Journal of Manufacturing Processes最新文献

Micro-convex structure-assisted laser-induced backside dry etching for high-efficiency fabrication of micro-groove array on fused silica toward manipulation of wetting characteristics

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-26 DOI: 10.1016/j.jmapro.2025.04.063

Hu Huang , Hong An , Yongfeng Qian , Zhiyu Zhang , Minqiang Jiang , Jiwang Yan

Fused silica has garnered significant attention in various industrial applications. The creation of surface micro/nanostructures on fused silica can impart specialized functional properties, such as extreme wettability, ultra-high transmittance, and excellent antimicrobial characteristics. Nonetheless, existing technologies for achieving surface micro/nanopatterning of fused silica, including wire electrical discharge machining, chemical etching, and electrochemical deposition, generally suffer from long processing cycles and high energy consumption, which are inconsistent with the development concept of the low-carbon economy. In this study, a novel approach termed micro-convex structure-assisted laser-induced backside dry etching (MCSALIBDE) is proposed for high-efficiency fabrication of micro-groove array on fused silica surfaces. The synergistic effect of enhanced heat conduction and plasma explosion in a confined space facilitates material ablation. The impact of processing parameters including scanning pitch and peak laser power intensity on the morphological and topographical features of the MCSALIBDE-processed fused silica surfaces is studied. Notably, the micro-groove structures with a depth-to-width ratio of 1.19 is fabricated. Furthermore, the wetting behaviors of the MCSALIBDE-processed fused silica surfaces before and after annealing treatment are studied. Experimental findings reveal that the MCSALIBDE-processed fused silica surfaces exhibit enhanced hydrophilicity compared to the untreated one. After annealing treatment, a transformation from hydrophilicity to superhydrophobicity is observed. The superhydrophobic surfaces produced under different laser parameters exhibit distinct adhesion and droplet bouncing behaviors. This work offers an in-depth understanding of the high-efficiency fabrication of micro/nanostructures on fused silica surfaces and as well the modulation of their wetting characteristics, heralding innovations in surface engineering for diverse applications.

{"title":"Micro-convex structure-assisted laser-induced backside dry etching for high-efficiency fabrication of micro-groove array on fused silica toward manipulation of wetting characteristics","authors":"Hu Huang , Hong An , Yongfeng Qian , Zhiyu Zhang , Minqiang Jiang , Jiwang Yan","doi":"10.1016/j.jmapro.2025.04.063","DOIUrl":"10.1016/j.jmapro.2025.04.063","url":null,"abstract":"<div><div>Fused silica has garnered significant attention in various industrial applications. The creation of surface micro/nanostructures on fused silica can impart specialized functional properties, such as extreme wettability, ultra-high transmittance, and excellent antimicrobial characteristics. Nonetheless, existing technologies for achieving surface micro/nanopatterning of fused silica, including wire electrical discharge machining, chemical etching, and electrochemical deposition, generally suffer from long processing cycles and high energy consumption, which are inconsistent with the development concept of the low-carbon economy. In this study, a novel approach termed micro-convex structure-assisted laser-induced backside dry etching (MCSALIBDE) is proposed for high-efficiency fabrication of micro-groove array on fused silica surfaces. The synergistic effect of enhanced heat conduction and plasma explosion in a confined space facilitates material ablation. The impact of processing parameters including scanning pitch and peak laser power intensity on the morphological and topographical features of the MCSALIBDE-processed fused silica surfaces is studied. Notably, the micro-groove structures with a depth-to-width ratio of 1.19 is fabricated. Furthermore, the wetting behaviors of the MCSALIBDE-processed fused silica surfaces before and after annealing treatment are studied. Experimental findings reveal that the MCSALIBDE-processed fused silica surfaces exhibit enhanced hydrophilicity compared to the untreated one. After annealing treatment, a transformation from hydrophilicity to superhydrophobicity is observed. The superhydrophobic surfaces produced under different laser parameters exhibit distinct adhesion and droplet bouncing behaviors. This work offers an in-depth understanding of the high-efficiency fabrication of micro/nanostructures on fused silica surfaces and as well the modulation of their wetting characteristics, heralding innovations in surface engineering for diverse applications.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 236-251"},"PeriodicalIF":6.1,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143875049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A data-efficient sequential learning framework for melt pool defect classification in Laser Powder Bed Fusion

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-25 DOI: 10.1016/j.jmapro.2025.03.118

Ahmed Shoyeb Raihan , Austin Harper , Israt Zarin Era , Omar Al-Shebeeb , Thorsten Wuest , Srinjoy Das , Imtiaz Ahmed

Ensuring the quality and reliability of Metal Additive Manufacturing (MAM) components is crucial, especially in the Laser Powder Bed Fusion (L-PBF) process, where melt pool defects such as keyhole, balling, and lack of fusion can significantly compromise structural integrity. This study presents SL-RF+ (Sequentially Learned Random Forest with Enhanced Sampling), a novel Sequential Learning (SL) framework for melt pool defect classification designed to maximize data efficiency and model accuracy in data-scarce environments. SL-RF+ utilizes an RF classifier combined with Least Confidence Sampling (LCS) and Sobol sequence-based synthetic sampling to iteratively select the most informative samples, refining the model’s decision boundaries with minimal labeled data. Results demonstrate that SL-RF+ achieves an accuracy of 83.3%, outperforming the traditional RF model (78.8%) with significantly fewer labeled samples in melt pool defect classification. Moreover, SL-RF+ improves precision (83.1%), recall (76.9%), and F1-score (78.9%), surpassing the baseline model in all key performance metrics. Notably, SL-RF+ achieves competitive classification performance with fewer than 150 sequentially added samples, whereas the traditional RF model requires all 275 labeled samples to reach similar accuracy levels. By prioritizing high-uncertainty regions in the process parameter space, this framework efficiently captures complex defect patterns, ultimately achieving superior classification performance without the need for extensive labeled datasets. While this study utilizes pre-existing experimental data, SL-RF+ shows strong potential for real-world applications in pure sequential learning settings, where data is acquired and labeled incrementally, mitigating the high costs and time constraints of sample acquisition.

{"title":"A data-efficient sequential learning framework for melt pool defect classification in Laser Powder Bed Fusion","authors":"Ahmed Shoyeb Raihan , Austin Harper , Israt Zarin Era , Omar Al-Shebeeb , Thorsten Wuest , Srinjoy Das , Imtiaz Ahmed","doi":"10.1016/j.jmapro.2025.03.118","DOIUrl":"10.1016/j.jmapro.2025.03.118","url":null,"abstract":"<div><div>Ensuring the quality and reliability of Metal Additive Manufacturing (MAM) components is crucial, especially in the Laser Powder Bed Fusion (L-PBF) process, where melt pool defects such as keyhole, balling, and lack of fusion can significantly compromise structural integrity. This study presents SL-RF+ (Sequentially Learned Random Forest with Enhanced Sampling), a novel Sequential Learning (SL) framework for melt pool defect classification designed to maximize data efficiency and model accuracy in data-scarce environments. SL-RF+ utilizes an RF classifier combined with Least Confidence Sampling (LCS) and Sobol sequence-based synthetic sampling to iteratively select the most informative samples, refining the model’s decision boundaries with minimal labeled data. Results demonstrate that SL-RF+ achieves an accuracy of 83.3%, outperforming the traditional RF model (78.8%) with significantly fewer labeled samples in melt pool defect classification. Moreover, SL-RF+ improves precision (83.1%), recall (76.9%), and F1-score (78.9%), surpassing the baseline model in all key performance metrics. Notably, SL-RF+ achieves competitive classification performance with fewer than 150 sequentially added samples, whereas the traditional RF model requires all 275 labeled samples to reach similar accuracy levels. By prioritizing high-uncertainty regions in the process parameter space, this framework efficiently captures complex defect patterns, ultimately achieving superior classification performance without the need for extensive labeled datasets. While this study utilizes pre-existing experimental data, SL-RF+ shows strong potential for real-world applications in pure sequential learning settings, where data is acquired and labeled incrementally, mitigating the high costs and time constraints of sample acquisition.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 201-210"},"PeriodicalIF":6.1,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143869871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Directed energy deposition on sheet metal forming for reinforcement structures

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-25 DOI: 10.1016/j.jmapro.2025.03.120

Fan Chen , Rujing Zha , Jihoon Jeong , Shuheng Liao , Jian Cao

While incremental forming processes can inexpensively create complex geometries from sheet metal, they struggle with adding sharp out of plane features for stiffness enhancement. With the implementation of directed-energy deposition (DED), an additive manufacturing process that locally deposits metal onto metallic substrates, reinforcement structures can be formed on the sheet metal. Furthermore, a design engineer may take advantage of the high residual stresses of DED to directly alter shapes in the substrate metal sheet. This hybrid forming-deposition process, as well as the application of local reinforcement, requires a good understanding of the process mechanism to predict expected shapes and minimize undesired deformations. In this work, numerical approaches are applied to evaluate heat transfer, thermal stress, and buckling of thin sheets under the stresses of deposition. These results are compared to analogous experiments conducted on an open-architecture laser-powder DED machine. The results of the thermal-mechanical analysis resemble the deformation trends observed in the experiments. However, the small-displacement formulation in the simulation used for ease of convergence does not fully capture the magnitude of the observed deformations. Nevertheless, the simulations effectively illustrate the effect of different scan strategies on the final deformed shape of the sheet metal.

{"title":"Directed energy deposition on sheet metal forming for reinforcement structures","authors":"Fan Chen , Rujing Zha , Jihoon Jeong , Shuheng Liao , Jian Cao","doi":"10.1016/j.jmapro.2025.03.120","DOIUrl":"10.1016/j.jmapro.2025.03.120","url":null,"abstract":"<div><div>While incremental forming processes can inexpensively create complex geometries from sheet metal, they struggle with adding sharp out of plane features for stiffness enhancement. With the implementation of directed-energy deposition (DED), an additive manufacturing process that locally deposits metal onto metallic substrates, reinforcement structures can be formed on the sheet metal. Furthermore, a design engineer may take advantage of the high residual stresses of DED to directly alter shapes in the substrate metal sheet. This hybrid forming-deposition process, as well as the application of local reinforcement, requires a good understanding of the process mechanism to predict expected shapes and minimize undesired deformations. In this work, numerical approaches are applied to evaluate heat transfer, thermal stress, and buckling of thin sheets under the stresses of deposition. These results are compared to analogous experiments conducted on an open-architecture laser-powder DED machine. The results of the thermal-mechanical analysis resemble the deformation trends observed in the experiments. However, the small-displacement formulation in the simulation used for ease of convergence does not fully capture the magnitude of the observed deformations. Nevertheless, the simulations effectively illustrate the effect of different scan strategies on the final deformed shape of the sheet metal.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"144 ","pages":"Pages 339-349"},"PeriodicalIF":6.1,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143869258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Machine learning-based early detection of malicious G-code manipulations in 3D printing

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-25 DOI: 10.1016/j.jmapro.2025.04.012

Hala Ali, Alberto Cano, Irfan Ahmed

The increased adoption of 3D printing across various critical manufacturing sectors has made it a fruitful target for adversaries, particularly through the manipulation of G-code instructions that control the operations of 3D printers. Simple modifications to these instructions could significantly impact the integrity of 3D-printed objects. While side-channel analysis during printing is a common detection method, identifying potential malicious G-code before printing can save time and resources. Existing work relies on primitive encryption and hashing techniques and cannot distinguish between benign and malicious G-code instructions. It assumes that G-code files are benign and uses them as a reference model, focusing only on the integrity checking of G-code during storage and transmission. This paper introduces a novel automated approach to efficiently differentiate between benign and subtly manipulated G-code caused by filament, thermodynamic, and Z-profile attacks without requiring a reference model. As the first study leveraging recent advancements in Machine Learning (ML), we address several challenges in dataset generation, feature engineering, G-code segmenting and labeling, and ML classifier selection. We generate diverse G-code datasets to identify the optimal dataset characteristics and conduct a comprehensive formal analysis to extract the most suitable features. Efficient labeling strategies are employed at both layer and command levels, using the Multiple Instance Learning (MIL) paradigm for the former. We adopt the Bidirectional Long Short-Term Memory (Bi-LSTM) model enhanced by an attention mechanism and focal loss function for layer classification. Meanwhile, the Random Forest (RF) algorithm and Multilayer Perceptron (MLP) neural network model are used for command classification. All classifiers are designed to handle the imbalanced dataset. Experimental evaluation demonstrates the efficacy of our approach. The Bi-LSTM model achieves F1 scores up to 91.3% in detecting filament attacks, while the RF algorithm performs better in detecting nuanced thermodynamic and Z-profile changes at the command level, achieving F1 scores between 81.6% and 99.3%.

{"title":"Machine learning-based early detection of malicious G-code manipulations in 3D printing","authors":"Hala Ali, Alberto Cano, Irfan Ahmed","doi":"10.1016/j.jmapro.2025.04.012","DOIUrl":"10.1016/j.jmapro.2025.04.012","url":null,"abstract":"<div><div>The increased adoption of 3D printing across various critical manufacturing sectors has made it a fruitful target for adversaries, particularly through the manipulation of G-code instructions that control the operations of 3D printers. Simple modifications to these instructions could significantly impact the integrity of 3D-printed objects. While side-channel analysis during printing is a common detection method, identifying potential malicious G-code before printing can save time and resources. Existing work relies on primitive encryption and hashing techniques and cannot distinguish between benign and malicious G-code instructions. It assumes that G-code files are benign and uses them as a reference model, focusing only on the integrity checking of G-code during storage and transmission. This paper introduces a novel automated approach to efficiently differentiate between benign and subtly manipulated G-code caused by filament, thermodynamic, and Z-profile attacks without requiring a reference model. As the first study leveraging recent advancements in Machine Learning (ML), we address several challenges in dataset generation, feature engineering, G-code segmenting and labeling, and ML classifier selection. We generate diverse G-code datasets to identify the optimal dataset characteristics and conduct a comprehensive formal analysis to extract the most suitable features. Efficient labeling strategies are employed at both layer and command levels, using the Multiple Instance Learning (MIL) paradigm for the former. We adopt the Bidirectional Long Short-Term Memory (Bi-LSTM) model enhanced by an attention mechanism and focal loss function for layer classification. Meanwhile, the Random Forest (RF) algorithm and Multilayer Perceptron (MLP) neural network model are used for command classification. All classifiers are designed to handle the imbalanced dataset. Experimental evaluation demonstrates the efficacy of our approach. The Bi-LSTM model achieves F1 scores up to 91.3% in detecting filament attacks, while the RF algorithm performs better in detecting nuanced thermodynamic and Z-profile changes at the command level, achieving F1 scores between 81.6% and 99.3%.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 211-235"},"PeriodicalIF":6.1,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143875048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fabrication and characterization of 3D micro-coils with hybrid manufacturing methods

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-24 DOI: 10.1016/j.jmapro.2025.04.040

Hakan Alaboz , Andras Kovacs , Daniel Johannes Förster , Lucas Werling , Jajnabalkya Guhathakurta , Julien Petit , Morgan Madec , Luc Hébrard , Ulrich Mescheder

This study presents novel fabrication steps and hybrid approaches that combine contemporary methods with modified manufacturing strategies for fabricating 3D micro-coils on tubular surfaces. Although several methods have been developed in the last decade, most of them still rely on complicated multi-step lithography processes or sophisticated devices. In this article, rapid, simple, and straightforward approaches are presented for fabricating micro-coils directly on curved surfaces. The hybrid fabrication methods in this study provide an alternative solution to reduce the complexity in fabrication steps, provide better control over aspect ratios and the homogeneity of thin film coatings on tubular surfaces. Micro-coils from copper and conductive polymer with outer diameter of 2.5 mm and line width of 40 μm and a separation of 110 μm were designed, produced, and characterized. Magnetic fields inside the coils made of copper and conductive polymer were measured as 11 and 2.5 μT, respectively. The developed coating and structuring methods will open new avenues not only in application fields such as analytical characterization methods of fluidic samples with miniaturized nuclear magnetic resonance (μNMR) systems, but also for 3D sensor fabrication on curved surfaces.

{"title":"Fabrication and characterization of 3D micro-coils with hybrid manufacturing methods","authors":"Hakan Alaboz , Andras Kovacs , Daniel Johannes Förster , Lucas Werling , Jajnabalkya Guhathakurta , Julien Petit , Morgan Madec , Luc Hébrard , Ulrich Mescheder","doi":"10.1016/j.jmapro.2025.04.040","DOIUrl":"10.1016/j.jmapro.2025.04.040","url":null,"abstract":"<div><div>This study presents novel fabrication steps and hybrid approaches that combine contemporary methods with modified manufacturing strategies for fabricating 3D micro-coils on tubular surfaces. Although several methods have been developed in the last decade, most of them still rely on complicated multi-step lithography processes or sophisticated devices. In this article, rapid, simple, and straightforward approaches are presented for fabricating micro-coils directly on curved surfaces. The hybrid fabrication methods in this study provide an alternative solution to reduce the complexity in fabrication steps, provide better control over aspect ratios and the homogeneity of thin film coatings on tubular surfaces. Micro-coils from copper and conductive polymer with outer diameter of 2.5 mm and line width of 40 μm and a separation of 110 μm were designed, produced, and characterized. Magnetic fields inside the coils made of copper and conductive polymer were measured as 11 and 2.5 μT, respectively. The developed coating and structuring methods will open new avenues not only in application fields such as analytical characterization methods of fluidic samples with miniaturized nuclear magnetic resonance (μNMR) systems, but also for 3D sensor fabrication on curved surfaces.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 190-200"},"PeriodicalIF":6.1,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143869870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A non-intrusive interval analysis method for chatter stability of uncertain milling systems

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-24 DOI: 10.1016/j.jmapro.2025.04.047

Jiachang Tang , Taolin Zhang , Tao Liu , Zhanzhan Zhang , Lixiong Cao , Qishui Yao

A non-intrusive interval analysis method is proposed to evaluate the chatter stability of uncertain milling systems. First, the interval theory is introduced to establish the interval milling system dynamics model. Second, a novel contraction and expansion sampling strategy is proposed to obtain the upper and lower bounds of the stability lobes diagram (SLD) in the solution process inspired by the simplex method. Subsequently, a convergence mechanism is proposed based on the edge detection technique to improve the convergence rate of the bounds calculation of the SLDs. The mean square error and structural similarity of the two iteration results were calculated to control the iteration. In addition, the proposed method is non-intrusive. It can be employed by selecting the corresponding chatter stability analysis method according to the characteristics of different milling systems, without being limited to a specific analysis method. Finally, the accuracy and efficiency are verified by the examples of one and two degrees of freedom milling systems, and the feasibility of the proposed method is validated through experiments.

{"title":"A non-intrusive interval analysis method for chatter stability of uncertain milling systems","authors":"Jiachang Tang , Taolin Zhang , Tao Liu , Zhanzhan Zhang , Lixiong Cao , Qishui Yao","doi":"10.1016/j.jmapro.2025.04.047","DOIUrl":"10.1016/j.jmapro.2025.04.047","url":null,"abstract":"<div><div>A non-intrusive interval analysis method is proposed to evaluate the chatter stability of uncertain milling systems. First, the interval theory is introduced to establish the interval milling system dynamics model. Second, a novel contraction and expansion sampling strategy is proposed to obtain the upper and lower bounds of the stability lobes diagram (SLD) in the solution process inspired by the simplex method. Subsequently, a convergence mechanism is proposed based on the edge detection technique to improve the convergence rate of the bounds calculation of the SLDs. The mean square error and structural similarity of the two iteration results were calculated to control the iteration. In addition, the proposed method is non-intrusive. It can be employed by selecting the corresponding chatter stability analysis method according to the characteristics of different milling systems, without being limited to a specific analysis method. Finally, the accuracy and efficiency are verified by the examples of one and two degrees of freedom milling systems, and the feasibility of the proposed method is validated through experiments.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 142-157"},"PeriodicalIF":6.1,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143863432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spatiotemporal prediction and mechanisms of molten pool instability in variable polarity plasma arc robotic welding via CNN-LSTM

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-23 DOI: 10.1016/j.jmapro.2025.04.052

Fan Jiang , Penglin Xiang , Jingbo Liu , Shujun Chen , Shibo Li , Lipeng Guo

This study proposes a method for spatiotemporal prediction of molten pool states via an end-to-end CNN-LSTM model, addressing the dynamic and complex manufacturing scenarios under variable polarity plasma arc (VPPA) robotic welding. The model utilizes CNN to extract spatial features from molten pool images and employs LSTM to extract temporal features in image sequences of the molten pool. This enables early warning of transition from stability to instability of the molten pool states. Experimental results show that when predicting molten pool states at a 1.5 s prediction time using a 0.5 s image sequence sample, the CNN-LSTM model achieves a prediction accuracy of 99.21 %, with a false negative rate of only 0.72 %. In real manufacturing scenarios, the model predicts molten pool that was not part of the training data, achieving a prediction accuracy of 90.61 %. The prediction accuracy was improved to 96.43 % by fine-tuning the model with data not included in the training process. Grad-CAM visualization analysis reveals that the CNN-LSTM model primarily focuses on the rear wall region of the molten pool during the prediction of molten pool states. Insufficient molten metal supply in this region is identified as the key cause of molten pool instability. The proposed model demonstrates well performance in prediction accuracy, false negative rate, and applicability. It provides a robust method for enhancing the intelligence and reliability of VPPA robotic welding processes.

{"title":"Spatiotemporal prediction and mechanisms of molten pool instability in variable polarity plasma arc robotic welding via CNN-LSTM","authors":"Fan Jiang , Penglin Xiang , Jingbo Liu , Shujun Chen , Shibo Li , Lipeng Guo","doi":"10.1016/j.jmapro.2025.04.052","DOIUrl":"10.1016/j.jmapro.2025.04.052","url":null,"abstract":"<div><div>This study proposes a method for spatiotemporal prediction of molten pool states via an end-to-end CNN-LSTM model, addressing the dynamic and complex manufacturing scenarios under variable polarity plasma arc (VPPA) robotic welding. The model utilizes CNN to extract spatial features from molten pool images and employs LSTM to extract temporal features in image sequences of the molten pool. This enables early warning of transition from stability to instability of the molten pool states. Experimental results show that when predicting molten pool states at a 1.5 s prediction time using a 0.5 s image sequence sample, the CNN-LSTM model achieves a prediction accuracy of 99.21 %, with a false negative rate of only 0.72 %. In real manufacturing scenarios, the model predicts molten pool that was not part of the training data, achieving a prediction accuracy of 90.61 %. The prediction accuracy was improved to 96.43 % by fine-tuning the model with data not included in the training process. Grad-CAM visualization analysis reveals that the CNN-LSTM model primarily focuses on the rear wall region of the molten pool during the prediction of molten pool states. Insufficient molten metal supply in this region is identified as the key cause of molten pool instability. The proposed model demonstrates well performance in prediction accuracy, false negative rate, and applicability. It provides a robust method for enhancing the intelligence and reliability of VPPA robotic welding processes.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 116-132"},"PeriodicalIF":6.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143859948","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

High-quality machining of thin copper plate based on error proofing method

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-23 DOI: 10.1016/j.jmapro.2025.04.061

Di Wu

The response characteristics of target materials under dynamic compression have been a focused topic in the fields of shock wave physics and explosion dynamics. Thin copper plate with high-quality surface is frequently applied as flyer in detonation or plate impact experiments to study these characteristics. Nevertheless, existing mechanical machining method is hard to achieve high-quality surface for copper flyer due to ductile property and residual stress. To settle this problem, a technology based on error proofing is proposed to realize thinning, surface shape error control and improving surface quality in turn. Firstly, an isopotential control method is proposed in electrochemical lapping process to suppress stray current corrosion, which lays the foundation for high-efficiency low-stress thinning with MRR = 1 μm /min. Secondly, a chemical mechanical lapping method considering pressure and speed distribution is applied to control surface shape error deterministically with acceptable roughness S_a ≤ 350 nm (measure size: 0.36 mm × 0.27 mm). Then electrochemical mechanical polishing at low pressure (P = 0.27 psi) is applied to obtain surface roughness of S_a ≤ 5 nm with maintained flatness. Finally, a high-quality machining technology for thin copper plate is developed. By the low-stress machining method, copper plate (Φ100 mm × 3 mm) achieves flatness of PV 2 μm and moderate roughness S_a 4.2 nm.

目标材料在动态压缩下的响应特性一直是冲击波物理和爆炸动力学领域的重点课题。具有高质量表面的薄铜板经常被用作引爆或板冲击实验中的飞行器来研究这些特性。然而，由于延展性和残余应力的影响，现有的机械加工方法很难获得高质量的铜飞边表面。为了解决这一问题，我们提出了一种基于误差校正的技术，以实现减薄、控制表面形状误差和提高表面质量。首先，在电化学研磨工艺中提出了抑制杂散电流腐蚀的等电位控制方法，为实现 MRR = 1 μm /min 的高效低应力减薄奠定了基础。其次，采用考虑压力和速度分布的化学机械研磨方法，确定性地控制表面形状误差，可接受的粗糙度 Sa ≤ 350 nm（测量尺寸：0.36 mm × 0.27 mm）。然后，在低压（P = 0.27 psi）下进行电化学机械抛光，以获得表面粗糙度 Sa ≤ 5 nm 并保持平面度。最后，开发出一种高质量的薄铜板加工技术。通过低压加工方法，铜板（Φ100 mm × 3 mm）的平面度达到 PV 2 μm，中等粗糙度 Sa 4.2 nm。

{"title":"High-quality machining of thin copper plate based on error proofing method","authors":"Di Wu","doi":"10.1016/j.jmapro.2025.04.061","DOIUrl":"10.1016/j.jmapro.2025.04.061","url":null,"abstract":"<div><div>The response characteristics of target materials under dynamic compression have been a focused topic in the fields of shock wave physics and explosion dynamics. Thin copper plate with high-quality surface is frequently applied as flyer in detonation or plate impact experiments to study these characteristics. Nevertheless, existing mechanical machining method is hard to achieve high-quality surface for copper flyer due to ductile property and residual stress. To settle this problem, a technology based on error proofing is proposed to realize thinning, surface shape error control and improving surface quality in turn. Firstly, an isopotential control method is proposed in electrochemical lapping process to suppress stray current corrosion, which lays the foundation for high-efficiency low-stress thinning with <em>MRR</em> = 1 μm /min. Secondly, a chemical mechanical lapping method considering pressure and speed distribution is applied to control surface shape error deterministically with acceptable roughness <em>S</em><sub>a</sub> ≤ 350 nm (measure size: 0.36 mm × 0.27 mm). Then electrochemical mechanical polishing at low pressure (<em>P</em> = 0.27 psi) is applied to obtain surface roughness of <em>S</em><sub>a</sub> ≤ 5 nm with maintained flatness. Finally, a high-quality machining technology for thin copper plate is developed. By the low-stress machining method, copper plate (Φ100 mm × 3 mm) achieves flatness of PV 2 μm and moderate roughness <em>S</em><sub>a</sub> 4.2 nm.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 172-189"},"PeriodicalIF":6.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143863430","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Effect of anisotropic property on machining response of selective laser melted Ti6Al4V alloys in high-speed milling

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-23 DOI: 10.1016/j.jmapro.2025.04.066

Dejian Liu , Chenbing Ni , Youqiang Wang , Lida Zhu , Wei Lu , Xingbao Huang

Selective laser melted (SLMed) technology provides an advanced manufacturing method for many complicated and sophisticated components due to the advantage of near-net shape and high-efficiency. However, post-machining SLMed parts is greatly essential for obtaining high-quality surface due to their poor surface integrity. This paper studies the effect of anisotropic property on machining response of SLMed Ti6Al4V alloy fabricated by three laser scanning strategies (0°, 67.5° and 90°). Especially, the effects of needle-like martensitics α′ and melt-pool boundary are considered in this study. High-speed milling experiments are conducted on the top and front surfaces of samples for studying the relationship among anisotropic microstructure, mechanical property and machinability. The cutting force, surface and chip characteristic are applied to evaluate the anisotropy of machinability. The results show that the cutting forces of top surface are larger than those of front surface, which leads to higher surface quality and lower surface roughness of front surface. This can be associated with the distinct melt-pool boundary of similar block grain boundary (top surface) and columnar grain boundary (front surface). The laser scanning strategy changes the distribution of needle-like martensitics α′ and further influences the evolution of microstructure, mechanical property and machinability. The surface formation experiences a coordinated deformed process of melt-pool boundary, plastic flow, grain distortion, dislocation accumulation and entanglement induced by complex thermo-mechanical coupled effect. This paper systematically revealed the underlying mechanism of the surface integrity during milling SLMed Ti6Al4V alloy with different laser scanning strategy and machined surface. The researched results can provide in-depth insights for improving the surface quality and performance of additive manufactured (AMed) parts by post machining technology.

{"title":"Effect of anisotropic property on machining response of selective laser melted Ti6Al4V alloys in high-speed milling","authors":"Dejian Liu , Chenbing Ni , Youqiang Wang , Lida Zhu , Wei Lu , Xingbao Huang","doi":"10.1016/j.jmapro.2025.04.066","DOIUrl":"10.1016/j.jmapro.2025.04.066","url":null,"abstract":"<div><div>Selective laser melted (SLMed) technology provides an advanced manufacturing method for many complicated and sophisticated components due to the advantage of near-net shape and high-efficiency. However, post-machining SLMed parts is greatly essential for obtaining high-quality surface due to their poor surface integrity. This paper studies the effect of anisotropic property on machining response of SLMed Ti6Al4V alloy fabricated by three laser scanning strategies (0°, 67.5° and 90°). Especially, the effects of needle-like martensitics α′ and melt-pool boundary are considered in this study. High-speed milling experiments are conducted on the top and front surfaces of samples for studying the relationship among anisotropic microstructure, mechanical property and machinability. The cutting force, surface and chip characteristic are applied to evaluate the anisotropy of machinability. The results show that the cutting forces of top surface are larger than those of front surface, which leads to higher surface quality and lower surface roughness of front surface. This can be associated with the distinct melt-pool boundary of similar block grain boundary (top surface) and columnar grain boundary (front surface). The laser scanning strategy changes the distribution of needle-like martensitics α′ and further influences the evolution of microstructure, mechanical property and machinability. The surface formation experiences a coordinated deformed process of melt-pool boundary, plastic flow, grain distortion, dislocation accumulation and entanglement induced by complex thermo-mechanical coupled effect. This paper systematically revealed the underlying mechanism of the surface integrity during milling SLMed Ti6Al4V alloy with different laser scanning strategy and machined surface. The researched results can provide in-depth insights for improving the surface quality and performance of additive manufactured (AMed) parts by post machining technology.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 158-171"},"PeriodicalIF":6.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143863433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Particle-on-demand electrohydrodynamic printing from a reciprocating tip

IF 6.1 1区工程技术 Q1 ENGINEERING, MANUFACTURING

Journal of Manufacturing Processes

Pub Date : 2025-04-23 DOI: 10.1016/j.jmapro.2025.04.011

Ji-hun Jeong , Seong Jae Kim , Sanha Kim , A. John Hart

While inkjet printing has revolutionized manufacturing of graphics and decorations, flexible electronics, and has enabled new additive manufacturing (AM) technologies, direct micro-scale deposition of metals remains challenging. Here, we present a particle-on-demand electrohydrodynamic printing approach, using a reciprocating tip mechanism that enables particles to be fed and ejected individually from an oil-coated membrane, and then printed to a target substrate. We examine the mechanism of printing using high-speed imaging and study the limiting mechanisms via controlled experiments with a range of particle sizes and materials, and then extract representative scaling laws for the ejection behavior. Based on this understanding, we demonstrate printing of two-dimensional patterns of stainless steel microparticles over a wide size range (50–700 μm particle diameter). With envisioned improvements to the tip geometry and particle-fluid interaction, and via parallelization, this particle-on-demand approach would be a versatile addition to high-resolution printing technologies for metals, including for manufacturing of intricate miniature components.

虽然喷墨打印已经彻底改变了图形和装饰品、柔性电子产品的制造，并实现了新的增材制造（AM）技术，但金属的直接微尺度沉积仍然具有挑战性。在这里，我们提出了一种颗粒按需电流体动力打印方法，该方法采用往复式针尖机制，可将颗粒从涂有油的膜上单独送入和喷出，然后打印到目标基底上。我们利用高速成像技术检查了打印机制，并通过对一系列颗粒大小和材料的控制实验研究了限制机制，然后为喷射行为提取了具有代表性的缩放定律。在此基础上，我们演示了大尺寸范围（50-700 微米颗粒直径）的不锈钢微颗粒二维图案打印。通过对针尖几何形状和微粒与流体相互作用的改进以及并行化，这种按需打印微粒的方法将成为金属高分辨率打印技术的多功能补充，包括用于制造复杂的微型部件。

{"title":"Particle-on-demand electrohydrodynamic printing from a reciprocating tip","authors":"Ji-hun Jeong , Seong Jae Kim , Sanha Kim , A. John Hart","doi":"10.1016/j.jmapro.2025.04.011","DOIUrl":"10.1016/j.jmapro.2025.04.011","url":null,"abstract":"<div><div>While inkjet printing has revolutionized manufacturing of graphics and decorations, flexible electronics, and has enabled new additive manufacturing (AM) technologies, direct micro-scale deposition of metals remains challenging. Here, we present a particle-on-demand electrohydrodynamic printing approach, using a reciprocating tip mechanism that enables particles to be fed and ejected individually from an oil-coated membrane, and then printed to a target substrate. We examine the mechanism of printing using high-speed imaging and study the limiting mechanisms via controlled experiments with a range of particle sizes and materials, and then extract representative scaling laws for the ejection behavior. Based on this understanding, we demonstrate printing of two-dimensional patterns of stainless steel microparticles over a wide size range (50–700 μm particle diameter). With envisioned improvements to the tip geometry and particle-fluid interaction, and via parallelization, this particle-on-demand approach would be a versatile addition to high-resolution printing technologies for metals, including for manufacturing of intricate miniature components.</div></div>","PeriodicalId":16148,"journal":{"name":"Journal of Manufacturing Processes","volume":"145 ","pages":"Pages 133-141"},"PeriodicalIF":6.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143859750","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0