Al Ain Men’s campus (AAM) Mechanical Engineering faculty, Dr. Hussien AlWedyan, has added another piece to his academic portfolio with the publication of his latest paper in the Carl Hanser Verlag, Muchen Materials Testing Journal which is a well-recognized industrial Journal in Germany.
The publication of “Prediction and controlling of roundness during the BTA deep hole drilling process: Experimental investigations and fuzzy modeling” brings the total of his publications to 22. The Journal is the only German-English language journal dealing with all aspects of material and component testing in industrial application, in test laboratories and research. The magazine provides first-hand information on non-destructive, destructive, optical, physical and chemical test procedures. It contains peer-reviewed exclusive articles by experts with international reputations.
In this collaborative research work, Dr. AlWedyan designed first order Sugeno-fuzzy models by using the cutting parameters as input data and the roundness errors as output data. The relation between the input and the output is created to find the influence of the input parameters on the output surface quality in terms of roundness errors. Hence, the best cutting condition in deep hole drilling is designated to improve the output. A scheme is recommended to precisely create the relationship between the different cutting parameters using subtractive clustering procedure based on the first order Sugeno fuzzy model. Minimum error model with lesser numbers of rules for roundness error is established by enumerative exploration of the clustering parameters. The resulted model with best clustering factors is then attuned by using adaptive neuro-fuzzy inference system (ANFIS).
The fuzzy model expected a roundness error of 0.998 μm if a work piece rotational frequency of 21.1 Hz, a feed rate of 0.0944 mm × rev-1 and a tool usage equal to 3.08 were used. The same cutting parameters were used in the experimental validation and the roundness result was equal to 3.59 μm.
The difference in numbers is not that important. The most important thing is that with this complex, vague and unpredictable process, fuzzy modeling was able to tackle this ambiguity and was able to provide us with the best cutting parameters to get the minimum roundness error of the workpiece. For workpiece frequency equal to 23.3 Hz, feed rate equal to 0.100 mm × rev-1 and tool usage equal to 1, the roundness resulting from the experiment was 9.049 μm, with the same parameters, the model forecasted the roundness between 5 and 7 μm. Finally, combining ANFIS with subtractive clustering techniques achieved excellent results.