Prediction of surface roughness in CNC end milling by machine vision system using artificial neural network based on 2D Fourier transform

This paper presents a system for automated, non-contact, and flexible prediction of surface roughness of end-milled parts through a machine vision system which is integrated with an artificial neural network (ANN). The images of milled surface grabbed by the machine vision system could be extracted using the algorithm developed in this work, in the spatial frequency domain using a two-dimensional Fourier transform to get the features of image texture (major peak frequency F ₁, principal component magnitude squared value F ₂, and the average gray level G _a). Since F1 is the distance between the major peak and the origin, it is a robust measure to overcome the effect of lighting of the environment. The periodically occurring features such as feed marks and tool marks present in the gray-level image can be easily observed from the principal component magnitude squared value F _2. The experimental machining variables speed S, feedrate F, depth of cut D, and the response extracted image variables F ₁, F ₂, and G _a could be used as input data, and the response surface roughness R _a measured by Surfcorder SE-1100 (traditional stylus method) could be used as output data of an ANN ability to construct the relationships between input and output variables. The ANN was trained using the back-propagation algorithm developed in this work due to its superior strength in pattern recognition and reasonable speed. Using the trained ANN, the experimental result had shown that the surface roughness of milled parts predicted by machine vision system over a wide range of machining conditions could be got with a reasonable accuracy compared with those measured by traditional stylus method. Compared with the stylus method, the constructed machine vision system is a useful method for prediction of the surface roughness faster, with a lower price, and lower environment noise in manufacturing process. Experimental results have shown that the proposed machine vision system can be implemented for automated prediction of surface roughness with accuracy of 97.53%. The results are encouraging that machine vision system can be extended to many real-time industrial prediction applications.     

Investigation for optimal parametric combination for achieving better surface finish during turning of Al/SiC-MMC

This paper presents an experimental investigation of the influence of cutting conditions on surface finish during turning of Al/SiC-MMC. In this study, the Taguchi method, a powerful tool for experiment design,is used to optimise cutting parameters for effective turning of Al/SiC-MMC using a fixed rhombic tooling system. An orthogonal L₂₇(3¹³) array is used for 3³ factorial design and analysis of variance (ANOVA) is employed to investigate the influence of cutting speed, feed and depth of cut on the surface roughness height R _a and R _t respectively. The influence of the interaction of cutting speed/feed on the surface roughness height R _a and R _t and the effect of cutting speed on cutting speed/feed two factor cell total interaction for surface roughness height R _a and R _t are analysed through various graphical representations. Taking significant cutting parameters into consideration and using multiple linear-regression, mathematical models relating to surface roughness height R _a and R _t are established to investigate the influence of cutting parameters during turning of Al/SiC-MMC. Confirmation test results established the fact that the mathematical models are appropriate for effectively representing machining performance criteria, e.g. surface roughness heights during turning of Al/SiC-MMC.     

An integrated optimization of cutting parameters and tool path generation in ultraprecision raster milling

This paper provides a new methodology for the integrated optimization of cutting parameters and tool path generation (TPG) based on the development of prediction models for surface roughness and machining time in ultraprecision raster milling (UPRM). The proposed methodology simultaneously optimizes the cutting feed rate, the path interval, and the entry distance in the feed direction to achieve the best surface quality in a given machining time. Cutting tests are designed to verify the integrated optimization methodology. The experimental results show that, in the fabrication of plane surface, the changing of entry distance improves surface finish about 40 nm (R _a ) and 200 nm (R _t ) in vertical cutting and decreases about 8 nm (R _a ) and 35 nm (R _t ) in horizontal cutting with less than 2 s spending extra machining time. The optimal shift ratio decreases surface roughness about 7 nm (R _a ) and 26 nm (R _t ) in the fabrication of cylinder surfaces, while the total machining time only increases 2.5 s. This infers that the integrated optimization methodology contributes to improve surface quality without decreasing the machining efficiency in ultraprecision milling process.     

Optimum selection of abrasive flow machining conditions during fine finishing of Al/15 wt% SiC-MMC using Taguchi method

Abrasive flow machining (AFM) is gaining widespread application finishing process on difficult to reach surfaces in aviation, automobile, and tooling industry. Al/SiC_p-MMC is a promising material in these industries. Here, AFM has been used to finish conventionally machined cylindrical surface of Al/15 wt% SiC_p-MMC workpiece. This paper presents the utilization of robust design-based Taguchi method for optimization of AFM parameters. The influences of AFM process parameters on surface finish and material removal have been analyzed. Taguchi experimental design concept, L₁₈ (6¹?×?3⁷) mixed orthogonal array is used to determine the S/N ratio and optimize the AFM process parameters. Analysis of variance and F test values also indicates the significant AFM parameters affecting the finishing performance. The mathematical models for R _a, R _t, ΔR _a, and ΔR _t and material removal are established to investigate the influence of AFM parameters. Conformation test results verify the effectiveness of these models and optimal parametric combination within the considered range. Scanning electron micrographs testifies the effectiveness of AFM process in fine finishing of Al/15 wt% SiC_p-MMC.     

Investigation of wire electrical discharge machining characteristics of Al6063/SiCp composites

In this investigation, the effect of wire electrical discharge machining (WEDM) parameters such as pulse-on time (T _ON), pulse-off time (T _OFF), gap voltage (V) and wire feed (F) on material removal rate (MRR) and surface roughness (R _a) in metal matrix composites (MMCs) consisting of aluminium alloy (Al6063) and silicon carbide (SiC_p) is discussed. The Al6063 is reinforced with SiC_p in the form of particles with 5%, 10% and 15% volume fractions. The experiments are carried out as per design of experiments approach using L₉ orthogonal array. The results were analysed using analysis of variance and response graphs. The results are also compared with the results obtained for unreinforced Al6063. From this study, it is found that different combinations of WEDM process parameters are required to achieve higher MRR and lower R _a for Al6063 and composites. Generally, it is found that the increase in volume percentage of SiC resulted in decreased MRR and increased R _a. Regression equations are developed based on the experimental data for the prediction of output parameters for Al6063 and composites. The results from this study will be useful for manufacturing engineers to select appropriate WEDM process parameters to machine MMCs of Al6063 reinforced with SiC_p at various proportions.     

Grinding-aided electrochemical discharge machining of particulate reinforced metal matrix composites

A grinding-aided electrochemical discharge machining (G-ECDM) process has been developed to improve the performance of the conventional ECDM process in machining particulate reinforced metal matrix composites (MMCs). The G-ECDM process functions under a combined action of electrochemical dissolution, spark erosion, and direct mechanical grinding. The tool electrode has a coating containing a hard reinforcement phase of diamond particles. The MMC employed in this study was Al₂O₃ particulate reinforced aluminum 6061 alloy. The material removal mechanism of this hybrid process has been analyzed. The results showed that the grinding action can effectively remove re-cast material deposited on the machining surface. The surface roughness (R _a) measured for the G-ECDM specimen was ten times smaller than that of the specimen machined without grinding aid (i.e., ECDM alone). Moreover, the material removal rate (MRR) of G-ECDM was about three times higher than that of ECDM under the experimental conditions of this study. The voltage waveform and crater distribution were also analyzed, and the experimental results showed that the G-ECDM process operates in a stable condition. The relative importance of the various processing parameters on MRR was established using orthogonal analysis. The results showed that MRR is influenced by the machining parameters in the order of duty cycle?>?current?>?electrolyte concentration. This study showed that the G-ECDM process is superior to the ECDM process for machining particulate reinforced MMCs, where a higher machining efficiency and a better surface quality can be obtained.     

Turning titanium metal matrix composites (Ti-MMCs) with carbide and CBN inserts

Despite excellent mechanical and physical features of titanium metal matrix composite (Ti-MMC), hard and abrasive ceramic particles within the matrix structure, as well as high price, may lead to severe difficulties on machining and machinability of Ti-MMCs. Review of literature denotes that only limited studies are available on machining Ti-MMCs with commercial cutting tools under various cutting conditions and cutting tools/inserts. Furthermore, limited studies are available on machinability attributes of Ti-MMC under various cutting conditions used. Therefore, to remedy the lack of knowledge observed, this work intends to report turning Ti-MMCs with carbide, and cubic boron nitride (CBN) inserts under various cutting conditions. The mean values of surface roughness (R_a) and directional cutting forces, as well as flank wear (V_B) were studied as the machinability attributes. The microstructural evaluations were conducted to discover the wear modes. Furthermore, the statistical tools were used to present the factors governing machining attributes studied. Adhesion, abrasion, and oxidation were observed as the principle wear modes on the flank sides of the tested inserts. According to experimental results and statistical analysis, the R_a and V_B could be controlled by cutting parameters only when CBN inserts were used. Despite the inset used, factors governing both responses were not identical. Although average cutting forces were directly affected by cutting parameters used, however, the relatively low correlation of determination (R²) of directional cutting forces can be attributed to effects of cutting speed, elevated temperature in the cutting zone as well as rapid tool wear which are all correlated to others.     

Surface roughness and roundness of bearing raceway machined by floating abrasive polishing and their effects on bearing’s running noise

As the two most important indexes of bearing raceway, surface roughness and roundness have significant influence on bearing noise. Some researchers have carried out studies in this field, however, reason and extent of the influence of raceway surface geometric characteristics on bearing running noise are not perfectly clear up to now. In this paper, the raceway of 6309 type bearing＇s inner and outer ring is machined by floating abrasive polishing adopting soft abrasive pad. Surface roughness parameters, arithmetical mean deviation of the profile Ra, the point height of irregularities Rz, maximum height of the profile Rmax and roundness fof raceways, are measured before and after machining, and the change rules of the measured results are studied. The study results show that the floating abrasive polishing can reduce the surface geometric errors of bearing raceway evidently. The roundness error is reduced by 25%, Rm~x value is reduced by 35.5%, Rz value is reduced by 22% and Ra value is reduced by 5%. By analyzing the change of the geometrical parameters and the shape difference of the raceway before and after machining, it is found that the floating abrasive polishing method can affect the roundness error mainly by modifying the local deviation of the raceway＇s surface profile. Bearings with different raceway surface geometrical parameter value are assembled and the running noise is tested. The test results show that Ra has a little, Rmax and Rz have a measurable, and the roundness error has a significant influence on the running noise. From the viewpoint of controlling bearings＇ running noise, raceway roundness error should be strictly controlled, and for the surface roughness parameters, R,n~x and Rz should be mainly controlled. This paper proposes an effective method to obtain the low noise bearing by machining the raceway with floating abrasive polishing after super finishing.     

Application of Taguchi method for determining optimum surface roughness in turning of high-alloy white cast iron

This paper focused on optimizing the cutting conditions for the average surface roughness (R_a) obtained in machining of high-alloy white cast iron (Ni-Hard) at two different hardness levels (50 HRC and 62 HRC). Machining experiments were performed at the CNC lathe using ceramic and cubic boron nitride (CBN) cutting tools on Ni-Hard materials. Cutting speed, feed rate and depth of cut were chosen as the cutting parameters. Taguchi L₁₈ orthogonal array was used to design of experiment. Optimal cutting conditions was determined using the signal-to-noise (S/N) ratio which was calculated for R_a according to the “the-smaller-the-better” approach. The effects of the cutting parameters and tool materials on surface roughness were evaluated by the analysis of variance. The statistical analysis indicated that the parameters that have the biggest effect on R_a for Ni-Hard materials with 50 HRC and 62 HRC are the cutting speed and feed rate, respectively. Additionally, the optimum cutting conditions for the materials with 50 HRC and 62 HRC was found at different levels.     

The ability of precision hard turning to increase rolling contact fatigue life

In this paper, precision hard turning is proposed for the finishing of the AISI 52100 bearing components to improve rolling contact fatigue life. This finishing process induces a homogenous microstructure at surface and subsurface layers. Fatigue life tests performed on a twin-disk machine show that rolling contact fatigue life increases as R_a value decreases. The bearing components reached 0.32 million cycles for R_a=0.25 μm and 5.2 million cycles for R_a=0.11 μm. In comparison, the bearing components achieved 1.2 million cycles with grinding (R_a=0.2 μm) and 3.2 million cycles with grinding followed by honing (R_a=0.05 μm) respectively.     

A multivariate surface roughness modeling and optimization under conditions of uncertainty

Correlated responses can be written in terms of principal component scores, but the uncertainty in the original responses will be transferred and will influence the behavior of the regression function. This paper presents a model building strategy that consider the multivariate uncertainty as weighting matrix for the principal components. The main objective is to increase the value of R² predicted to improve model’s explanation and optimization results. A case study of AISI 52100 hardened steel turning with Wiper tools was performed in a Central Composite Design with three-factors (cutting speed, feed rate and depth of cut) for a set of five correlated metrics (R_a, R_y, R_z, R_q and R_t). Results indicate that different modeling methods conduct approximately to the same predicted responses, nevertheless the response surface to Weighted Principal Component – case b – (WPC1^b) presented the highest predictability.     

Simulation of surface generated during abrasive flow finishing of Al/SiCp-MMC using neural networks

Abrasive flow machining (AFM) is a multivariable finishing process which finds its use in difficult to finish surfaces on difficult to finish materials. Near accurate prediction of generated surface by this process could be very useful for the practicing engineers. Conventionally, regression models are used for such prediction. This paper presents the use of artificial neural networks (ANN) for modeling and simulation of response characteristics during AFM process in finishing of Al/SiC_p metal matrix composites (MMCs) components. A generalized back-propagation neural network with five inputs, four outputs, and one hidden layer is designed. Based upon the experimental data of the effects of AFM process parameters, e.g., abrasive mesh size, number of finishing cycles, extrusion pressure, percentage of abrasive concentration, and media viscosity grade, on performance characteristics, e.g., arithmetic mean value of surface roughness (R _a, micrometers), maximum peak–valley surface roughness height (R _t, micrometers), improvement in R _a (i.e., ΔR _a), and improvement in R _t (i.e., ΔR _t), the networks are trained for finishing of Al/SiC_p-MMC cylindrical components. ANN models are compared with multivariable regression analysis models, and their prediction accuracy is experimentally validated.     

Surface roughness measurement in turning carbon steel AISI 1045 using wiper inserts

Surface roughness of the workpiece is an important parameter in machining technology. Wiper inserts have emerged as a significantly class of cutting tools, which are increasingly being utilized in last years. This study considers the influence of the wiper inserts when compared with conventional inserts on the surface roughness obtained in turning. Experimental studies were carried out for the carbon steel AISI 1045 because of its great application in manufacturing industry. Surface roughness is represented by different amplitude parameters (R_a, R_zD, R_3z, R_q, R_t, R_a/R_q, R_q/R_t, R_a/R_t). With wiper inserts and high feed rate it is possible to obtain machined surfaces with R_a < 0.8 μm (micron). Consequently it is possible to get surface quality in workpiece of mechanics precision without cylindrical grinding operations.     

Investigations on machined metal surfaces through the stylus type and optical 3D instruments and their mathematical modeling with the help of statistical techniques

The measurement of roughness on machined metal surfaces is of considerable importance to manufacturing industries as the roughness of a surface has a significant influence on its quality and function of products. In this paper, an experimental approach for surface roughness measurement has been based on the comparison of roughness values taken from the stylus and optical type instruments on the machined metal surfaces (turning, grinding and milling) is presented.Following this experimental study, all measured surface roughness parameters have been analyzed by using Statistical Package for Social Science (SPSS 15.0) statistically and mathematical models for the two most important and commonly used roughness parameters R_a and R_z have been developed so that R_a = R_a (F, P, C) and R_z = R_z (F, P, C, M), whereas F expresses feed, P periodicity, C contrast and M the type of material. The statistical results from numerous tests showed that there has been a correlation between the surface roughness and the properties of the surface topography and there have been slight differences among three measurement instruments on machined metal surfaces in this experimental study.     

Investigation of machining parameters for the multiple-response optimization of micro electrodischarge milling

This paper investigated the influence of three micro electrodischarge milling process parameters, which were feed rate, capacitance, and voltage. The response variables were average surface roughness (R _a ), maximum peak-to-valley roughness height (R _y ), tool wear ratio (TWR), and material removal rate (MRR). Statistical models of these output responses were developed using three-level full factorial design of experiment. The developed models were used for multiple-response optimization by desirability function approach to obtain minimum R _a , R _y , TWR, and maximum MRR. Maximum desirability was found to be 88%. The optimized values of R _a , R _y , TWR, and MRR were 0.04, 0.34 μm, 0.044, and 0.08 mg min^?1, respectively for 4.79 μm s^?1 feed rate, 0.1 nF capacitance, and 80 V voltage. Optimized machining parameters were used in verification experiments, where the responses were found very close to the predicted values.     

Comparison of different laser-assisted electrochemical methods based on surface morphology characteristics

This paper presents two different laser electrochemical machining (LECM) methods: point-by-point method and scan method. Experiments about the two LECM methods were performed on the aluminum alloy plates placed in a shallow container filled with NaNO₃ electrolyte 2 mm above. Then, the scanning electron microscopy and the optical profiling system were used to analyze machining quality and morphology characteristics of two different machining methods. For the point-by-point method, the shape accuracy at the starting point, finishing point, and node of line cross is lower. Compared with the point-by-point method, the better etching morphology was gained by the scan method in the experiments. The influence of the scan speed on machining quality and groove width was investigated. The better etching surface quality and the higher etching efficiency were gained when the scan speed varied from 0.1 to 0.15 mm/s. These results demonstrate that laser electrochemical machining by the scan method is a promising method to achieve complex profiles.     

Parametric optimization of CNC turning process for hybrid metal matrix composite

Machining of hybrid metal matrix composite is difficult as the particulates are abrasive in nature and they behave like a cutting edge during machining resulting in quick tool wear and induces vibration. An attempt was made in this experimental study to evaluate the machining characteristics of hybrid metal matrix composite, and a mathematical model was developed to predict the responses, namely surface finish, intensity of vibration and work-tool interface temperature for known cutting condition while machining was performed in computer numerical control lathe. Design of experiments approach was used to conduct the trials; response surface methodology was employed to formulate a mathematical model. The experimental study inferred that the vibration in V _x, V _y, and V _z were 41.59, 45.17, and 26.45 m/s², respectively, and surface finish R _a, R _q, and R _z were 1.76, 3.01, and 11.94 μm, respectively, with work-tool interface temperature ‘T’ of 51.74 °C for optimal machining parameters, say, cutting speed at 175 m/min, depth of cut at 0.25 mm and feed rate at 0.1 mm/rev during machining. Experimental results were in close conformity with response surface method overlay plot for responses.     

Improvement of grinding performance and investigation of cutting parameters using a grinding mechanism with secondary rotational axis

“Grinding Mechanism having Advanced Secondary Rotational Axis” (GMASRA) is one of the newer plane surface grinding methods that have an uncommon abrasion mechanism. Unlike conventional methods, in GMASRA, there are two rotations of a wheel. The first rotation is the same as in conventional grinding methods, which is the circumferential rotation. The other rotation is the newly developed axial rotation, where the wheel rotates around itself perpendicular to its radial axis. In this study, the effects of certain cutting parameters on arithmetical mean deviation of the assessed profile (the R_a parameter) were investigated. Particularly, the effects of cutting parameters on R_a in the GMASRA grinding process were examined. The selected cutting parameters were the depth of cut, the number of axial revolutions of the wheel, and the stepover distance of the wheel. Five wheels with different properties were chosen. Additionally, GMASRA was modeled using the Taguchi orthogonal test design. In this orthogonal design, the depth of cut, the spindle speed, and the type of grinding wheel were chosen as the control factors. The effect of the specified control factors on the surface roughness was demonstrated using an analysis of variance (ANOVA) test. Results show that GMASRA produced better R_a values than the conventional method. R_a values were very close to each other in every part of the ground workpieces. According to the modeling results, the spindle speed had the highest effect on R_a, followed by the depth of cut and the type of grinding wheel. GMASRA is also very cost effective and can be adapted to most milling machines and CNC milling machines.     

Strength modeling and optimizing ultrasonic welded parts of ABS-PMMA using artificial intelligence methods

This study analyzes variations in metal removal rate (MRR) and quality performance of roughness average (R _a) and corner deviation (CD) depending on parameters of wire electrical discharge machining (WEDM) process in relation to the cutting of pure tungsten profiles. A hybrid method including response surface methodology (RSM) and back-propagation neural network (BPNN) integrated simulated annealing algorithm (SAA) were proposed to determine an optimal parameter setting. The results of 18 experimental runs via a Taguchi orthogonal table were utilized to train the BPNN to predict the MRR, R _a, and CD properties. Simultaneously, RSM and SAA approaches were individually applied to search for an optimal setting. In addition, analysis of variance was implemented to identify significant factors for the processing parameters. Furthermore, the field-emission scanning electron microscope images show that a lot of built-edge layers were presented on the finishing surface after the WEDM process. Finally, the optimized result of BPNN with integrated SAA was compared with that obtained by an RSM approach. Comparisons of the results of the algorithms and confirmation experiments show that both RSM and BPNN/SAA methods are effective tools for the optimization of parameters in WEDM process.     

Enhancing AFM process productivity through improved fixturing

Abrasive flow machining (AFM) is a non-conventional finishing process that deburrs and polishes by forcing an abrasive laden media across the workpiece surface. The process embraces a wide range of applications from critical aerospace and medical components to high-production volumes of parts. One serious limitation of this process is its low productivity in terms of rate of improvement in surface roughness. Limited efforts have hitherto been directed towards enhancing the productivity of this process with regard to better quality of workpiece surface. This paper discusses improved fixturing as a technique for productivity enhancement in terms of surface roughness (R _a). A rotating centrifugal-force-generating (CFG) rod is used inside the cylindrical workpiece which provides the centrifugal force to the abrasive particles normal to the axis of workpiece. The effect of the key parameters on the performance of process has been studied. The results shows that for a given improvement in R _a value, the processing time can be reduced by as much as 70–80%. It is seen that the significant process parameters are revolutions per minute of CFG rod, extrusion pressure and abrasive mesh size.