Machinability analysis of LM26 aluminium metal matrix composites reinforced with graphite and fly ash using the AWJM process

Abstract

The incorporation of various reinforcements in aluminium composites markedly improves their thermal, mechanical, and wear characteristics when compared to composites with a single reinforcement. Nevertheless, heightened hardness presents challenges in machining, rendering traditional methods less efficient for obtaining high-quality cuts. This study explores the application of Abrasive Water Jet Machining (AWJM) on aluminium alloy LM26 hybrid composites that are reinforced with graphite and fly ash, which were manufactured through stir casting. AWJM is assessed as a viable machining technique to tackle the challenges presented by these advanced composites. The analysis focuses on key process parameters, including transverse speed, stand-off distance, and abrasive flow rate, to evaluate their impact on essential output responses such as surface roughness, material removal rate (MRR), and overcut error. A systematic experimental design utilizing an L27 orthogonal array is implemented to thoroughly examine the influences of various parameters. Response Surface Methodology (RSM) serves as a vital tool for statistical analysis and optimization, facilitating an in-depth exploration of parameter interactions and their influence on machining performance. The results offer important perspectives on enhancing AWJM parameters for hybrid aluminium composites, leading to better machining precision and efficiency. The ideal machining parameters were determined to be an abrasive flow rate of 440 g/min, a traverse speed of 300 mm/min, and a stand-off distance of 1.50 mm, leading to a peak MRR of 1.201 g/sec and a minimum surface roughness (Ra) of 2.018 µm. The investigation highlights AWJM’s proficiency in processing high-hardness materials while preserving favorable surface qualities. The findings enhance manufacturing methods for reinforced aluminium composites, facilitating their wider use in aerospace, automotive, and other high-performance sectors.

Introduction

Aluminium-based Metal Matrix Composites (MMC) have attracted significant attention across various industries because of their lightweight nature, exceptional wear resistance, strength, and low density. These properties render them suitable alternatives to monolithic materials in aerospace, structural, electronics, and automotive applications¹. At present, B4 C, TiC, Si3 N, SiC, GNT, TiO2, Al2O3, graphite, fly ash, and red mud are frequently utilised as reinforcement materials in aluminium MMC^2–4. The researchers investigated the mechanical properties and structural relationships of carbon nanotube (CNT) reinforced aluminium composites created through powder metallurgy. The findings demonstrate that intergranular carbon nanotubes improve both elongation and strength, with the microstructure exhibiting a consistent distribution of carbon nanotubes^5,6.

Cao et al⁷. developed AA5052-based composites reinforced with carbon fibre to enhance wear and mechanical characterisation. The findings demonstrated a notable enhancement in composite hardness and ultimate tensile strength, with increases of 46.8% and 16.9%, respectively. Additionally, wear tests revealed a significant decrease in composite wear loss, estimated at around 70%. The examination into the wear and mechanical characteristics of B4 C-reinforced AA5083 aluminium composite reveals that both hardness and wear properties improve with increased reinforcement⁸. The LM26 aluminium alloy was used as the base matrix, reinforced with porcelain powder in varying weight percentages ranging from 0 to 8%, in 2% intervals. The incorporation of porcelain particles led to enhancements in hardness, tensile strength, flexural strength, and compressive properties. Notably, the composite containing 6 wt.% porcelain exhibited the highest resistance to wear⁹. Recent advancements in sustainable reinforcement materials, including green-synthesized nanoparticles, have further expanded the potential of aluminium MMCs¹⁰.

Ravi Kumar et al.¹¹ conducted fabrication and analysis of aluminium alloy LM6 reinforced with red mud and cenosphere composites. The findings validated enhanced mechanical properties alongside corrosion resistance and a reduction in density. Pawar et al.¹² analyzed the wear characteristics of unreinforced LM26 and SiC and Ni-Gr reinforced aluminium composites utilizing the Taguchi method. The investigation results indicate that the percentages of SiC and Ni-Gr, along with the load, are the most significant influencing factors, contributing 59.21% and 45.11% to the coefficient of friction and wear damage, respectively. Due to its superior mechanical properties, thermal conductivity, and damping capacity, graphite has great potential. Compared to conventional aluminium alloys, graphite reinforced aluminium composites have better mechanical and tribological properties¹³.

Fly ash particles serve as a significant reinforcement in the production of aluminum-based composites due to their various benefits, including enhanced hardness, cost-effectiveness, and improved wear and abrasion resistance while reducing overall costs. This is relevant for multiple domains, including brake pads, pans, and valve covers^14–16. Aluminium MMCs demonstrate superior mechanical and physical properties when compared to aluminium alloys, making them a preferred choice for a variety of applications¹⁷. The inclusion of reinforcement restrictions in AMMCs meets industry requirements, yet it complicates the machining process. Nonetheless, unconventional machining processes are capable of cutting intricate-shaped profiles with remarkable precision¹⁸.

Innovative machining techniques are being adopted more frequently to tackle fabrication challenges, including the creation of intricate shapes, the machining of high-strength materials, the attainment of high precision, the improvement of surface quality, the facilitation of miniaturization, the reduction of production time, and the minimization of waste. Among these, AWJM has garnered considerable interest from engineers for its exceptional cutting quality and broad operational capabilities¹⁹. AWJM represents a versatile machining technique applicable to a wide range of materials. A high-speed water jet system employs abrasive particles to erode the workpiece via a nozzle. The effectiveness is contingent upon factors such as jet velocity, nozzle diameter, standoff distance, abrasive flow rate, and traverse speed²⁰. NareshBabu et al.²¹ examine and enhance the process parameters affecting the cut surface quality of brass 360 material in AWJM. The findings indicate that the surface quality of AWJM machining is predominantly influenced by the effects of pump pressure parameters. Liao et al.²² analyzed a dual method in AWJM machining, utilizing two distinct nozzles and two different abrasives: garnet and stainless steel ball. The initial cutting process employs a nozzle with garnet abrasive to remove material, while the subsequent process utilizes a nozzle with stainless steel balls for surface modification of the machined surface. The secondary process eliminates scratches and particles from the initial cutting process, enhancing the overall surface quality.

Gnanavel Babu et al.¹⁸ analyzed input process parameters in the AWJM process for hybrid composites, revealing that traverse speed and jet water pressure have a significant impact on surface quality and kerf taper angle, whereas abrasive flow and mesh size exert minimal influence. The machining of Inconel 617 material for high-pressure boiler components is conducted using AWJM. The output performance metrics of MRR, circularity, parallelism, and perpendicularity errors were assessed with varying input parameters. The optimal process parameter settings for achieving superior performance are identified as a standoff distance of 4.5 mm, a water pressure of 180 MPa, a table feed rate of 65 mm/min, and an abrasive flow rate of 0.24 kg/min²³. Uthayakumar et al.²⁴ investigated the machining of nickel-based super alloys using the AWJM process, focusing on three input parameters: standoff distance, traverse speed, and water pressure, each evaluated at three different levels. The study indicates that water pressure is the primary factor influencing surface quality and morphology.

Al et al. (2020) developed a second-order model using RSM to predict cutting forces in SiCp/Al composite machining, finding that cutting force increases with depth of cut and feed rate, while higher cutting speeds reduce it, with minimal prediction error. Laghari et al.²⁵ reviewed sustainable machining practices for MMCs, focusing on lubrication, cooling techniques, and economic considerations. The study emphasized optimizing machinability factors such as surface integrity and tool wear for Al, Ti, Mg, and Cu-based MMCs. Laghari et al.²⁶ explored advancements in turning SiCp/Al MMCs, addressing tool wear, chip formation, and modeling techniques, while suggesting future research directions. Laghari and Li²⁷ optimized cutting forces for SiCp/Al 45% and 50% composites using RSM, confirming model accuracy with ANOVA and residual analysis. Laghari et al.²⁵ further developed a genetic programming (GP) model, which outperformed RSM, achieving over 40% better prediction accuracy.

Kang et al.²⁸ performed experiments under varying processing conditions to analyse AWJM featuring blasted dimples of different sizes and geometries. The findings indicate that, among all the input processing parameters in this process, air pressure and the variation of process time are the most influential and controllable factors affecting the dimple dimension. It was also noted that air pressure and standoff distance significantly influence the definition of dimple geometry. Azarsa et al.²⁹ utilized AWJM processes to fabricate fins in the maximum quantity per unit length of Al6061-T6 material, to enhance the heat transfer process. Square channels were precisely cut using an AWJM micro nozzle. Through careful analysis of multiple parameters, it was determined that reduced nozzle traverse speeds, increased water pressure, and elevated abrasive flow rates effectively minimized levelling at a specified offset. The development of Amino Thermosetting Plastic (ATP) for abrasive jet polishing of aluminium alloy involved both theoretical and experimental analysis. The findings suggest that ATP abrasives effectively polish aluminium alloy, achieving commendable surface quality and fatigue properties³⁰.

The application of metal masks in AWJM requires a thorough investigation to minimize the feature width in Al6061-T6 and borosilicate glass materials. For Al 6061-T6, the results indicated a reduction of 2–3 times in thickness, along with an 11% decrease in centerline surface roughness and a 44% reduction in centerline waviness when compared to machining without masks. The waviness and roughness in borosilicate glass are significantly high, which limits the applicability of this technology in certain microfluidic applications. This necessitates the use of post-processing techniques to enhance the surface finish³¹. Hu et al.³² developed a multi-jet machining technique that utilized water transformed into a dry abrasive air jet to machine masked channels and dimples on cemented carbon and silicon wafer surfaces, allowing for the quantification of roughness and erosion rate. They carried out multiple comparative studies under both wet and dry conditions. The findings indicate that in wet conditions, the sliding grinding effect enhances surface finish and decreases the erosion rate. This study focuses on the reinforcement of aluminium alloy LM26 with graphite and fly ash particles through the stir casting process. The fabricated composite undergoes machining via the Abrasive Water Jet process, and the effects are examined thoroughly.

Materials and methods

Fabrication of hybrid composite

Aluminium alloy LM26 exhibits outstanding mechanical and tribological characteristics, attributed to the favorable properties of LM series alloys. It is primarily utilized in the automotive industry, particularly for engine pistons. The aluminium alloy LM26 has been placed in the stir casting furnace and heated for melting. The mechanical agitator, treated with ceramic, has been employed to blend the molten alloy. The stirrer speed was regulated and maintained at 400 to 410 rpm for 15 min. Fly ash and graphite reinforcements are measured by weight percentage and preheated to 440 °C³³. A small amount of magnesium is added to enhance wettability. The preheated graphite and fly ash reinforcements are systematically introduced into the furnace containing molten LM26 alloy. To prevent the formation of blowholes, hexachloroethane tablets have been added systematically. The reinforced molten aluminium alloy was introduced into the preheated 10 mm-thick and 100 mm square die, where it subsequently solidified at room temperature.

The fly ash and graphite reinforcements are combined in varying proportions to create four distinct composite samples, as detailed in Table 1. Following fabrication, the mechanical characterization for all samples are documented in Table 1. Microstructural images are analyzed to confirm the even distribution of fly ash and graphite particles within the aluminium alloy LM26^34,35, as illustrated in Fig. 1. These inclusions significantly influence composite performance. Graphite is expected to improve wear resistance and reduce friction due to its lubricating nature. Fly ash’s impact on mechanical strength, hardness, and thermal conductivity varies depending on its particle size, morphology, and chemical composition. Fly ash enhances composite hardness through its load-bearing capacity, minor solid solution hardening, and grain refinement, resulting in a cost-effective reinforcement. Following the evaluation and examination of sample 3, it demonstrates superior mechanical properties in comparison to the other samples^34,35. Consequently, the composites of sample 3 (Aluminium alloy LM26 85% + Fly ash 7.5% + Graphite 7.5%) are selected for the machinability study. The selected sample microhardness is 163.3 VHN with a tensile strength of 238 N/mm².

Table 1.

Fabricated Composites with various proportions and their properties.

S.No	Name of samples	Sample Proportions (Wt. %)			Hardness Value (VHN)	Tensile Strength (N/mm2)
		Aluminium alloy LM26	Graphite	Fly Ash
1	Sample 1	100	0	0	122.2	201
2	Sample 2	85	5	10	141	204
3	Sample 3	85	7.5	7.5	163.3	238
4	Sample 4	85	10	5	148.7	223

Fig. 1.

SEM image of AHMMC (Aluminium Alloy LM26 85% + Graphite + Fly ash) Composites.

AWJM represents an advanced and straightforward method for machining intricate, non-uniform shapes from a range of hard materials. The process operates on the principle of mechanical energy for material removal, utilizing a mixture of abrasives and water to create an abrasive slurry. High-speed abrasive slurry strikes the workpiece, resulting in the removal of material from its surface. The AWJM process enables the rapid cutting of challenging materials that are difficult to machine and electrically nonconductive, while maintaining precise dimensions. Therefore, it can be utilized for cutting, drilling, and cleaning hard materials. Currently, the machining process uses 80-mesh garnet abrasives. Figure 2a illustrates the specimen of the AWJM cut hybrid composite, while the experimental setup for AWJM is depicted in Fig. 2b.

Fig. 2.

(a) Machined composite specimen, (b) AWJM experimental setup (LM26 + Graphite + Fly ash) (c) Surface roughness tester.

Measurement of AWJM parameters

Conducting a trial and error works for this cast sample, and experimentation is planned afterward to incorporate three factors at three levels with varying proportions, as detailed in Table 2. L27 Taguchi designs are widely used for robust optimization and quality improvement by identifying control factors that reduce variation caused by noise. The experiments conducted were streamlined according to the experimental design 27 trials aimed at optimizing effectiveness and accuracy, as presented in Table 3.

Table 2.

Process parameters and levels.

S. No	Machining process parameters	Level 1	Level 2	Level 3
1	Abrasive flow rate (g/min)	240.0	340.0	440.0
2	Traverse speed (mm/min)	300.0	350.0	400.0
3	Standoff distance (mm)	1.5	2.5	3.5

Table 3.

RSM experiments and responses.

S.No	Traverse feed (mm/min)	Abrasive flow rate (g/min)	Standoff distance (mm)	MRR (gm/sec)	SR (μm)
1	300	240	1.5	1.219	2.128
2	300	240	2.5	1.228	2.383
3	300	240	3.5	1.263	2.272
4	300	340	1.5	1.199	2.112
5	300	340	2.5	1.202	2.223
6	300	340	3.5	1.292	2.219
7	300	440	1.5	1.250	2.053
8	300	440	2.5	1.204	2.105
9	300	440	3.5	1.187	2.178
10	350	240	1.5	1.321	2.658
11	350	240	2.5	1.316	2.893
12	350	240	3.5	1.311	2.701
13	350	340	1.5	1.392	2.571
14	350	340	2.5	1.404	2.681
15	350	340	3.5	1.381	2.603
16	350	440	1.5	1.336	2.461
17	350	440	2.5	1.349	2.552
18	350	440	3.5	1.343	2.511
19	400	240	1.5	1.439	2.991
20	400	240	2.5	1.398	3.104
21	400	240	3.5	1.374	3.082
22	400	340	1.5	1.464	2.901
23	400	340	2.5	1.444	3.056
24	400	340	3.5	1.433	2.975
25	400	440	1.5	1.519	2.857
26	400	440	2.5	1.482	2.951
27	400	440	3.5	1.493	2.891

The density of the developed composite was predicted based on the fundamental principle established by Archimedes of Syracuse. Following the machining of each sample, the weight loss is determined using a precision weighing machine. The stopwatch serves as a tool for assessing the duration of operations.

1

The removal rate of material as well as the periphery was anticipated as per the method expressed in Eq. (1).

The quality of the machined surface was assessed utilizing a surface roughness tester. The surface roughness was characterized by the Ra values presented in this study, as illustrated in Table 3.

Design of experiments and RSM

RSM investigates the relationships between various descriptive factors and one or multiple responses. It was developed by George Box and Wilson in 1951. The primary objective of the RSM is to utilise an advanced experimental design to attain the best results. The methodology is employed to examine and define the relationships between AWJM parameters and their variables. Numerical methods like RSM can be utilized to enhance the production of exceptional materials through the optimization of process variables. To elaborate on the improvement of the RSM, the application of a fitting plan of analysis has gained widespread utilization. Unlike traditional approaches, the interaction between functional components can be determined through mathematical methods.

The standard representation of a second-order regression equation (Eq. 2) for the proposed model aimed at predicting the response Y can be expressed as:

2

where,

Y is the expected response, b₀ is the model constant;

a₁, a₂ and a₃ are freelance variables;

b₁, b₂, and b₃ are linear coefficients;

b₁₂, b₁₃, and b₂₃ are vector coefficients.

and b₁₁, b₂₂, and b₃₃ are the quadratic coefficients.

The analyses are carried out with three factors and three levels through Box-Behnken Design using by the way of using Minitab 19 environment. The second-order polynomial regression equations are used to establish the outcome of MRR and Ra by using the input AWJM parameters as per Eq. (2).

The soft computing technology was utilized to evaluate the responses of the surface model and to analyze the outcome of MRR and Ra Eq. (3) and Eq. (4) in the coded factor after eradicating the slightest important factors.

3

4

Results and discussion

Response surface methodology

RSM is a statistical and mathematical tool used to model and optimize processes influenced by multiple variables. It helps in understanding the interactions between input factors and their effect on a desired response. In machining studies, such as micro-turning, RSM is used to develop empirical models that link process parameters (like traverse feed, Abrasive flow and Standoff Distance) to outcomes such as MRR and Surface Roughness. These models assist in identifying optimal settings and improving process efficiency. Experimentation and analysis are two fundamental components of the scientific method, both essential for drawing meaningful inferences from research. RSM is a widely recognized and effective optimization technique used to address complex problems involving multiple variables. RSM comprises a set of mathematical and statistical tools that assist in modeling and analyzing systems. The goal is to identify optimal conditions for the desired response [Astahk]. Statistics, as a scientific discipline, is integral to achieving reliable and targeted outcomes in such analyses. The core principle of RSM is to conduct a series of structured experiments to efficiently explore the relationships between variables and determine the optimal response.RSM is particularly useful when investigating problems where several independent variables simultaneously affect a dependent variable or response. In this study, mathematical models were developed using RSM based on experimental data. These models are empirical and are designed to examine the relationships between a set of input factors and observed outcomes, specifically, MRR and Surface Roughness, in this context. The development of such models is critical for optimizing machining processes by linking machining responses to the influencing variables.

Perturbation plots are used to show how the response changes as each factor moves away from its reference (or center) point, while all other factors are held constant. These plots help visualize the sensitivity of the response to individual factors, giving insight into which variables have the most significant effect on the outcome. The “normal path” refers to the standard direction or progression of the experimental design from the center point. Perturbation from this path represents deviation in one factor at a time, showing how robust or sensitive the response is to small changes. These plots identify the most influential factors affecting the responses, allowing researchers to concentrate on the variables with the greatest impact. Perturbation in the normal path or state can be visualized from Fig. 3. A steep curve indicates high sensitivity, while a flat line shows minimal effect. Together, RSM and perturbation analysis support data-driven decisions for process optimization.

Fig. 3.

(a) Perturbation in the normal path, (b) Perturbation in the normal path.

Analysis of variance (ANOVA) on MRR

ANOVA was employed to evaluate the significance of the regression models, individual coefficients, and lack of fit. Table 3 shows ANOVA results for the quadratic model and related terms for Material Removal Rate (MRR), along with model adequacy indicators such as R², Adjusted R², and Predicted R². Higher R² values, closer to 1, indicate good model fit. According to Lee et al. (2008), model terms are significant when the p-value is below 0.05; in this case, the p-value is 0.0001, confirming statistical significance. The model’s F-value of 33.4 supports this, while the Adequate Precision of 18.15 (above the threshold of 4) indicates a strong signal, as per Design-Expert software (Stat-Ease Inc., 2008). Among the input parameters, traverse speed and its associated interactions were identified as having the most significant influence on the responses, compared to abrasive flow rate and standoff distance.

According to the findings presented in Table 4, the traverse speed, the rate of abrasive flow, and the interaction between these two factors hold significant importance. Notably, traverse speed accounts for a substantial 90.65%, while the abrasive flow contributes 1.96%. The interactive effect of traverse speed and abrasive flow has an involvement of 4.20%. The remaining individual, interactive, and second-order terms show minimal contribution to the overall results. Thus, it is referred to as negligible in terms of MRR. The coefficient of the developed regression equation, R2, was found to be 0.9470. The developed equation demonstrates a closer alignment with the experimental data. The anticipated R2 of 0.85096 aligned well with the adjusted R2 of 0.9189.

Table 4.

ANOVA on MRR.

Source	Sum of Squares	df	Mean Square	F-value	p-value		PCR
Model	0.2456	9	0.0273	33.74	< 0.0001	Significant
A-Traverse speed	0.2227	1	0.2227	275.31	< 0.0001		90.65
B-Abrasive flow rate	0.0048	1	0.0048	5.94	0.0261		1.96
C-Standoff distance	0.0002	1	0.0002	0.2640	0.6140		0.09
AB	0.0103	1	0.0103	12.77	0.0023		4.20
AC	0.0032	1	0.0032	3.96	0.0630		1.30
BC	0.0002	1	0.0002	0.2680	0.6114		0.09
A2	0.0009	1	0.0009	1.07	0.3158		0.35
B2	0.0028	1	0.0028	3.48	0.0794		1.16
C2	0.0005	1	0.0005	0.6009	0.4489		0.20
Residual	0.0137	17	0.0008
Cor Total	0.2593	26
S = 0.0284, R2 = 0.9470, R2Adju = 0.9189, R2Pred = 0.8509, Adeq Precision = 18.1504

To enhance the MRR, it is advisable to maintain higher values for traverse speed, standoff distance, and medium abrasive flow. The anticipated MRR was derived from the generated equation, with the conditions of progression outlined in Eq. (3).

ANOVA on surface roughness

Table 5 shows the ANOVA results for the quadratic model related to surface roughness. The model is statistically significant (p = 0.0001) and shows no lack of fit (p = 0.1033 > 0.05). A high F-value of 160.2 confirms its significance. The Adequate Precision ratio is 39.8639, well above the desirable threshold of 4, indicating a strong signal. Figures 4 and 5 present the normal probability plots of residuals for MRR and surface roughness, respectively. These plots show that the residuals closely align with a straight line, suggesting the model is well-fitted and statistically valid (Montgomery, 2001). ANOVA and model summaries identify traverse speed as the most influential parameter, exerting a greater effect on both MRR and surface roughness than abrasive flow rate and stand-off distance.

Table 5.

ANOVA on Ra.

Source	Sum of Squares	df	Mean Square	F-value	p-value		PCR
Model	3.08	9	0.3419	160.27	< 0.0001	Significant
A-Traverse speed	2.83	1	2.83	1325.70	< 0.0001		91.91
B-Abrasive flow rate	0.1518	1	0.1518	71.15	< 0.0001		4.93
C-Standoff distance	0.0272	1	0.0272	12.76	0.0023		0.88
AB	0.0001	1	0.0001	0.0375	0.8487		0.00
AC	0.0026	1	0.0026	1.22	0.2840		0.08
BC	0.0004	1	0.0004	0.1860	0.6717		0.02
A2	0.0113	1	0.0113	5.29	0.0343		0.38
B2	0.0001	1	0.0001	0.0688	0.7963		0.00
C2	0.0556	1	0.0556	26.04	< 0.0001		1.80
Residual	0.0363	17	0.0021
Cor Total	3.11	26
S = 0.0462, R2 = 0.9884, R2Adju = 0.9822, R2Pred = 0.9691, Adeq Precision = 39.8639

Fig. 4.

Normal probability plot on MRR.

Fig. 5.

Normal probability plot surface roughness.

Variance analysis is typically employed to identify the most significant machining parameters that influence the responses. The experimental evaluation of values identified traverse speed, abrasive flow, standoff distance, and the second order terms of traverse speed and standoff as significant factors influencing surface roughness. This study indicates that the maximum involvement of jet traverse speed reached 91.91%, while abrasive flow accounted for 4.93% and standoff distance contributed 0.88%. Additionally, the second-order terms for traverse speed and standoff were 0.38% and 1.80%, respectively, as detailed in Table 5. The mixed factor effect did not show significance concerning the surface roughness.

An increased quantity of abrasive particles offers multiple cutting edges, enhancing the MRR and influencing surface roughness (Babu et al., 2017). The findings from the ANOVA indicate that the optimal surface finish was achieved with the minimum standoff distance^36,37. The increased value of the standoff distance facilitates the expansion of the jet’s diameter, enhancing the exposure of the external surface traction from surrounding elements³⁸. Consequently, augmenting the standoff distance leads to a simultaneous rise in the jet diameter while reducing the kinetic energy. From this evidence, the recommendation is to keep the standoff distance minimal to achieve the best surface texture. The optimal surface performance was achieved with a low standoff and reduced traverse speed^36,37.

Since the model’s response p-value is less than 0.0001, the equation is significant. The established model has 33.74 and 160.27 F-values for MRR and Ra. The calculated F-values exceed the standard F-value of 2.896, indicating that the model is significant within the confidence interval. Use the lack of fit approach to determine if a model fits the objective configuration. The model’s lack of fit value should be negligible to fit the design. The analysis reveals a “Lack of Fit” of 0.4876 for MRR and 0.071 for Ra, both exceeding 0.05. This suggests that the lack of fit is not significant, indicating that the model is deemed appropriate.

Acceptable precision measures signal-to-noise ratio. A ratio over 4 is best. MRR 18.1504 and Ra 39.8639 indicate a good signal. Predicted MRR and Ra, R₂ values of 0.9470 and 0.9884 match adjusted R₂ values of 0.9189 and 0.9822, respectively, with a difference of less than 0.2.

ANOVA results show that MRR and Ra points closely align with ideal lines, indicating strong correlation between predicted and actual values. This confirms the accuracy of the models within the desired range.

Parameters affect MRR

The interactive effects of traverse feed and abrasive flow will be considered, as illustrated in Fig. 5. The current outcome of material removal stands at 17% (1.482 gm/sec), which is improved with a higher traverse feed of 400 mm/min and an increased flow of abrasives at 440 gm/min, all maintained at a standoff distance of 2.5 mm. The relationship between standoff distance and abrasive flow is illustrated in Fig. 6. The MRR is expected to increase by 3%, reaching a value of 1.36 gm/sec, when utilizing a higher standoff distance of 2.7 mm and an increased abrasive flow rate of 350 gm/min, with a transverse feed rate set at 350 mm/min. (Fig. 7).

Fig. 6.

Traverse speed and Abrasive flow on MRR.

Fig. 7.

Abrasive flow and Standoff distance on MRR.

The interactive effects of transverse feed and standoff distance will be considered, as illustrated in Fig. 8. The MRR is expected to increase by 17% (1.486 g/sec) when utilizing a greater standoff distance of 1.5 mm and achieving a maximum traverse speed of 400 mm/min, with the abrasive flow set at 340 g/min. This analysis reveals that increasing the transverse feed does not result in any observable impact. A reduction in standoff distance from 2.7 mm to 1.5 mm results in a decrease in MRR from 1.486 to 1.360, indicating an approximate 8% reduction in the rate of material removed, which significantly impacts the overall MRR.

Fig. 8.

Traverse speed and standoff distance on MRR.

When the transverse feed is raised from 300 mm/min to 400 mm/min at standoff distances of 1.5 and 2.5 mm, the MRR escalates from 1.235 g/sec to 1.486 g/sec, reflecting an increase of 17%. The analysis indicates that the main influencing factor in MRR is the traverse feed, with standoff distance being the next significant factor, as illustrated in Fig. 8.

Increased traverse speed boosts intermolecular interactions and jet kinetic energy. The increased kinetic energy of the AWJM increases the strike rate of multipoint abrasive particles in the cutting zone, increasing material removal.

The variation in standoff distance influences the kinetic impact, leading to a reduction in jet blows and, simultaneously, a decrease in material removal³⁹. The heightened rate of abrasive flow in the jet enhances the cutting boundaries, subsequently leading to an increase in the MRR⁴⁰. The interplay between reduced traverse speed and increased abrasive flow leads to a decrease in the MRR. This is attributed to the effects of larger spreads, a slower flow rate, and an increased standoff distance with abrasives. Consequently, this combination adversely affects the working efficiency, ultimately resulting in a diminished MRR. Increased traverse speed and abrasive flow at the cutting zone will improve plane ploughing and MRR. The increased volume of the abrasive jet that rotates the workpiece in the jet tends to distract and diminish the MRR⁴¹. Generally, an increase in traverse rate leads to a rise in the MRR, whereas a greater quantity of traverse rate combined with standoff tends to reduce the MRR⁴². The outputs of MRR are fundamentally influenced by the abrasive water jet effect and how composite particle reinforcement is presented. The configuration and dimensions of the nozzle, along with its placement and orientation, typically elevate the temperature in the cutting area, while the standoff distance also affects the MRR. A rise in temperature within the cutting area of the composite leads to a decrease in strain rate and tensile strength, as the material becomes softer due to thermal effects. Thus, the essential energy of the abrasive particles will be concentrated, improving material removal⁴³. The significant rate of reinforcement in the composite concurrently enhances the brittleness. This work primarily aims to sustain the individual and interactive effects of traverse feed and standoff distance at a minimal level while ensuring a higher rate of MRR.

Influence of cutting parameters on surface roughness

Figures 9, 10, 11, show contour graphs showing how machining parameters affect surface roughness. The result was achieved by considering traverse speed and abrasive flow rate while maintaining a moderate standoff distance. Figure 6 shows the interactive effect of traverse speed and abrasive flow rate at 2.5 mm standoff.

Fig. 9.

Traverse speed and Abrasive flow on Ra.

Figure 9 shows that as traverse speed interacts with abrasive flow, material surface roughness increases from lower to higher levels. Nozzle configuration and dimensions greatly affect product surface quality. Generally, a higher traverse speed results in faster machining times, which leads to incomplete material removal and consequently increases surface roughness⁴⁴. The influence of traverse speed and standoff distance on surface roughness, along with the examination of abrasive flow and standoff distance, is illustrated in Figs. 10 and 11. The analysis of this interactive effect reveals that the surface roughness rises from 2.24 µm to 3.09 µm as the traverse speed and abrasive flow are increased from lower to higher levels, maintaining a standoff distance of 2.5 mm.

Fig. 10.

Traverse speed and Standoff distance on Ra.

Fig. 11.

Abrasive flow and Standoff distance on Ra.

It can be expected that a rise in traverse speed facilitates a reduced area of overlap during machining and minimizes the amount of abrasive particles acting on the workpiece, which certainly influences the surface roughness⁴⁵ and⁴⁶. Therefore, a reduced jet traverse speed was essential to attain an enhanced surface quality. The increased traverse speed primarily influences the machining characteristics,due to the inconsistent machining speed in the material, there is a potential for increased surface roughness. The reduction in jet traverse speed led to a smoother machined surface (Chithirai Pon⁴⁷).

The interaction between traverse speed and standoff distance, with an abrasive flow of 340 g/min, reveals that as traverse speed increases across various standoff distances, the surface roughness value remains consistent at 2.48 µm. Notably, at the mid-point standoff distance of 2.5 mm, improved responses are observed in alignment with the investigation’s objectives. It was noted that maintaining a small to moderate standoff distance contributes to improved surface quality. An improved surface quality was achieved through lower standoff and traverse speed^36,37.

The interactive analysis indicates that the traverse speed directly influenced the increase in surface roughness values from 2.5 µm to 3.09 µm. The reduced jet traverse speed of 300 mm/min facilitates the penetration of excess abrasive particles into the workpiece. The extensive presence of abrasive particles provides numerous cutting edges and enhances the MRR; however, it has an impact on surface quality⁴⁰. Furthermore, an analysis reveals an interactive effect between abrasive flow and standoff distance at a traverse speed of 350 mm/min. The findings indicate that at a midpoint standoff distance of 2.5 mm, varying levels of abrasive flow result in a decrease in surface roughness from 2.78 µm to 2.59 µm, exhibiting a parabolic curve nature.

The ANOVA results indicate that improved surface roughness was achieved with a reduced standoff distance^36,37. A greater standoff distance typically enables the jet to increase its diameter, enhancing its interaction with the outer surface influenced by the surrounding environment³⁸. Consequently, an increase in standoff distance leads to a larger jet diameter while decreasing the kinetic energy and the impact on the material surface, which in turn results in increased surface roughness. From this perspective, increased standoff results in a divergence of the jet path, lead to a decrease in kinetic energy while simultaneously enhancing surface roughness. The scattering rate of abrasive particles occurs swiftly at lower positions, resulting in a rapid surface attack on the part, which facilitates material removal in the form of chips and enhances surface quality³⁹. There is also a possibility of ploughing, and a period during which grooves form as a result of the material’s plastic deformation⁴¹. The standoff distances generally do not exert any physical influences on the surface roughness of the workpiece. The enhancement in standoff will lead to improved contact performance throughout the machining process, resulting in superior surface quality.

Composite manufacturing increases surface roughness because reinforcement materials improve bonding and hardness¹⁸. Increased reinforcement material creates voids and cavities on the surface, improving surface roughness. Water jet abrasive particle size and flow affect composite strength, plastic deformation, and cutting temperature, affecting surface roughness⁴⁰.

Generally, an increased amount of reinforcement enhances the brittleness of the material, which may lead to easier erosion from the surface and a decline in surface quality, ultimately resulting in crack formation. The atmosphere around the jet core plane increases water jet volume and surface roughness. AWJM’s machined surface pattern has three zones: the initial damage zone, the soft zone, and the hard part, as the jet moves from top to bottom. The jet’s broad and shallow angle and upward deflection harm the initial and coarse zone⁴⁴. Delamination, ploughing, and heat-affected zones affect the composition of AWJM machining materials, which determines their surface integrity. Cleaning ductile materials removes particles easily.

Multi-objective optimization – desirability

Optimization involves identifying the most effective solutions from a range of possibilities. Multi-objective optimization analysis uses desirability to derive an objective function D(X). The function’s main goal is to convert output variable values into a non-quantitative excellence measure.

Initially, Yi is transformed from the range of 0 ≤ di ≤ 1 into the format of the required function, with Yi being defined in Eq. (2) of the target and target function. This study involved conducting multi-point optimization to estimate the desired value utilising the Design Expert soft computing methodology. The processes were conducted according to the parametric conditions outlined in Table 3, utilising the specified input(s) and output(s). The primary objective of this optimization is to classify a collection of variables that enhance MRR and reduce Ra. Additionally, Table 6 presents the objectives, impacts, constraints, and process data related to inputs and outputs.

Table 6.

Variables and responses with importance.

Name	Goal	Lower Limit	Upper Limit	Lower Weight	Upper Weight	Importance
A:Traverse speed	is in range	300	400	1	2.18776	3
B:Abrasive flow rate	maximize	240	440	1.15495	1	3
C:Standoff distance	is in range	1.5	3.5	1	1	3
MRR	is in range	1.175	1.805	4.7863	1	5
Surface roughness	minimize	2.053	3.104	1	1.04713	5

The bar chart shows the process variables’ optimal values and MRR and Ra outputs. Figure 12 shows that the desirable attractive data closeness is 0 to 1, with importance 1 indicating the greatest variable contact.

Fig. 12.

Desirability rate of input & output Parameters.

The statistical and graphical analyses are used to obtain the goals of factors and outputs. The goal of this process is to maximum abrasive flow and within the range of traverse speed and standoff distance have to be worked out. As recommended by⁴⁸ for multiple outputs (s), it is suggested to perform statistical analysis first,otherwise, it may not be possible to expose a practicable region. The arithmetic optimization tool detects a point or points that maximize MRR and minimizes Ra.

The interactive effect of traverse speed and abrasive flow, along with a lower rate of standoff distance, is illustrated in both the MRR and surface roughness responses, as depicted in Figs. 13 and 14. The diagrammatic flow illustrates that in MRR, the trend is nearly parabolic, while Ra follows a diagonal pattern. This suggests that a lower traverse speed combined with a higher abrasive flow and reduced standoff distance results in improved performance for both responses.

Fig. 13.

Traverse feed and Abrasive flow on MRR.

Fig. 14.

Traverse feed and Abrasive flow on Ra.

In graphical optimization involving multiple responses, it is essential for the requirements to concurrently satisfy the proposed criteria, as illustrated in the overlay plot presented in Fig. 15. The diagrammatic perspective allows for the identification of feasible optimal performance solutions. Consequently, the numerical optimization is addressed initially, followed by the execution of the diagrammatic search. The parameters for the AWJM process were determined to achieve the highest MRR and the lowest surface roughness in the hybrid composite material process; with a traverse speed of 300 mm/min, abrasive flow of 440 g/min, and a standoff distance of 1.50 mm, the results are 1.201 g/sec and 2.018 µm. The optimized process parameters of AWJM, along with corresponding responses achieving 100% desirability, are presented in Table 7. The verification results indicated that exceptional performance was achieved, with deviations of less than 5% from the measured values.

Fig. 15.

Overlay Plot on multiple responses.

Table 7.

Optimized parameters with suitable responses.

Number	Traverse speed	Abrasive flow rate	Standoff distance	MRR	Surface roughness	Desirability
1	300.000	440.000	1.500	1.201	2.018	1.000	Selected
2	300.084	440.000	1.589	1.201	2.040	1.000
3	300.408	440.000	1.620	1.202	2.050	1.000
4	300.012	440.000	1.641	1.200	2.051	1.000
5	300.035	440.000	1.568	1.201	2.034	1.000

Conclusions

This research refined the parameters of AWJM for the processing of hybrid aluminium alloy LM26 composites, which are reinforced with graphite and fly ash, utilizing RSM.

• (i)The examination revealed that traverse speed serves as the predominant element affecting both MRR and surface roughness, representing over 90% of machining efficacy.

• (ii) The identified influencing parameters suggest that the stand-off distance, in conjunction with traverse speed, plays a crucial role in achieving both the maximum MRR and the minimal surface roughness. It was noted that keeping a reduced traverse speed facilitates efficient material erosion while maintaining surface quality. The findings indicated that reducing both traverse speed and stand-off distance results in optimal machining performance.

• (iii) Mathematical models were created and validated using ANOVA to forecast the maximum MRR and minimum surface roughness, thereby affirming the dependability of the optimization procedure.

• (iv) The optimal machining parameters were determined to be a traverse speed of 300 mm/min, an abrasive flow rate of 440 g/min, and a stand-off distance of 1.50 mm, leading to a peak MRR of 1.201 g/sec and a minimum surface roughness (Ra) of 2.018 µm. Furthermore, it was observed that surface roughness is affected by material reinforcement, micro-defects, and cavities resulting from uncut chips.

• (v) The findings provide key insights into optimizing AWJM parameters for machining hybrid aluminum composites, improving precision and efficiency in aerospace, automotive, and other high-performance industries.

Author contributions

C. P, S.S, D.M., and R.K. All the authors contributed equally, and all authors reviewed the manuscript.

Data availability

The data used to support the findings of this study are included within the article.

Declarations

Competing interests

The authors declare no competing interests.