Combination of Machine-Learning Algorithms for Fault Prediction in High-Precision Foundries

Foundry is one of the activities that has contributed to evolve the society, however, the manufacturing process is carried out in the same manner as it was many years ago. Therefore, several defects may appear in castings when the production process is already finished. One of the most difficult defect to detect is the microshrinkage: tiny porosities that appear inside the casting. Another important aspect that foundries have to control are the attributes that measure the faculty of the casting to withstand several loads and tensions, also called mechanical properties. Both cases need specialised staff and expensive machines to test the castings and, in the second one, also, destructive inspections that render the casting invalid. The solution is to model the foundry process to apply machine learning techniques to foresee what is the state of the casting before its production. In this paper we extend our previous research and we propose a general method to foresee all the defects via building a meta-classifier combining different methods and without the need for selecting the best algorithm for each defect or available data. Finally, we compare the obtained results showing that the new approach allows us to obtain better results, in terms of accuracy and error rates, for foretelling microshrinkages and the value of mechanical properties.

[1]  Yoseba K. Penya,et al.  Enhanced Foundry Production Control , 2010, DEXA.

[2]  Matt Brown,et al.  Invited talk , 2007 .

[3]  Junyan Yang,et al.  Intelligent fault diagnosis of rolling element bearing based on SVMs and fractal dimension , 2007 .

[4]  P. Tuyls.  Invited Talk 2 , 2010 .

[5]  Igor Santos,et al.  Enhancing fault prediction on automatic foundry processes , 2010, 2010 World Automation Congress.

[6]  Wenfei Fan,et al.  Keys with Upward Wildcards for XML , 2001, DEXA.

[7]  Arvinder Kaur,et al.  Comparative analysis of regression and machine learning methods for predicting fault proneness models , 2009, Int. J. Comput. Appl. Technol..

[8]  J. Fernández-Carrasquilla,et al.  Estudio de una fundicin nodular mediante mecnica de la fractura , 1999 .

[9]  Thomas G. Dietterich.  Multiple Classifier Systems , 2000, Lecture Notes in Computer Science.

[10]  David J. C. MacKay,et al.  Neural network model of creep strength of austenitic stainless steels , 2002 .

[11]  Steven R Schmid Kalpakjian,et al.  Manufacturing Engineering and Technology , 1991 .

[12]  Paulo J. G. Lisboa,et al.  The Use of Artificial Neural Networks in Decision Support in Cancer: a Systematic Review , 2005 .

[13]  Yoseba K. Penya,et al.  N-grams-based File Signatures for Malware Detection , 2009, ICEIS.

[14]  J. P. Oria,et al.  Ultrasonic sensing classification of foundry pieces applying neural networks , 1998, AMC'98 - Coimbra. 1998 5th International Workshop on Advanced Motion Control. Proceedings (Cat. No.98TH8354).

[15]  Yoseba K. Penya,et al.  Towards noise and error reduction on foundry data gathering processes , 2010, 2010 IEEE International Symposium on Industrial Electronics.

[16]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[17]  Ludmila I. Kuncheva,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2004 .

[18]  R. Gonzaga-Cinco,et al.  Dependencia de las propiedades mecánicas y de la composición química en la fundición de grafito esferoidal , 2006 .

[19]  Subhash C. Bagui,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2005, Technometrics.

[20]  Stephen R. Garner,et al.  WEKA: The Waikato Environment for Knowledge Analysis , 1996 .

[21]  Yoseba K. Penya,et al.  Optimising Machine-Learning-Based Fault Prediction in Foundry Production , 2009, IWANN.

[22]  Johannes Fürnkranz,et al.  An Evaluation of Grading Classifiers , 2001, IDA.

[23]  Jiri Matas,et al.  On Combining Classifiers , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[24]  R. Gonzaga-Cinco,et al.  Mecanical properties dependency on chemical composition of spheroidal graphite cast iron , 2006 .

[25]  Yoseba K. Penya,et al.  Machine-learning-based mechanical properties prediction in foundry production , 2009, 2009 ICCAS-SICE.

[26]  David H. Wolpert,et al.  Stacked generalization , 1992, Neural Networks.

[27]  J. Sertucha,et al.  Influencia de las condiciones de moldeo y las características de los moldes sobre la formación de defectos de contracción en piezas de fundición esferoidal , 2007 .

[28]  Pei Zhang,et al.  Optimizing Casting Parameters of Ingot Based on Neural Network and Genetic Algorithm , 2008, 2008 Fourth International Conference on Natural Computation.

[29]  C W Lung,et al.  Mechanical properties of metals , 1999 .

[30]  Tom Fawcett,et al.  Analysis and Visualization of Classifier Performance: Comparison under Imprecise Class and Cost Distributions , 1997, KDD.

[31]  Juan M. Corchado,et al.  Distributed Computing, Artificial Intelligence, Bioinformatics, Soft Computing, and Ambient Assisted Living, 10th International Work-Conference on Artificial Neural Networks, IWANN 2009 Workshops, Salamanca, Spain, June 10-12, 2009. Proceedings, Part II , 2009, IWANN.

[32]  David J. Hand,et al.  Advances in intelligent data analysis , 2000 .

[33]  Yoseba K. Penya,et al.  Mechanical properties prediction in high-precision foundry production , 2009, 2009 7th IEEE International Conference on Industrial Informatics.

[34]  M Perzyk,et al.  Detection of causes of casting defects assisted by artificial neural networks , 2003 .