The energy management strategy plays a critical role in scheduling the operations and enhancing the overall efficiency for electric vehicles. This paper proposes an effective model predictive control-based (MPC) energy management strategy to simultaneously control the battery thermal management system (BTMS) and the cabin air conditioning (AC) system for electric vehicles (EVs). We aim to improve the overall energy efficiency, while retaining soft constraints from both BTMS and AC systems. It is implemented by optimizing the operation and discharging schedule to avoid peak load and by directly utilizing the regenerative power instead of recharging. Compared to the systematic performance without any control coordination between BTMS and AC, results reveal that there are a 4.3% reduction for the recharging energy, and a 6.5% improvement for the overall energy consumption that gained from the MPC-based energy management strategy. Overall the MPC-based energy management is a promising solution to enhance the efficiency for electric vehicles.
{"title":"A Model Predictive Control-Based Energy Management Strategy Considering Electric Vehicle Battery Thermal and Cabin Climate Control","authors":"Yuan Liu, Jie Zhang","doi":"10.1115/detc2020-22318","DOIUrl":"https://doi.org/10.1115/detc2020-22318","url":null,"abstract":"\u0000 The energy management strategy plays a critical role in scheduling the operations and enhancing the overall efficiency for electric vehicles. This paper proposes an effective model predictive control-based (MPC) energy management strategy to simultaneously control the battery thermal management system (BTMS) and the cabin air conditioning (AC) system for electric vehicles (EVs). We aim to improve the overall energy efficiency, while retaining soft constraints from both BTMS and AC systems. It is implemented by optimizing the operation and discharging schedule to avoid peak load and by directly utilizing the regenerative power instead of recharging. Compared to the systematic performance without any control coordination between BTMS and AC, results reveal that there are a 4.3% reduction for the recharging energy, and a 6.5% improvement for the overall energy consumption that gained from the MPC-based energy management strategy. Overall the MPC-based energy management is a promising solution to enhance the efficiency for electric vehicles.","PeriodicalId":415040,"journal":{"name":"Volume 11A: 46th Design Automation Conference (DAC)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121083832","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}
Mojtaba Arezoomand, Elliott J. Rouse, J. Austin-Breneman
A key assumption of new product development is that user requirements and related preferences do not vary on time scales of the process length. However, prior work has identified cases in which user preferences for product attributes can vary with time. This study proposes a method, Design for Dynamic User Preferences, which adapts reinforcement learning (RL) algorithms for designing physical systems whose functionality changes with user feedback. An illustrative case comprised of the design of a variable stiffness prosthetic ankle is presented to evaluate the potential usefulness of the framework. Lifetime user satisfaction for static and dynamic design strategies are compared over simulated user preferences under a number of conditions. Results suggest RL-based strategies outperform static strategies for cases with dynamic user preferences despite significantly less initial information. Within RL methods, upper-confidence bound policies led to higher user satisfaction on average. This study suggests that further investigation into RL-based design strategies is warranted for situations with possibly dynamic preferences.
{"title":"Theoretical Framework for Design for Dynamic User Preferences","authors":"Mojtaba Arezoomand, Elliott J. Rouse, J. Austin-Breneman","doi":"10.1115/detc2020-22460","DOIUrl":"https://doi.org/10.1115/detc2020-22460","url":null,"abstract":"\u0000 A key assumption of new product development is that user requirements and related preferences do not vary on time scales of the process length. However, prior work has identified cases in which user preferences for product attributes can vary with time. This study proposes a method, Design for Dynamic User Preferences, which adapts reinforcement learning (RL) algorithms for designing physical systems whose functionality changes with user feedback. An illustrative case comprised of the design of a variable stiffness prosthetic ankle is presented to evaluate the potential usefulness of the framework. Lifetime user satisfaction for static and dynamic design strategies are compared over simulated user preferences under a number of conditions. Results suggest RL-based strategies outperform static strategies for cases with dynamic user preferences despite significantly less initial information. Within RL methods, upper-confidence bound policies led to higher user satisfaction on average. This study suggests that further investigation into RL-based design strategies is warranted for situations with possibly dynamic preferences.","PeriodicalId":415040,"journal":{"name":"Volume 11A: 46th Design Automation Conference (DAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129667716","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}
Graphs can be used to represent many engineering systems and decisions because of their ability to capture discrete compositional and relational information. In this article, improved methods for effectively representing and generating all graphs in a space defined by certain complex specifications are presented. These improvements are realized through enhancements to the original perfect matching-inspired approach utilizing a component catalog definition to capture the graphs of interest. These enhancements will come in many forms, including more efficient graph enumeration and labeled graph isomorphism checking, expansion of the definition of the component catalog, and the effective inclusion of new network structure constraints. Several examples are shown, including improvements to the original case studies (with up to 971× reduction in computational cost) as well as graph problems in common system architecture design patterns. The goal is to show that the work presented here and tools developed from it can play a role as the domain-independent architecture decision support tool for a variety of graph enumeration-based engineering design challenges.
{"title":"Enhancements to the Perfect Matching Approach for Graph Enumeration-Based Engineering Challenges","authors":"Daniel R. Herber","doi":"10.1115/detc2020-22774","DOIUrl":"https://doi.org/10.1115/detc2020-22774","url":null,"abstract":"\u0000 Graphs can be used to represent many engineering systems and decisions because of their ability to capture discrete compositional and relational information. In this article, improved methods for effectively representing and generating all graphs in a space defined by certain complex specifications are presented. These improvements are realized through enhancements to the original perfect matching-inspired approach utilizing a component catalog definition to capture the graphs of interest. These enhancements will come in many forms, including more efficient graph enumeration and labeled graph isomorphism checking, expansion of the definition of the component catalog, and the effective inclusion of new network structure constraints. Several examples are shown, including improvements to the original case studies (with up to 971× reduction in computational cost) as well as graph problems in common system architecture design patterns. The goal is to show that the work presented here and tools developed from it can play a role as the domain-independent architecture decision support tool for a variety of graph enumeration-based engineering design challenges.","PeriodicalId":415040,"journal":{"name":"Volume 11A: 46th Design Automation Conference (DAC)","volume":"241 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133432310","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}
Glen Williams, N. Meisel, T. Simpson, Christopher McComb
The widespread growth of additive manufacturing, a field with a complex informatic “digital thread”, has helped fuel the creation of design repositories, where multiple users can upload distribute, and download a variety of candidate designs for a variety of situations. Additionally, advancements in additive manufacturing process development, design frameworks, and simulation are increasing what is possible to fabricate with AM, further growing the richness of such repositories. Machine learning offers new opportunities to combine these design repository components’ rich geometric data with their associated process and performance data to train predictive models capable of automatically assessing build metrics related to AM part manufacturability. Although design repositories that can be used to train these machine learning constructs are expanding, our understanding of what makes a particular design repository useful as a machine learning training dataset is minimal. In this study we use a metamodel to predict the extent to which individual design repositories can train accurate convolutional neural networks. To facilitate the creation and refinement of this metamodel, we constructed a large artificial design repository, and subsequently split it into sub-repositories. We then analyzed metadata regarding the size, complexity, and diversity of the sub-repositories for use as independent variables predicting accuracy and the required training computational effort for training convolutional neural networks. The networks each predict one of three additive manufacturing build metrics: (1) part mass, (2) support material mass, and (3) build time. Our results suggest that metamodels predicting the convolutional neural network coefficient of determination, as opposed to computational effort, were most accurate. Moreover, the size of a design repository, the average complexity of its constituent designs, and the average and spread of design spatial diversity were the best predictors of convolutional neural network accuracy.
{"title":"Deriving Metamodels to Relate Machine Learning Quality to Design Repository Characteristics in the Context of Additive Manufacturing","authors":"Glen Williams, N. Meisel, T. Simpson, Christopher McComb","doi":"10.1115/detc2020-22518","DOIUrl":"https://doi.org/10.1115/detc2020-22518","url":null,"abstract":"\u0000 The widespread growth of additive manufacturing, a field with a complex informatic “digital thread”, has helped fuel the creation of design repositories, where multiple users can upload distribute, and download a variety of candidate designs for a variety of situations. Additionally, advancements in additive manufacturing process development, design frameworks, and simulation are increasing what is possible to fabricate with AM, further growing the richness of such repositories. Machine learning offers new opportunities to combine these design repository components’ rich geometric data with their associated process and performance data to train predictive models capable of automatically assessing build metrics related to AM part manufacturability. Although design repositories that can be used to train these machine learning constructs are expanding, our understanding of what makes a particular design repository useful as a machine learning training dataset is minimal. In this study we use a metamodel to predict the extent to which individual design repositories can train accurate convolutional neural networks. To facilitate the creation and refinement of this metamodel, we constructed a large artificial design repository, and subsequently split it into sub-repositories. We then analyzed metadata regarding the size, complexity, and diversity of the sub-repositories for use as independent variables predicting accuracy and the required training computational effort for training convolutional neural networks. The networks each predict one of three additive manufacturing build metrics: (1) part mass, (2) support material mass, and (3) build time. Our results suggest that metamodels predicting the convolutional neural network coefficient of determination, as opposed to computational effort, were most accurate. Moreover, the size of a design repository, the average complexity of its constituent designs, and the average and spread of design spatial diversity were the best predictors of convolutional neural network accuracy.","PeriodicalId":415040,"journal":{"name":"Volume 11A: 46th Design Automation Conference (DAC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134520200","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}
Jennifer Bracken, Christopher McComb, T. Simpson, K. Jablokow
As additive manufacturing (AM) increases in popularity, many companies seek to identify which parts can be produced via AM. This has led to new areas of research known as “part filtering”, “part selection”, or “part identification” for AM. Numerous methods have been proposed to quantify the suitability of a design to be made with AM, and each has its own benefits and drawbacks. This paper reviews popular methods of part filtering and elaborates on the advantages and disadvantages of the various approaches. The approaches for part filtering, and the example methods, are categorized and sorted along a continuum of opportunistic and restrictive methods in order to clarify use cases for various part filtering techniques. The approaches are also examined through the lens of specificity of process, as some are designed to be process agnostic, while others are customized for a specific AM technology or even a specific AM system. Finally, current gaps that exist in the part filtering research literature are discussed to help identify necessary and promising directions for future investigation.
{"title":"A Review of Part Filtering Methods for Additive Manufacturing","authors":"Jennifer Bracken, Christopher McComb, T. Simpson, K. Jablokow","doi":"10.1115/detc2020-22448","DOIUrl":"https://doi.org/10.1115/detc2020-22448","url":null,"abstract":"\u0000 As additive manufacturing (AM) increases in popularity, many companies seek to identify which parts can be produced via AM. This has led to new areas of research known as “part filtering”, “part selection”, or “part identification” for AM. Numerous methods have been proposed to quantify the suitability of a design to be made with AM, and each has its own benefits and drawbacks. This paper reviews popular methods of part filtering and elaborates on the advantages and disadvantages of the various approaches. The approaches for part filtering, and the example methods, are categorized and sorted along a continuum of opportunistic and restrictive methods in order to clarify use cases for various part filtering techniques. The approaches are also examined through the lens of specificity of process, as some are designed to be process agnostic, while others are customized for a specific AM technology or even a specific AM system. Finally, current gaps that exist in the part filtering research literature are discussed to help identify necessary and promising directions for future investigation.","PeriodicalId":415040,"journal":{"name":"Volume 11A: 46th Design Automation Conference (DAC)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122722579","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}
Additive Manufacturing (AM) is a revolutionary development that is being viewed as a core technology for fabricating current and future engineered products. While AM has many advantages over subtractive manufacturing processes, one of the primary limitations of AM is to swiftly evaluate precise part build orientations. Current algorithms are either computationally expensive or provide multiple alternative orientations, requiring additional decision tradeoffs. To hasten the process of finding accurate part build orientation, a data-driven predictive model is introduced by mapping standard machining features to build orientation angles. A combinatory learning algorithm of classification and regression is utilized for the prediction of build orientation. The framework uses 54,000 voxelized standard tessellated language (STL) files as input to train the classification algorithm for eighteen standard machining features using a nine-layer 3D Convolutional Neural Network (CNN). Additionally, a multi-machining feature dataset of 1000 voxelized STL files are evaluated in parallel by performing quaternion rotations to obtain build orientation angles based on minimization of support structure volume. A regression model is then developed to establish a relationship between the machining features and orientation angles to predict optimal build orientation for new parts.
{"title":"Predicting Build Orientation of Additively Manufactured Parts With Mechanical Machining Features Using Deep Learning","authors":"Aliakbar Eranpurwala, S. E. Ghiasian, K. Lewis","doi":"10.1115/detc2020-22043","DOIUrl":"https://doi.org/10.1115/detc2020-22043","url":null,"abstract":"\u0000 Additive Manufacturing (AM) is a revolutionary development that is being viewed as a core technology for fabricating current and future engineered products. While AM has many advantages over subtractive manufacturing processes, one of the primary limitations of AM is to swiftly evaluate precise part build orientations. Current algorithms are either computationally expensive or provide multiple alternative orientations, requiring additional decision tradeoffs. To hasten the process of finding accurate part build orientation, a data-driven predictive model is introduced by mapping standard machining features to build orientation angles. A combinatory learning algorithm of classification and regression is utilized for the prediction of build orientation. The framework uses 54,000 voxelized standard tessellated language (STL) files as input to train the classification algorithm for eighteen standard machining features using a nine-layer 3D Convolutional Neural Network (CNN). Additionally, a multi-machining feature dataset of 1000 voxelized STL files are evaluated in parallel by performing quaternion rotations to obtain build orientation angles based on minimization of support structure volume. A regression model is then developed to establish a relationship between the machining features and orientation angles to predict optimal build orientation for new parts.","PeriodicalId":415040,"journal":{"name":"Volume 11A: 46th Design Automation Conference (DAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123306860","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}
R. Yingling, Anand Balu Nellippallil, Matthew Register, Travis Hannan, J. Simmons, A. Shores, R. Prabhu
Hydrocephalus is a condition that affects humans and animals in which excess cerebrospinal fluid (CSF) builds up within the ventricles of the brain, causing an increase in intracranial pressure. The CSF can be released using a ventriculoperitoneal shunt, which effectively removes the fluid from the ventricles of the brain to the peritoneal cavity. In canines, hydrocephalus is sometimes a fatal condition complicated by shunt failure due to obstructions. The medical procedure is also expensive and has a high failure rate over the long term. In this paper, we present a systematic framework to carry out the multi-objective design exploration of canine shunts for managing hydrocephalus. We demonstrate the efficacy of the framework by designing a shunt prototype to meet specific goals of meeting the CSF flow rate target, minimizing shear stress on the shunt, and minimizing shunt weight. The shunt design variables considered for the problem include the inner diameter, inlet hole diameter, and the distance from the inlet holes to the outlet. A multi-objective design problem is formulated using the systematic framework to explore the combination of shunt design variables that best satisfy the conflicting goals defined. The framework and associated design constructs are generic and support the formulation and decision-based design of similar biomedical devices for different health conditions.
{"title":"Multi-Objective Design Exploration of a Canine Ventriculoperitoneal Shunt for Hydrocephalus","authors":"R. Yingling, Anand Balu Nellippallil, Matthew Register, Travis Hannan, J. Simmons, A. Shores, R. Prabhu","doi":"10.1115/DETC2020-22696","DOIUrl":"https://doi.org/10.1115/DETC2020-22696","url":null,"abstract":"\u0000 Hydrocephalus is a condition that affects humans and animals in which excess cerebrospinal fluid (CSF) builds up within the ventricles of the brain, causing an increase in intracranial pressure. The CSF can be released using a ventriculoperitoneal shunt, which effectively removes the fluid from the ventricles of the brain to the peritoneal cavity. In canines, hydrocephalus is sometimes a fatal condition complicated by shunt failure due to obstructions. The medical procedure is also expensive and has a high failure rate over the long term.\u0000 In this paper, we present a systematic framework to carry out the multi-objective design exploration of canine shunts for managing hydrocephalus. We demonstrate the efficacy of the framework by designing a shunt prototype to meet specific goals of meeting the CSF flow rate target, minimizing shear stress on the shunt, and minimizing shunt weight. The shunt design variables considered for the problem include the inner diameter, inlet hole diameter, and the distance from the inlet holes to the outlet. A multi-objective design problem is formulated using the systematic framework to explore the combination of shunt design variables that best satisfy the conflicting goals defined. The framework and associated design constructs are generic and support the formulation and decision-based design of similar biomedical devices for different health conditions.","PeriodicalId":415040,"journal":{"name":"Volume 11A: 46th Design Automation Conference (DAC)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116708313","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}
Products must often endure unpredictable and challenging conditions while fulfilling their intended functions. Game-theoretic methods make it possible for designers to design solutions that are robust against complicated conditions, however, these methods are often specific to the problems they investigate. This work introduces the Game-Augmented Robust Simulated Annealing Teams (GARSAT) framework, a game-theoretic agent-based architecture that generates solutions robust to variation, and models problems with elementary information, making it easily extendable. The platform was used to generate designs under consideration of a multidimensional attack. Designs were produced under various adversarial settings and compared to designs generated without considering adversaries to validate the model. The process successfully created robust designs able to withstand multiple combined conditions, and the effects of the adversarial settings on the designs were explored.
{"title":"Stochastic Stackelberg Games for Agent-Driven Robust Design","authors":"Sean C. Rismiller, J. Cagan, Christopher McComb","doi":"10.1115/detc2020-22153","DOIUrl":"https://doi.org/10.1115/detc2020-22153","url":null,"abstract":"\u0000 Products must often endure unpredictable and challenging conditions while fulfilling their intended functions. Game-theoretic methods make it possible for designers to design solutions that are robust against complicated conditions, however, these methods are often specific to the problems they investigate. This work introduces the Game-Augmented Robust Simulated Annealing Teams (GARSAT) framework, a game-theoretic agent-based architecture that generates solutions robust to variation, and models problems with elementary information, making it easily extendable. The platform was used to generate designs under consideration of a multidimensional attack. Designs were produced under various adversarial settings and compared to designs generated without considering adversaries to validate the model. The process successfully created robust designs able to withstand multiple combined conditions, and the effects of the adversarial settings on the designs were explored.","PeriodicalId":415040,"journal":{"name":"Volume 11A: 46th Design Automation Conference (DAC)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126432403","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：