Application of GONNS to predict constrained optimum surface roughness in face milling of high-silicon austenitic stainless steel

Surface roughness is a technical requirement for machined products and one of the main product quality specifications. In the present research, a genetically optimized neural network system (GONNS) is proposed for prediction of constrained optimal cutting conditions in face milling of a high-silicon austenitic stainless steel (UNS J93900) in order to minimize surface roughness. In order to attain minimum operation numbers and decrease the cost of machining, an experimental scheme was arranged by using Taguchi method. The considered parameters were cutting speed, feed, depth of cut, and engagement. Cutting force components and surface roughness were measured, and then analysis of variance is performed. The results show that the feed is the dominant factor affecting the surface roughness. Backpropagation artificial neural network was utilized to create predictive models of surface roughness and cutting forces exploiting the experimental data, and the genetic algorithm was employed to find the constrained optimum of surface roughness. Finally, in order to validate the method, an experiment with the obtained optimal cutting condition was carried out, and the results were compared with the predicted value of surface roughness. The corresponding results show the capability of GONNS to predict constrained surface roughness.     

Time series modeling and spectrum analysis for chatter mode in endmilling dynamics

To analyze the chatter mechanics, the power spectrum of a time series method was analyzed by considering cutting and structural dynamics. In this study, several time series models such as AR (burg, least square, yule walker, geometric lattice, instrument variable), ARX (arx, iv4), ARMAX, ARMA, Box Jenkins, Output Error were modeled and compared with one another. Finally, it was proved that arx, armax and iv4 are more desirable and reliable algorithms than the others for the calculation of the chatter mode in the endmilling operation. The cutting forces Fx and Fz have more powerful effects on the chatter than Fy in the sense that there is no shifting or pseudo mode in the power spectrum.     

The design and optimization of micro polycrystalline diamond ball end mill for repairing micro-defects on the surface of KDP crystal

Micro-milling is a promising approach to repair the micro-defects on the surface of KH₂PO₄ (KDP) crystal. The geometrical parameters of micro ball end mill will greatly influence the repairing process as a result of the soft brittle properties of KDP crystal. Two types of double-edged micro ball end mills were designed and a three-dimensional finite element (FE) model was established to simulate the micro milling process of KDP crystal, which was validated by the milling experiments. The rake angle of −45°, the relief angle of 45° and the cutting edge radius of 1.5–2 μm were suggested to be the optimal geometrical parameters, whereas the rake angle of −25° and the relief angle of 9° were optimal just for micro ball end mill of Type I, the configuration with the rake angles ranging from 0° to 35°, by fully considering the cutting force, and the stress–strain distribution over the entire tool and the cutting zone in the simulation. Moreover, the micro polycrystalline diamond (PCD) ball end mills adopting the obtained optimal parameters were fabricated by wire electro-discharge machining (WEDM) and grinding techniques, with the average surface roughness Ra of tool rake face and tool flank face ∼0.10 μm, and the cutting edge radius of the tool ∼1.6 μm. The influence of tool's geometrical parameters on the finished surface quality was verified by the cutting experiments, and the tool with symmetric structure was found to have a better cutting performance. The repairing outlines with Ra of 31.3 nm were processed by the self-fabricated tool, which could successfully hold the growth of unstable damage sites on KDP crystal.     

Optimum surface roughness prediction in face milling by using neural network and harmony search algorithm

This paper presents a new approach to determine the optimal cutting parameters leading to minimum surface roughness in face milling of X20Cr13 stainless steel by coupling artificial neural network (ANN) and harmony search algorithm (HS). In this regard, advantages of statistical experimental design technique, experimental measurements, analysis of variance, artificial neural network and harmony search algorithm were exploited in an integrated manner. To this end, numerous experiments on X20Cr13 stainless steel were conducted to obtain surface roughness values. A predictive model for surface roughness was created using a feed forward neural network exploiting experimental data. The optimization problem was solved by harmony search algorithm. Additional experiments were performed to validate optimum surface roughness value predicted by HS algorithm. The obtained results show that the harmony search algorithm coupled with feed forward neural network is an efficient and accurate method in approaching the global minimum of surface roughness in face milling.     

Spectral analysis of malfunction mode in endmilling

For the investigation of the chatter modes, the power spectrum of the parametric time series model was adopted and analyzed at several mixed conditions of different revolution. This paper describes a methodology for an application of several time series such asAR (forward-backward, burg, least square, Yule Walker, geometric lattice, instrumental variable),ARX (least square, instrumental variable),ARMAX, ARMA, Box Jenkins, Output Error. To estimate the chatter mode using their spectral analysis their results were compared with one another. As a result, it was proven that several time series methods can be used for chatter mode estimation. Among them, theARX, ARMAX and instrumental variable methods (iv4) are more desirable and reliable than the other algorithm for the exact calculation of the chatter mode in endmilling. Among three cutting forces, the z direction cutting force,Fz, has more powerful characteristics of chatter occurring than the cutting forces,Fx andFy, in the sense that weak mode is calculated exactly and there is no shifted or pseudo mode in the estimated power spectra of endmilling forces.     

Optimizing end-milling parameters for surface roughness under different cooling/lubrication conditions

The effect of cutting parameters on average surface roughness (Ra) in the different cooling/lubrication conditions, including minimal quantity lubrication, wet and dry cutting, was analyzed in this study. Orthogonal arrays were applied in the design of experiments, and Ti6Al4V end-milling experiments were performed on the Daewoo machining center. The white light interferometer (Wyko NT9300) was used to obtain the 3D profile of machined surface and calculate Ra values. Then, exponential model and quadratic model were proposed to fit the experimental data of surface roughness, respectively. Exponential fit model was employed to determine the significant cutting parameters on average surface roughness. Quadratic fit model was used to optimize the cutting parameters when cutting tool and material removal rate were given. The optimal average surface roughnesses were estimated according to the quadratic model. Finally, the verification experiments were performed, and the experimental results showed good agreement with the estimated results.     

Accuracy Optimisation for Mould Surface Profile Milling

In the profile milling process for mould surfaces, for any type of curve combination, when implementing the milling process on a CNC machine, the surface can be produced from straight lines or curves in a piecewise or continuous milling process. This work employs the features of a CNC machine: 1. To divide the various sizes of curve at different slopes. 2. To use with different milling spindle speeds. 3. To use cutting feeds for actual milling experimentation. The results of the profile dimension accuracy and profile surface roughness from these experiments are then used in a neural network to establish an experiment result and a model for the milling variables. The neural network is composed of a number of functional nodes. Once the milling parameters (spindle speed, feed speed and milling angle) are given, the milling processing performance (the surface roughness, the surface profile-error) can be accurately predicted by the net-work developed. The optimal milling processing parameters can be searched for by a simulation annealing (SA) optimis-ation algorithm with a performance index to obtain a satisfactory mould surface. The experiment is then used to show that improved conditions for mould profile processing can indeed be obtained.     

Soft computing models and intelligent optimization system in electro-discharge machining of SiC/Al composites

In this paper, a multi-variable regression model, a back propagation neural network (BPNN) and a radial basis neural network (RBNN) have been utilized to correlate the cutting parameters and the performance while electro-discharge machining (EDM) of SiC/Al composites. The four cutting parameters are peak current (I_p), pulse-on time (T_on), pulse-off time (T_off), and servo voltage (S_v); the performance measures are material remove rate (MRR) and surface roughness (Ra). By testing a large number of BPNN architectures, 4-5-1 and 4-7-1 have been found to be the optimal one for MRR and Ra, respectively; and it can predict them with 10.61 % overall mean prediction error. As for RBNN architectures, it can predict them with 12.77 % overall mean prediction error. The multivariable regression model yields an overall mean prediction error of 13.93 %. All of these three models have been used to study the effect of input parameters on the material remove rate and surface roughness, and finally to optimize them with genetic algorithm (GA) and desirability function. Then, an intelligent optimization system with graphical user interface (GUI) has been built based on these multi-optimization techniques, in which users can obtain the optimized cutting parameters under the desired surface roughness (Ra).     

An investigation of the machinability of PA 6/nano-CaCO3 composite

Nowadays, polymer nanocomposites have attracted manufacturers’ attention because of their good mechanical, thermal, and physical properties. Over the past decade, the requirement of the direct machining of polymer nanocomposites has increased due to the production of most polymer nanocomposites using the extrusion method in simple cross-section and the increased demand for personalized products. In this paper, the effect of milling parameters (spindle speed and feed per tooth) and nano-CaCO₃ content on the machinability properties of PA 6/nano-CaCO₃ composites was studied by analyzing variance. Harmony search-based neural network (HSNN) was then utilized to create predictive models of surface roughness and total cutting forces from the experimental data. The results revealed that the nano-CaCO₃ content on PA 6 decreased the cutting forces significantly, but did not have a significant effect on surface roughness. Moreover, the results for modeling total cutting forces and surface roughness showed that HSNN is effective, reliable, and authoritative in modeling the surface roughness formation and total cutting force mechanism for end-milling of PA 6/nano-CaCO₃ composites.     

Integrated ANN-LWPA for cutting parameter optimization in WEDM

In this paper, we present a new approach to determinate cutting parameters in wire electrical discharge machining (WEDM), integrated artificial neuron network (ANN), and wolf pack algorithm based on the strategy of the leader (LWPA). The cutting parameters considered in this paper are pulse-on, current, water pressure, and cutting feed rate. Models of the effects of the four parameters on machining time (T_p), machining cost (C_p), and surface roughness (Ra) are mathematically constructed. An ANN-LWPA integration system with multiple fitness functions is proposed to solve the modelling problem. By using the proposed approach, this study demonstrates that T_p, C_p, and Ra can be estimated at 164.1852 min, 239.5442 RMB, and 1.0223 μm in single objective optimization, respectively. For example, as for Ra, integrated ANN-LWPA has optimized the Ra value by the reduction of 0.1337 μm (11.6 %), 0.3377 μm (24.8 %), and 0.105 μm (10.3 %) compared to experimental data, regression model, and ANN model, respectively. Consequently, the ANN-LWPA integration system boasts some advantages over decreasing the value of fitness functions by comparison with the experimental regression model, ANN model, and conventional LWPA result. Moreover, the proposed integration system can be also utilized to obtain multiple solutions by uniform design-based exploration. Therefore, in order to solve complex machining optimization problems, an intelligent process scheme could be integrated into the numeric control system of WEDM.     

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.     

On the prediction of surface roughness in the hard turning based on cutting parameters and tool vibrations

This research work concerns the elaboration of a surface roughness model in the case of hard turning by exploiting the response surface methodology (RSM). The main input parameters of this model are the cutting parameters such as cutting speed, feed rate, depth of cut and tool vibration in radial and in main cutting force directions. The machined material tested is the 42CrMo4 hardened steel by Al₂O₃/TiC mixed ceramic cutting tool under different conditions. The model is able to predict surface roughness of Ra and Rt using an experimental data when machining steels. The combined effects of cutting parameters and tool vibration on surface roughness were investigated while employing the analysis of variance (ANOVA). The quadratic model of RSM associated with response optimization technique and composite desirability was used to find optimum values of cutting parameters and tool vibration with respect to announced objectives which are the prediction of surface roughness. The adequacy of the model was verified when plotting the residuals values. The results indicate that the feed rate is the dominant factor affecting the surface roughness, whereas vibrations on both pre-cited directions have a low effect on it. Moreover, a good agreement was observed between the predicted and the experimental surface roughness. Optimal cutting condition and tool vibrations leading to the minimum surface roughness were highlighted.     

The evaluation of the effects of control factors on surface roughness in the drilling of Waspaloy superalloy

In this study, the effects of cutting tool type (Ct), cutting speed (Vc), feed rate (f) and drill bit angle (A) on the average surface roughness (Ra) were investigated in the drilling of Waspaloy superalloy with coated and uncoated solid carbide drills. Experimental studies were performed in the orthogonal array of L18 (2¹ × 3³) by using Taguchi method. A second order predictive equation was developed with Linear Regression Analysis and coefficient of correlation for Ra calculated as R² = 96.9%. The most effective parameters on Ra were determined as A, f, Vc and Ct with 49.44%, 15.0%, 14.45% and 13.47% contribution ratios, respectively. Ra surface roughness values increased with the increasing tool wear. In this study, the chip formation and tool wear were also evaluated. Three types chip formation such as spiral chip, string chip and short chip were observed in the drilling of Waspaloy with solid carbide drills.     

Sustainability analyses of cutting edge radius on specific cutting energy and surface finish in side milling processes

The aim of this work is to determine the influence of cutting edge radius on the specific cutting energy and surface finish in a mechanical machining process. This was achieved by assessing the direct electrical energy demand during side milling of aluminium AW6082-T6 alloy and AISI 1018 steel in a dry cutting environment using three different cutting tool inserts. The specific energy coefficient was evaluated as an index of the sustainable milling process. The surface finish of the machined parts was also investigated after machining. It was observed that machining with the 48.50-μm cutting edge radius insert resulted in lower specific cutting energy requirements when compared with the 68.50 and 98.72-μm cutting edge radii inserts, respectively. However, as the ratio of the undeformed chip thickness to cutting edge radius is less than 1, the surface roughness increases. The surface roughness values gradually decrease as the ratio of undeformed chip thickness to cutting edge radius (h/r _e) tends to be 1 and at minimum surface roughness values when the ratio of h/r _e equalled to 1. However, the surface roughness values increased as h/r _e becomes higher than 1. This machining strategy further elucidates the black box and trade-offs of ploughing and rubbing characteristics of micro machining and optimization strategy for minimum energy and sustainable manufacture.     

Cutting force analysis in reaming of ZL102 aluminium cast alloys by PCD reamer

The cutting forces are studied for the reaming of casting aluminium alloy ZL102 (ZAlSi12) using a four-flute PCD reamer on DMG DMU 50 linear five-axis CNC machine. Kistler Dynamometer is used to measure the cutting force (X-, Y- and Z-axes) and torque. Three-coordinate measuring machine is used for measuring cylindricity of machined hole. The cutting force signatures are diagnosed by using Fast Fourier Transform (FFT) algorithm. FFT analysis suggests that fundamental frequency is only determined by cutting speed. The frequency for the thrust signal of cutting force is zero which illustrates that the thrust is constant when cutting speed and cutting feed are fixed. The relationship between the cutting force and cutting parameter are investigated by using FFT filtering. High value of cutting feed and the low cutting speed provide the higher thrust. The optimized machining parameters are provided relative to cutting force (Fx and Fy) and the cylindricity of reamed holes.     

Optimization of surface roughness in end milling Castamide

Castamide is vulnerable to humidity up to 7%; therefore, it is important to know the effect of processing parameters on Castamide with and without humidity during machining. In this study, obtained quality of surface roughness of Castamide block samples prepared in wet and dry conditions, which is processed by using the same cutting parameters, were compared. Moreover, an artificial neural network (ANN) modeling technique was developed with the results obtained from the experiments. For the training of ANN model, material type, cutting speed, cutting rate, and depth of cutting parameters were used. In this way, average surface roughness values could be estimated without performing actual application for those values. Various experimental results for different material types with cutting parameters were evaluated by different ANN training algorithms. So, it aims to define the average surface roughness with minimum error by using the best reliable ANN training algorithm. Parameters as cutting speed (V _c), feed rate (f), diameter of cutting equipment, and depth of cut (a _p) have been used as the input layers; average surface roughness has been also used as output layer. For testing data, root mean squared error, the fraction of variance (R ²), and mean absolute percentage error were found to be 0.0681%, 0.9999%, and 0.1563%, respectively. With these results, we believe that the ANN can be used for prediction of average surface roughness.     

Optimisation of parameters affecting surface roughness of Co28Cr6Mo medical material during CNC lathe machining by using the Taguchi and RSM methods

This study involves modelling of experimental data of surface roughness of Co28Cr6Mo medical alloy machined on a CNC lathe based on cutting parameters (spindle rotational speed, feed rate, depth of cut and tool tip radius). In order to determine critical states of the cutting parameters variance analysis (ANOVA) was applied while optimisation of the parameters affecting the surface roughness was achieved with the Response Surface Methodology (RSM) that is based on the Taguchi orthogonal test design. The validity of the developed models necessary for estimation of the surface roughness values (Ra, Rz), was approximately 92%. It was found that for Ra 38% of the most effective parameters is on the tool tip radius, followed by 33% on the feed rate whereas for Rz tool tip radius occupied 43% with the feed being at 33% rate. To achieve the minimum surface roughness, the optimum values obtained for spindle rpm, feed rate, depth of cut and tool tip radius were respectively, 318 rpm, 0.1 mm/rev, 0.7 mm and 0.8 mm.     

Process investigation on meso-scale hard milling of ZrO2 by diamond coated tools

This paper investigates the feasible machining of zirconium oxide (ZrO₂) ceramics, in the hard state, via milling by diamond coated miniature tools (from here on briefly indicated as meso-scale hard milling). The workpiece material is a fully sintered yttria stabilized tetragonal zirconia polycrystalline ceramic (Y-TZP). Diamond coated WC mills, 2 mm in diameter, four flutes and large corner radius (0.5 mm) are chosen as cutting tools, and experiments are conducted on a state-of-the-art micro milling machine centre. The influence of cutting parameters, including axial depth of cut (a_p) and feed per tooth (f_z), on the achievable surface quality is studied by means of a one-factor variation experimental design. Further tests are also conducted to monitor the process performance, including surface roughness, tool wear and machining accuracy, over the machining time. Mirror quality surfaces, with average surface roughness Ra below 80 nm, are obtained on the machined samples; the SEM observations of the surface topography reveal a prevailing ductile cutting appearance. Tool wear initiates with delamination of the diamond coating and progresses with the wear of the WC substrate, with significant effect on the cutting process and its performance. Main applications of this research include three dimensional surface micro structuring and superior surface finishing.     

Prediction and analysis of surface roughness characteristics of a non-ferrous material using ANN in CNC turning

Surface roughness, an indicator of surface quality is one of the most-specified customer requirements in a machining process. For efficient use of machine tools, optimum cutting parameters (speed, feed, and depth of cut) are required. So it is necessary to find a suitable optimization method which can find optimum values of cutting parameters for minimizing surface roughness. The turning process parameter optimization is highly constrained and non-linear. In this work, machining process has been carried out on brass C26000 material in dry cutting condition in a CNC turning machine and surface roughness has been measured using surface roughness tester. To predict the surface roughness, an artificial neural network (ANN) model has been designed through feed-forward back-propagation network using Matlab (2009a) software for the data obtained. Comparison of the experimental data and ANN results show that there is no significant difference and ANN has been used confidently. The results obtained conclude that ANN is reliable and accurate for predicting the values. The actual R _a value has been obtained as 1.1999???m and the corresponding predicted surface roughness value is 1.1859???m, which implies greater accuracy.     

Study of face milling of hardened AISI D3 steel with a special design of carbide tools

This paper studies the impact of a special carbide tool design on the process viability of the face milling of hardened AISI D3 steel (with a hardness of 60 HRC), in terms of surface quality and tool life. Due to the advances in the manufacturing of PVD AlCrN tungsten carbide coated tools, it is possible to use them in the manufacturing of mould and die components. Experimental results show that surface roughness (R_a) values from 0.1 to 0.3 μm can be obtained in the workpiece with an acceptable level of tool life. These outcomes suggest that these tools are suitable for the finishing of hardened steel parts and can compete with other finishing processes. The tool performance is explained after a tool wear characterization, in which two wear zones were distinguished: the region along the cutting edge where the cutting angle (κ) is maximum (κ_max) for a given depth of cut, and the zone where the cutting angle is minimum (κ?=?0) that generates the desired surface. An additional machining test run was made to plot the topography of the surface and to measure dimensional variations. Finally, for the parameters optimal selection, frequency histograms of R_a distribution were obtained establishing the relationship between key milling process parameters (V_c and f_z), surface roughness and tool wear morphology.