Page 168 - Data Science Algorithms in a Week

152 Djordje Cica and Davorin Kramar

Figure 1. Experimental setup.

In this research, three levels of diameter of the nozzle Dn, distance between the
impact point of the jet and the cutting edge s, pressure of the jet p, cutting speed vc, and
feed f, were used as the variables for cutting forces modeling (Table 1). Depth of cut was
fixed to 2 mm. With the cutting parameters defined and according to their levels, in total
27 experiments were realized as shown in Table 2.

Table 1. Design factors and their levels

Level
Machining parameters
1 2 3
Diameter of the nozzle Dn [mm] 0.25 0.3 0.4
Distance between the impact point of the jet and the cutting edge s [mm] 0 1.5 3
Pressure of the jet p [MPa] 50 90 130
Cutting speed vc [m/min] 46 57 74
Feed f [mm/rev] 0.2 0.224 0.25

ARTIFICIAL NEURAL NETWORKS BASED MODELING

Artificial Neural Networks Trained by Backpropagation Algorithm

In recent years, ANN have attracted attention of many researchers as an effective
modeling tool for a wide range of linear or nonlinear engineering problems that cannot be
solved using conventional methods. An ANN is comprised of a series of information