Application of instance-based learning for cast iron casting defects prediction

• Autorzy: Sika Robert , Szajewski Damian , Hajkowski Jakub , Popielarski Paweł

Identyfikatory

DOI

10.24425/mper.2019.131450

• Języki publikacji: EN

Abstrakty

EN

The paper presents an example of Instance-Based Learning using a supervised classification method of predicting selected ductile cast iron castings defects. The test used the algorithm of k-nearest neighbours, which was implemented in the authors’ computer application. To ensure its proper work it is necessary to have historical data of casting parameter values registered during casting processes in a foundry (mould sand, pouring process, chemical composition) as well as the percentage share of defective castings (unrepairable casting defects). The result of an algorithm is a report with five most possible scenarios in terms of occurrence of a cast iron casting defects and their quantity and occurrence percentage in the casts series. During the algorithm testing, weights were adjusted for independent variables involved in the dependent variables learning process. The algorithms used to process numerous data sets should be characterized by high efficiency, which should be a priority when designing applications to be implemented in industry. As it turns out in the presented mathematical instance-based learning, the best quality of fit occurs for specific values of accepted weights (set #5) for number k = 5 nearest neighbours and taking into account the search criterion according to “product index”.

Słowa kluczowe

EN

soft modelling; instance-based learning; k-nearest neighbours algorithm; cast iron casting defects; computer application

• Wydawca: Production Engineering Committee of the Polish Academy of Sciences ; Polish Association for Production Management
• Czasopismo: Management and Production Engineering Review
• Rocznik: 2019
• Tom: Vol. 10, No. 4
• Strony: 101--107
• Opis fizyczny: Bibliogr. 35 poz., rys., tab.

Twórcy

autor

Sika Robert

• Poznan University of Technology, Faculty of Mechanical Engineering and Management, Piotrowo 3, 60-965 Poznań, Poland

autor

Szajewski Damian

• Poznan University of Technology, Faculty of Mechanical Engineering and Management, Poland

autor

Hajkowski Jakub

• Poznan University of Technology, Faculty of Mechanical Engineering and Management, Poland

autor

Popielarski Paweł

• Poznan University of Technology, Faculty of Mechanical Engineering and Management, Poland

Bibliografia

• [1] Ignaszak Z., Popielarski P., Hajkowski J., Problem of Acceptability of Internal Porosity in SemiFinished Cast Product as New Trend “Tolerance of Damage”, Present in Modern Design Office, Defect and Diffusion Forum, 326–328, 612–619, 2012.
• [2] HamrolA., ZerbstS.,BozekM.,GrabowskaM.,Weber M., Analysis of the conditions for effective use of numerically controlled machine tools, Lec. Not. in Mech. Eng., Springer, 201519, 3–12, 2018.
• [3] Ivanov V., Dehtiarov I., Pavlenko I., Liaposhchenko O., Zaloga V., Parametric optimization of fixtures for multiaxis machining of parts, [in:] Hamrol A., Kujawińska A., Barraza M. [Eds], Advances in Manufacturing II, Lec. Not. in Mech. Eng., Springer, pp. 335–347, 2019.
• [4] Kujawińska A., Vogt K., Diering M., Rogalewicz M., Waigaonkar M.D., Organization of visual inspection and its impact on the effectiveness of inspection, Advances in Manufacturing, Lec. Not. in Mech. Eng., Springer, 201519, 899–909, 2018.
• [5] Rogalewicz M., Kujawińska A., Piłacińska M., Selection of data mining method for multidimensional evaluation of the manufacturing process state, Management and Production Engineering Review, Production Engineering Committee PAN, 3, 2, 27–35, 2012.
• [6] Ignaszak Z., Hajkowski J., Contribution to the Identification of Porosity Type in AlSiCu High-PressureDie-Castings by Experimental and Virtual Way, Arch. of Found. Eng., 15, 1, 143–151, 2015.
• [7] Grabowska M., Takala J., Assessment of quality management system maturity, Lec. Not. in Mech. Eng., Springer, 201519, 889–898, 2018.
• [8] Presentation at the French Embassy in the Germany, “Industry of the future“, http://www.ambafrance-de.org/Vorstellung-des-neuen-franzosischenPlans-Industrie-du-Futur-in-der-Botschaft, accessed: 05.06.2019, (2015).
• [9] Varela M.L.R., Putnik G.D., Manupati V.K., Rajyalakshmi G., Trojanowska J., Machado J., Collaborative manufacturing based on cloud, and on other I4.0 oriented principles and technologies: a systematic literature review and reflections, Management and Production Engineering Review, 9, 3, 90–99, 2018.
• [10] Rojko A., Industry 4.0 Concept: Background and Overview, Internat. J. of Interact. Mob. Technol. (iJTM), 111, 5, 77–90, 2017.
• [11] Reis S.M., Kenett R., Assessing the value of information of data-centric activities in the chemical processing industry 4.0, AIChE – Proc. Sys. Eng., 64, 11, 3868–3881, 2018.
• [12] Perzyk M., Kozłowski J., Methodology of Fault Diagnosis in Ductile Iron Melting Process, Arch. of Found. Eng., 16, 4, 101–108, 2016.
• [13] Vijayaram T.R., Sulajman S., Hamouda A.M.S., AhmadM.H.M.,Foundry quality control aspects and prospects to reduce scrap rework and rejection in metal casting manufacturing industries, J. of Mater. Process. Technol., Elsevier, 178, 1, 39–43, 2006.
• [14] Ignaszak Z., Sika R., Rogalewicz M., Contribution to the Assessment of the Data Acquisition Effectiveness in the Aspect of Gas Porosity Defects Prediction in Ductile Cast Iron Castings, Arch. of Found. Eng., 18, 1, 35–40, 2018.
• [15] Khan S., Finkelstein L., Mathematical modelling in the analysis and design of hard and soft measurement systems, Measurement, Elsevier, 46, 8, 2936– 2941, 2013.
• [16] Stekh Y., Lobur M., Shvarts M., Some methods for improving the accuracy of prediction recommendations, Bulletin of Lviv Polytechnic National University, Series: Computer Systems Design Theory and Practice, 882, 46–49, 2017.
• [17] Mahanta B.K., Chakraborti N., Evolutionary Data Driven Modeling and Multi Objective Optimization of Noisy Data Set in Blast Furnace Iron Making Process, Steel Research Int., Wiley, 89, 8, 1–11 (1800121), 2018, doi: 10.1002/srin.201800121.
• [18] Zdobytskyi A., Lobur M., Iwaniec M., Breznitskyi V., Optimization of the Structural Characteristics of the Robotic System Holder, 15th International Conference on the Experience of Designing and Application of CAD Systems (CADSM), Polyana (Svalyava), Ukraine, IEEE, 2019.
• [19] Grzegorzewski P., Kochański A., From Data to Reasoning, Part I – Theory in Soft Modeling in Industrial Manufacturing, Grzegorzewski P., Kochański A., Kacprzyk J. [Eds], Soft Modeling in Industrial Manufacturing, 2018.
• [20] Gweon H., Schonlau M., Steiner S.H., The k conditional nearest neighbor algorithm for classification and class probability estimation, Neural Computation, 17, 3, 731–40, 2019.
• [21] Sanodiya R., Saha S., Mathew K., A kernel semisupervised distance metric learning with relative dis tance: Integration with a MOO approach, Expert Systems with Applications, 125, 233–248, 2019.
• [22] Ryoo J., Arunachalam M., Khanna R., Kandemir M.T., Efficient K nearest Neighbor Algorithm Implementations for Throughput-Oriented Architectures, 19th Int’l Symposium on Quality Electronic Design, IEE, pp. 144–150, 2018.
• [23] Juan L., An Improved K-Nearest Neighbor Algorithm Using Tree Structure and Pruning Technology, Intell. Autom. and Soft Comp., 25, 1, 35–48, 2019.
• [24] Sika R., Ignaszak Z., Data Acquisition procedures for A&DM systems dedicated for foundry industry, Advances in Design, Simulation and Manufacturing II, pp. 692–701, 2019.
• [25] Comac E., Arslan A., A new training method for support vector machines: Clustering k-NN support vector machines, Expert Systems with Applications, Elsevier, 35, 564–568, 2018.
• [26] Balan K.P., Metallurgical Failures Analysis, Elsevier 2018.
• [27] Thacker K.B., Analysis of parameters for casting ductile iron pipe, International Journal of Engineering Research and General Science, 3, 1, 496–503, 2015.
• [28] Rogalewicz M., Sika R., Popielarski P., Wytyk G., Forecasting of steel consumption with use of nearest neighbors method, Modern Technol. Manufact. (MTeM 2017 – AMaTUC), 137, 01010-1–01010-6, 2017.
• [29] Henon G., Mascre C., Blanc G., Investigation of the cast products quality, International Committee of Foundry Technical Associations, France, 1986.
• [30] PN-85 H-83105, Casting Division and terminology of defects, Poland, 1985.
• [31] Baler J., Koppen M., Manual casting defects, IKOErbsloh, 2004.
• [32] All you need to know about sorting in Postgres, https://madusudanan.com/blog/all-you-need-toknow-about-sorting-in-postgres, accessed: 09.05.2019.
• [33] Hills M., Klint P., Vinju, J. J., Enabling PHP software engineering research in Rascal, Science of Comp. Programm, Elsevier, 134, 37–46, 2016.
• [34] Advanced features of PHP7, 2017, https://etrixtech.com/advanced-features-of-php-7, accessed: 09.05.2019.
• [35] Horwat W., JavaScript 2.0: Evolving a Language for Evolving Systems, 2005.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2020).