For carbon fiber-reinforced plastic (CFRP) composites, controlling the interior fiber distribution and orientation during the manufacturing process is a common approach to optimal the structural performance of fabricated parts. However, few studies have been conducted to investigate the fiber alignment during the additive manufacturing of CFRP composites. This study proposes a magnetic field controlled (MFC) method to control the fiber orientation during the fused filament fabrication (FFF) of nickel-coated carbon fiber (NCF) reinforced polymer composites. Firstly, a theoretical analysis model is established to explore the suitable magnetic field intensity for fiber rotation. Secondly, a customized FFF system with MFC components is implemented, and a polylactic acid matrix composite containing 10 wt. % NCF is printed to validate the feasibility of the proposed approach. The microstructure of the printed samples is examined to assess the effectiveness of the method. Finally, uniaxial tensile tests are performed to investigate the impact of fiber orientation adjustment on the mechanical properties. The experimental results reveal that the MFC method can effectively align the interior fiber orientation of CFRP composites, leading to a significant increase in the tensile strength (approximately 8.8 %) and Young's modulus (around 10.5 %) of the printed samples.
{"title":"Investigation of Fiber Orientation of Fused Filament Fabricated CFRP Composites via an External Magnetic Field","authors":"Haoran Zhang, Kaifeng Wang","doi":"10.1115/1.4065354","DOIUrl":"https://doi.org/10.1115/1.4065354","url":null,"abstract":"\u0000 For carbon fiber-reinforced plastic (CFRP) composites, controlling the interior fiber distribution and orientation during the manufacturing process is a common approach to optimal the structural performance of fabricated parts. However, few studies have been conducted to investigate the fiber alignment during the additive manufacturing of CFRP composites. This study proposes a magnetic field controlled (MFC) method to control the fiber orientation during the fused filament fabrication (FFF) of nickel-coated carbon fiber (NCF) reinforced polymer composites. Firstly, a theoretical analysis model is established to explore the suitable magnetic field intensity for fiber rotation. Secondly, a customized FFF system with MFC components is implemented, and a polylactic acid matrix composite containing 10 wt. % NCF is printed to validate the feasibility of the proposed approach. The microstructure of the printed samples is examined to assess the effectiveness of the method. Finally, uniaxial tensile tests are performed to investigate the impact of fiber orientation adjustment on the mechanical properties. The experimental results reveal that the MFC method can effectively align the interior fiber orientation of CFRP composites, leading to a significant increase in the tensile strength (approximately 8.8 %) and Young's modulus (around 10.5 %) of the printed samples.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140690949","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. McGregor, Sagar Patel, Kevin Zhang, Adam Yu, M. Vlasea, Stewart McLachlin
Additive manufacturing (AM) enables new possibilities for the design and manufacturing of complex metal architectures. Incorporating lattice structures into complex part geometries can enhance strength-to-weight and surface area-to-volume ratios for valuable components, particularly in industries such as medical devices and aerospace. However, lattice structures and their interconnections may result in unsupported down-skin surfaces, potentially limiting their manufacturability by metal AM technologies, such as laser powder bed fusion (LPBF). This study aimed at examining the correlation between down-skin surface area and the manufacturability of lattice structures fabricated using LPBF. Image processing algorithms were used to analyze down-skin surface areas of seven unique lattice designs and to devise quantitative metrics (such as down-skin surface area, discrete surface count, surface inter-connectivity, down-skin ratio, over-print/under-print volumes, etc.) to evaluate LPBF manufacturability. The seven lattice designs were subsequently manufactured using maraging steel via LPBF, and then examined using imaging using X-ray micro-computed tomography (XCT). The geometric accuracy of the lattice designs was compared with XCT scans of the manufactured lattices by employing a voxel-based image comparison technique. The results indicated a strong relationship between down-skin surface area, surface interconnectivity, and the manufacturability of a given lattice design. The digital manufacturability evaluation workflow was also applied to a medical device design, further affirming its potential industrial utility for complex geometries.
{"title":"A manufacturability evaluation of complex architectures by laser powder bed fusion additive manufacturing","authors":"M. McGregor, Sagar Patel, Kevin Zhang, Adam Yu, M. Vlasea, Stewart McLachlin","doi":"10.1115/1.4065315","DOIUrl":"https://doi.org/10.1115/1.4065315","url":null,"abstract":"\u0000 Additive manufacturing (AM) enables new possibilities for the design and manufacturing of complex metal architectures. Incorporating lattice structures into complex part geometries can enhance strength-to-weight and surface area-to-volume ratios for valuable components, particularly in industries such as medical devices and aerospace. However, lattice structures and their interconnections may result in unsupported down-skin surfaces, potentially limiting their manufacturability by metal AM technologies, such as laser powder bed fusion (LPBF). This study aimed at examining the correlation between down-skin surface area and the manufacturability of lattice structures fabricated using LPBF. Image processing algorithms were used to analyze down-skin surface areas of seven unique lattice designs and to devise quantitative metrics (such as down-skin surface area, discrete surface count, surface inter-connectivity, down-skin ratio, over-print/under-print volumes, etc.) to evaluate LPBF manufacturability. The seven lattice designs were subsequently manufactured using maraging steel via LPBF, and then examined using imaging using X-ray micro-computed tomography (XCT). The geometric accuracy of the lattice designs was compared with XCT scans of the manufactured lattices by employing a voxel-based image comparison technique. The results indicated a strong relationship between down-skin surface area, surface interconnectivity, and the manufacturability of a given lattice design. The digital manufacturability evaluation workflow was also applied to a medical device design, further affirming its potential industrial utility for complex geometries.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140699998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper introduces a novel hole edge inspection and metrology technology by edge diffractometry, which occurs when light interacts with the hole edge. The proposed method allows for simultaneous characterization of hole part error and edge roughness conditions. Edge diffraction occurs as light bends at a sharp edge. Such a diffractive fringe pattern, the so-called interferogram, is directly related to edge geometry and roughness. Image-based diffractometry inspection technology was developed to capture the diffractive fringe patterns. The collected fringe patterns were analyzed through statistical feature extraction methods, and numerical results such as roundness index, concentricity, and via edge roughness (VER) were obtained. Through-focus scanning optical microscopy (TSOM) was also utilized to perform three-dimensional characterization of the hole features along the depth direction. As a result, the proposed method could characterize hole part error and evaluate its roughness conditions. This study showed the potential to be adapted for automatic optical inspection for advancing microelectronics and semiconductor packaging technology.
{"title":"Hole Edge Metrology and Inspection by Edge Diffractometry","authors":"Kuan Lu, ChaBum Lee","doi":"10.1115/1.4065314","DOIUrl":"https://doi.org/10.1115/1.4065314","url":null,"abstract":"\u0000 This paper introduces a novel hole edge inspection and metrology technology by edge diffractometry, which occurs when light interacts with the hole edge. The proposed method allows for simultaneous characterization of hole part error and edge roughness conditions. Edge diffraction occurs as light bends at a sharp edge. Such a diffractive fringe pattern, the so-called interferogram, is directly related to edge geometry and roughness. Image-based diffractometry inspection technology was developed to capture the diffractive fringe patterns. The collected fringe patterns were analyzed through statistical feature extraction methods, and numerical results such as roundness index, concentricity, and via edge roughness (VER) were obtained. Through-focus scanning optical microscopy (TSOM) was also utilized to perform three-dimensional characterization of the hole features along the depth direction. As a result, the proposed method could characterize hole part error and evaluate its roughness conditions. This study showed the potential to be adapted for automatic optical inspection for advancing microelectronics and semiconductor packaging technology.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140698863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pee-Yew Lee, Chen-Yu Li, Yi-Hong Bai, Hung Ji Huang, Chun-Jen Weng, Yung-Sheng Lin
Abstract In fluoride-assisted galvanic replacement reaction (FAGRR), metallic dendrites are formed simultaneously with hydrogen gas. However, the presence of hydrogen bubbles impedes the reduction of metallic ions to form metallic dendrites. This study investigates the FAGRR approach to manufacturing Ag dendrites where ethanol is incorporated into a AgNO3 reaction solution. The findings of this study demonstrate the efficacy of ethanol as an antifoaming agent in enhancing the deposition of the Ag dendrites during the FAGRR process. The antifoaming effect of ethanol becomes more intense at higher concentrations of AgNO3. The introduction of ethanol into FAGRR can significantly improve the processing efficiency and yield in the limited time for the manufacturing science and engineering.
{"title":"Effects of antifoaming agents on manufacturing silver dendrites through fluoride-assisted galvanic replacement reaction","authors":"Pee-Yew Lee, Chen-Yu Li, Yi-Hong Bai, Hung Ji Huang, Chun-Jen Weng, Yung-Sheng Lin","doi":"10.1115/1.4065277","DOIUrl":"https://doi.org/10.1115/1.4065277","url":null,"abstract":"\u0000 Abstract In fluoride-assisted galvanic replacement reaction (FAGRR), metallic dendrites are formed simultaneously with hydrogen gas. However, the presence of hydrogen bubbles impedes the reduction of metallic ions to form metallic dendrites. This study investigates the FAGRR approach to manufacturing Ag dendrites where ethanol is incorporated into a AgNO3 reaction solution. The findings of this study demonstrate the efficacy of ethanol as an antifoaming agent in enhancing the deposition of the Ag dendrites during the FAGRR process. The antifoaming effect of ethanol becomes more intense at higher concentrations of AgNO3. The introduction of ethanol into FAGRR can significantly improve the processing efficiency and yield in the limited time for the manufacturing science and engineering.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140736530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Modeling multi-track laser-directed energy deposition (LDED) is different from single-track deposition. There is a temporal variation in the deposition geometry and integrity in a multi-track deposition which is not well understood. This paper employs an analytical model for power attenuation and powder catchment in the melt pool in conjunction with a robust fully-coupled metallurgical-thermomechanical finite element (FE) model iteratively to simulate the multi-track deposition. The novel hybrid analytical-numerical approach incorporates the effect of pre-existing tracks on melt pool formation, powder catchment, geometry evolution, dilution, residual stress, and defect generation. CPM 9V steel powder was deposited on the H13 tool steel substrate for validating the model. The deposition height is found to be a function of the track sequence but reaches a steady-state height after a finite number of tracks. The height variation determines the waviness of the deposited surface and, therefore, the effective layer height. The inter-track spacing (I) plays a vital role in steady-state height evolution. A larger value of I facilitates faster convergence to the steady-state height but increases the surface waviness. The FE model incorporates the effects of differential thermal contraction, volume dilation, and transformation-induced plasticity. It predicts the deposition geometry and integrity as a function of inter-track spacing and powder feed rate. The insufficient remelting of the substrate or the preceding track can induce defects. A method to predict and mitigate these defects has also been presented in this paper.
{"title":"Hybrid Analytical-numerical Modeling of Surface Geometry Evolution and Deposition Integrity in a Multi-track Laser-directed Energy Deposition Process","authors":"Chaitanya Vundru, Gourhari Ghosh, Ramesh Singh","doi":"10.1115/1.4065274","DOIUrl":"https://doi.org/10.1115/1.4065274","url":null,"abstract":"\u0000 Modeling multi-track laser-directed energy deposition (LDED) is different from single-track deposition. There is a temporal variation in the deposition geometry and integrity in a multi-track deposition which is not well understood. This paper employs an analytical model for power attenuation and powder catchment in the melt pool in conjunction with a robust fully-coupled metallurgical-thermomechanical finite element (FE) model iteratively to simulate the multi-track deposition. The novel hybrid analytical-numerical approach incorporates the effect of pre-existing tracks on melt pool formation, powder catchment, geometry evolution, dilution, residual stress, and defect generation. CPM 9V steel powder was deposited on the H13 tool steel substrate for validating the model. The deposition height is found to be a function of the track sequence but reaches a steady-state height after a finite number of tracks. The height variation determines the waviness of the deposited surface and, therefore, the effective layer height. The inter-track spacing (I) plays a vital role in steady-state height evolution. A larger value of I facilitates faster convergence to the steady-state height but increases the surface waviness. The FE model incorporates the effects of differential thermal contraction, volume dilation, and transformation-induced plasticity. It predicts the deposition geometry and integrity as a function of inter-track spacing and powder feed rate. The insufficient remelting of the substrate or the preceding track can induce defects. A method to predict and mitigate these defects has also been presented in this paper.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140737311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Changheon Han, Heebum Chun, Jiho Lee, Fengfeng Zhou, Huitaek Yun, ChaBum Lee, M. Jun
Semiconductors play an indispensable role in data collection, processing, and analysis, ultimately enabling more agile and productive operations. Given the importance of wafers in semiconductor fabrication, the purity of a wafer is essential to maintain the integrity of the overall manufacturing process. To tackle this issue, this study proposes a novel Automated Visual Inspection (AVI) framework for scrutinizing semiconductor wafers from scratch, capable of both identifying defective wafers and pinpointing the location of defects through autonomous data annotation. Initially, this proposed methodology leveraged a texture analysis method known as Gray Level Co-occurrence Matrix (GLCM) that categorized wafer images—captured via a stroboscopic imaging system—into distinct scenarios for clear and noisy wafer inspection. GLCM approaches further allowed for a complete separation of noisy wafers into defective and normal wafers as well as the extraction of defect images from noisy defective wafers, which were then used for training a Convolutional Neural Network (CNN) model. Consequently, the CNN model excelled in localizing defects on noisy defective wafers, achieving an F1 score exceeding 0.901. In clear wafers, a background subtraction technique represented defects as clusters of white points. The quantity of these white points not only determined the defectiveness of clear wafers but also pinpointed locations of defects on clear wafers. Lastly, the application of a CNN further enhanced performance, robustness, and consistency irrespective of variations in the ratio of white point clusters. This technique demonstrated accuracy in localizing defects on clear wafers, yielding an F1 score greater than 0.993.
{"title":"Hybrid Semiconductor Wafer Inspection Framework via Autonomous Data Annotation","authors":"Changheon Han, Heebum Chun, Jiho Lee, Fengfeng Zhou, Huitaek Yun, ChaBum Lee, M. Jun","doi":"10.1115/1.4065276","DOIUrl":"https://doi.org/10.1115/1.4065276","url":null,"abstract":"\u0000 Semiconductors play an indispensable role in data collection, processing, and analysis, ultimately enabling more agile and productive operations. Given the importance of wafers in semiconductor fabrication, the purity of a wafer is essential to maintain the integrity of the overall manufacturing process. To tackle this issue, this study proposes a novel Automated Visual Inspection (AVI) framework for scrutinizing semiconductor wafers from scratch, capable of both identifying defective wafers and pinpointing the location of defects through autonomous data annotation. Initially, this proposed methodology leveraged a texture analysis method known as Gray Level Co-occurrence Matrix (GLCM) that categorized wafer images—captured via a stroboscopic imaging system—into distinct scenarios for clear and noisy wafer inspection. GLCM approaches further allowed for a complete separation of noisy wafers into defective and normal wafers as well as the extraction of defect images from noisy defective wafers, which were then used for training a Convolutional Neural Network (CNN) model. Consequently, the CNN model excelled in localizing defects on noisy defective wafers, achieving an F1 score exceeding 0.901. In clear wafers, a background subtraction technique represented defects as clusters of white points. The quantity of these white points not only determined the defectiveness of clear wafers but also pinpointed locations of defects on clear wafers. Lastly, the application of a CNN further enhanced performance, robustness, and consistency irrespective of variations in the ratio of white point clusters. This technique demonstrated accuracy in localizing defects on clear wafers, yielding an F1 score greater than 0.993.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140738804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The high cost of collecting and annotating wafer bin maps (WBMs) necessitates few-shot WBM classification, i.e., classifying WBM defect patterns using a limited number of WBMs. Existing few-shot WBM classification algorithms mainly utilize meta learning methods that leverage knowledge learned in several episodes. However, meta-learning methods require a large amount of additional real WBMs, which can be unrealistic. To help train a network with a few real WBMs while avoiding this challenge, we propose the use of simulated WBMs to pre-train a classification model. Specifically, we employ transfer learning by pre-training a classification network with sufficient amounts of simulated WBMs and then fine-tuning it with a few real WBMs. We further employ ensemble learning to overcome the overfitting problem in transfer learning by fine-tuning multiple sets of classification layers of the network. A series of experiments on a real dataset demonstrate that our model outperforms the meta-learning methods that are widely used in few-shot WBM classification. Additionally, we empirically verify that transfer and ensemble learning, the two most important yet simple components of our model, reduce the prediction bias and variance in few-shot scenarios without a significant increase in training time.
{"title":"Few-shot Classification of Wafer Bin Maps Using Transfer Learning and Ensemble Learning","authors":"Hyeonwoo Kim, Heegeon Yoon, Heeyoung Kim","doi":"10.1115/1.4065255","DOIUrl":"https://doi.org/10.1115/1.4065255","url":null,"abstract":"\u0000 The high cost of collecting and annotating wafer bin maps (WBMs) necessitates few-shot WBM classification, i.e., classifying WBM defect patterns using a limited number of WBMs. Existing few-shot WBM classification algorithms mainly utilize meta learning methods that leverage knowledge learned in several episodes. However, meta-learning methods require a large amount of additional real WBMs, which can be unrealistic. To help train a network with a few real WBMs while avoiding this challenge, we propose the use of simulated WBMs to pre-train a classification model. Specifically, we employ transfer learning by pre-training a classification network with sufficient amounts of simulated WBMs and then fine-tuning it with a few real WBMs. We further employ ensemble learning to overcome the overfitting problem in transfer learning by fine-tuning multiple sets of classification layers of the network. A series of experiments on a real dataset demonstrate that our model outperforms the meta-learning methods that are widely used in few-shot WBM classification. Additionally, we empirically verify that transfer and ensemble learning, the two most important yet simple components of our model, reduce the prediction bias and variance in few-shot scenarios without a significant increase in training time.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140749530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Defective chips in wafer bin maps (WBMs) form different spatial patterns depending on the root causes of process failures. Therefore, the identification of defect patterns in WBMs can help practitioners identify the root causes. Previous studies have focused on wafer-level classification even though chip-level classification can provide additional information regarding defect locations and defect sizes. Chip-level classification is more challenging than wafer-level classification because existing chip-level classification methods require chip-level labels, which are laborious to collect. We propose a method for chip-level defect classification using only wafer-level labels based on weakly supervised semantic segmentation. We first train a classification network using wafer-level labels and extract class activation maps (CAMs), which are visualizations of the discriminative regions. We then generate chip-level pseudo-labels using the extracted CAMs and use these labels to train a segmentation network, which predicts chip-level defect types. Experimental results verify effectiveness of the proposed method.
{"title":"Classification of Chip-level Defect Types in Wafer Bin Maps Using Only Wafer-level Labels","authors":"Hyuck Lee, Hyeonwoo Kim, Heeyoung Kim","doi":"10.1115/1.4065226","DOIUrl":"https://doi.org/10.1115/1.4065226","url":null,"abstract":"\u0000 Defective chips in wafer bin maps (WBMs) form different spatial patterns depending on the root causes of process failures. Therefore, the identification of defect patterns in WBMs can help practitioners identify the root causes. Previous studies have focused on wafer-level classification even though chip-level classification can provide additional information regarding defect locations and defect sizes. Chip-level classification is more challenging than wafer-level classification because existing chip-level classification methods require chip-level labels, which are laborious to collect. We propose a method for chip-level defect classification using only wafer-level labels based on weakly supervised semantic segmentation. We first train a classification network using wafer-level labels and extract class activation maps (CAMs), which are visualizations of the discriminative regions. We then generate chip-level pseudo-labels using the extracted CAMs and use these labels to train a segmentation network, which predicts chip-level defect types. Experimental results verify effectiveness of the proposed method.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140754106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The primary goal of the paper is to monitor the health of the spindle in machine tools to ensure optimal performance and reduce costly downtimes. Spindle health monitoring is essential to detect wear and cracks in spindle bearings, which can be challenging due to their gradual development and hidden locations. The proposed approach combines physics-based modeling and data-driven techniques to monitor spindle health effectively. In Part I and Part II of the paper, mathematical models of bearing faults and spindle imbalance are integrated into the digital model of the spindle. This allows for simulating the operation of the spindle both with and without faults. The integration of fault models enables the generation of vibrations at sensor locations along the spindle shaft. The generated vibration data from the physics-based model are used to train a recurrent neural network-based (RNN) fault detection algorithm. The RNN learns from the labeled vibration spectra to identify different fault conditions. Bayesian optimization is used to automatically tune the hyperparameters governing the accuracy and efficiency of the learning models during the training process. The RNN classifiers are further fine-tuned using a small set of experimentally-collected data for the generalization of the model on real-world data. Once the RNN classifier is trained, it can distinguish between different types of damages and identify their specific locations on the spindle assembly. The proposed algorithms achieved an accuracy of 98.43% on experimental data sets that were not used in training the network.
{"title":"A Physics-based Model-data-driven Method for Spindle Health Diagnosis, Part III: Model Training and Fault Detection","authors":"Chung-Yu Tai, Yusuf Altintas","doi":"10.1115/1.4065227","DOIUrl":"https://doi.org/10.1115/1.4065227","url":null,"abstract":"\u0000 The primary goal of the paper is to monitor the health of the spindle in machine tools to ensure optimal performance and reduce costly downtimes. Spindle health monitoring is essential to detect wear and cracks in spindle bearings, which can be challenging due to their gradual development and hidden locations. The proposed approach combines physics-based modeling and data-driven techniques to monitor spindle health effectively. In Part I and Part II of the paper, mathematical models of bearing faults and spindle imbalance are integrated into the digital model of the spindle. This allows for simulating the operation of the spindle both with and without faults. The integration of fault models enables the generation of vibrations at sensor locations along the spindle shaft. The generated vibration data from the physics-based model are used to train a recurrent neural network-based (RNN) fault detection algorithm. The RNN learns from the labeled vibration spectra to identify different fault conditions. Bayesian optimization is used to automatically tune the hyperparameters governing the accuracy and efficiency of the learning models during the training process. The RNN classifiers are further fine-tuned using a small set of experimentally-collected data for the generalization of the model on real-world data. Once the RNN classifier is trained, it can distinguish between different types of damages and identify their specific locations on the spindle assembly. The proposed algorithms achieved an accuracy of 98.43% on experimental data sets that were not used in training the network.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140752742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mathematical modeling of bearing faults, worn tool holder taper contact interface, and unbalance are presented and integrated into a digital dynamic model of spindles in Part I of this paper. These faults lead to changes in preload and dynamic stiffness over time, consequently resulting in observable vibrations. This paper predicts the vibrations of a spindle at a particular measurement location by simulating the presence of a specific fault or multiple faults during spindle rotation. The vibration spectra generated by the digital spindle model at the spindle speed and its harmonics, the changes in the natural frequencies, and dynamic stiffnesses are correlated to faults with experimental validations. The simulated vibration spectrums are later used in training an artificial neural network for fault condition monitoring presented in Part III of the paper.
{"title":"A Physics-based Model-data-driven Method for Spindle Health Diagnosis, Part II: Dynamic Simulation and Validation","authors":"Chung-Yu Tai, Yusuf Altintas","doi":"10.1115/1.4065221","DOIUrl":"https://doi.org/10.1115/1.4065221","url":null,"abstract":"\u0000 Mathematical modeling of bearing faults, worn tool holder taper contact interface, and unbalance are presented and integrated into a digital dynamic model of spindles in Part I of this paper. These faults lead to changes in preload and dynamic stiffness over time, consequently resulting in observable vibrations. This paper predicts the vibrations of a spindle at a particular measurement location by simulating the presence of a specific fault or multiple faults during spindle rotation. The vibration spectra generated by the digital spindle model at the spindle speed and its harmonics, the changes in the natural frequencies, and dynamic stiffnesses are correlated to faults with experimental validations. The simulated vibration spectrums are later used in training an artificial neural network for fault condition monitoring presented in Part III of the paper.","PeriodicalId":507815,"journal":{"name":"Journal of Manufacturing Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140786514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}