Application of digital image magnification for surface roughness evaluation using machine vision

In this work, a machine vision system has been utilized to capture the images and then the quantification of the surface roughness of machined surfaces (ground, milled and shaped) is done by the application of regression analysis. Subsequently, original images have been magnified using Cubic Convolution interpolation technique and improved (edge enhancement) through Linear Edge Crispening algorithm. Based on the surface image features, a parameter called G_a has been estimated using regression analysis, for the original images and for the magnified quality improved images. Finally, a comparison has been carried to establish correlation between magnification index, G_a and surface roughness.     

An in-process surface recognition system based on neural networks in end milling cutting operations

An in-process based surface recognition system to predict the surface roughness of machined parts in the end milling process was developed in this research to assure product quality and increase production rate by predicting the surface finish parameters in real time. In this system, an accelerometer and a proximity sensor are employed as in-process surface recognition sensors during cutting to collect the vibration and rotation data, respectively. Using spindle speed, feed rate, depth of cut, and the vibration average per revolution (VAPR) as four input neurons, an artificial neural networks (ANN) model based on backpropagation was developed to predict the output neuron-surface roughness R_a values. The experimental results show that the proposed ANN surface recognition model has a high accuracy rate (96–99%) for predicting surface roughness under a variety of combinations of cutting conditions. This system is also economical, efficient, and able to be implemented to achieve the goal of in-process surface recognition by retrieving the weightings (which were generated from training and testing by the artificial neural networks), predicting the surface roughness R_a values while the part is being machined, and giving feedback to the operators when the necessary action has to be taken.     

Surface roughness measurement on machined surfaces using angular speckle correlation

Surface roughness is of great importance in fields, such as tribology, semiconductor technology and medicine. Stylus techniques, in which a stylus is drawn along the surface and the vertical movement of the stylus is recorded, have been used traditionally in measuring surface roughness. Non-contact methods, such as optical ones, have the advantage that they can be used for the in-process measurement of surface roughness. In this paper, results of the measurement of surface roughness using angular speckle-correlation on machined surfaces are presented. Surfaces of approximately 1.6 < R_a < 6.3 μm have been measured, the surfaces being classified in the same manner as when using a stylus instrument. ASC is a technique that also allows in-process measurement of the roughness of surfaces on machined surfaces. A new technique to achieve increased repeatability by using an angle detection unit is also presented in this paper.     

A vision system for surface roughness characterization using the gray level co-occurrence matrix

Computer vision technology has maintained tremendous vitality in many fields. Several investigations have been performed to inspect surface roughness based on computer vision technology. This work presents a new approach for surface roughness characterization using computer vision and image processing techniques. A vision system has been introduced to capture images for surfaces to be characterized and a software has been developed to analyze the captured images based on the gray level co-occurrence matrix (GLCM).Three standard specimens and 10 machined samples with different roughness values have been characterized by the presented approach. Three-dimensional plots of the GLCMs for various captured images have been introduced, compared and discussed. In addition, some statistical parameters (maximum occurrence of the matrix, maximum occurrence position and standard deviation of the matrix) have been calculated from the GLCMs and compared with the arithmetic average roughness R_a. Furthermore, a new parameter called maximum width of the matrix is introduced to be used as an indicator for surface roughness.     

A new approach to cutting state monitoring in end-mill machining

A new cutting state monitoring approach is proposed for the real-time predicting of the machining trouble and the surface quality of the machined products. In this approach, the relationships among the mechanical model of cutting process and its corresponding time series model, the surface roughness of the machined workpiece are evaluated through theoretical analysis and experimental investigation. It is therefore revealed that there is the linear relationships among the AR parameter a₁, the stiffness k₃ of cutting model and surface roughness P_z, and consequently the cutting process state can be estimated by only monitoring time series parameter a₁ of vibration signal measured during machining operation. In particular, it was found that the variation in the surface roughness of P_z=3–5 μm can be fully monitored.     

Research on the fractal of surface topography of grinding

Fractal theory is widely used in analysing the topography of machined surfaces. In this paper, the formula that describes the relation between the fractal dimension D and R_a or R_q or S_m of surface roughness of different ground surfaces is obtained by measuring ground surfaces and researching the fractal features of them. Using a computer, a theoretical basis is built for the fractal simulation of the ground surface.     

Anticorrosion property enhancement of Al–Li alloy 2A97 by lowering machined surface roughness

In this study, after face-milling Al–Li alloy 2A97 under dry machining condition, the machined surface roughness S_a and S_z are dealt with statistical analyses. Then, the supplementary face-milling trials are conducted to verify the analyses and gain the effect of the feed rate per tooth (f_z) on S_a and S_z. Lastly, the electrochemical impedance spectroscopy tests are conducted on the machined surfaces to determine their corrosion resistance. The variance analyses show that although f_z and the depth of cut are more important than the width of cut and the cutting speed, each of the four parameters is individually insignificant to S_a and S_z. Therefore, the quadratic regression models that take the interaction effects of the parameters into consideration are established. The validity of the models is proved with the comparisons between the measured and calculated S_a and S_z. The equivalent circuit models are proposed on the basis of the shape characteristics of the impedance spectra for machined surfaces. The estimations from the circuit models reveal the superiority in the corrosion resistances of face-milled surfaces with the optimized surface roughness.     

An element-free Galerkin method approach for estimating sensitivity of machined surface parameters

The main objective of this study is to implement a parameter sensitivity analysis method to be used in the search of optimal machining conditions with respect to surface quality. Presently, the element-free Galerkin (EFGM) approximating functions are used to evaluate the properties of machined surfaces with cutting parameters when turning AISI 4140 steel using arbitrary sets of experimental values and the EFGM approximation functions, based on the moving least-squares method, in order to obtain the sensitivities through proper local derivations. This method shows the sensitivity of each surface parameter for each input variable. The variables investigated were cutting speed (v_c), depth of cut (a_p), feed rate (f) and the surface roughness (R_a). The sensitivity results showed that the feed rate has the highest influence on surface roughness when turning AISI 4140 steel followed by cutting speed and depth of cut.     

Surface integrity of finished turned Ti–6Al–4V alloy with PCD tools using conventional and high pressure coolant supplies

 Surfaces generated when machining Ti–6Al–4V alloy with PCD tools using conventional and high pressure coolant supplies was investigated. Longer tool life was recorded when machining Ti–6Al–4V with high-pressure coolant supplies and the recorded surface roughness R_a values were well below the tool rejection criterion (1.6 μm) for all cutting conditions investigated. The micro-structure of the machined surfaces were examined on a scanning electron microscope. Micrographs of the machined surfaces show that micro-pits and re-deposited work material were the main damages to the surfaces. Micro-hardness analysis showed hardening of the top machined surfaces when machining with conventional coolant while softening of the subsurface layer was observed when machining under high-pressure coolant supplies. The later is probably due to lower heat generated, with the consequent tempering action when machining with PCD tools with high-pressure coolant supplies. The microstructure below the machined surfaces had minimal or no plastic deformation when machining with conventional and high-pressure coolant supplies.     

On the use of fractal methods for the tool flank wear characterization

Fractal theory is widely used to analyze the topography of machined surfaces, but the relationship between fractal dimensions and tool flank wear has hardly been reported. In this paper, the fractal dimensions of tool flank wear are described based on the surface roughness R_a rather than the conventional worn width VB to evaluate tool wear, thus providing better fractal identification in evaluating tool performance.     

Intelligent tool path generation for milling of free surfaces using neural networks

The presented paper has an intention to show how with the help of Artificial Neural Network (ANN), the prediction of milling tool-path strategy could be made in order to establish which milling path strategy or their sequence will show the best results (will be the most appropriate) at free surface machining, according to set technological aim. In our case the best possible surface quality of machined surface was taken as the primary technological aim. Configuration of used Neural Network (NN) is presented, and the whole procedure is shown on an example of mould, for producing light switches. The verification of machined surface quality, according to average mean roughness, R_a, is also being done, and compared with the NN predicted results.     

Surface wavelength content based clustering using neural networks for manufacturing process mapping

Manufacturing processes produce a unique texture on the surface that serves as a fingerprint of the process. It is possible to provide feedback to the process by studying the surfaces carefully. Analytical techniques such as Fourier analysis and digital filters are commonly used to characterize surface profiles. Parameters extracted from filtered profiles are monitored to detect variations in the process. This requires the development of an inference engine to map metrology parameters to manufacturing process parameters. This paper presents an artificial neural network (ANN) based inference engine for providing process feedback with surface finish input. Parameters such as R_a and W_a as well as advanced wavelet based features are extracted from surface finish data collected from a crankshaft manufacturing line and fed as input to the neural network. This input is then clustered using a competitive neural network trained in unsupervised mode. The resulting clusters are analyzed and discussed. The network is then tested with new data to evaluate the quality of the clusters previously generated and to demonstrate the applicability of this technique for detecting process variations.     

Developments in Australia's surface roughness measurement system

Roughness standards in Australia are established, maintained and disseminated by the Melbourne Branch of the National Measurement Laboratory (NML), Commonwealth Scientific and Industrial Research Organisation (CSIRO).A Taylor Hobson Talysurf 3 stylus instrument has been upgraded by replacement of the original valve amplification with solid-state electronics and the implementation of Windows 95 based software for profile analysis. The amplification is calibrated using gauge blocks wrung onto an optical flat or with master groove standards, calibrated using an interference microscope with a helium–neon laser light source. Measurements of step height or peak-to-valley roughness height in the range 0.1–50 μm are possible.The stylus instrument interfaces directly with a PC via a high speed analogue-to-digital card. Software has been developed to analyse grooves and steps and to characterise surfaces by calculation of various parameters such as arithmetic mean deviation (R_a). Typical uncertainties are better than 4% at a confidence level of 95% and are calculated automatically by the software according to the ISO “Guide to the Expression of Uncertainty in Measurement”.Technical and administrative procedures are discussed in the context of accreditation within the National Association of Testing Authorities, the Australian laboratory accreditation organisation.The facility has participated informally in a regional Asia–Pacific Metrology Programme intercomparison in which three surface roughness and two groove standards were circulated and the results are discussed.Using historical data, alternate traceability routes, and various artefact chains, the integrity of surface texture measurement at the new facility has been evaluated and verified.     

Experimental investigation into cutting forces and active grain density during abrasive flow machining

It is important to know cutting force components and active grain density during abrasive flow machining (AFM) as this information could be used to evaluate the mechanism involved in AFM. The results show that cutting force components and active grain density govern the surface roughness produced during AFM process. In this paper, an attempt has been made to study the influence of these two parameters, namely cutting force and active grain density, on the surface roughness. This study will help in developing a more realistic theoretical model.The present paper highlights a suitable two-component disc dynamometer for measuring axial and radial force components during AFM. The influence of three controllable variables (extrusion pressure, abrasive concentration and grain size) on the responses (material removal, reduction in surface roughness (R_a value), cutting forces and active grain density) are studied. The preliminary experiments are conducted to select the ranges of variables by using single-factor experimental technique. Five levels for abrasive concentration and six levels for extrusion pressure and abrasive grain size were used. A statistical 2³ full factorial experimental technique is used to find out the main effect, interaction effect and contribution of each variable to the machined workpiece surface roughness. The machined surface textures are studied using a scanning electron microscope.     

Artificial Neural Networks for Surface Roughness Prediction when Face Milling Al 7075-T7351

In this work, different artificial neural networks (ANN) are developed for the prediction of surface roughness (R _a ) values in Al alloy 7075-T7351 after face milling machining process. The radial base (RBNN), feed forward (FFNN), and generalized regression (GRNN) networks were selected, and the data used for training these networks were derived from experiments conducted using a high-speed milling machine. The Taguchi design of experiment was applied to reduce the time and cost of the experiments. From this study, the performance of each ANN used in this research was measured with the mean square error percentage and it was observed that FFNN achieved the best results. Also the Pearson correlation coefficient was calculated to analyze the correlation between the five inputs (cutting speed, feed per tooth, axial depth of cut, chip’s width, and chip’s thickness) selected for the network with the selected output (surface roughness). Results showed a strong correlation between the chip thickness and the surface roughness followed by the cutting speed.     

Micromilling characteristics and electrochemically assisted reconditioning of polycrystalline diamond tool surfaces for ultra-precision machining of high-purity SiC

The production of extremely thick silicon carbide (SiC) has recently become possible with the advent of a specific chemical vapor deposition process. Ultra-precision machining of high-purity SiC has been performed by using a polycrystalline diamond (PCD) micromilling tool to investigate the machining characteristics. Results indicate that a high-quality surface (R_a = 1.7 nm) can be obtained when the removed chips are thin enough to achieve ductile mode machining. Micron-sized wells and groove structures with nanometer-scale surface roughness were successfully machined by using the PCD tool. In addition, a new electrochemically assisted surface reconditioning process has been proposed to remove the contaminant material adhered onto the PCD tool surfaces after prolonged machining.     

Modification of surface finish produced by hard turning using superfinishing and burnishing operations

In this study the surface finish produced by hard turning of a 41Cr4 low-alloy steel quenched to about 60 HRC hardness, using mixed Al₂O₃-TiC ceramic inserts, was subsequently modified by superfinishing and multipass burnishing operations. In the case of hard turning surfaces were produced by conventional and Wiper cutting tool inserts. The main goal of this study was to examine how additional abrasive and non-removal technological operations change 2D and 3D roughness parameters and enhance service properties of the machined surfaces. It was documented that both superfinishing and burnishing operations allow to obtain smoother surfaces with lower surface roughness and better bearing characteristics.     

Non-abrasive polishing of glass

Supersmooth surfaces are often needed in high- and new-technology products. The traditional polishing method to obtain supersmooth surfaces is abrasive polishing. In this paper, a new kind of polishing method, non-abrasive polishing (NAP), is presented. The polishing wheel of NAP is made of ice that is frozen deionized water, so there is no abrasive in the polishing wheel. Compared with traditional polishing methods, NAP can obtain Ångström-order surface roughness values free from microscratches and subsurface cracks. Based on adhesion theory, the material removal mechanism of NAP is presented in the paper. The paper also analyzes the polishing equipment and processing technique of NAP. The best surface roughness, R_a=0.48 nm, of K9 glass is obtained using NAP. According to the relationship between surface roughness and adhesion, the surface roughness fluctuation phenomenon is explained in the paper. The surface roughness fluctuation phenomenon is expected to be avoided by measuring the fractal dimension of the polished surface.     

A Taguchi and experimental investigation into the optimal processing conditions for the abrasive jet polishing of SKD61 mold steel

This study introduces an abrasive jet polishing (AJP) technique in which the pneumatic air stream carries not only abrasive particles, but also an additive of either pure water or pure water with a specified quantity of machining oil. Taguchi design experiments are performed to identify the optimal AJP parameters when applied to the polishing of electrical discharge machined SKD61 mold steel specimens. A series of experimental trials are then conducted using the optimal AJP parameters to investigate the respective effects of the additive type and the abrasive particle material and diameter in achieving a mirror-like finish of the polished surface. The Taguchi trials indicate that when polishing is performed using pure water as an additive, the optimal processing parameters are as follows: an abrasive material to additive ratio of 1:2, an impact angle of 30°, a gas pressure of 4 kg/cm², a nozzle-to-workpiece height of 10 mm, a platform rotational velocity of 200 rpm, and a platform travel speed of 150 mm/s. Applying these processing parameters, it is found that the optimal polishing effect is attained using #8000SiC abrasive particles and a 1:1 mixture of water-solvent machining oil and pure water. The experimental results show that under these conditions, the average roughness of the electrical discharge machined SKD61 surface is reduced from an original value of R_a=1.03 μm (R_max: 7.74 μm) to a final value of R_a=0.13 μm (R_max: 0.90 μm), corresponding to a surface roughness improvement of approximately 87%.     

Prediction and control of surface roughness in CNC lathe using artificial neural network

In this study, a neural network approach is presented for the prediction and control of surface roughness in a computer numerically controlled (CNC) lathe. Experiments have been performed on the CNC lathe to obtain the data used for the training and testing of a neural network. The parameters used in the experiment were reduced to three cutting parameters which consisted of depth of cutting, cutting speed, and feed rate. Each of the other parameters such as tool nose radius, tool overhang, approach angle, workpiece length, workpiece diameter and workpiece material was taken as constant. A feed forward multi-layered neural network was developed and the network model was trained using the scaled conjugate gradient algorithm (SCGA), which is a type of back-propagation. The adaptive learning rate was used. Therefore, the learning rate was not selected before training and it was adjusted during training to minimize training time. The number of iterations was 8000 and no smoothing factor was used. R_a, R_z and R_max were modeled and were evaluated individually. One hidden layer was used for all models while the numbers of neurons in the hidden layer of the R_a model were five and the numbers of neurons in the hidden layers of the R_z and R_max models were ten. The results of the neural network approach were compared with actual values. In addition, inasmuch as the control of surface roughness is proposed, a control algorithm was developed in the present investigation. The desired surface roughness was entered into the control system as a reference value and the controller determined the cutting parameters for these surface roughness values. A new surface roughness value was determined by sending the cutting parameters to the observer (ANN block). The obtained surface roughness was fed back to the comparison unit and was compared with the reference value and the difference surface roughness was then sent to the controller. The iteration was continued until the difference was reduced to a certain value of surface roughness which could be permitted for machining accuracy. When the surface roughness reached the permitted value, these cutting parameters were sent to the CNC turning system as input values. In conclusion, both the surface roughness values corresponding to the cutting parameters and suitable cutting parameters for a certain surface roughness can be determined prior to a machining operation using the ANN and control algorithm.