Application of the Gradient-Boosting with Regression Trees to Predict the Coefficient of Friction on Drawbead in Sheet Metal Forming DOI Open Access
Sherwan Mohammed Najm, Tomasz Trzepieciński, Salah Eddine Laouini

и другие.

Materials, Год журнала: 2024, Номер 17(18), С. 4540 - 4540

Опубликована: Сен. 15, 2024

Correct design of the sheet metal forming process requires knowledge friction phenomenon occurring in various areas drawpiece. Additionally, at drawbead is decisive to ensure that flows desired direction. This article presents results experimental tests enabling determination coefficient and using a specially designed tribometer. The test material was DC04 carbon steel sheet. were carried out for different orientations samples relation rolling direction, heights, lubrication conditions average roughnesses countersamples. According aim this work, Features Importance analysis, conducted Gradient-Boosted Regression Trees algorithm, used find influence several parameter features on friction. advantage gradient-boosted decision trees their ability analyze complex relationships data protect against overfitting. Another there no need prior processing. best authors’ knowledge, effectiveness analyzing has not been previously studied. To improve accuracy model, five MinLeafs applied regression tree, together with 500 ensembles utilized learning learned nodes, noting MinLeaf indicates minimum number leaf node observations. least-squares-boosting technique, often known as LSBoost, train group trees. analysis shown (dry lubricated conditions) had most significant friction, 56.98%, followed by height, 23.41%, sample width, 11.95%. surface roughness rollers orientation have smallest impact value 6.09% 1.57%, respectively. dispersion deviation observed testing dataset from indicate model’s predict values R2 = 0.972 mean-squared error MSE 0.000048. It qualitatively found order optimal (the lowest) it necessary control (use lubricant) height.

Язык: Английский

Tribological Performance of Borided Tool Steel with Minimum Bio-Lubrication for Sheet Metal Forming Applications DOI
Cesar David Resendiz-Calderon, Orlando Soriano‐Vargas, Julio A. Cao-Romero-Gallegos

и другие.

Wear, Год журнала: 2025, Номер unknown, С. 205748 - 205748

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0

Tribocorrosion behavior of aluminum alloys in 3.5 wt.% NaCl solution DOI Creative Commons
Jin Du,

Linlan Hu,

Meng Chen

и другие.

Journal of Materials Research and Technology, Год журнала: 2025, Номер unknown

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0

APPLICATION OF THE TENSILE BENDING TEST TO DETERMINE THE TRIBOLOGICAL BEHAVIOR OF DC01 STEEL SHEETS DOI
Marek Szewczyk, Krzysztof Szwajka, Tomasz Trzepieciński

и другие.

Tribologia, Год журнала: 2025, Номер 311(1), С. 67 - 76

Опубликована: Март 31, 2025

The article presents the results of analysis changes in coefficient friction occurring on edge punch during sheet metal plastic processing. Experimental studies were conducted using method bending and stretching. A tribometer developed by authors was used. The research material consisted 0.8 mm thick DC01 low-carbon steel sheets. evolution tests studied dry conditions lubrication surface with S100 Plus S300 oil (Naftochem). Countersamples made 145Cr6 tool steel, additionally modified applying anti-wear coatings obtained showed that depended coating used, nature this change conditions. In analyzed conditions, a decrease values Sa Ssk parameters observed, an increase Sku parameter relation to delivery state.

Язык: Английский

Процитировано

0

Analysis of the friction performance of deep-drawing steel sheets using network models DOI Creative Commons
Sherwan Mohammed Najm, Tomasz Trzepieciński, Omar M. Ibrahim

и другие.

The International Journal of Advanced Manufacturing Technology, Год журнала: 2024, Номер 132(7-8), С. 3757 - 3769

Опубликована: Апрель 16, 2024

Abstract This article presents the results of pilot studies on lubrication blankholder zone in sheet metal forming using a pressurised lubricant. The authors invented method and built special tribometer for pressure-assisted lubrication. approach reduces friction processes compared to conventional Moreover, artificial neural network combined with force-directed Fruchterman-Reingold graph algorithm Spearman’s correlation was used first time analyse relationships between process parameters output (the coefficient resulting surface roughness metal). experimental tests were conducted utilising strip drawing four grades steel sheets known be outstanding deep-drawing quality. Different oils, oil pressures contact used. Artificial models determine these test where every parameter is represented by one node, all nodes are connected edges each other. R Software version 4.2.3 construct ‘qgraph’ ‘networktools’ packages. It found that conditions had highly significant negative (COF) moderately final roughness. However, initial as-received COF positive metal. most related strength coefficient, ultimate tensile (dry or lubrication). coefficients showed strong kinematic viscosity conditions.

Язык: Английский

Процитировано

2

Analysis of the Frictional Performance of AW-5251 Aluminium Alloy Sheets Using the Random Forest Machine Learning Algorithm and Multilayer Perceptron DOI Open Access
Tomasz Trzepieciński, Sherwan Mohammed Najm, Omar M. Ibrahim

и другие.

Materials, Год журнала: 2023, Номер 16(15), С. 5207 - 5207

Опубликована: Июль 25, 2023

This paper is devoted to the determination of coefficient friction (COF) in drawbead region metal forming processes. As test material, AW-5251 aluminium alloys sheets fabricated under various hardening conditions (AW-5251-O, AW-5251-H14, AW-5251-H16 and AW-5251H22) were used. The tested using a simulator with different countersample roughness orientations specimens relation sheet rolling direction. A was designed model when passed through forming. experimental tests carried out dry lubrication surfaces three lubricants: machine oil, hydraulic engine oil. Based on results tests, value COF determined. Random Forest (RF) learning algorithm artificial neural networks (ANNs) used identify parameters affecting COF. R statistical package software version 4.1.0 for running RF network. relative importance inputs analysed 12 activation functions ANNs nine loss RF. it concluded that samples cut along direction greater than transverse However, COF’s most relevant input oil viscosity (0.59), followed by average counter sample Ra (0.30) yield stress Rp0.2 strength K (0.05 0.06, respectively). hard sigmoid function had poorest R2 (0.25) nRMSE (0.30). ideal run found after training testing (R2 = 0.90 ± 0.028). values 1.1 between 105 190 resulted decreased dropped 9–35 105–190 Rp0.2, gap 110 130 added. low 9–35, 0.95–1.25. interaction other inputs, which produces relatively limited range reduced values, least relevant. setting 190, 0.95 1.25, 9 35.

Язык: Английский

Процитировано

5

Analysis of the Lubrication Performance of Low-Carbon Steel Sheets in the Presence of Pressurised Lubricant DOI Open Access
Tomasz Trzepieciński, Valmir Dias Luiz, Krzysztof Szwajka

и другие.

Advances in Materials Science, Год журнала: 2023, Номер 23(2), С. 64 - 76

Опубликована: Июнь 1, 2023

Abstract In sheet metal forming processes, friction increases the force parameters of process and produces a deterioration in quality surface components. The basic way to reduce unfavourable impact is lubricate with commercial oils. This article presents results experimental studies analysis variance (ANOVA) DC01 low-carbon steel sheets using strip drawing test. For these tests, special device was built containing countersamples flat made 145Cr6 covered protective AlTiN coating. Lubricants different viscosities were fed into contact zone under forced pressure. effect pressure on value coefficient also determined. predicted R² 0.9227 reasonable agreement adjusted 0.9411 confirming that ANOVA model reliable. It found increasing lubricant had beneficial reducing friction. higher pressure, more effectively pressurised oil reduced

Язык: Английский

Процитировано

3

EXPERIMENTAL DETERMINATION OF THE DRAW BEAD COEFFICIENT OF FRICTION OF CuZn SHEETS IN SHEET METAL FORMING PROCESSES DOI Creative Commons
Tomasz Trzepieciński, Valmir Dias Luiz, Marcin Szpunar

и другие.

Acta Metallurgica Slovaca, Год журнала: 2023, Номер 29(3), С. 123 - 129

Опубликована: Сен. 25, 2023

In sheet metal forming, draw beads are used to limit the flow of in specific areas stamping die. The value coefficient friction at bead determines achievement desired resistance displacement. This article presents results experimental tests for determining on a using specially developed tribotester. test material consisted CuZn37, CuZn30 and CuZn10 brass sheets various states hardening. Investigations were carried out with different roughness countersamples sliding speeds. addition, under conditions dry lubrication surface LAN-46 machine oil. relationships between process parameters analysed analysis variance. It was found that decreases increasing mean countersamples. Lubrication reduced by about 6.2-29.8% depending grade tested speed.

Язык: Английский

Процитировано

2

Mechanical Alloying of Aluminium Alloys DOI
Rayappa Shrinivas Mahale,

Krishnamurthy Goggi,

Shamanth Vasanth

и другие.

Advances in chemical and materials engineering book series, Год журнала: 2024, Номер unknown, С. 89 - 126

Опубликована: Фев. 27, 2024

Mechanical alloying is a solid-state powder processing technique that has revolutionized the development of advanced materials, including aluminium alloys. This chapter provides concise overview key aspects mechanical applied to alloys, exploring its principles, advantages, and potential applications. The process involves subjecting elements intense deformation using high-energy ball mill. results in repeated welding, fracturing, cold welding particles, leading creation homogenous fine-grained microstructure. mechanisms encompass diffusion deformation-induced mixing, facilitating uniform dispersion within matrix. several distinct advantages over conventional methods.

Язык: Английский

Процитировано

0

Application of the Gradient-Boosting with Regression Trees to Predict the Coefficient of Friction on Drawbead in Sheet Metal Forming DOI Open Access
Sherwan Mohammed Najm, Tomasz Trzepieciński, Salah Eddine Laouini

и другие.

Materials, Год журнала: 2024, Номер 17(18), С. 4540 - 4540

Опубликована: Сен. 15, 2024

Correct design of the sheet metal forming process requires knowledge friction phenomenon occurring in various areas drawpiece. Additionally, at drawbead is decisive to ensure that flows desired direction. This article presents results experimental tests enabling determination coefficient and using a specially designed tribometer. The test material was DC04 carbon steel sheet. were carried out for different orientations samples relation rolling direction, heights, lubrication conditions average roughnesses countersamples. According aim this work, Features Importance analysis, conducted Gradient-Boosted Regression Trees algorithm, used find influence several parameter features on friction. advantage gradient-boosted decision trees their ability analyze complex relationships data protect against overfitting. Another there no need prior processing. best authors’ knowledge, effectiveness analyzing has not been previously studied. To improve accuracy model, five MinLeafs applied regression tree, together with 500 ensembles utilized learning learned nodes, noting MinLeaf indicates minimum number leaf node observations. least-squares-boosting technique, often known as LSBoost, train group trees. analysis shown (dry lubricated conditions) had most significant friction, 56.98%, followed by height, 23.41%, sample width, 11.95%. surface roughness rollers orientation have smallest impact value 6.09% 1.57%, respectively. dispersion deviation observed testing dataset from indicate model’s predict values R2 = 0.972 mean-squared error MSE 0.000048. It qualitatively found order optimal (the lowest) it necessary control (use lubricant) height.

Язык: Английский

Процитировано

0