检索结果-内蒙古大学图书馆

20th CIRP Conference on Modeling of Machining Operations in Mons, CIRP CMMO 2025

作者： Rivière-Lorphèvre, Edouard Van Hee, Martin Beuscart, Thomas Ducobu, François Machine Design and Production Engineering Lab UMONS Research Institute for Materials Science and Engineering University of Mons Place du Parc 20 Mons7000 Belgium

The simulation of cutting forces in milling is a prerequisite for reliable process modeling. Among the different approaches used in the literature, the mechanistic models show good compromise between good precision and reasonable simulation time. The main challenge for these models is the identification of their parameters, such as the cutting coefficients, for a given tool/material couple. This paper presents a method based on an inverse analysis to identify these parameters. At first, the preprocessing of the signal is made to automatically detect the time during which the cutter is actively engaged in the workpiece. The consistency of the input data is also checked to reject outliers. Then, to avoid the classical pitfalls of this approach such as the non-uniqueness of solution, the identification is made on a whole database of results using an iterative method. The use of optimization algorithm allows the identification of parameters having nonlinear effect on the results such as cutter runout. A set of 57 milling tests in Ti6Al4V alloys have been used to demonstrate the effectiveness of this method. It allows a reduction of 5 to 20% of the root mean square error between the model and the measurements as compared to the use of coefficients identified on a single cutting test. © 2025 Elsevier B.V.. All rights reserved.

关键词： Titanium alloys

来源：评论

学校读者我要写书评

暂无评论

Image proce bing with deep-learning and transfer learning for cutting tool degradation monitoring 1

Image proce bing with deep-learning and transfer learning fo...

引用

1st International Conference on Digital Sovereignty, ICDS 2024

作者： Colantonio, Lorenzo Equeter, Lucas Giovannelli, Hugo Dehombreux, Pierre Mahmoudi, Saïd Ducobu, François Machine Design and Production Engineering Lab University of Mom Place du pare 20 Mons7000 Belgium Computer Science Unit Place du parc 20 Mons7000 Belgium

Monitoring the degradation of cutting tools is of utmost importance in the manufacturing world. Tools with substantial wear fail to produce high-quality parts in terms of geometry, residual stre b, and surface finish. Furthermore, replacing tools in a non-optimal manner can lead to increased production costs and downtime. Therefore, monitoring the condition of the tool is e bential to avoid these additional costs and ensure good production quality. This article explores various cla bification models, specifically VGG19, EfficientNetV2, and Vision Transformers. These models cla bify the state of tools using their images. Using transfer learning, a comparison of the best-performing artificial intelligence-based image analysis models is conducted to identify those most suitable for monitoring cutting tools. A comparative analysis of their generalizability, performance and explainability is realized. The model with the best performance is VGG19 with an accuracy of 94%, followed by EfficientNetV2 and ViT with an accuracy of 87%. A full comparison of these results is carried out. © 2025 The Author(s).

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Feed rate optimisation scheme in robotic machining operations for dynamic error compensation 20

Feed rate optimisation scheme in robotic machining operation...

引用

20th CIRP Conference on Modeling of Machining Operations in Mons, CIRP CMMO 2025

作者： Dambly, Valentin Olivier, Bryan Rivière-Lorphèvre, Édouard Ducobu, François Verlinden, Olivier UMONS Research Institute for Materials Science and Engineering Machine Design Production and Engineering Lab Place du parc 20 Mons Belgium UMONS Research Institute for Materials Science and Engineering Theoretical Mechanics Dynamics and Vibrations Lab Place du parc 20 Mons Belgium

The manufacturing sector demands shift towards parts with more complex geometries with the need for flexibility in production, driving interest to robotic machining. This advancing technology brings advantages like affordability, versatility, and ease of implementation, making it well-suited for agile production environments. Nevertheless, robotic machining struggles with accuracy issues due to the inherent flexibility of robots, which results in deviations and vibrations. The positioning error along a robotic machining trajectory is composed of two contributions: the steady state error and the transient. Initially generated from CAM software, the trajectory is considered as a path with a speed profile. It is then discretised in elementary sections, modelled with Hermite splines and connected by nodes. To address, offline, the lack of accuracy, an updated trajectory is computed by iteratively replacing these nodes space based on the error estimated from the dynamics simulation, strongly reducing the steady state error. However, in transient sections, the error reduction is not sufficient. This research focusses on the impact of the feedrate modification in transient areas, typically the entrance and exit of tool in the workpiece. Specific speed profiles are defined for these sections by applying linear segments with parabolic blends expressed in terms of the curvilinear abscissa. An optimisation scheme is proposed to update their feedrate considering the node repositioning necessary to compensate tool-Tip deviation. The investigation of the feedrate update is based on the results from virtual machining simulator including the robot dynamical model, responsible for steady-state and transient errors respectively and a cutter-workpiece engagement module. © 2025 The Authors.

关键词： Error compensation

来源：评论

学校读者我要写书评

暂无评论

Image processing with deep-learning and transfer learning for cutting tool degradation monitoring 12

Image processing with deep-learning and transfer learning fo...

引用

12th CIRP Global Web Conference, CIRPe 2024

Monitoring the degradation of cutting tools is of utmost importance in the manufacturing world. Tools with substantial wear fail to produce high-quality parts in terms of geometry, residual stress, and surface finish. Furthermore, replacing tools in a non-optimal manner can lead to increased production costs and downtime. Therefore, monitoring the condition of the tool is essential to avoid these additional costs and ensure good production quality. This article explores various classification models, specifically VGG19, EfficientNetV2, and Vision Transformers. These models classify the state of tools using their images. Using transfer learning, a comparison of the best-performing artificial intelligence-based image analysis models is conducted to identify those most suitable for monitoring cutting tools. A comparative analysis of their generalizability, performance and explainability is realized. The model with the best performance is VGG19 with an accuracy of 94%, followed by EfficientNetV2 and ViT with an accuracy of 87%. A full comparison of these results is carried out. © 2025 The Author(s).

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Identification of Cutting Coefficients from Multiple Milling Tests

引用

Procedia CIRP 2025年 133卷 37-42页

作者： Edouard Rivière-Lorphèvre Martin Van Hee Thomas Beuscart François Ducobu Machine Design and Production Engineering Lab UMONS Research Institute for Materials Science and Engineering University of Mons Place du Parc 20 7000 Mons Belgium

关键词：

来源：评论

学校读者我要写书评

暂无评论

Application of material constitutive and friction models parameters identified with AI and ALE to a CEL orthogonal cutting model 19

Application of material constitutive and friction models par...

引用

19th CIRP Conference on Modeling of Machining Operations, CMMO 2023

作者： Ducobu, François Palanisamy, Nithyaraaj Kugalur Arrazola, Pedro-José Rivière-Lorphèvre, Edouard Machine Design and Production Engineering Lab Research Institute for Science and Material Engineering University of Mons Mons7000 Belgium Mechanical and Manufacturing Department Faculty of Engineering Mondragon Unibertsitatea Loramendi 4 Arrasate-Mondragón20500 Spain

The identification of input parameters for funite element modelling of the cutting process is still a complex task as the experimental testing equipment cannot reach its combined levels of strains, strain rates and temperatures. Inverse identification using Artificial Intelligence method provides a relevant alternative. In this paper, material constitutive and friction models parameters identified with an Efficient Global Optimization algorithm and an ALE orthogonal cutting model are introduced in a CEL model. Assessment of the differences in the results due to the formulation and dependence of parameters identification to the finite element model are then performed. © 2023 Elsevier B.V.. All rights reserved.

关键词： Finite element method

来源：评论

学校读者我要写书评

暂无评论

Editorial

引用

Procedia CIRP 2025年 133卷 1-1页

作者： Ducobu, Fran ois Lauwers, Bert Machine Design and Production Engineering Lab Research Institute for Materials Science and Engineering University of Mons Mons7000 Belgium Department of Mechanical Engineering KU Leuven Leuven3001 Belgium

来源：评论

学校读者我要写书评

暂无评论

Simulation of chip formation for varying uncut chip thickness during vibration-assisted drilling with the CEL method 20

Simulation of chip formation for varying uncut chip thicknes...

引用

20th CIRP Conference on Modeling of Machining Operations in Mons, CIRP CMMO 2025

作者： Schumski, L. Ducobu, F. Langenhorst, L. Sölter, J. Karpuschewski, B. Leibniz Institute for Materials Engineering IWT Badgasteiner Straße 3 Bremen28359 Germany University of Bremen MAPEX Center for Materials and Processes Bibliothekstraße 1 Bremen28359 Germany Machine Design and Production Engineering Lab 20 Place du Parc MonsB-7000 Belgium

Vibration-assisted drilling is particularly efficient for long-chipping and difficult-to-machine materials due to its special kinematics with interrupted cutting. In order to optimize the process, a deeper understanding of the occurring external and internal material loads is required. For this purpose, a model is developed and simulation results are compared with experimental data on the materials Ti6Al4V and AISI 4140. Based on a kinematic simulation, individual segments on the main cutting edge are considered and the local chip formation is modelled as an orthogonal cut using the CEL method. Special consideration is given to the process kinematics, where, in addition to the entry and exit of the tool cutting edge, a varying uncut chip thickness occurs. Drilling without vibration assistance is also modelled in this way. The resulting drilling torques and thrust forces are compared with experimental data. The experiments are carried out on tubes with different internal diameters to represent the different segments of the main cutting edge. In addition, experimental results on drilling a rod and into solid material quantify the effects of the chisel edge and the minor cutting edges. The results show that both drilling and vibration-assisted drilling can be modelled using orthogonal cutting models with acceptable computation times. © 2025 The Author(s).

关键词： Ternary alloys

来源：评论

学校读者我要写书评

暂无评论

An experimental methodology to improve the robotic drilling of aluminium alloys

引用

CIRP Annals 2025年

作者： Ducobu, François Beuscart, Thomas Dambly, Valentin Rivière-Lorphèvre, Edouard Ortiz-de-Zarate, Gorka Arrazola, Pedro-José Machine Design and Production Engineering Lab Research Institute for Science and Material Engineering UMONS Belgium Faculty of Engineering Department of Mechanics and Industrial Production Mondragon Unibertsitatea Spain

In aeronautics, mastering the drilling process is crucial to meet strict quality standards. While robots provide a cost-effective solution for drilling large, complex parts, challenges arise in metallic materials due to their low stiffness. This study establishes a systematic experimental method to improve hole quality during robotic drilling of aluminium alloys. Holes were drilled in plates in four positions and 64 postures using a 6 mm tool to analyse their influence on a new defect index. A static deflection compensation method was developed to improve hole quality, addressing static deflection issues narrowing the gap between robotic drilling and CNC machining. © 2025 CIRP

关键词： Aluminum alloys

来源：评论

学校读者我要写书评

暂无评论

Mechanistic modeling of cutting forces in milling of unidirectional Glass Fiber Reinforced Polymer (UD-GFRP) 20

Mechanistic modeling of cutting forces in milling of unidire...

引用

20th CIRP Conference on Modeling of Machining Operations in Mons, CIRP CMMO 2025

作者： Nutte, Matthias Rivière-Lorphèvre, Edouard Dambly, Valentin Arrazola, Pedro-José Lazoglu, Ismail Granjon, Aurélie Ducobu, François Machine Design and Production Engineering Lab Research Institute for Science and Material Engineering UMONS Mons7000 Belgium Mechanical and Manufacturing Department Faculty of Engineering Mondragon Unibertsitatea Loramendi 4 Arrasate-Mondragón20500 Spain Manufacturing and Automation Research Center Koc University Istanbul34450 Turkey Sobelcomp SPRL Rue de l'économie 13 Loncin 4431 Belgium

Mechanistic models, established on experimental data, are recognized for their precision and simplicity in implementation. In this context of milling, the modeling of cutting forces, specifically radial and tangential components, has been extensively explored for unidirectional carbon fiber-reinforced polymer (UD-CFRP) parts. These models account for the anisotropic nature of fiber-reinforced polymers (FRPs), with cutting coefficients that vary based on the instantaneous cutting angle. However, due to the inherent flexibility of these materials, primarily attributed to their low thickness, axial cutting forces must also be incorporated to accurately represent the phenomena occurring during the machining process. This study develops a model for calculating cutting forces in three spatial directions for UD-GFRPs (unidirectional Glass Fiber-reinforced polymers). Building on an existing model from the literature for UD-CFRP, this new version introduces several modifications. First, the model validated for CFRPs must be extended to GFRPs. Secondly, the model must be extended from two dimensions to three dimensions. Finally, to ensure that the model includes only the actual contributions of the trimming operation, machining tests will be performed using only the rolling teeth. The cutting coefficients are modelled as a periodic function with a fundamental period of π. A first-order Fourier series was chosen for this purpose. Fourier series parameters are identified from the milling forces measured during a set of slotting operations at three feed rates and four fiber orientations. While the average levels of cutting forces are accurately modelled for both slotting and shoulder milling (with 50% immersion in up milling), the modeling of force amplitudes still requires improvement. © 2025 The Author(s).

关键词： Slotting

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：