Effect of Surface-Textured AlSiTiN Coating Parameters on the Performance of Ball-End Milling Cutter in Titanium Alloy Milling

Abstract

In the high-speed milling of titanium alloys, the combined application of surface texture and coatings can significantly enhance the performance of cemented carbide tools. Investigating the synergistic effect of surface texture and AlSiTiN coating on tool performance is crucial for advancing the development of their integrated preparation process. Therefore, in this study, a cemented carbide ball-end milling cutter is taken as the research object, and a surface-textured AlSiTiN coating is applied to the rake face. The effects of texture and coating parameters on the milling performance of titanium alloys are analyzed, and a regression model is developed to optimize the relevant parameters. The results indicate that the surface texture effectively reduces the actual contact area between the tool and the chip, serves as a storage space for chips, and enhances the wear resistance of the AlSiTiN coating. The coating thickness significantly affects milling force, milling temperature, and surface wear. An increase in coating thickness improves the hardness and integrity of the coating surface, and it also strengthens the adhesion of the texture to the coating. Additionally, precise control of the laser power plays a key role in reducing the milling temperature, while both the number of scans and the scanning speed significantly influence surface wear. Furthermore, maintaining an appropriate distance from the edge is crucial for enhancing the surface roughness of the workpiece. The optimized parameters for surface texture and coating preparation are as follows: coating thickness (h) = 3.0 µm, laser power (p) = 40 W, scanning speed (v) = 1590 µm/min, number of scans (n) = 6, texture diameter (d) = 42 µm, texture spacing (l) = 143 µm, and distance from the edge (l₁) = 104 µm. The optimized milling performance of the milling cutter shows a significant improvement.

1. Introduction

Titanium alloy, a metal material commonly used in aerospace and biomedical applications, possesses high strength, high hardness, low elastic modulus, and poor thermal conductivity—characteristics that significantly affect its machinability [1]. High-speed milling, as a high-performance processing technology for difficult-to-machine materials, is widely applied in titanium alloy machining [2]. However, during the high-speed milling of titanium alloy, issues such as high milling temperatures and significant tool wear may arise, which can adversely affect both milling efficiency and quality [3]. As a result, enhancing the performance of tools in the milling of titanium alloys has become a significant challenge in contemporary manufacturing industries [4]. In efforts to enhance the milling performance of titanium alloys, surface texturing and coating technologies have been shown to significantly extend tool life and reduce wear. Studies suggested that the combined application of these technologies can further optimize cutting conditions and improve the machining quality during titanium alloy milling [5]. Thus, investigating the synergistic preparation of surface textures and coatings on cutting tools is of critical importance for enhancing the milling performance of titanium alloys.

Currently, in the study of synergistic preparation of surface textures and coatings on cutting tool, researchers have employed a method in which coatings are first deposited on the substrate surface, followed by surface texturing, to investigate the combined beneficial effects of these features on tool performance. Meng et al. [6] prepared micro-groove textured TiAlN coating tools. The experimental results show that the micro-texture increases the bonding area between the coating and the substrate, the texture morphology increases the anchoring effect on the coating, and it also improves the residual stress in the coating deposition process. Zhang et al. [7] applied a Ti55Al45N coating onto the surface of cemented carbide tools, which is followed by the fabrication of nano-textures. Their cutting experiments demonstrated that nano-textured, coated tools significantly decreased milling forces, reduced milling temperatures, lowered the friction coefficient, and mitigated tool wear. Deba Kumar Sarma et al. [8] introduced micro-texture morphologies on the flank surfaces of coated tools and conducted cutting experiments. The results demonstrated that compared to conventional coated tools, the micro-textured coated tools exhibited reduced feed resistance during cutting and improved wear resistance on the flank surface. Similarly, Rajbongshi et al. [9] prepared micro-textures with varying morphologies on the flank faces of coated tools and performed dry cutting experiments on AISI D2 steel.

In the high-speed cutting of difficult-to-machine materials, as the friction between the chip and the tool increases, the temperature at the tool–chip contact area on the rake face rises, leading to an increase in the cutting force acting on the rake face. Consequently, the rake face experiences significant wear, resulting in reduced machining quality of the workpiece [10,11]. To address this issue, researchers in related fields have proposed applying surface textures and coatings to the rake face of the tool to enhance its cutting performance. Arulkirubakaran et al. [12] created different types of surface textures on the rake faces of uncoated and coated (TiN and TiAlN) tungsten–cobalt tools and conducted titanium alloy cutting experiments. The results indicated that compared to other tools, the TiAlN-coated tool with surface texture effectively reduces friction between the rake face and the chip, and the surface texture oriented perpendicular to the chip flow direction demonstrates superior cutting performance. After depositing an AlCrN coating on the rake face of a YS8 cemented carbide tool, Jiang [13] prepared linear and circular micro-textures, respectively, and investigated the cutting performance of the tool when machining AISI 660 austenitic stainless steel. The results demonstrated that both types of micro-textures effectively reduce cutting force and cutting temperature as well as minimize rake face wear. Among them, the linear micro-textured AlCrN-coated tool exhibited the most significant effect. Zhang [14] developed micro-groove and crater-type micro-textures on the rake face of TiXCo3-coated ball-end milling cutters and investigated their impact on the milling performance of aluminum alloys. The study demonstrated that the introduction of micro-textures significantly reduces tool wear on the rake face and enhances the milling performance of the coated tool. Notably, the crater-type micro-texture, covering 30% of the rake face area, was particularly effective in reducing both milling force and temperature. Bibeye Jahaziel Ronadson et al. [15] investigated the effect of surface texture on the rake face of TiN-WS₂-coated tools by comparing them with non-textured TiN-WS₂ coated tools, non-coated textured tools, and conventional tools. A dry cutting test on titanium alloy was conducted. The experimental results demonstrated that compared to the conventional tool, the TiN-WS₂ coated tool with surface texture exhibited a 14% reduction in cutting force, a 47% reduction in the coefficient of friction, and a 43% increase in tool life, significantly enhancing cutting performance. Yang et al. [16] fabricated micro-textures on the rake face of AlTiN-coated YG8 ball-end milling cutters and investigated the influence of micro-texture and cutting parameters on the milling performance of the tool. The experimental results indicated that the application of micro-textures on the rake face of the coated tool effectively reduces tool wear and improves the frictional interaction between the coated tool and the chip, enhancing the cutting performance of the tool and the quality of the workpiece. Moreover, the optimal milling parameters for surface-textured AlTiN-coated tools were also identified.

In summary, numerous studies have demonstrated that the application of surface texture on the rake face of coated tools can significantly enhance the cutting conditions and improve the cutting performance when machining difficult-to-cut materials. However, current research primarily focuses on the individual effects of surface texture on coated tools with limited exploration of the synergistic effects of both coating and surface texture on tool performance. A review of the relevant literature [17] suggested that investigating the synergistic effects of surface texture and coating parameters on tool cutting performance is crucial for advancing surface-textured coating technology. Therefore, the aim of this study is to examine the synergistic effects of surface texture and coating on the milling performance of cemented carbide cutters during the high-speed milling of titanium alloy materials.

AlSiTiN coating exhibits high hardness, excellent mechanical properties, and strong resistance to friction and wear at elevated temperatures, making it widely used in the cutting of difficult-to-machine materials with coated tools [18,19]. In line with the aforementioned research objectives, this study selects an AlSiTiN coating to be deposited onto the surface of cemented carbide ball-end milling cutters. After the coating deposition, surface texture is applied to the rake face, and a titanium alloy milling performance test platform is established to investigate the synergistic effects of coating parameters and surface texture parameters on milling force, milling temperature, surface wear, and workpiece surface roughness. Based on the experimental data, a regression model for evaluating milling performance indices and their related parameters is developed, and the optimal parameter combination is determined through optimization using the artificial bee colony algorithm. The novelty presented in this paper arises from the limited research on the synergistic effects of coating and surface texture, as highlighted in the literature. The primary motivation behind this novelty is to provide a deeper theoretical foundation and technical support for the application of tool surface textured coating technology in the high-speed milling of titanium alloys.

2. Experimental Design of Surface-Textured AlSiTiN-Coated Cemented Carbide Ball-End Cutters for Milling Titanium Alloy

2.1. Preparation of Surface-Textured AlSiTiN-Coated Cemented Carbide Ball-End Milling Cutters

The tool used in this experiment is a ball-end milling cutter with both the blade and tool holder made of YG8 cemented carbide. The dimensions of the ball-end milling cutter and its blade are presented in Figure 1a,b.

Table 1. Properties of YG8 cemented carbide.

Parameter	Value	Parameter	Value
Density/(kg/m3)	14,500	Hardness/HRA	89
Elastic modulus/GPa	640	Thermal conductivity/(W/m·K)	75.4
Yield strength/MPa	2600	Poisson ratio	0.22

provides the material properties of the cemented carbide ball-end milling cutter.

Based on previous research conducted by the team, it has been demonstrated that selecting a micro-pit texture can enhance the frictional performance of the tool surface [20]. Therefore, after depositing the coating, micro-pit textures were prepared on the rake face of the ball-end milling cutter using a Zhengtian fiber laser marking system. Following the completion of surface texture, the ball-end milling cutter underwent post-processing using an ultrasonic cleaner with acetone. The process of coating deposition and surface texture preparation is illustrated in Figure 1.

An orthogonal experiment was designed to examine the effects of coating and surface texture parameters. Based on the 27 parameter sets determined by the research team in preliminary work [21], surface-textured AlSiTiN-coated carbide ball-end milling cutters were prepared, and milling tests were conducted on titanium alloy workpieces. The orthogonal test design is presented in

Table 2. Orthogonal test table of seven factors and three levels of coating and surface texture parameters.

	Factor	Coating Thickness h (μm)	Laser Power p (W)	Scanning Speed v (mm/s)	Scanning Time n (time)	Texture Diameter d (μm)	Texture Distance l (μm)	Distance from Blade l1 (μm)
Level
1		2.6	35	1500	6	40	130	90
2		2.8	40	1600	7	50	150	100
3		3.0	45	1700	8	60	170	110

.

2.2. Construction of a Milling Performance Test Platform for Surface-Textured AlSiTiN-Coated Cemented Carbide Ball-End Milling Cutters

Ti6Al4V (TC4), a titanium alloy with a dual-phase α + β structure, exhibits high strength at both room and elevated temperatures along with good plasticity and toughness. Compared to other titanium alloys, TC4 is particularly well suited for high-speed machining [22]. Thus, TC4 was selected as the material for the test workpiece. The workpiece material used in the test is titanium alloy Ti6Al4V (TC4), as shown in Figure 2a. The optimal milling condition is achieved when the machining angle between the axis of the ball-end milling cutter and the machined surface of the workpiece is 15° [23]. Therefore, the milling fixture for the ball-end milling cutter was clamped and fixed at this 15° angle. The experimental device was a VDL-1000E three-axis CNC milling machine. Based on the pre-selected parameters [24], the milling speed was set to v_c = 100 m/min, the feed per tooth is f_z = 0.08 mm/tooth, the milling depth a_p = 0.3 mm, and the milling width a_e = 0.3 mm. Horizontal milling was performed along the width of the workpiece with unidirectional down milling under dry cutting conditions. Each parameter set corresponds to a milling stroke of 9 mm, and all experimental cutting parameters are kept consistent throughout.

The milling forces in three directions were recorded using a Kistler rotary dynamometer. The experimental setup is illustrated in Figure 2e,f, and the corresponding milling force signals are shown in Figure 2g. After collecting the milling force data, the mean values of the three components were calculated, and the resultant force was determined using vector analysis. The milling temperature test used the Haikang H-26 temperature measurement thermal imager to measure the milling temperature. The milling temperature results were collected by HIKMICRO Studio software V1.1.0. The temperature measurement acquisition device is shown in Figure 2b, and the temperature measurement acquisition method and acquisition interface are shown in Figure 2h.

Tool wear was measured using a super-depth-of-field microscope in the tool–chip contact area of the rake face on a ball-end milling cutter, as shown in Figure 2c. The wear region was divided into five equal sections, and the tool wear width KBi for each section was recorded. The average of these measurements was then taken as the tool wear value. The method for measuring tool wear is shown in Figure 2i.

The surface quality of the workpiece was assessed using the surface roughness parameter Ra. The surface roughness was measured using a TR240 surface roughness tester, as shown in Figure 2d. During the measurement, three points were taken perpendicular to the feed direction for each specimen, ensuring consistent positioning across all samples. The locations of the measurement points are illustrated in Figure 2j.

3. Analysis of Experimental Results on Milling Titanium Alloy with Surface-Textured AlSiTiN-Coated Cemented Carbide Ball-End Milling Cutter

3.1. Analysis of Milling Force Results for Surface-Textured AlSiTiN-Coated Cemented Carbide Ball-End Milling Cutter

The calculation results after milling force data acquisition are shown in

Table 3. Calculation results of milling force.

Number of Tests	Milling Force F/(N)	Number of Tests	Milling Force F/(N)	Number of Tests	Milling Force F/(N)
1	233.1	10	171.6	19	194.4
2	186.6	11	219.9	20	197.1
3	195.9	12	199.3	21	147.7
4	174.5	13	166.9	22	176.2
5	213.6	14	171.9	23	159.1
6	222.8	15	227.2	24	169.6
7	216.1	16	213.3	25	142.1
8	183.0	17	154.1	26	160.7
9	181.5	18	187.8	27	173.8

.

Range and variance analyses were conducted on the milling force data with significant findings and the influence trends of each parameter on the milling force illustrated in Figure 3. The influence mechanism of coating and texture preparation parameters on milling force is illustrated in Figure 4. In the subsequent analysis, this influence mechanism is discussed in detail.

The coating thickness has the most significant effect on the milling force with a p-value of 0.044 obtained from the one-way ANOVA. As the coating thickness increases from 2.6 to 3 μm, the milling force decreases by approximately 30 N. This is because under the same laser parameters, a thicker coating results in a relatively shallower surface texture, enhancing the coating’s surface integrity. As the coating thickness increases, the ability of the coating to conduct heat to its interior decreases. Consequently, the laser energy absorbed per unit area of the coating is reduced, leading to a slower cooling rate in the molten pool formed on the surface. This reduction in cooling rate results in smaller grain sizes of the compounds within the coating, increasing the surface hardness and enhancing the wear resistance. These improvements reduce friction in the chip–tool contact area, which in turn leads to a corresponding reduction in milling force [19].

The texture diameter has the most significant effect on milling force among the texture geometric parameters with a p-value of 0.099 obtained from the one-way ANOVA. As the texture diameter increases from 40 to 60 μm, the milling force initially decreases and then increases. Specifically, when the texture diameter reaches 50 μm, the milling force decreases by approximately 30 N, while at 60 μm, the milling force increases by about 30 N. This is because an increase in texture diameter reduces the contact area on the rake face of milling cutter, leading to a decrease in the friction force. Additionally, surface textures facilitate chip storage. As the texture diameter grows, the wear on the tool’s coating caused by chips diminishes, improving the coating’s integrity and further reducing friction on the rake face. The combined effect of these factors results in a reduction in milling force [25]. When the diameter of the micro-texture becomes too large, the overlap between the heat-affected zones of adjacent laser points increases, which reduces the cooling rate of the molten pool. This slower cooling leads to grain growth and a subsequent reduction in the density of the coating [26]. In addition, an excessively large texture diameter may indirectly reduce the laser energy per unit processing point, leading to shallower texture pits and decreased chip storage capacity. Both factors contribute to increased surface friction on the milling cutter, raising the milling force.

The scanning times significantly affect the milling force in laser preparation parameters with the p-value from the one-way ANOVA being 0.215. As the scanning times increase from 6 to 8, the milling force rises by approximately 20 N. This can be attributed to the fact that each pulse of laser energy induces a quenching effect in the laser-affected zone [14]. Each pass of the laser scanning increases the hardness of the coating on the milling cutter, reducing friction on the rake face and consequently lowering the milling force. However, as the scanning times increase, cracks may form in the re-solidified coating splatter, degrading the performance of the coating and leading to an increase in the milling force.

It is evident that the coating parameters and laser texture preparation parameters must be controlled within an optimal range to minimize the milling force.

3.2. Analysis of Milling Temperature for Surface-Textured AlSiTiN-Coated Cemented Carbide Ball-End Milling Cutter

The average milling temperature of each ball-end milling cutter within 40 min is shown in

Table 4. Milling temperature acquisition data.

Number of Tests	Milling Temperature/°C	Number of Tests	Milling Temperature/°C	Number of Tests	Milling Temperature/°C
1	207	10	237.4	19	209.2
2	225.2	11	227.7	20	209.9
3	219.4	12	227.6	21	218.5
4	238.4	13	219.8	22	210.9
5	216.4	14	222.1	23	229.4
6	232.9	15	215.3	24	239.6
7	236.4	16	209	25	218.4
8	226.3	17	226	26	226.9
9	214.6	18	209.2	27	213.3

.

Range and variance analyses of the milling temperature results were conducted with significant findings, and the influence trends of each parameter on the milling temperature are illustrated in Figure 5. The influence mechanism of coating and texture preparation parameters on the milling temperature is shown in Figure 6. The mechanism illustrated in Figure 6 is further explained in the following analysis.

Coating thickness has the most significant effect on milling temperature with a p-value of 0.094 obtained from the one-way ANOVA. As the coating thickness increases from 2.6 to 3 μm, the milling temperature decreases by approximately 10 °C. This is because the AlSiTiN coating possesses excellent wear resistance and thermal stability [27]. As previously analyzed, increasing the coating thickness enhances the wear resistance of the coating. This improvement reduces friction between the rake face of the ball-end milling cutter and the chip, decreasing the amount of frictional heat generated during the milling process.

Laser power also significantly affects milling temperature with a p-value of 0.183 obtained from the one-way ANOVA. As the laser power increases from 35 to 45 W, the milling temperature rises by approximately 12 °C. This can be attributed to the fact that higher laser power expands the molten pool in the surface texture area. The increased disturbance in the molten pool causes rapid bubble formation, which then sprays out of the liquid surface, resulting in droplet splashing and an increased amount of molten material splashing around the surface texture. During the laser processing of cemented carbide surfaces, Chu, Wang, and Won [28,29,30] also observed that higher laser power leads to an increase in surface splash. This condensed spatter can easily break into abrasive particles during milling, causing abrasive wear on the coating surface. The friction from these particles generates additional heat during the milling process, leading to a rise in the milling temperature.

The influence of scanning speed and the scanning times on milling temperature is also notable. The p-values obtained from one-way ANOVA are 0.167 and 0.181, respectively. As the scanning speed increases from 1500 to 1700 mm/min, the milling temperature rises by approximately 11 °C. This effect can be attributed to changes in scanning speed, which influence the duration of laser exposure on the surface of specimen. As the scanning speed increases, the interaction time of the laser with the coating decreases, resulting in shallower textures. This decreases chip storage in the pits and leads to increased frictional heat generated between the ball-end milling cutter and the chips. After the number of scans increased from 6 to 8, the milling temperature decreased by about 10 °C and then increased by about 6 °C. This can be explained by the repeated engraving at the same laser point, which deepens the texture and reduces the accumulation of chips on the surface of the ball-end milling cutter, leading to a decrease in milling temperature. However, excessive scanning may degrade the coating integrity, causing the milling temperature to rise again. However, excessive scanning times can introduce defects in the coating, compromising its wear resistance and thermal stability, which in turn increase the milling temperature.

By controlling the coating and laser texture preparation parameters within an optimal range, the milling temperature can be effectively reduced.

3.3. Analysis of Surface Wear for Surface-Textured AlSiTiN-Coated Cemented Carbide Ball-End Milling Cutter

The wear values of the ball-end milling cutter after milling are presented in

Table 5. Experimental results of surface wear of ball-end milling cutter.

Number of Tests	KB Average Value/μm	Number of Tests	KB Average Value/μm	Number of Tests	KB Average Value/μm
1	272.74	10	239.08	19	225.08
2	291.9	11	268.13	20	289.79
3	311.94	12	279.24	21	251.01
4	314.47	13	270.83	22	245.58
5	319.29	14	265.08	23	244.72
6	307.4	15	242.97	24	208.83
7	330.21	16	251.27	25	241.62
8	263.4	17	299.25	26	253.62
9	300.72	18	300.03	27	249.59

, while Figure 7 illustrates the wear morphology in the tool–chip contact area of the rake face. Based on the observed wear patterns, it is evident that the predominant forms of wear on the surface coating of the ball-end milling cutter are adhesive wear and abrasive wear. The analysis indicates that due to the combined effects of extrusion and the elevated temperature of the titanium alloy, some titanium alloy chips adhere to the surface of the ball-end milling cutter. This adhesion increases the shear stress on the coating during milling. When the shear stress exceeds the bonding strength between the coating and the substrate, the coating is torn, losing its protective function, which results in adhesive wear. In addition, since the hardness of titanium alloy is lower than that of cemented carbide, the surface of the titanium alloy workpiece is damaged and torn first during the milling process. As the milling progresses, chips are formed, which continuously shear the coating surface and create furrows. As wear progresses, sections of the coating surface may peel off, forming abrasive particles that embed into the worn surface, leading to secondary micro-grinding.

Range and variance analyses were performed on the wear value of the ball-end milling cutter. The significant findings and the influence trends of each parameter on surface wear are presented in Figure 8. The influence mechanism of coating and texture preparation parameters on tool wear is shown in Figure 9. In the following analysis, the influence mechanism in Figure 9 is explained in detail.

The scanning times and scanning speed have the most significant effect on surface wear, with p-values of 0.012 and 0.027, respectively, obtained from one-way ANOVA. As the scanning speed increases from 1500 to 1700 mm/min and the number of scanning passes increases from 6 to 8, the average surface wear increases by approximately 30 μm in both cases. The underlying mechanism is as follows: under the influence of laser irradiation and tool friction, the titanium (Ti) element in the surface and intermediate layers of the coating is oxidized to form a small amount of TiO₂. Studies have shown that excessive TiO₂ content leads to the volumetric expansion of the coating, resulting in compressive stress and local stress concentration on the coating surface, which weakens the adhesive strength of the coating and reduces its wear resistance [31,32]. However, the Al element on the coating surface undergoes an oxidation reaction under the influence of laser and the friction on rake face, forming an Al₂O₃ film. This film reduces friction and wear during the cutting process. At the same time, it prevents the oxidation of the Ti element within the coating, thus protecting its internal structure. Therefore, optimizing scanning times and speed can enhance the processing of surface texture, promote the formation of the Al₂O₃ film, and reduce surface wear. Conversely, an increase in scanning speed and times may shorten the surface texturing time and reduce the Al₂O₃ film thickness, leading to a higher TiO₂ content in the coating’s internal compounds and a decrease in surface hardness and the bonding strength of the film. Both surface hardness and the film-to-substrate bonding strength are critical factors influencing the wear resistance of the ball-end milling cutter [33].

The effect of coating thickness on surface wear is significant with a p-value of 0.01 obtained from one-way ANOVA. When the coating thickness increases from 2.6 to 3 μm, the surface wear decreases by approximately 40 μm. The underlying reason for this trend is that the AlSiTiN coating can effectively enhance tool life, and thus, a thicker coating provides better protection for the milling cutter. As the coating thickness increases, the relative depth of the surface texture decreases, the cooling rate after laser treatment increases, the surface grain size of the coating becomes finer, the surface hardness improves, and the wear resistance of the coating is enhanced. Moreover, the reduced relative depth of the surface texture minimizes the oxidation effect of the laser on the internal structure of the coating, which helps to decrease the formation of TiO₂. This also improves the adhesion between the surface texture, the coating, and the substrate, reducing adhesive wear on the rake face of the ball-end milling cutter. The combined effects of these factors significantly reduce surface wear.

3.4. Analysis of Surface Roughness of Titanium Alloy Workpieces

The surface roughness of the workpiece, measured using a surface roughness meter after milling with ball-end milling cutters, is presented in

Table 6. Surface roughness test results.

Number of Tests	Ra Average Value (μm)	Number of Tests	Ra Average Value (μm)	Number of Tests	Ra Average Value (μm)
1	0.287	10	0.223	19	0.305
2	0.221	11	0.305	20	0.306
3	0.290	12	0.194	21	0.304
4	0.284	13	0.241	22	0.207
5	0.251	14	0.317	23	0.327
6	0.239	15	0.349	24	0.244
7	0.279	16	0.292	25	0.152
8	0.312	17	0.260	26	0.286
9	0.313	18	0.455	27	0.352

.

Range and variance analyses of the workpiece surface roughness were conducted, and the significant findings along with the influence trends of each parameter on surface roughness are presented in Figure 10.

The distance from the blade has the most significant effect on the surface roughness of the workpiece. The p-value obtained from the one-way ANOVA is 0.162. The results indicate that as the distance from the edge increases from 90 to 110 μm, the surface roughness of the workpiece increases by 0.05 μm. This effect is attributed to the fact that the influence of blade distance on surface roughness primarily depends on the extent of the textured area engaged in the milling process. The surface texture on the coated tool, as previously mentioned, reduces the contact area with the chip and facilitates chip storage. When the distance from the blade is smaller, a larger portion of the textured surface engages in the milling process, leading to a reduction in friction for the ball-end milling cutter and a decrease in milling temperature. Since the heat generated during milling directly impacts the surface roughness of the workpiece, a decrease in milling temperature results in an increase in surface roughness [34]. Similarly, as the distance from the blade increases, the area of the texture involved in milling decreases, resulting in a reduction in surface quality.

The laser power has a significant impact on the surface roughness of the workpiece among the preparation parameters with a p-value of 0.233 obtained from the one-way ANOVA. When the laser power increases from 35 to 40 W, the surface roughness of the workpiece increases by 0.066 μm. However, when the laser power is further increased to 45 W, no significant change in the surface quality of the workpiece is observed. This is because laser power is the primary factor controlling the energy delivered to the coating surface, and variations in power directly affect the amount of energy acting on the surface. As laser power increases, the energy imparted to the coating per unit time rises, leading to a decrease in the cooling rate of the coating surface. This results in larger grain sizes within the compound, an increase in intergranular defects, reduced hardness, greater surface friction on the ball-end milling cutter, and elevated heat generation during the milling process [35]. Additionally, as laser power increases, the extent of coating ablation intensifies, leading to a higher deposition of cladding around the texture, which aggravates vibrations during the milling process [36]. Consequently, as laser power increases, the surface roughness decreases, leading to an improvement in surface quality. Once the laser power reaches a certain threshold, the depth of the surface texture increases, and the remelted metal at the bottom becomes difficult to accumulate on the surface of the coating. As a result, no significant change in the surface quality of the workpiece is observed after milling [37].

The number of scanning passes also significantly affects the surface roughness of the workpiece with a p-value of 0.243 obtained from the one-way ANOVA. When the number of scanning passes increased from 6 to 7, the surface roughness of the workpiece increased by 0.047 μm. However, when the number of scanning passes increased to 8, no significant change in the surface quality of the workpiece was observed. This can be attributed to the heat accumulation on the surface caused by repeated laser scanning. Similar to laser power, this accumulated heat affects the laser energy delivered to the surface of the coating [38]. Therefore, the mechanism by which scanning passes influence the surface quality of the workpiece is similar to that of laser power.

3.5. Optimization of Surface Texture and Coating Parameters of Cemented Carbide Ball-End Milling Cutter

3.5.1. Establishment of Regression Model

The relationship between each influencing factor and the experimental index is determined based on the empirical formula commonly used in milling performance studies [39]. A multiple regression model is established for coating thickness (h), laser power (p), scanning speed (v), number of scanning passes (n), texture diameter (d), texture spacing (l), and distance from the edge (l₁), with milling force, milling temperature, tool surface wear, and machined surface quality as the dependent variables [40]. The resulting regression models for milling force F, milling temperature T, workpiece surface roughness Ra, and tool wear KB are as follows:

$$F={543.081}^{-1}{h}^{81.2444}{p}^{2.0287}{v}^{-3.6957\times {10}^{-3}}{n}^{1.7581}{d}^{-0.1306}{l}^{0.3174}{l}_1^{-0.0966}⊲$$

(1)

$$T=110.726h^{-1.146}{p}^{1.09}{v}^{0.05}{n}^{-2.07}{d}^{0.1757}{l}^{-0.1169}{l}_1^{0.0217}$$

(2)

$$KB=10.1927h^{-5.0821}{p}^{1.2893}{v}^{0.0561}{n}^{2.707}{d}^{0.3221}{l}^{0.0542}{l}_1^{-0.1456}$$

(3)

$$Ra=0.3886h^{-0.0856}{p}^{-3.4776\times {10}^{-3}}{v}^{-1.669\times {10}^{-4}}{n}^{0.0232}{d}^{-2.41\times {10}^{-4}}{l}^{1.017\times {10}^{-3}}{l}_1^{2.611\times {10}^{-3}}$$

(4)

To evaluate the significance of the regression models, it is essential to consider both the p-value and the F-value [41]. Referring to the significance test results in

Table 7. Significance test of different evaluation index regression models.

Optimization Object	F-Statistic	p-Values
Milling force	2.8762	0.0417
Milling temperature	3.048	0.039
Workpiece surface roughness	2.8836	0.0377
Tool wear	3.098	0.036

, it is evident that the statistical parameters F and p-values for the four regression models meet these criteria. This confirms that the proposed regression models for milling performance are statistically significant and effective. The residuals of the regression models are analyzed using the Shapiro–Wilk test to assess the normality of the residuals for all four models. The results indicate that the residuals of all four regression models follow a normal distribution. The Quantile–Quantile plots of the residuals for the four models are presented in Figure 11.

3.5.2. Parameter Optimization Results

The artificial bee colony (ABC) algorithm [42,43] is an optimization method based on swarm intelligence, which is inspired by the foraging behavior of bees. Researchers have demonstrated that this algorithm outperforms or is comparable to other optimization algorithms when applied to multi-modal and multi-dimensional numerical optimization problems [44]. As a result, it has been widely used in optimizing process parameters in machining performance studies [45,46]. The artificial bee colony (ABC) algorithm is used to optimize the coating and surface texture parameters, integrating the established regression model into the algorithm within MATLAB. The optimization flow chart of the algorithm is shown in Figure 12a. In the algorithm, the colony size is set to 100, the maximum number of iterations is 100, and the nectar period restriction is limited to 10. To improve the speed and accuracy of the algorithm’s optimization, the ranges of coating and laser parameters corresponding to each evaluation index were applied as constraints. The optimization results are as follows: when the coating thickness (h) = 3.0 μm, the laser power (p) = 40 W, the scanning speed (v) = 1590 mm/min, the scanning times (n) = 6, the texture diameter (d) = 42 μm, texture distance (l) = 143 μm, and the distance from the blade (l₁) = 104 μm, the milling force, milling temperature, workpiece surface roughness, and tool wear are minimized.

3.5.3. Optimization Results Validation

The surface-textured AlSiTiN-coated cemented carbide ball-end milling cutter was re-prepared using the optimized parameter combination and conduct cutting tests again with the same cutting parameters. As shown in Figure 12b, the milling performance of the optimized milling cutter is significantly better than that of the unoptimized milling cutter. Specifically, the milling force is reduced by 17% compared to the average milling force of the previous 27 milling cutter groups, the milling temperature is reduced by 5% compared to the average milling temperature of the previous 27 milling cutter groups, the wear of the milling cutter is reduced by 14% compared to the average wear of the previous 27 milling cutter groups, and the surface roughness of the workpiece is reduced by 30% compared to the average surface roughness of the previous 27 milling cutter groups, verifying the accuracy of the optimization results.

4. Conclusions

(1) The AlSiTiN coating on the surface of ball-end milling cutter reduces thermal adhesion between the rake face and the chip. Surface texture on the coating surface of the ball-end milling cutter enhances heat dissipation from the rake face, reduces the actual contact area between the rake face and chip, and improves the ability of the coating to store chips. This effectively reduces the milling force, milling temperature, and surface wear, leading to improved workpiece surface quality. In addition, the surface texture of the coating on the ball-end milling cutter improves the wear resistance of the coating surface and delays the peeling of the coating.

(2) By establishing a milling test platform for titanium alloy using surface-textured AlSiTiN-coated cemented carbide ball-end milling cutters, the influence mechanisms of various coating and texture parameters on the milling performance of the ball-end milling cutter were analyzed. The results indicate that the coating thickness significantly affects the milling force, milling temperature, and surface wear. Increasing the coating thickness enhances surface hardness and strengthens the anchoring effect of the surface texture on the coating. Additionally, the increase in coating thickness not only allows the surface texture to fully exert its anti-wear and anti-friction properties but also enhances the surface integrity of the coating. Moreover, laser power has a significant impact on milling temperature. Both the number of scans and scanning speed substantially influence surface wear, while the distance from the blade plays a critical role in determining the surface roughness of the workpiece.

(3) Based on the milling test results, a regression model was developed to map the relationship between milling performance evaluation indices and the parameters of the surface texture and coating. The coating and surface texture parameters were subsequently optimized using the artificial bee colony algorithm. The results show that when the coating thickness (h) = 3.0 μm, laser power (p) = 40 W, scanning speed (v) = 1590 μm/min, scanning times (n) = 6, texture diameter (d) = 42 μm, texture distance (l) = 143 μm, and distance from the edge (l₁) = 104 μm, the milling performance of the optimized milling cutter is significantly superior to that of the unoptimized milling cutter. Specifically, the milling force is reduced by 17% compared to the average force of the first 27 milling cutter groups, the milling temperature is reduced by 5% relative to the average temperature of the first 27 milling cutter groups, surface wear is decreased by 14% compared to the average surface wear of the first 27 milling cutter groups, and the surface roughness of the workpiece is reduced by 30% compared to the average surface roughness of the first 27 milling cutter groups.

(4) In this study, a surface-textured coating was applied to the rake face of a cemented carbide ball-end milling cutter, which significantly enhanced the milling cutter’s lifespan and improved the surface quality of titanium alloy workpieces. This research contributes to the advancement of tool surface-texturing coating technology. Future research should focus on further investigating the evolution of surface micro-characteristics during the preparation of surface-textured coating tools with the goal of precisely controlling the influence of these micro-characteristics on the cutting performance of titanium alloys.