Selective characteristics of the applied methods and approaches are presented in the first part of the paper. Besides T-S fuzzy logic systems, the Mamdani fuzzy logic reasoning [1] and feed forward back propagation (FFBP) neural networks are discussed. Since the general basis of the considered methods is known, the discussion focuses on methods for their automatic design. Such an approach reflects the main goal of the research conducted, i.e. the research aimed at revealing the capability of FL<T-S> systems to generalise the knowledge provided in the form of measured data.

In the next part of the paper several tests are depicted. Both, the classification task as well as the approximation task were considered. The Takagi-Sugeno systems were applied for tool wear diagnostics. Here, a problem of classifying drill wear during multi-spindle drilling was analysed. Next, an example of thermal deformation diagnostics of a grinding machine was studied. The FL<T-S> fuzzy logic system was used to model the relationship between temperatures measured in selected points of the machine and the undesirable changes in the workpiece diameter. Finally, Takagi-Sugeno systems were applied for positioning error detection in the high speed feed axis. In this case the measured temperatures and the position of the table determined in the computer numerical controlled (CNC) machine were the inputs to the systems. The FL<T-S> systems were to estimate the error of positioning, i.e. the difference between the real position and the position determined in the CNC. As it may be noticed, knowing such an error allows the position in order to achieve desired accuracy of machining to be easily corrected.

It is important to recall that the results of tests of Takagi-Sugeno fuzzy logic systems were compared to the results achieved with, above mentioned, FFBP neural networks and Mamdani fuzzy logic systems. Such an approach allowed the emphasis of the advantages and disadvantages of all methods considered and the wide assessment of the applicability of Takagi-Sugeno fuzzy logic systems for machine tool diagnostics.

1

E.H. Mamdani, "Advances in the linguistic synthesis of fuzzy controllers", Int. Journal of Man-Machine Studies, 8, 1976.

2

A. Sokolowski, "On some aspects of fuzzy logic application in machine monitoring and diagnostics", Journal of Engineering Applications of Artificial Intelligence, 17, 429-437, 2004. doi:10.1016/j.engappai.2004.04.016

3

T. Takagi, M. Sugeno, "Fuzzy identification of systems and its applications to modelling and control", IEEE Trans. on Systems, Man and Cybernetics, 15, 1985.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £78 +P&P)