Prediction of optimum CNC cutting conditions using artificial neural network models for the best wood surface quality, low energy consumption, and time savings

Çakıroğlu, E. O., Demir, A., Aydın, İ., and Büyüksarı, Ü. (2022). "Prediction of optimum CNC cutting conditions using artificial neural network models for the best wood surface quality, low energy consumption, and time savings," BioResources 17(2), 2501-2524.

Abstract

This study aimed to predict the CNC cutting conditions for the best wood surface quality, energy, and time savings using artificial neural network (ANN) models. In the CNC process, walnut, and ash wood were used as materials, while three different cutting tool diameters (3 mm, 6 mm, and 8 mm), spindle speed (12000 rpm, 15000 rpm, and 18000 rpm), and feed rate (3 m/min, 6 m/min, and 9 m/min) were determined as cutting conditions. After the cutting processes were completed with the CNC machine, energy consumption and processing time were determined for all groups. Surface roughness and wettability tests were performed on the processed wood samples, and their surface qualities were determined. The experimentally obtained data were analysed in ANN, and the models with the best performance were obtained. By using these prediction models, optimum cutting conditions were determined. Using the findings of the study, the optimum cutting condition values can be determined for walnut and ash wood with the smoothest and best wettable surface. Furthermore, in CNC processes using such materials, minimum energy consumption and shorter processing time can be obtained with optimum cutting conditions.

Download PDF

Full Article

Prediction of Optimum CNC Cutting Conditions Using Artificial Neural Network Models for the Best Wood Surface Quality, Low Energy Consumption, and Time Savings

Evren Osman Çakıroğlu,^a Aydın Demir,^b İsmail Aydın,^b and Ümit Büyüksarı ^c,*

DOI: 10.15376/biores.17.2.2501-2524

Keywords: Cutting conditions; Artificial neural network; CNC machine; Surface quality; Energy consumption; Processing timeDOI: 10.15376/biores.17.2.2501-2524

Contact information: a: Department of Materials and Material Processing, Artvin Vocational School, Artvin Çoruh University, Artvin, Turkey; b: Department of Forest Industry Engineering, Faculty of Forestry, Karadeniz Technical University, Trabzon, Turkey; c: Department of Wood Mechanics and Technology, Faculty of Forestry, Düzce University, Düzce, Turkey;

* Corresponding author: umitbuyuksari@duzce.edu.tr

INTRODUCTION

The furniture industry is a sector in which solid wood and wood-based panels are consumed in very high quantities to supply a fast-growing market worldwide. In particular, with the use of computer numerical control (CNC) router machines in the furniture industry, production quantities have increased rapidly, while the production costs and labour have decreased (Pelit et al. 2021). These machines are highly preferred in processes such as patterning, milling, drilling, and grooving, and their integration with other automation systems is very flexible. These machines, which increase productivity and reduce time loss, also improve the surface quality of the processed materials (Sutcu 2013; Koc et al. 2017; Sofuoglu 2017).

The most accurate determination of the parameters describing the surface quality of materials such as surface roughness and wettability is extremely significant for the successful application of wood finishing processes such as painting, coating and varnishing (Qin et al. 2015; Sofuoglu 2017; Martha et al. 2020). The most important factors affecting the surface roughness are classified as the factors originating from the material, such as wood species, density, hardness, moisture, and factors originating from the process applied to the material such as cutting tool diameter, tool shape, radius of the tool nose, spindle speed, step-over, depth of cut, and feed rate (Bajić et al. 2008). Wettability, which is sensitive to the interactions between the wood surface and liquid substances such as water, adhesive, paint, varnish, and coating, is traditionally determined by the contact angle, and the small angles indicate more wettable surfaces (Gindl and Tschegg 2002; Gindl et al. 2004; Rathke and Sinn 2013; Fang et al. 2016). The most important factors affecting the wettability of wood surface are wood species, moisture, fibre direction, polarity, pH, surface roughness, wood aging, and processing (Mohammed-Ziegler et al. 2004; Cao et al. 2005; Unsal et al. 2011; Qin et al. 2015).

Table 1. The Comparison of Some Optimization Studies with this Study

Before performing the operations on CNC machines, cutting conditions such as step-over, spindle speed, cutter plunge speed, feed rate, tool diameter, depth of cut, and tool strategy must be set correctly. Otherwise, problems may occur in the surface quality of the processed wood and wood-based panels (Koc et al. 2017; Bal and Akcakaya 2018). The necessity of determining the optimum CNC cutting condition values that give the best surface quality for wood materials used in the furniture industry has recently been an important research issue in the literature. Therefore, many researchers have focused on this issue in their studies. The comparison of some of these studies which investigated on solid wood with this study is presented in Table 1.

Table 1 shows that only surface roughness values are used as output variables in the parameter optimization studies related to CNC-processed solid wood in the literature. Furthermore, it has been determined that factors such as energy consumption and processing time, which are very important in terms of environment and cost, are not used in the optimization studies. Therefore, this study aimed to predict optimum cutting conditions depending on the surface roughness, wettability, energy consumption, and processing time of the solid wood processed with CNC machine.

EXPERIMENTAL

Materials

In this study, walnut (Juglans regia L.) and ash (Fraxinus excelsior L.) wood, which are widely used in the furniture industry, were used unlike the studies in the literature (Table 1). The wood samples were conditioned in an air-conditioning chamber until they reached a moisture content of 12% ± 1% before the cutting process with a CNC machine. A four-axis CNC milling machine (Megatron 2128, Bursa, Turkey) with a spindle power of 9 kW and a maximum spindle speed of 24000 was used for cutting operations. The toolpath strategy was chosen as offset and double flute straight milling cutters in three different diameters (3 mm, 6 mm, and 8 mm) were preferred. The offset strategy was applied constantly in tangential directions of wood samples. The CNC processing of wood samples and the used cutting tools are shown in Fig. 1.

Fig. 1. CNC processing of wood samples and the used cutting tools

Spindle speed and feed rate were determined not only by the most frequently used values in the literature, but also by choosing homogeneous parameter ranges for a successful ANN modelling. Consequently, three spindle speed (12000, 15000, and 18000 rpm) and feed rate (3, 6, and 9 m/min) were used for CNC processing. The depth of cut was 3 mm. The energy consumption of each sample was determined using a wattmeter immediately after processing with the CNC machine, and the total processing times were recorded using a stopwatch. In the energy consumption measurements, a value was taken from the wattmeter just before the CNC process starts, and another value was taken right after the CNC process was finished, and the difference between these two values (last wattmeter value – first wattmeter value) gave the energy consumption of that group. Similarly, the processing time of the groups was determined by using a stopwatch. Afterwards, the samples were sized to 50 mm x 50 mm and test specimens were obtained for surface roughness and wettability measurements. Five test specimens were prepared to represent each group and the test phase was started.

Methods

Surface roughness measurements

The surface roughness measurements were performed according to DIN 4768 (1990) standard to determine the surface quality of wood materials. The R_a (arithmetic mean) values of the measurement parameters were determined perpendicular to the fibres of the wood specimens in the Mitutoyo Surftest SJ-301 test device (Kawasaki, Japan) according to the DIN 4798 (1990) standard. The device with detector nose radius of 5 µm was set as evaluating length of 12.5 mm, cut-off length of 2.5 mm, resolution of 350 µm. Ten measurements were made for each group.

Contact angle measurements

The wettability of the wood surfaces was determined by measuring the contact angles between the surface of the wood specimens and the droplets of distilled water. Using the DSA100 Drop Shape Analysis System (KRUSS GmbH, Hamburg, Germany) equipped with image analysis software, a total of ten drops of 5 µL volume were randomly dropped onto the specimen surfaces. The contact angle values of the specimens were calculated five seconds after the droplets were deposited on the surface.

Artificial neural network analysis

Artificial neural network (ANN) analyses were carried out using the energy consumption and processing time values measured immediately after the CNC process, and the surface roughness and contact angle values obtained from the tests. As a result of the ANN analysis, the prediction models with the best performance were used both to predict the surface roughness, wettability, energy consumption and processing time values of the cutting conditions that were not used in the experimental studies, and to determine the optimum cutting condition values that give the best surface quality and energy-time savings for walnut and ash wood. The CNC cutting conditions such as spindle speed, feed rate, and cutting tool diameter were main variables in ANN modelling of this study. The data obtained from experimental studies were modelled using the MATLAB Neural Network Toolbox. The experimental data were randomly grouped as training data, validation data, and testing data for each test. Training data were presented to the network during training, and the network is adjusted according to its error. Validation data is used to measure network generalization and to halt training when generalization stops improving. Testing data has no effect on training and so provide an independent measure of network performance during and after training. Each network was trained with 38 data (about 70% of total data) and was subsequently validated with 8 experimental data (about 15% of total data) and tested with 8 experimental data (about 15% of total data).

The data sets used in the prediction models are given in Tables 3 and 4. The Levenberg Marquardt algorithm (trainlm) was chosen as the training algorithm. This algorithm typically requires more memory but less time. Training automatically stops when generalization stops improving, as indicated by an increase in the mean square error (MSE) of the validation samples. The MSE calculated by Eq. 1 was preferred as the performance function,

(1)

where ti is the actual output (targeted values), tdi is the neural network output (predicted values), and N is the total number of training patterns.

The feed forward and backpropagation multilayer ANN were used to determine prediction models. In ANN analysis trials, the transfer (activation) function was used the hyperbolic tangent sigmoid function (tansig) in the hidden layer whilst it was used the linear transfer function (purelin) in the output layer. The layers in which these activation functions are used are shown in Fig. 2.

Fig. 2. The network structures of the surface roughness (a), wettability (b), energy consumption (c) and processing time (d) prediction models

Table 2. Connection Weights and Bias Values of the Prediction Models

The actual (measured) values were compared with the prediction values obtained from ANN analyses after the testing process. The performances of the prediction models were determined by using the root mean square error (RMSE) calculated by Eq. 2 and the mean absolute percent error (MAPE) calculated by Eq. 3,

(2)

(3)

where t_i is the actual output values, td_i is the neural network predicted values, and N is the number of objects.

The connection weights (w) and biases (b) of the surface roughness and wettability, energy consumption, and processing time prediction models were given in Table 2. Moreover, the neurons and biases were denoted by n and b in Table 2.

RESULTS AND DISCUSSION

Experimental and ANN Analysis Results

The experimentally obtained data and the prediction values obtained from ANN models of these data are given in Tables 3 and 4 according to wood species. In addition, training, validation and testing data sets used in ANN analysis are indicated.

The performance values of the surface roughness, wettability, energy consumption and processing time prediction models are given in Table 5.

The MAPE values for the surface roughness were 4.71% for training, 6.88% for validation, and 9.60% for testing, whilst the values for the wettability were 1.64% for training, 2.02% for validation, and 3.86% for testing. The values for the energy consumption were 1.11% in training phase, 1.62% in validation phase, and 2.33% in testing phase, while the values for the processing time were 1.14% in training phase, 1.97% in validation phase, and 1.53% in testing phase (Table 5). The MAPE value, which is frequently used by researchers to evaluate ANN model performances, is expected to be below 10% (Antanasijević et al. 2013; Tiryaki et al. 2016). It has been demonstrated that the prediction performance of ANN models is high with these values lower than 10% (Yadav and Nath 2017). It is stated in the literature that it is extremely important to calculate RMSE values as well as MAPE values in order to determine the performance of prediction models (Kucukonder et al. 2016). In this study, the RMSE values of the surface roughness prediction model for training, validation, and testing phase were 0.32, 0.42, and 0.45 whilst the values of the wettability were 1.98, 2.23, and 3.98, respectively. The RMSE values of the energy consumption prediction model for training, validation, and testing phase were 1.07, 2.77, and 2.22 while the values of the processing time were 0.37, 1.07, and 0.51, respectively (Table 5). Taspınar and Bozkurt (2014) stated that the low RMSE values obtained from the ANN analyses are an indicator of the successful performance of the prediction models. The MAPE and RMSE values obtained from the study proved that the ANN models used for prediction and optimization are reliable and can give satisfactory accurate results.

The MSE changes of the ANN prediction models depending on iteration are shown in Fig. 3. The best validation performances of the surface roughness, wettability, energy consumption and processing time prediction models were realized in the 6th, 7th, 37th, and 14th iterations, respectively. After these iterations, the training phases of the networks were stopped. The MSE values after this stage are given in Table 5.

Table 3. Experimental and ANN Analysis Results for Walnut Wood

Table 4. Experimental and ANN Analysis Results for Ash Wood

Table 5. The Performance Values of the Prediction Models

Fig. 3. MSE changes at each iteration for the surface roughness (a), wettability (b), energy consumption (c) and processing time (d) prediction models

The error histograms of the prediction models are shown in Fig. 4. The blue, green, and red bars denote the training data, validation data and testing data respectively. The error values were calculated as the difference between the experimental and the predicted values. The orange line marks the zero-error line. The largest portion of data in the prediction models coincided with the zero-error line.

Fig. 4. Error histograms of the surface roughness (a), wettability (b), energy consumption (c) and processing time (d) prediction models

Regression analysis between predicted values and measured values is often used to evaluate the validity and accuracy of networks. The estimation accuracy of models increases when the Pearson correlation coefficients approach to 1 (Ozsahin 2012). This indicates that there is a perfect fit between the real values and the predicted values. The diagrams showing the relationships between calculated values and real values are presented in Figs. 5 and 6. The Pearson correlation coefficients of the surface roughness prediction model were 0.96325, 0.88323, 0.90291, and 0.94874 respectively for the training, validation and testing data sets and for all data sets. These values for the wettability prediction model were 0.96301, 0.95880, 0.91437, and 0.94492, respectively. These values for the energy consumption prediction model were 0.99977, 0.99925, 0.99950, and 0.99956, respectively. These values for the energy consumption prediction model were 0.99992, 0.99926, 0.99958 and 0.99979, respectively. These values showed that the developed models have a good performance and supported the predictive use of ANNs.

Fig. 5. Regression plots of the surface roughness (a) and wettability (b) prediction models


(a)	(b)

Fig. 6. Regression plots of the energy consumption (a) and processing time (b) prediction models

Optimization Results

Thanks to ANN models with low error values (MAPE and RMSE) and high performance, output values can be predicted with high accuracy for intermediate input values that are not used in experiments (Varol et al. 2018). In this study, the surface roughness, wettability, energy consumption, and processing time values were determined by the ANN prediction models for different spindle speed and feed rate values and were shown in Figs. 7 through 10 according to the wood species and tool diameter.

Fig. 7. Effects of CNC cutting conditions on the wood surface roughness

Figure 7 shows that the surface roughness prediction values vary considerably according to the wood species and cutting tool diameter. The smoothest surfaces were obtained from high spindle speed values in both wood species of wood samples processed with 3 mm cutting tool. Furthermore, similar changes were observed in the walnut wood processed with 6 mm cutting tool and the ash wood processed with 8 mm cutting tool. Suresh et al. (2012) stated that with increasing spindle speed, the temperature in the cutting zone increases, causing the material surface to soften, and therefore the surface roughness of the material decreases. In addition, high spindle speed causes less vibration by removing less material from the cutting tooth, and therefore, the surface roughness is reduced (De Deus et al. 2015).

Fig. 8. Effects of CNC cutting conditions on the wettability of wood surface

Higher spindle speed causes higher tooth passing frequencies, shortening the plane area, reducing the chip thickness, and as a result smoother surfaces are obtained (Sarikaya and Gullu 2014). The same relationship between spindle speed and surface roughness was found in many studies (Prakash and Palanikumar 2011; Sofuoglu 2017; Sedlecký et al. 2018; Hazir and Koc 2019). The lowest surface roughness prediction values were obtained from low feed rate values, especially for ash wood. Both the vibration and temperature between the workpiece and the cutting tool increase, depending on the increase in the feed rate, and therefore the surface roughness increases (Suresh et al. 2012).

Fig. 9. Effects of cutting conditions on the energy consumption of CNC process

The smoothest surfaces are obtained from low feed rate values (De Deus et al. 2015; Koc et al. 2017; Isleyen and Karamanoglu 2019). Moreover, the highest surface roughness values were predicted at 3 mm cutting tool diameter in both wood species. The smoothest surface of the samples also differed according to wood species and tool diameter. Although the most effective cutting conditions on the surface roughness of CNC machines are spindle speed and feed rate, tool geometry and tool diameter are some of the other effective conditions (Prakash and Palanikumar 2011). In this study, the surface roughness values of ash wood were generally found to be lower than walnut wood. Zhong et al. (2013) compared the surface roughness values of many wood species, including ash and walnut wood, and found that ash wood had smoother surfaces than walnut wood.

Fig. 10. Effects of cutting conditions on the processing time of CNC process

Figure 8 shows the prediction of the contact angle values indicating the wettability of the CNC machined wood sample surfaces and their differences according to the wood species and cutting tool diameter. Chemical residues and dust particles cause contamination of the surfaces of the materials due to their different wetting and adsorption properties, and therefore different contact angles can be obtained from these heterogeneous surfaces (Chau et al. 2009). In this study, the lowest contact angles with the best wettability were generally obtained from low spindle speed values. The lowest contact angle values were obtained from the lowest feed rate value for walnut wood whilst they were obtained from the highest feed rate value for ash wood. In the literature, Demir et al. (2021b) found that the contact angles of CNC-machined medium density fibreboard (MDF) samples increased with the increase in feed rate while Hosseinabadi et al. (2018) found that the contact angles of CNC-machined metal samples decreased with the increase in feed rate. These show that the type of material processed with CNC is an important factor in the effect of the feed rate. In this study, 3 mm cutting tool diameter gave the highest contact angle values for ash wood, while 8 mm cutting tool diameter gave the highest contact angle values for walnut wood. In general, it was determined that the contact angle values of ash wood were higher than walnut wood. The similar relationship was also seen in the surface roughness values. A close relationship was observed in the literature between the surface roughness and wettability of wood material surfaces (Ayrilmis 2011; Unsal et al. 2011). Moreover, wettability is known as one of the functions of surface roughness and the contact angles increased as the surface roughness of wood and wood-based panels increased (Mittal 2008; Akgul et al. 2012; Candan et al. 2012).

Figures 9 and 10 show that there was a similar relationship between energy consumption and processing time, spindle speed and feed rate in both wood species and all cutting tool diameters. As a result of the experiments and predictions, it has been determined that the feed rate is a very effective parameter in both energy consumption and processing time. Similarly, Camposeco-Negrete (2013) found that feed rate is the most significant factor for minimizing energy consumption and surface roughness. In this study, the energy consumption and processing time values decreased as the feed rate values increased. However, an accurate relationship with the change of spindle speed has not been determined. Bal and Dumankaya (2019) investigated the effects of spindle speed and feed rate on the energy consumption and total processing time of MDF processed in CNC machine and determined that these two properties decrease depending on the increase in the feed rate. However, no statistical difference was found in the processing time while an increase in energy consumption was observed due to the increase in spindle speed (Bal and Dumankaya 2019). Li et al. (2019) stated that the production time decreases with the increase in the feed rate, but the spindle speed is one of the main factors and it can increase or decrease the production time. It was observed that energy consumption and processing times decreased for both wood species with the increase in cutting tool diameter.

The values of optimum spindle speed and feed rate based on the surface roughness, wettability, energy consumption, and processing time values could be obtained thanks to prediction models. The optimum spindle speed and feed rate values which gave the smoothest surfaces, the lowest contact angle, minimum energy consumption and processing time are given in Table 6. In addition, the optimum values were predicted among the intermediate tool diameter values that were not used in the study.

Table 6 shows that the optimum spindle speed and feed rate values, which give the lowest values of both surface roughness and contact angle, differ according to wood species and tool diameter.

Table 6. Optimum CNC Parameter Results

Among all variables, the smoothest samples for walnut wood were obtained from 3 mm cutting tool diameter, 18000 rpm spindle speed and feed rate 3.8 m/min. In the ash wood, these values were 6 mm cutting tool diameter, 14800 rpm spindle speed and 3.4 m/min feed rate. In the parameter optimization studies on the surface roughness of solid wood samples processed with CNC machines in the literature, different values were found according to the specified parameter ranges, wood species, and prediction methods used. In our previous study, the optimum CNC machining parameter results were 2 mm cutting tool diameter, 10000 rpm spindle speed and 5 m/min feed rate for spruce wood, 4 mm cutting tool diameter, 12500 rpm spindle speed, and 5 m/min feed rate for beech wood (Demir et al. 2021a). While Hazir and Ozcan (2019) found these values as 8 mm cutting tool diameter, 17377 rpm spindle speed, and 2.012 m/min feed rate for cedar wood, Hazir and Koc (2019) in another study determined as 9.88 mm cutting tool diameter, 21000 rpm spindle speed, and 2.21 m/min feed rate for beech wood. Furthermore, Gürgen et al. (2022) found the optimum parameter results for pine wood as 17900 rpm spindle speed and 3 m/min feed rate. In this study, the samples with the highest wettability were obtained in walnut wood with a 4 mm cutting tool diameter, 14000 rpm spindle speed and 3 m/min feed rate while these values were determined as 3 mm cutting tool diameter, 12000 rpm spindle speed and 9 m/min feed rate in the ash wood.

Table 6 shows that the feed rate in the energy consumption and processing time was the same among all groups, and the spindle speed differed according to the wood species and tool diameter. The most energy-saving cutting conditions were 12000 rpm spindle speed for walnut wood, 14400 rpm spindle speed for ash wood, and 8 mm cutting tool diameter and 9 m/min feed rate for both wood species. The most time-saving cutting conditions were 15600 rpm spindle speed for walnut wood and 14400 rpm spindle speed for ash wood and 8 mm cutting tool diameter and 9 m/min feed rate for both wood species.

Apart from the CNC cutting conditions used in this study, there are many parameters such as tool geometry, tool strategy, step-over, and depth of cut. Therefore, based on this study, it is recommended to make optimization estimates for other parameters in future studies. Furthermore, in studies where similar wood species will be used, tool diameter, spindle speed and feed rate can be determined by utilizing the findings of this study.

CONCLUSIONS

The MAPE values of surface roughness, wettability, energy consumption and processing time prediction models were calculated as 9.60%, 3.86%, 2.33%, and 1.53%, whilst the RMSE values of these were 0.45, 3.98, 2.22, and 0.51 in the testing data sets, respectively. The Pearson correlation coefficient (R) values of these were 0.94874, 0.94492, 0.99956, and 0.99979 in all data sets, respectively. Although there was a complex and non-linear relationship between the input and output variables in the study, the performances of the models were validated with diagnostic tools and accurate, encouraging, and satisfactory results were obtained.
The output values corresponding to the intermediate cutting conditions not used in the study were also successfully predicted. Among all predicted groups, optimum cutting tool diameter, spindle speed and feed rate values, which gave the smoothest surface samples, were 3 mm, 18000 rpm and 3.8 m/min in walnut wood, while these values were 6 mm, 14800 rpm and 3.4 m/min in ash wood, respectively. The values, which gave the highest wettability samples, were 4 mm, 14000 rpm and 3 m/min in walnut wood, while these values were 3 mm, 12000 rpm and 9 m/min in ash wood, respectively.
The optimum cutting tool diameter and feed rate values, which minimized energy consumption and processing time, were 8 mm and 9 m/min for both wood species, respectively. The spindle speed values for energy consumption were 12000 rpm in the walnut wood and 14400 rpm in the ash wood. The values for processing time were 15600 rpm in the walnut wood and 14400 rpm in the ash wood.
In the case of using another cutting between 3 mm and 8 mm tool diameters, the cutting conditions for walnut and ash wood that gave the smoothest surface, lowest contact angle, minimum energy consumption and processing time can be determined from Table 6. Moreover, Figs. 7 to 10 can be used to determine surface roughness, contact angle values, energy consumption and processing time of walnut and ash wood processed with CNC at any spindle speed (between 12000 and 18000 rpm) and feed rate (between 3 m/min and 9 m/min) without conducting experimental research.
It is thought that the findings obtained from the study will make a significant contribution to both science and industry. The surface quality properties of wood species that are frequently used in the furniture industry such as walnut and ash can be predicted according to any cutting conditions, and energy and time losses can be minimized. Optimization studies using methods such as ANN that will minimize material losses and save time and energy in CNC cutting processes are extremely important. The accuracy of the prediction models decreases in the case of using other values outside the cutting conditions ranges used in the study and using different tree species. These are the main shortcomings of this study and similar optimization studies. Therefore, further optimization studies of CNC cutting conditions are needed in the literature in the future.

REFERENCES CITED

Akgül, M., Korkut, S., Çamlibel, O., Candan, Z., and Akbulut, T. (2012). “Wettability and surface roughness characteristics of medium density fiberboard panels from rhododendron (Rhododendron ponticum) biomass,” Maderas. Ciencia y Tecnología 14(2), 185-193. DOI: 10.4067/S0718-221X2012000200006

Antanasijević, D. Z., Pocajt, V. V., Povrenović, D. S., Ristić, M. Đ., and Perić-Grujić, A. A. (2013). “PM10 emission forecasting using artificial neural networks and genetic algorithm input variable optimization,” Science of the Total Environment 443, 511-519. DOI: 10.1016/j.scitotenv.2012.10.110

Ayrilmis, N. (2011). “Effect of fire retardants on surface roughness and wettability of wood plastic composite panels,” BioResources 6(3), 3178-3187. DOI: 10.15376/biores.6.3.3178-3187

Bajić, D., Lela, B., and Živković, D. (2008). “Modeling of machined surface roughness and optimization of cutting parameters in face milling,” Metalurgija 47(4), 331-334.

Bal, B. C., and Dumanoğlu, F. (2019). “Surface roughness and processing time of a medium density fiberboard cabinet door processed via CNC router, and the energy consumption of the CNC router,” BioResources 14(4), 9500-9508. DOI: 10.15376/biores.14.4.9500-9508

Bal, B. C., and Akcakaya, E. (2018). “The effects of step over, feed rate and finish depth on the surface roughness of fiberboard processed with CNC machine,” Furniture and Wooden Material Research Journal 1(2), 86-93. DOI: 10.33725/mamad.481278

Camposeco-Negrete, C. (2013). “Optimization of cutting parameters for minimizing energy consumption in turning of AISI 6061 T6 using Taguchi methodology and ANOVA,” Journal of Cleaner Production 53(16), 195-203. DOI: 10.1016/j.jclepro.2013.03.049

Candan, Z., Büyüksarı, U., Korkut, S., Unsal, O., and Çakıcıer, N. (2012). “Wettability and surface roughness of thermally modified plywood panels,” Industrial Crops and Products 36(1), 434-436. DOI: 10.1016/j.indcrop.2011.10.010

Cao, J. Z., Li, L.D., and Liu, Z. (2005). “Effect of ACQ-D treatment on the surface free energy of Chinese fir (Cunninghamia lanceolata),” Forestry Studies in China 7(4), 29-34.

Chau, T. T., Bruckard, W. J., Koh, P. T. L., and Nguyen, A. V. (2009). “A review of factors that affect contact angle and implications for flotation practice,” Advances in Colloid and Interface Science 150(2), 106-115. DOI: 10.1016/j.cis.2009.07.003

Demir, A., Cakiroglu, E. O., and Aydin, I. (2021a). “Determination of CNC processing parameters for the best wood surface quality via artificial neural network,” Wood Material Science & Engineering 1-8. (Early Access). DOI: 10.1080/17480272.2021.1929466

Demir, A., Birinci, A. U., and Ozturk, H. (2021b). “Determination of the surface characteristics of medium density fibreboard processed with CNC machine and optimisation of CNC process parameters by using artificial neural network,” CIRP Journal of Manufacturing Science and Technology 35, 929-942. DOI: 10.1016/j.cirpj.2021.10.005

Deus, P. R., de Sampaio Alves, M. C., Vieira, F. H. A. (2015). “The quality of MDF workpieces machined in CNC milling machine in cutting speeds, feedrate, and depth of cut,” Meccanica 50(12), 2899-2906. DOI: 10.1007/s11012-015-0187-z

DIN 4768 (1990). “Determination of values of surface roughness parameters Ra, Rz, Rmax using electrical contact (stylus) instruments, concepts and measuring conditions,” Deutsches Institut für Norming, Berlin, Germany.

Fang, Q., Cui, H. W., and Du, G. B. (2016). “Surface wettability, surface free energy, and surface adhesion of microwave plasma-treated Pinus yunnanensis wood,” Wood Science and Technology 50(2), 285-296. DOI: 10.1007/s00226-015-0793-x

Gindl, M., Reiterer, A., Sinn, G., and Stanzl-Tschegg, S. E. (2004). “Effects of surface ageing on wettability, surface chemistry, and adhesion of wood,” Holz Roh Werkst 62(4), 273-280. DOI: 10.1007/s00107-004-0471-4

Gindl, M., and Tschegg, S. (2002). “Significance of the acidity of wood to the surface free energy components of different wood species,” Langmuir 18(8), 3209-3212. DOI: 10.1021/la011696s

Gürgen, A., Çakmak, A., Yildiz, S., and Malkoçoğlu, A. (2022). “Optimization of CNC operating parameters to minimize surface roughness of Pinus sylvestris using integrated artificial neural network and genetic algorithm,” Maderas. Ciencia y Tecnología 24 (Early Access). DOI: 10.4067/S0718-221X2022005XXXXXX

Hazir, E., and Koc, K. H. (2019). “Optimization of wood machining parameters in CNC routers: Taguchi orthogonal array based simulated angling algorithm,” Maderas. Ciencia y Tecnología 21(4), 493-510. DOI: 10.4067/S0718-221X2019005000406

Hazir, E., and Ozcan, T. (2019). “Response surface methodology integrated with desirability function and genetic algorithm approach for the optimization of CNC machining parameters,” Arabian Journal for Science and Engineering 44(3), 2795-2809. DOI: 10.1007/s13369-018-3559-6

Hosseinabadi, H. N., Sajjady, S. A., and Amini, S. (2018). “Creating micro textured surfaces for the improvement of surface wettability through ultrasonic vibration assisted turning,” International Journal of Advanced Manufacturing Technology 96, 2825-2839. DOI: 10.1007/s00170-018-1580-2

İşleyen, Ü. K., and Karamanoğlu, M. (2019). “The influence of machining parameters on surface roughness of MDF in milling operation,” BioResources 14(2), 3266-3277. DOI: 10.15376/biores.14.2.3266-3277

Koc, K. H., Erdinler, E. S., Hazir, E., and Öztürk, E. (2017). “Effect of CNC application parameters on wooden surface quality,” Measurement 107, 12-18. DOI: 10.1016/j.measurement.2017.05.001

Küçükönder, H., Boyaci, S., and Akyüz, A. (2016). “A modeling study with an artificial neural network: developing estimation models for the tomato plant leaf area,” Turkish Journal of Agriculture and Forestry 40(2), 203-212 DOI: 10.3906/tar-1408-28

Li, C., Li, L., Tang, Y., Zhu, Y., and Li, L. (2019). “A comprehensive approach to parameters optimization of energy-aware CNC milling,” Journal of Intelligent Manufacturing 30(1), 123-138. DOI: 10.1007/s10845-016-1233-y

Martha, R., Dirna, F. C., Hasanusi, A., Rahayu, I. S., and Darmawan, W. (2020). “Surface free energy of 10 tropical woods species and their acrylic paint wettability,” Journal of Adhesion Science and Technology 34(2), 167-177. DOI: 10.1080/01694243.2019.1663009

Mittal, K. (2008). Contact Angle, Wettability, and Adhesion (Vol. 5), Brill Publications, Leiden, Netherlands.

Mohammed-Ziegler, I., Oszlánczi, Á., Somfai, B., Hórvölgyi, Z., Pászli, I., Holmgren, A., and Forsling, W. (2004). “Surface free energy of natural and surface-modified tropical and European wood species,” Journal of Adhesion Science and Technology 18(6), 687-713. DOI: 10.1163/156856104839338

Ozşahin, S. (2012). “The use of an artificial neural network for modelling the moisture absorption and thickness swelling of oriented strand board,” BioResources 7(1), 1053-1067. DOI: 10.15376/biores.7.1.1053-1067

Pelit, H., Korkmaz, M., and Budakçı, M. (2021). “Surface roughness of thermally treated wood cut with different parameters in CNC router machine,” BioResources 16(3), 5133-5147. DOI: 10.15376/biores.16.3.5133-5147

Prakash, S., and Palanikumar, K. (2011). “Modeling for prediction of surface roughness in drilling MDF panels using response surface methodology,” Journal of Composite Materials 45(16), 1639-1646. DOI: 10.1177/0021998310385026

Qin, Z., Chen, H., Gao, Q., Zhang, S., and Li, J. (2015). “Wettability of sanded and aged fast-growing poplar wood surfaces: I. Surface free energy,” BioResources 10(1), 1008-1023. DOI: 10.15376/biores.10.1.1008-1023

Rathke, J., and Sinn, G. (2013). “Evaluating the wettability of MUF resins and pMDI on two different OSB raw materials,” European Journal of Wood and Wood Products 71(3), 335-342. DOI: 10.1007/s00107-013-0675-6

Sarikaya, M., and Gullu, A. (2014). “Taguchi design and response surface methodology based analysis of machining parameters in CNC turning under MQL,” Journal of Cleaner Production 65, 604-616. DOI: 10.1016/j.jclepro.2013.08.040

Sedlecký, M., Kvietková, M. S., and Kminiak, R. (2018). “Medium-density fiberboard (MDF) and edge-glued panels (EGP) after edge milling-surface roughness after machining with different parameters,” BioResources 13(1), 2005-2021. DOI: 10.15376/biores.13.1.2005-2021

Sofuoglu, S. D. (2017). “Determination of optimal machining parameters of massive wooden edge glued panels which is made of Scots pine (Pinus sylvestris L.) using Taguchi design method,” European Journal of Wood and Wood Products 75(1), 33-42. DOI: 10.1007/s00107-016-1028-z

Stanojevic, D., Mandic, M., Danon, G., and Svrzic, S. (2017). “Prediction of the surface roughness of wood for machining,” Journal of Forestry Research 28(6), 1281-1283. DOI: 10.1007/s11676-017-0401-z

Suresh, R., Basavarajappa, S., Gaitonde, V. N., and Samuel, G. L. (2012). “Machinability investigations on hardened AISI 4340 steel using coated carbide insert,” International Journal of Refractory Metals and Hard Materials 33, 75-86. DOI: 10.1016/j.ijrmhm.2012.02.019

Sutcu, A. (2013). “Investigation of parameters affecting surface roughness in CNC routing operation on wooden EGP,” BioResources 8(1), 795-805. DOI: 10.15376/biores.8.1.795-805

Taşpınar, F., and Bozkurt, Z. (2014). “Application of artificial neural networks and regression models in the prediction of daily maximum PM10 concentration in Düzce, Turkey,” Fresenius Environmental Bulletin 23, 2450-2459.

Tiryaki, S., Bardak, S., and Aydın, A. (2016). “Modeling of wood bonding strength based on soaking temperature and soaking time by means of artificial neural networks,” International Journal of Intelligent Systems and Applications in Engineering 4, 153-157. DOI: 10.18201/ijisae.2016Special%20Issue-146964

Unsal, O., Candan, Z., and Korkut, S. (2011). “Wettability and roughness characteristics of modified wood boards using a hot-press,” Industrial Crops and Products 34(3), 1455-1457. DOI: 10.1016/j.indcrop.2011.04.024

Varol, T., Canakci, A., and Ozsahin, S. (2018). “Prediction of effect of reinforcement content, flake size and flake time on the density and hardness of flake AA2024-SiC nanocomposites using neural networks,” Journal of Alloys and Compounds 739, 1005-1014. DOI: 10.1016/j.jallcom.2017.12.256

Yadav, V., and Nath, S. (2017). “Forecasting of PM 10 using autoregressive models and exponential smoothing technique,” Asian Journal of Water, Environment and Pollution 14(4), 109-113. DOI: 10.3233/AJW-170041

Zhong, Z. W., Hiziroglu, S., and Chan, C. T. M. (2013). “Measurement of the surface roughness of wood-based materials used in furniture manufacture,” Measurement 46(4), 1482-1487. DOI: 10.1016/j.measurement.2012.11.041

Article submitted: December 9, 2021; Peer review completed: February 20, 2022; Revised version received: February 24, 2022; Accepted: February 27, 2022; Published: March 11, 2022.

DOI: 10.15376/biores.17.2.2501-2524