Page 167 - Data Science Algorithms in a Week

The Estimation of Cutting Forces in the Turning of Inconel 718 Assisted … 151

quantity lubrication). Furthermore, the effect of cutting parameters such as depth of cut,
feed and cutting speed on machining variables are also studied.
However, despite the fact that there are numerous applications of ANN in modeling
of the cutting forces reported in the literature, a review of the literature shows that no
work is reported by modeling these parameters under HPC conditions. This can be
explained by complex relations between large numbers of HPC process parameters, such
are pressure of the jet, diameter of the nozzle, cutting speed, feed, etc. that influence the
cutting forces and make it difficult to develop a proper estimation model. In this sense,
this paper presents ANN models for estimation of cutting forces in turning of Inconel 718
under HPC conditions. First, cutting forces were modeled by using conventional ANN
which uses backpropagation algorithm in its learning. In order to overcome the
limitations of traditional backpropagation algorithm, two bio-inspired computational
techniques, namely genetic algorithm (GA) and particle swarm optimization (PSO) were
also used as a training methods of ANN. The capacity modeling of ANN by using GA
and PSO has been compared to that of the conventional ANN.

EXPERIMENTAL DETAILS

The experiments were performed on machining nickel-based alloy Inconel 718
supplied as bars (145 mm diameter and 300 mm long) with hardness between 36 and 38
HRC. Machining experiments have been carried out on a conventional lathe, fitted with a
high-pressure plunger pump of 150 MPa pressure and 8 l/min capacity. Standard sapphire
orifices of 0.25, 0.3 and 0.4 mm diameter, commonly used in water jet cutting
applications, were set in a custom-made clamping device that enabled accurate jet
adjustments. The cooling lubricant jet was directed normal to the cutting edge at a low
angle (about 5-6º) with the tool rake face. The nozzle was located 22 mm away from the
tool tip in order to assure its use in the core zone of the jet and avoid variations in the
diameter of the jet and radial distribution of the pressure. The cutting tool inserts used in
the experiments were coated carbide cutting tools – SANDVIK SNMG 120408-23 with
TiAlN coating. Tool was mounted on a PSBNR 2020 K12 tool holder resulting in
positive rake angle (γ = 7º).
The cutting force components (main cutting force Fc, feed force Ff and passive force
Fp) were measured with a three-component dynamometer (Kistler 9259A). The
dynamometer was rigidly mounted on the lathe via a custom designed adapter for the tool
holder so that cutting forces could be accurately measured. Force signals obtained from
the dynamometer were amplified and then transferred to computer. The measurement
chain also included a charge amplifier (Kistler 5001), a data acquisition hardware and a
graphical programming environment for data analysis and visualization. The whole
measurement chain was statically calibrated. Experimental setup is shown on Figure 1.