John Kechagias

An experimental investigation of the surface roughness of parts produced by LOM process

Purpose – To investigate the influence of different process parameters of the laminated object manufacturing (LOM) process on the roughness of vertical surfaces along Z‐axis on ZX‐plane of parts produced by LOM.Design/methodology/approach – The process parameters tested were layer thickness, heater temperature, platform retract, heater speed, laser speed, feeder speed and platform speed. A typical test part has been used, and matrix experiments were carried out based on Taguchi design. Optimal process parameter values were identified and finally, a regression model was applied onto the experimental results, and compared with bibliography models, using arbitrary experiments.Findings – The statistical analysis of the experimental results showed that the surface roughness depends mainly on the heater temperature, layer thickness, and laser speed. Moreover, the regression model gave good predictions when heater temperature values were within the initial experimental area and inaccurate predictions when heater...

Investigation of LOM process quality using design of experiments approach

Purpose – To investigate laminated object manufacturing (LOM) process quality, using a design of experiments approach.Design/methodology/approach – The quality characteristics measured were in‐plane dimensional accuracy, actual layer thickness (ALT), and mean time per layer. The process parameters tested were nominal layer thickness (LT), heater temperature (HT), platform retract (PR), heater speed (HS), laser speed (LS), feeder speed (FS) and platform speed (PS). A typical test part has been used, and matrix experiments were carried out based on Taguchi design. Optimal process parameter values were identified and finally, additive and regression models were applied to the experimental results and tested using evaluation experiments.Findings – The statistical analysis of the experimental results shows that error in X direction was higher than error in Y direction. Dimensional accuracy in X direction depends mainly on the HS (89 percent) and HT (5 percent), and in Y direction on HS (50 percent), LT (31 per...

Process build‐time estimator algorithm for laminated object manufacturing

A method for estimating the build‐time required by the laminated object manufacturing (LOM) process is presented in this paper. The proposed algorithm – taking into account the real process parameters and the information included in the parts STL‐file – performs a minimum manipulation of the file, and calculates total volume, total surface area and flat areas involved in fine cross‐hatching. A number of experiments performed verify the applicability of the algorithm in process build‐time estimation. The time prediction estimates are within 7.6 per cent of the real build‐times for the LOM process. It is believed that, through specific minor adjustments, the algorithm could well be employed in process build‐time estimation for similar rapid prototyping processes.

Prediction of surface roughness magnitude in computer numerical controlled end milling processes using neural networks, by considering a set of influence parameters: An aluminium alloy 5083 case study

A mastering of surface quality issues during machining helps avoiding failure, enhances component integrity and reduces overall costs. Surface roughness significantly affects the quality performance of finished components. A number of parameters, both material and process oriented, influence at a different extend the surface quality of the finished product. Aluminium alloy 5083 component surface quality, achieved in side end milling, constitutes the subject of the present case study. The design of experiment method is employed: that is, 18 carbide two-flute end mill cutters – manufactured by a five-axis grinding machine – have been assigned to mill 18 pockets in finishing conditions – having different combinations of geometry and cutting parameters values, according to the L18 (21× 37) standard orthogonal array. Process performance is estimated using the statistical surface texture parameters Rα, Ry and Rz– measured during three different passes on the side surface of the pockets. The results indicate that process parameters – such as the cutting speed, the peripheral second relief angle and the core diameter – mostly influence surface texture. The experimental values are used to train a feed forward back-propagation artificial neural network for the prediction of the yield surface roughness magnitude.

Evaluation of Machinability in Turning of Engineering Alloys by Applying Artificial Neural Networks

The present paper investigates the influence of main cutting parameters on the machinability during turning process for three typical materials namely AISI D6 tool steel, Ti6Al4V ELI and CuZn39Pb3 brass, all three under dry cutting environment. Spindle speed, feed rate and depth of cut were selected for study whilst arithmetic surface roughness average (Ra) and main cutting force component (FC) were treated as quality objectives characterizing machinability. For the aforementioned materials a full factorial design of experiments was conducted to exploit main effects and interactions among parameters it terms of quality objectives. The results obtained from dry turning experiments were utilized as a data set to test, train and validate a feed-forward back propagation artificial neural network for machinability prediction regarding all three materials. The work presents the results obtained from the aforementioned experimental effort under an extensive state-of-the-art survey concerning neural network technology and implementation to machining optimization problems.

A parameter design of CNC plasma-arc cutting of carbon steel plates using robust design

An optimisation of the cutting parameters during CNC plasma-arc cutting of St37 mild steel plates is attempted using robust design. The process parameters tested were plate thickness, cutting speed, arc ampere, arc voltage, air pressure, pierce height, and torch standoff distance. An orthogonal matrix experiment [L18 (21 × 37) ] was conducted and the right bevel angle was measured and optimised according to the process parameters using an analysis of means and an analysis of variances. The results show that the arc ampere has an effect mainly on the bevel angle (50.89%), while the plate thickness and torch standoff distance also have an influence of 6.22 and 15.9% respectively. The other parameters have an F factor smaller than one, and thus their variations do not significantly affect the bevel angle in the experimental region. Finally, an additive model was applied on the experimental results to predict the optimum combination and was compared with actual values.

An investigation of surface texture parameters during turning of a reinforced polymer composite using design of experiments and analysis

The influence of cutting speed and feed rate during turning on the arithmetic mean roughness (Ra), the maximum peak to valley (Rt), and the fractal dimension (D) of a glass fibre polymer composite (Ertalon 66 GF-30) was experimentally investigated. Test specimens in the form of bars and a P20 cemented carbide cutting tool were used with the cutting depth kept constant during the experiment. Robust design using an orthogonal matrix experiment was conducted and the experimental results were analysed using an ANOM and an ANOVA analysis approach. Based on the statistical analysis of the experimental results it was found that the arithmetic mean roughness, the maximum peak to valley and the fractal dimension depend mainly on the feed rate parameter. Also, based on the interaction charts and evaluation experiments it was found that regression modelling applies only for the arithmetic mean roughness.

Prediction of Cutting Forces during Turning PA66 GF-30 Glass Fiber Reinforced Polyamide by Soft Computing Techniques

In the present paper the influence of the main cutting parameters on process performance during longitudinal turning of PA66 GF-30 Glass Fiber Reinforced Polyamide is investigated. The selected cutting parameters are cutting speed and feed-rate whilst depth of cut is kept constant. As outputs (responses), cutting force components Ft, FV and Fr were selected. Test specimens in the form of round bars and cemented carbide cutting tool were used during the experimental process. Fifteen experiments were conducted having all different combinations of cutting parameter values. Analysis of Variance (ANOVA), statistical approaches and soft computing techniques (artificial neural network) were applied in order to formulate stochastic models for relating the responses with main cutting parameters. The results obtained, indicate that the proposed soft computing techniques can be effectively used to predict the cutting force components (Ft, FV and Fr) thus; facilitating decision making during process planning since costly and time-consuming experimentation can be avoided.

An ANN Approach on the Optimization of the Cutting Parameters During CNC Plasma-Arc Cutting

The objective of the present study is to develop an Artificial Neural Network (ANN) in order to predict the bevel angle (response variable) during CNC plasma-arc cutting of St37 mild steel plates. The four (4) input parameters (plate thickness, cutting speed, arc ampere, and torch standoff distance) of the ANN was selected following the results (relative importance) of the Analysis Of Variance (ANOVA) performed based on seven (7) factors (plate thickness, cutting speed, arc ampere, arc voltage, air pressure, pierce height, and torch standoff distance) in a previous study. A multi-parameter optimization was carried out using the robust design. An L18 (21 × 37 ) Taguchi orthogonal array experiment was conducted and the right bevel angle was measured, aiming at the investigation of the influence of plasma-arc cut process parameters on right side bevel angle of St37 mild steel cut surface. The selection of quality characteristics, material, plate thickness and other process parameter levels and experimental limits was based on the experience and current needs of the Greek machining industry. A feed-forward backpropagation (FFBP) neural network was fitted on the experimental data. The results show that accurate predictions of the bevel angle can be achieved inside the experimental region, through the trained FFBP-ANN. The developed ANN model could be further used for the optimization of the cutting parameters during CNC plasma-arc cutting.Copyright

Prediction of Surface Texture Characteristics in Turning of FRPs Using ANN

The objective of the present study is to develop an artificial neural network (ANN) in order to predict surface texture characteristics for the turning performance of a fiber reinforced polymer (FRP) composite. Full factorial design of experiments was designed and conducted. The process parameters considered in the experiments were cutting speed and feed rate, whilst the depth of cut has been held constant. The corresponding surface texture parameters that have been studied are the Ra and Rt. A feed forward back propagation neural network was fitted on the experimental results. It was found that accurate predictions of performance can be achieved through the feed forward back propagation (FFBP) neural network developed for the surface texture parameters.