dc.description.abstract |
Plasma arc cutting is a non-traditional thermal cutting technology utilized in modern
manufacturing industries to create a variety of products at a lower cost. It can cut a variety of
electrically conductive materials. In Amhara Metal and Machine Industry technology and
Development Enterprise cutting of plate with CNC plasma arc cutting machine is difficult, due to
the combined effect of a continuous failure of the nozzle and getting low quality, this quality
phenomenon included high surface roughness and low material removal rate. This study aimed to
optimize process parameters like cutting current, cutting speed, gas pressure, and standoff
distance of a Plasma arc cutting machine that affect the material removal rate, surface roughness,
and nozzle diameter change after cutting performed. These process parameters have an influence
on the quality of the machined part and the effect of these process parameters on material
removal rate, surface roughness, and nozzle diameter change after cutting using a 60×30×10 mm
mild steel was investigated. The experiment was designed using Taguchi's L27 orthogonal array
to reduce the number of experimental runs, and it was carried out using an EDON CUT 100
CNC plasma arc cutting machine from Amhara Metal and Machine Industry Technology and
Development Enterprise. The parameters of the plasma arc cutting process were optimized
utilizing a hybrid Artificial Neural Network and Genetic Algorithm. For optimal weight, an
ANN model was created and trained with GA based on the input and output result values. This
objective function was further optimized by GA for obtaining an optimum result. From the
analysis of the result plasma arc cutting parameters 73.9764A cutting current, 444.325 mm/min
cutting speed, 0.7998 MPa gas pressure, and 3.1185 mm standoff distance with their response
value of 7.0032 g/sec material removal rate, 4.2062 µm surface roughness, and 1.3142 mm
nozzle diameter were obtained as an optimum value. Finally, a confirmation test was carried out
by cutting three specimens with the optimal process parameters. The measured responses
revealed that the 6.9280g/sec material removal rate, 4.3499 m surface roughness, and 1.3702 mm
nozzle diameter had an error of 0.05 to 0.14 from the predicted optimal value, and that the
responses with this optimal value were better than the previous cutting combination of the
enterprise.
Keywords: Plasma arc cutting machine, optimization of plasma arc cutting parameters, Artificial
Neural Network, Genetic Algorithm |
en_US |