusof. Neiroupravlenie i ego prilozheniia [Neurocontrol and its applications]. Book 2. Moscow, IPRZhR publ., 2000. 272 p.

[11] Popov E.V., Fominykh I.B., Kisel’ E.B., Shapot M.D. Staticheskie i dinamicheskie ekspertnye sistemy [Static and dynamic expert systems]. Moscow, Finansy i statistika publ., 1996. 320 p.

[12] Neilor K. Kak postroit’ svoiu ekspertnuiu sistemu [How to build your expert system]. Moscow, Energoatomizdat publ., 1991. 284 p.

[13] Iskusstvennyi intellekt. V 3-kh kn. Kn. 1. Sistemy obshcheniia i ekspertnye sistemy: Spravochnik [Artificial Intelligence. In 3 Vol. Vol. 1. The communication systems and expert systems: a handbook]. Ed. Popov E.V. Moscow, Radio i sviaz’ publ., 1990. 464 p.

[14] Iskusstvennyi intellekt. V 3-kh kn. Kn. 2. Modeli i metody: spravochnik [Artificial Intelligence. In 3 Vol. Vol. 2. Models and methods: a handbook]. Ed. Pospelov D.A. Moscow, Radio i sviaz’ publ., 1990. 304 p.

[15] Zakharov V.N., Khoroshevskii V.F. Iskusstvennyi intellekt: V 3-kh. Kn. 3. Programmnye i apparatnye sredstva: spravochnik [Artificial Intelligence. In 3 Vol. Vol. 3. Software and hardware: a handbook]. Moscow, Radio i sviaz’ publ., 1990. 304 p.

[16] Lor’er Zh.L. Sistemy iskusstvennogo intellekta [Artificial Intelligence systems]. Moscow, Mir publ., 1991. 342 p.