Optimization of process parameters in plasma arc cutting of commercial-grade aluminium plate

: Plasma arc cutting (PAC) has emerged as a versatile and e ﬃ cient method for the precision cutting of various materials, including commercial-grade aluminium plates. The optimization of process parameters is crucial for achieving high-quality cuts, minimizing material wastage, and enhancing overall productivity. This study aims to systematically investigate and optimize the key process parameters in PAC of commercial-grade aluminium plates. The experimental design involves the manipulation of parameters such as arc current, gas pressure, and work-piece thickness. A Design of Experiments approach, speci ﬁ - cally Taguchi ’ s orthogonal array, is employed to e ﬃ ciently explore the parameter space and identify the optimal combination of settings. The response variables considered for optimization include minimum surface roughness, minimum burr height, and maximum material removal rate (MRR). Analysis of variance is performed to get the percentage in ﬂ uence of each process parameter on the performance characteristic. The results obtained from the optimization process are expected to provide valuable insights into enhancing the e ﬃ ciency and precision of PAC for commercial-grade aluminium plates. Arc current is found to be the most signi ﬁ cant parameter in altering the surface roughness. The thickness of the material is the most signi ﬁ cant parameter in altering burr height. None of the parameters is found to be signi ﬁ cant in altering the MRR from Analysis of Variance analysis. From signal-to-noise ratio analysis and average performance graph, the optimum combination of processes in altering the bur height and MRR are found as arc current at 50 amp, the gas pressure at 5.4 bar, and the thickness of the workpiece at 6 mm.


Introduction
The promptitude and accelerated output in the manufacturing, machinery, and material engineering industries greatly depend upon the selection of cutting tools used in the relevance of economy and time consumption.The rapidity, mass manufacturing, and cost optimization are the major factors linking the growth and business in these industrial sectors.There are primarily three different types of thermal cutting techniques namely gas-cutting, laser cutting, and plasma arc cutting (PAC) techniques applied for cutting materials in different industrial usages like shipbuilding, railway wagons, manufacturing industries, material engineering, machine fabrication, etc. Considering different parameters like minimal slag, environmental factors, economy, safety, system complexity, maintenance, etc., PAC is the most advantageous [1].
PAC is one of the highly acclaimed cutting processes that use thermal energy.It has gained huge applications in cutting stainless steel, manganese steel, titanium alloys, copper, alloy materials, metals with high density and high melting points, and other difficult-to-machine alloys [2,3].The expense involved in the PAC method is two to three times less and efficiency is about 2.5 times higher than other oxy-fuel processes [4,5].PAC can also be used to cut complex shapes with accuracy and precision [4].PAC is commonly used in metal fabrication shops and manufacturing plants to cut and shape metals such as steel, aluminium, and stainless steel, automotive industry for cutting and shaping metal parts during the fabrication of vehicle components.Shipyards use PAC for the fabrication of ship components, including steel plates and sections used in the construction of hulls, decks, and other structural elements.Plasma cutting is employed in construction for cutting metal components used in buildings, bridges, and infrastructure projects.PAC plays a role in the metal recycling industry by efficiently cutting and separating metal components from larger structures.The PAC process is characterized by a plasma arc carrying a high current established between a tungsten electrode and the conductive metal piece.The electrode is connected to the negative terminal and the metal piece is connected to the positive terminal of the power supply.An arc is established between the constricted electrode nozzle and the workpiece.The plasma arc produced has a very high temperature, and high energy density sufficient to melt the workpiece and has a high velocity bearing the features of a plasma jet for removal of the molten material.In this process, an electric arc is passed through an inert gas blown at high speed out of a nozzle.During the processes, some of the inert gas converts to plasma.When this plasma falls on the surface of the metal to be cut, it instantly moves by melting and carrying away the molten metal from the cutting region [6].However, the quality of the cut varies depending on the arc current, the geometry of the torch, the type of inert gas, the gas flow rate, and the cutting speed [7].
It is feasible to find good number of research works to study the influence of process parameters of PAC and to obtain an optimum cutting condition.Due to high-temperature operations, the process of cutting by PAC on soft material like aluminium is treated as a challenging one.Improvement of the process depends on several quality characteristics namely, material removal rate (MRR), surface roughness, cutting width (kerf), kerf taper angle (conicity), edge roughness, heat-affected zone, burr formation [8], flatness, clean cut bevel angle [9], unevenness of cut surface, etc. [5].Despite a lot of experimental and theoretical works, there are still grey areas of the PAC process that need a better understanding of process mechanics and physics for improving its quality of performance.Researchers have tried to increase the energy density generated by the plasma arc system so that a higher cutting thickness can be achieved maintaining the quality of the cut.For that purpose, enormous investigations have been carried out on the PAC process by considering the process parameters.Cutting speed, cutting current, cutting height or standoff (the gap between the plasma torch and workpiece), and nature of the cut, cutting pressure, and flow of the plasma gas are obtained to be the most dominating parameters during the cutting of various materials [10,11].
Traditional designs of the experiment (DOE) approaches demand a greater quantity of experimentation when the process parameters and their consequence levels increase.The increased number of experimentations again demands time and resources.The DOE proposed by Taguchi is a powerful tool as it demands a lesser number of experimentations for the same process parameters and levels increase.Specially designed orthogonal arrays (OAs) are used in the Taguchi method to investigate the complete process parameters [12].This method allows a balanced comparison of levels of any process parameters by fractional factorial matrix.Each process parameters are to be assessed separately from all other parameters in Taguchi's DOE.This powerful tool for experiment design can also be used to optimize the cutting parameters of a machining process to evaluate the best possible combination for getting a better surface finish, a higher tool life, etc., and to identify the most effective parameters [13][14][15][16][17]. Several scientists have operated Taguchi's OA method for the optimization of process parameters in PAC operations.Charmarthi et al. [5] have evaluated the unevenness of Hardox-400 material plate of 12 mm thickness machined by the PAC process.The process parameters have been optimized by the authors using DOE to study the surface profile of the cut and constancy for the workpiece material.Das et al. [7] have conducted parametric optimization of process parameters for PAC of EN31 steel by combining the Taguchi method and grey relation analysis.Asiabanpour et al. [18] have used the design of experiments (DOE) to optimize the automated plasma-cutting process on a 0.25-inch-thick stainless steel sheet.The authors have used a response surface methodology approach followed by regression analysis to study the significant factors that influence the surface quality characteristics of the workpiece.To combine models obtained for each response and to balance the trade-offs between responses, the desirability function, and multi-response optimisation technique has been used.Bini et al. [19] have used a high tolerance plasma arc cutting system to cut mild steel of 15 mm thickness.The authors have performed response surface methodology and analysis of variance (ANOVA) to study the process parameters.The arc voltage was found to be the most significant parameter influencing the cutting quality.Chen et al. [20] have used Six Sigma and lean paradigms based on the Taguchi optimization technique in altering the roundness of holes made by the PAC process.For four controllable factors with three corresponding levels, L 9 Taguchi OA was used for the conduction of experiments.The optimized setting obtained by the authors was as follows: the smaller tip of the torch, medium range of feed rate, lower range of voltage, and higher range of current.Bhuvenesh et al. [21] have optimised the process parameters of a manual plasma arc cutter (Selco Genesis 90) to cut standard AISI 1017 Steel of 6 mm thickness.The authors have found an inverse relationship between surface roughness and material removal rate.Ismail and Taha [22] have used Taguchi's technique to optimise the machining process parameters of PAC during the cutting of two steel grades such as ASSAB618 and ASSABDF3.Arc current, scanning velocity, and carbon content of steel were selected as the process parameters.Arc current was found to be the most significant parameter influencing the surface hardness of the cutting surface.Many researchers have studied the machining and welding characteristics of Aluminium, Aluminium alloy, and Aluminium composites [23][24][25][26][27]. Adin [28] has studied the mechanical properties of Aluminium Alloys welded by laser welding and optimized the welding parameters using Taguchi and ANOVA methods.Adin [29,30] has investigated the Machining aerospace aluminium alloy with cryo-treated and untreated HSS cutting tools.
The above discussion reveals that although the plasma arc process has been used to study the cutting behaviour of various ferrous materials like steel, a negligible study has been found to study non-ferrous and soft materials like aluminium.The present article discusses the optimization of the machining process parameters in PAC of a commercial aluminium plate of grade 1050 applying Taguchi's OA technique.For optimization purposes, two distinct levels were employed for each parameter, including Arc current (C), Gas pressure (P), and Thickness of the workpiece (T).Taguchi's L8 experimental design has been used to derive an optimal configuration of PAC process parameters, aiming to achieve the best possible values for surface roughness, burr height, and Material Removal Rate (MRR).

Experimental investigation
All experiments have been performed on a manual air PAC system (Model-Powermax 105; Make -Hypertherm, Inc.).A hand torch (Torch -Duramax TM series 75°) has been used for cutting purpose hand torch.The details of the manual air plasma arc-cutting system are shown in Figure 1.In the figure, zone A is composed of channels through which secondary gas passes for cooling the torch.This secondary gas also helps in blowing out the molten metal from the cutting zone to create a fast and slag-free cut.Cool gas enters zone B, and a pilot arc is established between the electrode and the torch tip.This pilot arc heats and ionises the gas.The main cutting arc then transfers to the workpiece in zone C through the column of plasma gas.By forcing the plasma gas and electric arc through a small orifice, the torch delivers a high concentration of heat to a small area.
Commercial aluminium plates of grade 1050 are used as work material for the present study.Specimens of dimensions 152 mm × 50 mm × 6 mm and 152 mm × 40 mm × 10 mm are used for plasma cutting along the length of specimens.Material has been chosen based on the availability and accessibility of materials can be a practical consideration.Aluminium slabs are readily available or can be easily sourced within the constraints of the research environment, budget, and timeline.The cuts are performed in the middle of the workpiece as shown in Figure 2. The work clamp must be connected to the workpiece while cutting.The workpiece and the work clam must have good metal-tometal contact.To measure the surface roughness, the five highest peaks and five highest valleys of the surface texture (ten-point height method) within a sample length of 52 mm of the workpiece are considered.Peaks and valleys are measured with the help of the profile projector (Make -Banbros) from the datum parallel to the mean line of the workpiece.The uncut surface of the workpiece is used as a datum line for the present study.In the present study burr height is measured using Tool Maker's Microscope (Make -Banbros) of least count 0.01 mm.The experimental setup is shown in Figure 3.
To measure the surface roughness, the five highest peaks and five highest valleys of the surface texture (tenpoint height method) within a sample length of 52 mm of the workpiece are considered.Peaks and valleys are measured with the help of the profile projector (Make -Banbros) from the datum parallel to the mean line of the workpiece.The experimental setup for measuring surface roughness has been shown in Figure 3.The uncut surface of the workpiece is used as a datum line for the present study.In the present study, burr height is measured using Tool Maker's Microscope (Make -Banbros) of least count 0.01 mm as shown in Figure 4.
Material Removal Rate (MRR) is a crucial parameter that quantifies the volume of material removed per unit of time during a machining or material processing operation.The material removal is determined by weighing the workpiece before and after the machining process.The weight difference, combined with the material density, is used to calculate the volume removed.MRR has been assessed by measuring the reduction in weight of the Aluminium workpiece (expressed in g•s −1 ).The time required for cutting has been measured by a stopwatch.The weight of each  workpiece has been measured in electronics balance (Make -Digitron) before and after cutting.The difference between the two measurements gives the material removal rate.The parameters, viz., arc current (C), gas pressure (P), and thickness of workpiece (T) have been considered in the present study for optimisation of the machining process parameters.Two levels of each parameter have been taken into consideration.Table 1 shows the process parameters with their levels.In the present work, the responses like surface roughness, burr height, and MRR have been considered for analysis.The significance of the parameter has been studied using ANOVA, and the optimal setting of the parameters has been found by using Signal-to-Noise (S/N) ratio and average performance graph.An illustration of the experimental setup is provided in the flowchart presented in Figure 5.  Optimizing the plasma arc welding process parameters  5 3 Results and discussion

OAs and experimental design
As per Taguchi's experimental technique, a set of two levels is allotted to each process parameter.Therefore, each parameter has one degree of freedom (DOF) (level 1), and the interaction between two parameters namely, C × P, C × T, and P × T has one DOF.The DOFs are shown in Table 1.This gives a total of 7 DOF for three process parameters.So, Taguchi's OA L 8 has been designed for the present work.Using more than two levels for each parameter was not practical or feasible due to experimental constraints and the nature of the factors being studied.Two-level designs strike a balance between simplicity and informativeness.Before conducting the main experiment, the authors performed preliminary studies or pilot experiments.These small-scale trials helped in identifying influential parameters and refining the experimental design.The results of these preliminary studies have been taken as a reference for the selection of parameters for the main experiment.Taguchi's layout for OA 8 has been selected for the present analysis, and result obtained from the experimentation is shown in Table 2.

ANOVA
The intention of performing ANOVA is to investigate design parameters that significantly affect the quality characteristic.In the present study, ANOVA is carried out (shown in Table 3) for responses, namely, surface roughness, burr height, and MRR.The calculated F-ratio of the respective factors should be greater than 162.45 according to a 95% confidence level.The significant values are represented in bold as given in Table 3. From the ANOVA table, Arc current (C), Gas pressure (P), and Thickness of the workpiece (T) were found to be significant in altering surface roughness.Arc current (C) was observed to be the most significant parameter as compared to the other parameters namely, gas pressure (P) and thickness of the workpiece (T) for the parametric levels and confidence level considered in this study.Arc current (C) and thickness of the workpiece (T) was found to be significant, for the parametric levels and confidence levels considered in this study in altering burr height.The thickness of the workpiece (T) was observed to be the most significant parameter as compared to the other parameters namely, arc current (C) and gas pressure (P).No cutting parameter is found to be significant at a 95% confidence level in altering MRR.

Analysis of the S/N ratio
The S/N ratio is analysed for determining the performance characteristics of the parameters of the experiment.The S/N ratio is calculated for each trial number, and then, the average S/N ratio is determined for each  parameter at their respective level.The largest value of the S/N ratio gives the optimal settings of parameters at their respective level.Basically, there are three types of S/N ratios, namely, larger the better, smaller the better, and nominal the best.This analysis aims to explore the optimal parametric combination for minimum surface roughness, minimum burr height, and maximum MRR.Hence, in the present study, the smaller the better characteristics are used for the calculation of S/N ratios for surface roughness and burr height, and the larger the better characteristics are used for the calculation of S/N ratios for MRR.The S/N ratio is given by equation ( 1) The S/N ratios are calculated using the smaller the better characteristics.The MSD (mean square deviation) is given by equation ( 2   Optimizing the plasma arc welding process parameters  7 The calculated S/N ratios are shown in Table 4.The larger values of S/N ratios of the parameters at each level are shown in bold text in the tables.

Average performance graph
The average performance graph serves as a visual tool to assess the dominance of each parameter at different levels within the experimental design.The average performance graph is obtained by plotting the average values of all the parameters shown in Table 2.For this, the average values of all the responses of the parameters are considered level-wise.The average performance graph is obtained by plotting the average response values.
In the present study, the optimal settings for surface roughness and burr height will be given by the minimum values of the responses of the parameters level-wise, so the line connecting the minimum response values of all the parameters in the average performance graph will give the optimal settings.The optimal settings for MRR will be given by the maximum values of the responses of the parameters level-wise, so the line connecting the maximum response values of all the parameters in the average performance graph will give the optimal settings.Figure 6(a)-(c) shows the average performance graphs for PAC for responses, namely, surface roughness, burr height, and MRR.The S/N ratio (Table 4) the average performance graphs are shown in Figure 6(a)-(c), and the optimal settings of the influencing parameters in PAC are shown in Table 5.

Test of interaction
"Interaction" describes a condition in which the control of one factor upon the output responses is dependent on the  condition of another.To determine the presence of interaction between the parameters, interaction plots were made.The interaction between the Arc current (C) and gas pressure (P) (C × P), arc current (C) and thickness of the workpiece (T) (C × T), gas pressure (P) and thickness of the workpiece (T) (P × T) in altering surface roughness, burr height, and MRR are shown graphically in Figure 7(a)-(c).
No strong interaction between the parameters is found in altering surface roughness and MRR.An intense interaction between gas pressure (P) and the thickness of the workpiece (T) is found in altering burr height.
Confirmation tests are done to determine the optimum setting of the parameter using equations ( 3) and ( 4).Optimum performance without considering interaction effect (Y opt1 ) and Optimum performance with considering interaction effect (Y opt2 ) are calculated as follows: 1.Without considering interaction effect (P × T), i.e. for (C 2 , P 2 , T 1 ) from the data obtained from Table 1 (3) where Y opt1 is the optimum performance without considering the interaction effect, T̅ is the average result of eight trials = 3.211, C̅ 2 is the average result of C 2 = 2.073, P̅ 2 is the average result of P 2 = 3.148, and T̅ 1 is the average result of T 1 = 3.211.2. Considering interaction effect (P × T), i.e. for (C 2 , P 1 , T 1 ), from the data obtained from Table 1 where Y opt2 is the optimum performance considering interaction effect, and P T 1 1 is the average result of P 1 , T 1 = 1.755.
It is obtained that Y opt2 < Y opt1 , i.e. the optimum per- formance calculated without considering the interaction (P × T), is more than the optimum performance considering interaction (P × T).Hence, the optimum cutting condition considered is C 2 (50 amp), P 2 (5.4 bar), and T 1 (6 mm) in altering burr height which is like the result obtained from Table 5. Optimizing the plasma arc welding process parameters  9

Response vs Treatment condition plot
The response parameters are plotted against the treatment condition shown in Figure 8(a)-(c).It is observed that treatment condition 7 gives a minimum value of burr height and a maximum value of MRR.The level of the cutting parameters obtained for treatment 7 from Table 5 is C 2 = 50 Amp, P 2 = 5.4 bar, and T 1 = 6 mm in Figure 6.From Figure 8(a)-(c), it is observed that treatment condition 7 gives the minimum surface roughness.The level of the cutting parameters obtained for treatment 7 (from Table 5 is C 2 = 50 Amp, P 2 = 5.4 bar, and T 1 = 6 mm).But the optimum value obtained from the average performance graph and S/N ratio analyses is for the treatment condition 5 (C 2 = 50 Amp, P 1 = 4.3 bar, T 1 = 6 mm).This is because the S/N ratio represents the magnitude of the mean process compared to its variation [8].

Conclusions
The present work represents a parametric optimisation of PAC operation for different responses namely surface roughness, burr height, and MRR in commercial-grade aluminium plates (grade1050).Experimental analyses of the PAC operations in altering surface roughness, burr height, and MRR for the parametric levels and confidence levels considered in this study can be summarised as follows: • From ANOVA analysis, arc current is found to be the most significant parameter in altering surface roughness.The thickness of the workpiece is determined to be the most significant parameter in altering the burr height.However, none of the parameters is found to be significant in altering the MRR from ANOVA analysis.• The optimum combination of process parameters in altering surface roughness is found as arc current at 50 amp (level 2), gas pressure at 4.3 bar (level 1), and thickness of the workpiece at 6 mm (level 1).However, the optimum combination of processes in altering the bur height and MRR is found as arc current at 50 amp (level 2), the gas pressure at 5.4 bar (level 2), and the thickness of the workpiece at 6 mm (level 1).• A strong interaction between gas pressure (P) and workpiece thickness (T) is found in altering burr height.But the optimum performance calculated without considering the interaction (P × T) is more than the optimum performance considering interaction (P × T).Hence, the optimum cutting condition considered is C 2 (50 amp), P 2 (5.4 bar), and T 1 (6 mm) in altering burr height.No strong interaction between the parameters is found in altering surface roughness and MRR.• The percentage contribution of the arc current (C) is more than other parameters, namely, gas pressure (P) and workpiece thickness (T) in altering the process response.Surface roughness and MRR for the parametric level are considered in this study, whereas the percentage contribution of workpiece thickness (T) is more than the arc current (C) and gas pressure (P) in altering burr height.
• From the behaviour of the slopes of the response parameters vs treatment condition graphs, it is observed that treatment condition 7 is an optimum set of process parameters in altering burr height and MRR.However, for surface roughness, the optimum level of parameter obtained from the S/N ratio and average performance graph analysis is treatment condition 5.

Figure 2 :
Figure 2: Commercial-grade aluminium plate before and after plasma cutting.
) = MSD Sum of the square of the responses Total number of reponses .

Figure 7 :
Figure 7: (a)-(c) Interaction graph for C, P, and T in altering for surface roughness, burr height, and MRR.

Table 2 :
Taguchi's layout of L 8 with responses namely, surface roughness, burr height, metal removing rate

Table 4 :
Average values of S/N ratio for surface roughness, burr height, and MRR level wise in PAC

Table 5 :
Optimal setting of the parameters