Evaluation of neural models to estimate the roughness of advanced ceramics in surface grinding
by Mauricio Eiji Nakai; Hildo Guillardi Júnior; Paulo R. Aguiar; Eduardo Carlos Bianchi; Paulo Sérgio Da Silva
International Journal of Machining and Machinability of Materials (IJMMM), Vol. 17, No. 5, 2015

Abstract: There is an increasing trend for ceramic components to replace metal ones due to their excellent physical, chemical and mechanical properties. However, many of the characteristics that make ceramics so attractive also make them difficult to manufacture by traditional machining methods. The purpose of this study was to develop neural models based on acoustic emission and cutting power signals to estimate the roughness of advanced ceramics during the grinding process. Testing of alumina ceramic specimens was performed on a tangential surface grinder with a diamond wheel. The tests were performed using three cutting depths, 120 µm, 70 µm and 20 µm, a grinding wheel speed of 35 m/s and table speed of 2.3 m/s. Four neural models were studied: multilayer perceptron neural networks, radial basis function neural networks, general regression neural networks and adaptive neuro-fuzzy inference system. To better compare the performance of the neural models used in this study, an algorithm was developed to train all the possible combinations of inputs and parameters of each type of neural network. The results of the best models produced very low error values within the range of accuracy of the measuring instrument. Thus, it can be stated that these models achieved 100% accuracy in estimating workpiece roughness.

Online publication date: Wed, 25-Nov-2015

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Machining and Machinability of Materials (IJMMM):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com