An application of artificial neural network and particle swarm optimisation technique for modelling and optimisation of centreless grinding process
by Prosun Mandal; Subhas Chandra Mondal
International Journal of Productivity and Quality Management (IJPQM), Vol. 20, No. 3, 2017

Abstract: Centreless grinding operation is widely used in manufacturing industry for its high level of accuracy and micro-finishing of shaft, pin material. This paper presents an application of artificial neural network and particle swarm optimisation techniques for modelling and optimisation of centreless grinding operation for machining of C40 steel crane-hook-pin. Full factorial design is used taking three factors at three levels each and a total 33 or 27 number of experiments are done in all possible combination of factors. The performance of this particular trained neural network has been tested with the experimental data and found satisfactory. Thus the proposed ANN model is efficiently used for predicting surface roughness in centreless grinding operation. Particle swarm optimisation technique combining with response surface modelling are used to find optimal parameter settings. The optimal parameter corresponds to 12 rpm regulating wheel speed, 1/3rd opening of coolant valve opening and 20.002 µm depth of cut.

Online publication date: Sat, 04-Mar-2017

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 Productivity and Quality Management (IJPQM):
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