The Assessment of Cutting Ability of Nitride-Boron High Porous Wheels when Flat Grinding Steel 13Kh15N5AM3 Parts by Surface Macrogeometry Using Artificial Intelligence

The assessment of cutting ability of high porous wheels made from cubic boron nitride was conducted with the assistance of artificial intelligence systems: fuzzy logic and neural networks. The wheels’ cutting ability was evaluated by three indicators of flatness deviation: EFE_max, EFE_a and EFE_q, each being represented by two measures of position and dispersion, that is the median and the interquartile range. The cutting ability of eleven high porous wheels was analyzed. They differed by the type of cubic boron nitride (LKV 50, CBN 30), grain size (B76, B107, B126, B151), hardness (L, O, M), bonds (K27, S10), and pore-forming agent (KF40, KF25). High porous wheels LKV 50 B126 100 MVK27-KF40 and CBN30 В126 100 LVK27-KF25 that were rated «very good», were recommended for grinding parts made from steel 13Kh15N5AM3. In the study of high porous wheels with low cutting ability, the neural networks appeared to be more reliable than the fuzzy logic.

References

[1] Suslov A.G., Bez"iazychnyi V.F., Panfilov Iu.V. Inzheneriia poverkhnosti detalei [Engineering details surface]. Moscow, Mashinostroenie publ., 2008. 320 p.

[2] GOST 24642–81. Dopuski formy i raspolozheniia poverkhnostei. Osnovnye poniatiia i oboznacheniia [State Standard 24642–81. Basic norms of interchangeability. Tolerances of form and position. Terms and definitions]. Moscow, Standartinform publ., 1984. 68 p.

[3] Orlov A.I. Teoriia nechetkikh mnozhestv — chast’ teorii veroiatnostei [Fuzzy sets theory as the part of probability theory]. Nauchnyi zhurnal KubGAU [Scientific Journal of KubSAU]. 2013, no. 8(92), pp. 51–60.

[4] Leonenkov A.V. Nechetkoe modelirovanie v srede MATLAB i fuzzyTECH [Fuzzy modeling in MATLAB environment and fuzzyTECH]. Sankt-Peterburg, BKhV Peterburg publ., 2005. 736 p.

[5] Sistemy nechetkogo vyvoda [Fuzzy inference system]. Available at: http://nrsu.bstu.ru/chap27.html (accessed 15 March 2016).

[6] Khaikin S.B. Neironnye seti: polnyi kurs [Neural networks: a complete course]. Moscow, Vil’iams publ., 2006. 1104 p.

[7] Terekhov S.A. Lektsii po teorii i prilozheniiam iskusstvennykh neironnykh setei [Lectures on the theory and applications of artificial neural networks]. Available at: http://www.gotai.net/documents/doc-nn-007.aspx (accessed 11 March 2016).

[8] Quintana G., Garcia-Romeu M.L., Ciurana J. Surface roughness monitoring application based on artificial neural networks for ball-end milling operations. Journal of Intelligent Manufacturing, 2011, no. 22, pp. 607–617.

[9] Sick B. On-line and indirect tool wear monitoring in turning with artificial neural networks: a review of more than a decade of research. Mechanical systems and signal processing, 2002, vol. 16, pp. 487–546.

[10] Caydas U., Hascalik A. A study on surface roughness in abrasive waterjet machining process using artificial neural networks and regression analysis method. Journal of materials processing technology, 2008, no. 202, pp. 574–582.

[11] Nazar’eva V.A. Setevaia model’ vybora zernistosti shlifoval’nykh krugov s elementami ispol’zovaniia sistem iskusstvennogo intellekta [Network model of choice grain grinding wheels with the elements of the use of artificial intelligence systems]. STIN [Russian Engineering Research]. 2016, no. 2, pp. 37–40.

[12] GOST R 53923–2010. Krugi almaznye i iz kubicheskogo nitrida bora (el’bora) shlifoval’nye. Tekhnicheskie usloviia [ISO 22917:2004 Superabrasives – Limit deviations and run-out tolerances for grinding wheels with diamond or cubic boron nitride (MOD)]. Moscow, Standartinform publ., 2011. 27 p.

[13] Frezy i frezerovanie [Mills and milling]. Ed. Promptova A.I. Irkutsk, IrSTU publ., 2009. 172 p.

[14] GOST R 53922–2010. Poroshki almaznye i iz kubicheskogo nitrida bora (el’bora). Zernistost’ i zernovoi sostav shlifporoshkov. Kontrol’ zernovogo sostava [ISO 6106:2005 Abrasive products — checking the grit size of superabrasives (MOD)]. Moscow, Standartinform publ., 2010. 27 p.

[15] Soler Ya.I., Nguyen V.L. Selection of Synthesis Corundum Grain in Grinding Flat Parts from Hardened Steel 30ChGSA by the Macrogeometry Criterion, Applied Mechaniсs and Materials, 2015, vol. 788, pp. 95–101.

[16] Hollander M., Wolfe D.A. Nonparametric statistical methods. Wiley-Interscience, 1999. 787 p.

[17] Zaks L. Statisticheskoe otsenivanie [Statistical estimation]. Moscow, Statistika publ., 1976. 598 p.

[18] Uiller D., Chambers D. Statisticheskoe upravlenie protsessami [Statistical Process Control]. Moscow, Al’pina Biznes Buks, 2009. 469 p.

[19] GOST R ISO 5725-1–2002. Tochnost’ (pravil’nost’ i pretsizionnost’) metodov i rezul’tatov izmereniia. Ch. 1. Osnovnye poniatiia i opredeleniia [State Standard R ISO 5725-1–2002. Accuracy (trueness and precision) of measurement methods and results. Part 1. General principles and definitions]. Moscow, Standartinform publ., 2002. 24 p.

[20] GOST 24643–81. Dopuski formy i raspolozheniia poverkhnostei. Chislovye znacheniia [State Standard 24643–81. Basic norms of interchangeability. Tolerances of form and position. Numerical values]. Moscow, Standartinform publ., 1984. 14 p.

[21] Soler Ya.I., Kazimirov D.Yu. Selecting abrasive wheels for the plane grinding of airplane parts of the basis surface roughness. Russian engineering research, 2010, vol. 30, no. 3, pp. 251–261.

[22] Soler Ia.I., Nguen M.T. Poisk optimal’noi kharakteristiki nitridborovykh krugov pri ploskom shlifovanii detalei iz stali 06Kh14N6D2MVT-Sh po mikrorel’efu poverkhnosti v usloviiakh modelirovaniia nechetkoi logiki [Search for optimal grain size of nitride-boron wheels during flat grinding of parts made of 06H14N6D2MVT-SH steel on surface microrelief under conditions of fuzzy logic simulation]. Vestnik MGTU im. N. E. Baumana. Ser. Mashinostroenie [Herald of the Bauman Moscow State Technical University. Ser. Mechanical Engineering]. 2015, no. 6, pp. 96–111.

[23] Soler Ia.I., Nguen M.T. Optimizatsiia mikrorel’efa poverkhnosti ploskikh detalei iz korrozionno-stoikoi stali 13Kh15N5AM3 pri shlifovanii nitridborovymi krugami vysokoi poristosti [Optimization of surface microrelief flat parts made of corrosion-resistant steel 13H15N5AM3 grinding with community nitridborymi high porosity]. Fundamental’nye i prikladnye problemy tekhniki i tekhnologii [Fundamental and Applied Problems of Technics and technology]. 2015, no. 6(314), pp. 65–72.

[24] Ali Y.M., Zhang L.C. Surface roughness prediction of ground components using a fuzzy logic approach. Journal of Materials Processing Technology, 1999, vol. 89–90, pp. 561–568.

[25] Ali Y.M., Zhang L.C. A fuzzy model for predicting burns in surface grinding of steel. International Journal of Machine Tools and Manufacture, 2004, vol. 44(5), pp. 563–571.

[26] Viatchenin D.A. Nechetkie metody avtomaticheskoi klassifikatsii [Fuzzy methods of automatic classification]. Minsk, UP Tekhnoprint publ., 2004. 219 p.

[27] Kofman A. Vvedenie v teoriiu nechetkikh mnozhestv [Introduction to the theory of fuzzy sets]. Moscow, Radio i sviaz’ publ., 1982. 432 p.

[28] Harrington E.C. The desirability function. Industrial Quality Control, 1965, vol. 21, pp. 494–498.

[29] Mandrov B.I., Baklanov S.D., Baklanov D.D., Vlesko A.S., Putivskii A.N., Sukhinina S.D. Primenenie funktsii zhelatel’nosti Kharringtona pri ekstruzionnoi svarke listov iz polietilena marki PEND [Application features desirability Harrington with extrusion welding sheets of polyethylene HDPE grade]. Polzunovskii al’manakh [Polzunovsky almanac]. 2012, no. 1, pp. 62–64.

[30] Komartsova L.G., Maksimov A.V. Neirokomp’iutery [Neurocomputers]. Moscow, Bauman Press, 2004. 400 p.

[31] Vasenkov D.V. Metody obucheniia iskusstvennykh neironnykh setei [Methods for training of artificial neural networks]. Komp’iuternye instrumenty v obrazovanii [Computer Tools in Education]. 2007, no. 1, pp. 20–29.

[32] Oludele A., Olawale J. Neural Networks and Its Application in Engineering. Proceedings of Informing Science & IT Education Conference (InSITE), Macon, GA, USA, 2009, pp. 83–95.

[33] Rudoi G.I. Vybor funktsii aktivatsii pri prognozirovanii neironnymi setiami [Selecting the activation function in predicting neural networks]. Mashinnoe obuchenie i analiz dannykh [Journal of Machine Learning and Data Analysis]. 2011, no. 1, pp. 16–39.

[34] Kruglov V.V., Borisov V.V. Iskusstvennye neironnye seti. Teoriia i praktika [Artificial neural networks. Theory and practice]. Moscow, Goriachaia liniia-Telekom publ., 2001. 382 p.

[35] Sharstnev V.L., Vardomatskaia E.Iu. Analiz vozmozhnostei neironnykh setei dlia prognozirovaniia zadach legkoi promyshlennosti [Analysis of neural network forecasting capabilities for light industrial tasks]. Upravlenie ekonomicheskimi sistemami: elektronyi nauchnyi zhurnal [Economic systems management]. 2007, no. 9, pp. 3–7. Available at: http://uecs.mcnip.ru/modules.php?name=News&file=article&sid=67 (accessed 10 May 2016).