Neuro-fuzzy based predictive model for cutting force in CNC turning process of Al\u2013Si\u2013Cu cast alloy using modifier elements

Mohsen Marani,Chris K Mechefske,Mohammadjavad Zeinali,Saeed Farahany

doi:10.1007/s42452-020-03980-9

Mohsen Marani, Chris K Mechefske + Show 2 more

Open Access

https://doi.org/10.1007/s42452-020-03980-9

Copy DOI

Abstract

This study investigated the predicted cutting force model of a turning operation for Al–Si–Cu cast alloy modified with modifiers based on adaptive neuro-fuzzy inference system (ANFIS) approach. Feed rate, cutting speed and Silicon spacing were considered as the input parameters. A series of turning experiments were conducted at various feed rates and cutting speeds. The prediction result showed that the ANFIS model successfully predicted the cutting force value in terms of cutting speed, feed rate and Si spacing. A mathematical model was proposed to describe the cutting force changes during the machining of Al–Si–Cu cast alloy. Moreover, the addition of Bismuth into the base alloy decreased the cutting force compared to other refinement elements.

Highlights

In recent years, manufacturing industries have been seeking alloys with excellent machinability capability and suitable mechanical properties for general engineering industries
Among all neuro-fuzzy approaches, the adaptive neuro-fuzzy inference system (ANFIS) model with two Sigmoidal membership functions (MFs) for three inputs showed the best accuracy in terms of root mean square error (RMSE) and regression values
A mathematical model was proposed to describe the relationship between inputs, and the output

Summary

Introduction

In recent years, manufacturing industries have been seeking alloys with excellent machinability capability and suitable mechanical properties for general engineering industries. A material’s machinability determines its value in a range of applications [1]. Cutting force is one of the main machining parameters which play an important role during the machining process. Cutting force needs to be better understood and its measurement should be accurate to improve machinability of alloys [2, 3]. Aluminum-silicon alloys are extensively used in automobile, aerospace, and general engineering industries due to the low expansion coefficient mechanical properties, weldability, and machinability [4,5,6]. There is still a need to improve morphology of the silicon phase in the alloy

Methods

Results

Conclusion