Haneen A. F. Saymeh, Xiangying Zhang, T. Peng, Renzhong Tang, Zuoxu Wang, Pai Zheng
The dissemination of digital technologies enabled massive data generation and new forms of cooperation. Hence, a new type of innovation named digital innovation emerged. Digital innovation has evolved in different domains and contexts, one of which is called Smart Product-Service System (Smart PSS). Many research works have been done on the development of Smart PSS by leveraging various digital technologies and exploiting its digital capabilities along its lifecycle to generate personalized and innovative solutions. However, there are scarcely any works to explore how the digital innovation process can be manifested to fully exploit digital capabilities throughout Smart PSS development. Aiming to fill this gap, this research provides a theoretical basis for digital innovation in Smart PSS and investigates how it can be in line with the development of Smart PSS from an engineering lifecycle perspective. This work also provides a definition of Smart PSS from the context of digital innovation. A case study on smart office chair is employed to demonstrate the digital innovation process in the usage stage. This work can provide insight and guidance for Smart PSS development and further harness its digital innovation process.
{"title":"Exploration of the Digital Innovation Process in the Smart Product-Service System","authors":"Haneen A. F. Saymeh, Xiangying Zhang, T. Peng, Renzhong Tang, Zuoxu Wang, Pai Zheng","doi":"10.1115/detc2021-70848","DOIUrl":"https://doi.org/10.1115/detc2021-70848","url":null,"abstract":"\u0000 The dissemination of digital technologies enabled massive data generation and new forms of cooperation. Hence, a new type of innovation named digital innovation emerged. Digital innovation has evolved in different domains and contexts, one of which is called Smart Product-Service System (Smart PSS). Many research works have been done on the development of Smart PSS by leveraging various digital technologies and exploiting its digital capabilities along its lifecycle to generate personalized and innovative solutions. However, there are scarcely any works to explore how the digital innovation process can be manifested to fully exploit digital capabilities throughout Smart PSS development. Aiming to fill this gap, this research provides a theoretical basis for digital innovation in Smart PSS and investigates how it can be in line with the development of Smart PSS from an engineering lifecycle perspective. This work also provides a definition of Smart PSS from the context of digital innovation. A case study on smart office chair is employed to demonstrate the digital innovation process in the usage stage. This work can provide insight and guidance for Smart PSS development and further harness its digital innovation process.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"16 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73326073","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 rapid development of computer science and internet technology has enabled the prevailing digital transformation. Nowadays, more and more service companies have built up their smart service platforms. Due to the rapid growth in services, consumers may not get their personalized needs accurately. So companies have concentrated more on smart services than before, such as proactive recommending services. However, few studies focus on smart service recommendation, and few recommendation methods are suitable for service recommending. This paper proposes a smart service recommendation method based on user dynamic portrait and collaborative filtering. Theoretically, this method improves the accuracy of recommendation and reflects the change of users’ preference. Finally, we use the users’ data from a science and technology service platform to verify the effectiveness of the method.
{"title":"Service Recommendation Based on Dynamic User Portrait: An Integrated Approach","authors":"Yuqi Tang, Shanshan Li, Wenyan Song, Caibo Zhou, Zixuan Niu","doi":"10.1115/detc2021-68080","DOIUrl":"https://doi.org/10.1115/detc2021-68080","url":null,"abstract":"\u0000 The rapid development of computer science and internet technology has enabled the prevailing digital transformation. Nowadays, more and more service companies have built up their smart service platforms. Due to the rapid growth in services, consumers may not get their personalized needs accurately. So companies have concentrated more on smart services than before, such as proactive recommending services. However, few studies focus on smart service recommendation, and few recommendation methods are suitable for service recommending. This paper proposes a smart service recommendation method based on user dynamic portrait and collaborative filtering. Theoretically, this method improves the accuracy of recommendation and reflects the change of users’ preference. Finally, we use the users’ data from a science and technology service platform to verify the effectiveness of the method.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"102 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85332028","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}
Diagnosis of mechanical faults in the manufacturing systems is critical for ensuring safety and saving cost. With the development of data transmission and sensor technologies, the measuring systems can easily acquire multi-sensor and massive data. The traditional fault diagnosis methods usually depend on the features extracted by experts manually. The feature extraction process is usually time-consuming and laborious, which has a significant impact on the final results. Although Deep-Learning (DL) provides an end-to-end way to address the drawbacks of traditional methods, it is necessary to do deep research on an intelligent fault diagnosis method based on Multi-Sensor Data and Data Fusion. In this project, a novel intelligent diagnosis method based on Multi-Sensor Data Fusion and Convolutional Neural Network (CNN) is explored, which can automatically extract features from raw signals and achieve superior recognition performance. Firstly, a Multi-Signals-to-RGB-Image conversion method based on Principal Component Analysis (PCA) is applied to fuse multi-signal data into three-channel RGB images, which can eliminate the effect of handcrafted features and obtain the feature-level fused information. Then, the improved CNN with residual networks and the Leaky Rectified Linear Unit (LReLU) is defined and trained by the training samples, which can balance the relationship between computational cost and accuracy. After that, the testing data are fed into CNN to obtain the final diagnosis results. Two datasets, including the KAT bearing dataset and Gearbox dataset, are conducted to verify the effectiveness of the proposed method. The comprehensive comparison and analysis with widely used algorithms are also performed. The results demonstrate that the proposed method can detect different fault types and outperform other methods in terms of classification accuracy. For the KAT bearing dataset and Gearbox dataset, the proposed method’s average prediction accuracy is as high as 99.99% and 99.98%, which demonstrates that the proposed method achieves more reliable results than other DL-based methods.
制造系统中机械故障的诊断对于保证生产安全和节约成本至关重要。随着数据传输和传感器技术的发展，测量系统可以方便地获取多传感器和海量数据。传统的故障诊断方法通常依赖于专家人工提取的特征。特征提取过程通常耗时费力，对最终结果影响很大。尽管深度学习提供了一种端到端的方法来解决传统方法的不足，但有必要对基于多传感器数据和数据融合的智能故障诊断方法进行深入研究。本课题探索了一种基于多传感器数据融合和卷积神经网络(CNN)的智能诊断方法，该方法可以自动从原始信号中提取特征，并取得优异的识别性能。首先，采用基于主成分分析(PCA)的多信号-RGB图像转换方法，将多信号数据融合成三通道RGB图像，消除手工特征的影响，获得特征级融合信息;然后，定义带有残差网络和漏校正线性单元(Leaky Rectified Linear Unit, LReLU)的改进CNN，并使用训练样本进行训练，以平衡计算成本和准确率之间的关系。然后将测试数据输入CNN，得到最终的诊断结果。通过KAT轴承数据集和Gearbox数据集验证了该方法的有效性。并与常用算法进行了全面的比较和分析。结果表明，该方法能够检测出不同类型的故障，在分类精度上优于其他方法。对于KAT轴承数据集和Gearbox数据集，本文方法的平均预测准确率分别高达99.99%和99.98%，表明本文方法比其他基于dl的方法获得了更可靠的预测结果。
{"title":"Multi-Sensor Data Fusion for Rotating Machinery Fault Diagnosis Using Residual Convolutional Neural Network","authors":"Tingli Xie, Xufeng Huang, Seung-Kyum Choi","doi":"10.1115/detc2021-67406","DOIUrl":"https://doi.org/10.1115/detc2021-67406","url":null,"abstract":"\u0000 Diagnosis of mechanical faults in the manufacturing systems is critical for ensuring safety and saving cost. With the development of data transmission and sensor technologies, the measuring systems can easily acquire multi-sensor and massive data. The traditional fault diagnosis methods usually depend on the features extracted by experts manually. The feature extraction process is usually time-consuming and laborious, which has a significant impact on the final results. Although Deep-Learning (DL) provides an end-to-end way to address the drawbacks of traditional methods, it is necessary to do deep research on an intelligent fault diagnosis method based on Multi-Sensor Data and Data Fusion. In this project, a novel intelligent diagnosis method based on Multi-Sensor Data Fusion and Convolutional Neural Network (CNN) is explored, which can automatically extract features from raw signals and achieve superior recognition performance. Firstly, a Multi-Signals-to-RGB-Image conversion method based on Principal Component Analysis (PCA) is applied to fuse multi-signal data into three-channel RGB images, which can eliminate the effect of handcrafted features and obtain the feature-level fused information. Then, the improved CNN with residual networks and the Leaky Rectified Linear Unit (LReLU) is defined and trained by the training samples, which can balance the relationship between computational cost and accuracy. After that, the testing data are fed into CNN to obtain the final diagnosis results. Two datasets, including the KAT bearing dataset and Gearbox dataset, are conducted to verify the effectiveness of the proposed method. The comprehensive comparison and analysis with widely used algorithms are also performed. The results demonstrate that the proposed method can detect different fault types and outperform other methods in terms of classification accuracy. For the KAT bearing dataset and Gearbox dataset, the proposed method’s average prediction accuracy is as high as 99.99% and 99.98%, which demonstrates that the proposed method achieves more reliable results than other DL-based methods.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"75 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82549189","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 practice of treating phobias with Virtual Reality-based therapies is a well-established field. Understanding the level of realism required by the therapy to be most effective is an essential matter of study. This research aims to explore the effects of visual realism on the emotional response in subjects with social phobia when exposed to VR-based applications. Social phobias are triggered by the presence of other people, which translated into virtual environments, refers to avatars. Our hypothesis is that patients with social phobia experience different emotional response to humanlike avatars compared to people without social phobia. To try the hypothesis, a prototype-based survey is conducted. Three types of avatars are implemented with different levels of human likeness: low, medium, and high. The analysis of the collected data suggests that for people with social phobias the anxiety is lowest for avatars with high levels of human likeness. This result is in direct contrast with the uncanny valley effect theory. The research explores how we should design virtual environments to make them more effective in the treatment of phobias. Moreover, the research produces new knowledge about the perception of humanlike avatars in virtual reality.
{"title":"Influence of Realistic Virtual Environments and Humanlike Avatars on Patients With Social Phobia","authors":"Milena Stefanova, M. Pillan, A. Gallace","doi":"10.1115/detc2021-70265","DOIUrl":"https://doi.org/10.1115/detc2021-70265","url":null,"abstract":"\u0000 The practice of treating phobias with Virtual Reality-based therapies is a well-established field. Understanding the level of realism required by the therapy to be most effective is an essential matter of study.\u0000 This research aims to explore the effects of visual realism on the emotional response in subjects with social phobia when exposed to VR-based applications. Social phobias are triggered by the presence of other people, which translated into virtual environments, refers to avatars. Our hypothesis is that patients with social phobia experience different emotional response to humanlike avatars compared to people without social phobia.\u0000 To try the hypothesis, a prototype-based survey is conducted. Three types of avatars are implemented with different levels of human likeness: low, medium, and high. The analysis of the collected data suggests that for people with social phobias the anxiety is lowest for avatars with high levels of human likeness. This result is in direct contrast with the uncanny valley effect theory.\u0000 The research explores how we should design virtual environments to make them more effective in the treatment of phobias. Moreover, the research produces new knowledge about the perception of humanlike avatars in virtual reality.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89461989","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}
Zhangyue Shi, Chenang Liu, Chen Kan, Wenmeng Tian, Yang Chen
With the rapid development of the Internet of Things and information technologies, more and more manufacturing systems become cyber-enabled, which significantly improves the flexibility and productivity of manufacturing. Furthermore, a large variety of online sensors are also commonly incorporated in the manufacturing systems for online quality monitoring and control. However, the cyber-enabled environment may pose the collected online stream sensor data under high risks of cyber-physical attacks as well. Specifically, cyber-physical attacks could occur during the manufacturing process to maliciously tamper the sensor data, which could result in false alarms or failures of anomaly detection. In addition, the cyber-physical attacks may also illegally access the collected data without authorization and cause leakage of key information. Therefore, it becomes critical to develop an effective approach to protect online stream data from these attacks so that the cyber-physical security of the manufacturing systems could be assured. To achieve this goal, an integrative blockchain-enabled method, is proposed by leveraging both asymmetry encryption and camouflage techniques. A real-world case study that protects cyber-physical security of collected stream data in additive manufacturing is provided to demonstrate the effectiveness of the proposed method. The results demonstrate that malicious tampering could be detected in a relatively short time and the risk of unauthorized data access is significantly reduced as well.
{"title":"A Blockchain-Enabled Approach for Online Stream Sensor Data Protection in Cyber-Physical Manufacturing Systems","authors":"Zhangyue Shi, Chenang Liu, Chen Kan, Wenmeng Tian, Yang Chen","doi":"10.1115/detc2021-72023","DOIUrl":"https://doi.org/10.1115/detc2021-72023","url":null,"abstract":"\u0000 With the rapid development of the Internet of Things and information technologies, more and more manufacturing systems become cyber-enabled, which significantly improves the flexibility and productivity of manufacturing. Furthermore, a large variety of online sensors are also commonly incorporated in the manufacturing systems for online quality monitoring and control. However, the cyber-enabled environment may pose the collected online stream sensor data under high risks of cyber-physical attacks as well. Specifically, cyber-physical attacks could occur during the manufacturing process to maliciously tamper the sensor data, which could result in false alarms or failures of anomaly detection. In addition, the cyber-physical attacks may also illegally access the collected data without authorization and cause leakage of key information. Therefore, it becomes critical to develop an effective approach to protect online stream data from these attacks so that the cyber-physical security of the manufacturing systems could be assured. To achieve this goal, an integrative blockchain-enabled method, is proposed by leveraging both asymmetry encryption and camouflage techniques. A real-world case study that protects cyber-physical security of collected stream data in additive manufacturing is provided to demonstrate the effectiveness of the proposed method. The results demonstrate that malicious tampering could be detected in a relatively short time and the risk of unauthorized data access is significantly reduced as well.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"29 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76045574","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}
Many manufacturing enterprises have large collections of solid models and text-based assembly processes to support assembly operations. These data are often distributed across their extended enterprise. As these enterprises expand globally, there is often an increase in product and process variability which can often lead to challenges with training, quality control, and obstacles with change management to name a few. Thus, there is a desire to increase the consistency of assembly work instructions within and across assembly locations. The objective of this research is to retrieve existing 3d models of components and assemblies and their associated assembly work instructions. This is accomplished using 3d solid model similarity and text mining of assembly work instructions. Initially, a design study was conducted in which participants authored assembly work instructions for several different solid model assemblies. Next, a geometric similarity algorithm was used to compute similarity scores between solid models and latent semantic analysis is used to compute the similarity between text-based assembly work instructions. Finally, a correlation study between solid model-assembly instruction tuples is computed. A moderately strong positive correlation was found to exist between solid model similarity scores and their associated assembly instruction similarity scores. This indicates that designs with a similar shape have a similar assembly process and thus can serve as the basis for authoring new assembly processes. This aids in resolving differences in existing processes by linking three-dimensional solid models and their associated assembly work instructions.
{"title":"Computationally Assisted Retrieval and Reuse of 3D Solid Models and Assembly Work Instructions","authors":"R. S. Renu, Gregory M. Mocko","doi":"10.1115/detc2021-70480","DOIUrl":"https://doi.org/10.1115/detc2021-70480","url":null,"abstract":"\u0000 Many manufacturing enterprises have large collections of solid models and text-based assembly processes to support assembly operations. These data are often distributed across their extended enterprise. As these enterprises expand globally, there is often an increase in product and process variability which can often lead to challenges with training, quality control, and obstacles with change management to name a few. Thus, there is a desire to increase the consistency of assembly work instructions within and across assembly locations.\u0000 The objective of this research is to retrieve existing 3d models of components and assemblies and their associated assembly work instructions. This is accomplished using 3d solid model similarity and text mining of assembly work instructions.\u0000 Initially, a design study was conducted in which participants authored assembly work instructions for several different solid model assemblies. Next, a geometric similarity algorithm was used to compute similarity scores between solid models and latent semantic analysis is used to compute the similarity between text-based assembly work instructions. Finally, a correlation study between solid model-assembly instruction tuples is computed. A moderately strong positive correlation was found to exist between solid model similarity scores and their associated assembly instruction similarity scores. This indicates that designs with a similar shape have a similar assembly process and thus can serve as the basis for authoring new assembly processes. This aids in resolving differences in existing processes by linking three-dimensional solid models and their associated assembly work instructions.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"379 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78650962","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 proposes a new way of designing and fabricating conformal flexible electronics on free-form surfaces, which can generate woven flexible electronics designs conforming to free-form 3D shapes with 2D printed electronic circuits. Utilizing our recently proposed foliation-based 3D weaving techniques, we can reap unprecedented advantages in conventional 2D electronic printing. The method is based on the foliation theory in differential geometry, which divides a surface into parallel leaves. Given a surface with circuit design, we first calculate a graph-value harmonic map and then create two sets of harmonic foliations perpendicular to each other. As the circuits are processed as the texture on the surface, they are separated and attached to each leaf. The warp and weft threads are then created and manually woven to reconstruct the surface and reconnect the circuits. Notably, The circuits are printed in 2D, which uniquely differentiates the proposed method from others. Compared with costly conformal 3D electronic printing methods requiring 5-axis CNC machines, our method is more reliable, more efficient, and economical. Moreover, the Harmonic foliation theory assures smoothness and orthogonality between every pair of woven yarns, which guarantees the precision of the flexible electronics woven on the surface. The proposed method provides an alternative solution to the design and physical realization of surface electronic textiles for various applications, including wearable electronics, sheet metal craft, architectural designs, and smart woven-composite parts with conformal sensors in the automotive and aerospace industry. The performance of the proposed method is depicted using two examples.
{"title":"Computational Design and 3D Weaving of 2D-Printable Conformal Flexible Electronics Using Harmonic Foliation Theory","authors":"Qian Ye, Yang Guo, X. Gu, Shikui Chen","doi":"10.1115/detc2021-67811","DOIUrl":"https://doi.org/10.1115/detc2021-67811","url":null,"abstract":"\u0000 This paper proposes a new way of designing and fabricating conformal flexible electronics on free-form surfaces, which can generate woven flexible electronics designs conforming to free-form 3D shapes with 2D printed electronic circuits. Utilizing our recently proposed foliation-based 3D weaving techniques, we can reap unprecedented advantages in conventional 2D electronic printing. The method is based on the foliation theory in differential geometry, which divides a surface into parallel leaves. Given a surface with circuit design, we first calculate a graph-value harmonic map and then create two sets of harmonic foliations perpendicular to each other. As the circuits are processed as the texture on the surface, they are separated and attached to each leaf. The warp and weft threads are then created and manually woven to reconstruct the surface and reconnect the circuits. Notably, The circuits are printed in 2D, which uniquely differentiates the proposed method from others. Compared with costly conformal 3D electronic printing methods requiring 5-axis CNC machines, our method is more reliable, more efficient, and economical. Moreover, the Harmonic foliation theory assures smoothness and orthogonality between every pair of woven yarns, which guarantees the precision of the flexible electronics woven on the surface. The proposed method provides an alternative solution to the design and physical realization of surface electronic textiles for various applications, including wearable electronics, sheet metal craft, architectural designs, and smart woven-composite parts with conformal sensors in the automotive and aerospace industry. The performance of the proposed method is depicted using two examples.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"8 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85689266","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 presents the development of logic rules for evaluating the fitness of function models synthesized by an evolutionary algorithm. A set of 65 rules for twelve different function verbs are developed. The rules are abstractions of the definitions of the verbs in their original vocabularies and are stated as constraints on the quantity, type, and topology of flows connected to the functions. The rules serve as an objective and unambiguous basis of evaluating the fitness of function models developed by a genetic algorithm. The said algorithm and the rules are implemented in software code, which is used to both demonstrate and validate the efficacy of the rule-based approach of converging function model synthesis using GAs.
{"title":"Logic Rules for Automated Synthesis of Function Models Using Evolutionary Algorithms","authors":"Amaninder Singh Gil, Chiradeep Sen","doi":"10.1115/detc2021-70575","DOIUrl":"https://doi.org/10.1115/detc2021-70575","url":null,"abstract":"\u0000 This paper presents the development of logic rules for evaluating the fitness of function models synthesized by an evolutionary algorithm. A set of 65 rules for twelve different function verbs are developed. The rules are abstractions of the definitions of the verbs in their original vocabularies and are stated as constraints on the quantity, type, and topology of flows connected to the functions. The rules serve as an objective and unambiguous basis of evaluating the fitness of function models developed by a genetic algorithm. The said algorithm and the rules are implemented in software code, which is used to both demonstrate and validate the efficacy of the rule-based approach of converging function model synthesis using GAs.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88864995","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}
Kevontrez K. Jones, Zhuo Yang, H. Yeung, P. Witherell, Yan Lu
Laser powder-bed fusion is an additive manufacturing (AM) process that offers exciting advantages for the fabrication of metallic parts compared to traditional techniques, such as the ability to create complex geometries with less material waste. However, the intricacy of the additive process and extreme cyclic heating and cooling leads to material defects and variations in mechanical properties; this often results in unpredictable and even inferior performance of additively manufactured materials. Key indicators for the potential performance of a fabricated part are the geometry and temperature of the melt pool during the building process, due to its impact upon the underlining microstructure. Computational models, such as those based on the finite element method, of the AM process can be used to elucidate and predict the effects of various process parameters on the melt pool, according to physical principles. However, these physics-based models tend to be too computationally expensive for real-time process control. Hence, in this work, a hybrid model utilizing neural networks is proposed and demonstrated to be an accurate and efficient alternative for predicting melt pool geometries in AM, which provides a unified description of the melting conditions. The results of both a physics-based finite element model and the hybrid model are compared to real-time experimental measurements of the melt pool during single-layer AM builds using various scanning strategies.
{"title":"Hybrid Modeling of Melt Pool Geometry in Additive Manufacturing Using Neural Networks","authors":"Kevontrez K. Jones, Zhuo Yang, H. Yeung, P. Witherell, Yan Lu","doi":"10.1115/detc2021-71266","DOIUrl":"https://doi.org/10.1115/detc2021-71266","url":null,"abstract":"\u0000 Laser powder-bed fusion is an additive manufacturing (AM) process that offers exciting advantages for the fabrication of metallic parts compared to traditional techniques, such as the ability to create complex geometries with less material waste. However, the intricacy of the additive process and extreme cyclic heating and cooling leads to material defects and variations in mechanical properties; this often results in unpredictable and even inferior performance of additively manufactured materials. Key indicators for the potential performance of a fabricated part are the geometry and temperature of the melt pool during the building process, due to its impact upon the underlining microstructure. Computational models, such as those based on the finite element method, of the AM process can be used to elucidate and predict the effects of various process parameters on the melt pool, according to physical principles. However, these physics-based models tend to be too computationally expensive for real-time process control. Hence, in this work, a hybrid model utilizing neural networks is proposed and demonstrated to be an accurate and efficient alternative for predicting melt pool geometries in AM, which provides a unified description of the melting conditions. The results of both a physics-based finite element model and the hybrid model are compared to real-time experimental measurements of the melt pool during single-layer AM builds using various scanning strategies.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81476469","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}
Takumi Kuroyanagi, S. Yamada, Shigeki Hiramatsu, Hiroshi Unesaki, Shuichi Kondo, K. Aoyama
Herein, we have confirmed the importance of formulating product proposals and product-development processes equipped to cope effectively with uncertainty in the early design stage. The objective of this study was to derive the target performance and design priority order taking into account uncertainties in system design. Following the concept of set-based design, the approach adopted was to secure a set of solutions as design space that satisfy the target variables demands, dividing the design space into several clusters and evaluating each of the clusters, then gradually narrowing the cluster as the design progresses, and finally extracting the solution space that is desirable. Priority order of design was developed based on the strategy of increasing the degree of freedom of the subsequent process. The effectiveness of the proposed method was verified using the model of a plug-in hybrid vehicle. From the results, we confirmed the existence of a trade-off between design and target variables preference and development risk, that it is possible to determine the extent to which the solution space can be narrowed, that the shape of the solution space determines the design priorities, and we were able to derive a desirable design priority order according to the target performance.
{"title":"System Design Priority Order Considering Uncertainty in Early Stages","authors":"Takumi Kuroyanagi, S. Yamada, Shigeki Hiramatsu, Hiroshi Unesaki, Shuichi Kondo, K. Aoyama","doi":"10.1115/detc2021-68216","DOIUrl":"https://doi.org/10.1115/detc2021-68216","url":null,"abstract":"\u0000 Herein, we have confirmed the importance of formulating product proposals and product-development processes equipped to cope effectively with uncertainty in the early design stage. The objective of this study was to derive the target performance and design priority order taking into account uncertainties in system design.\u0000 Following the concept of set-based design, the approach adopted was to secure a set of solutions as design space that satisfy the target variables demands, dividing the design space into several clusters and evaluating each of the clusters, then gradually narrowing the cluster as the design progresses, and finally extracting the solution space that is desirable. Priority order of design was developed based on the strategy of increasing the degree of freedom of the subsequent process.\u0000 The effectiveness of the proposed method was verified using the model of a plug-in hybrid vehicle. From the results, we confirmed the existence of a trade-off between design and target variables preference and development risk, that it is possible to determine the extent to which the solution space can be narrowed, that the shape of the solution space determines the design priorities, and we were able to derive a desirable design priority order according to the target performance.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"32 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84646748","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：