检索结果-内蒙古大学图书馆

SENNA: Unified Hardware/Software Space Exploration for Parametrizable Neural Network Accelerators

ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS 2025年第2期24卷 1-26页

作者： Kwon, Jungyoon Min, Yemi Egger, Bernhard Seoul Natl Univ Dept Comp Sci & Engn Seoul South Korea

Parametrizable neural network accelerators enable the deployment of targeted hardware for specialized environments. Finding the best architecture configuration for a given specification, however, is challenging. A large number of hardware configurations have to be considered, and for each hardware instance, an efficient software execution plan needs to be found, leading to a vast search space. Prior work has tackled this problem by dividing the search into subproblems for individual layers of a network. There is no guarantee, however, that the overall best hardware configuration that delivers the desired end-to-end performance across the entire network is among the best individual layer configurations. This work presents SENNA, a unified hardware/software space exploration framework for parametrizable neural network accelerators. To guide the exploration toward the overall best configuration, SENNA employs a multi-objective genetic algorithm with a novel design space representation that encodes the configuration of hardware and software parameters in a single chromosome. Using the Parallel Island Model (PIM), each layer is represented by one or more individual islands each containing a separate population to simultaneously search for the best configuration across the entire network. A tailored gene migration technique enables the exchange of genes between the populations of different islands. SENNA is evaluated with three parametrizable architectures and four neural networks. The evaluation result demonstrates that SENNA achieves upto 1.92x EDP improvement compared to the State-of-the-Art. With equivalent evaluation budgets, SENNA shows 2.5x-9.3x speedup compared to an Oracle scheme and the State-of-the-Art.

关键词： HW/SW co-design reconfigurable accelerators design space exploration multi-objective genetic algorithm parallel island model

来源：评论

学校读者我要写书评

暂无评论

Dynamic modeling method and parameter calibration of subarea thin-layer elements in bolted connection area of diaphragm coupling

引用

PHYSICA SCRIPTA 2025年第3期100卷 036008-036008页

作者： Wen, Huabing Gao, Shuaishuai Jiangsu Univ Sci & Technol Sch Energy & Power Zhenjiang 212100 Peoples R China

Aiming at the finite element modeling problem on the dynamic characteristics of multi-layer elastic diaphragm bolted joints, and adopting the equivalent modeling method of subarea thin-layer elements, a parameter identification method that can accurately and efficiently identify the material parameters of each area of subarea thin-layer elements is proposed. Taking the diaphragm coupling in the marine propulsion shaft system as an example, the accuracy of the simplified model is verified by comparing the dynamics characteristics of the vibrational mode, driving point mechanical impedance, and vibration level drop between test and simulation. The variation of material parameters in each region under different bolt preloads was studied, and the effects of different material parameters on the dynamic properties were analyzed, in which the elastic modulus had a greater effect on the model dynamic characteristics, while Poisson's ratio and density had a smaller effect. The research results show that the subarea thin-layer element modeling method can effectively simplify the dynamic modeling method of the diaphragm coupling under the consideration of bolt preload, and the parameter identification based on the multi-objective genetic algorithm can accurately identify the material parameters of each area of the subarea thin-layer element, which can be used to analyze the dynamic characteristics of the multilayered elastic diaphragm bolted couplings.

关键词： subarea thin-layer element diaphragm coupling dynamic characteristics parameter identification multi-objective genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Joint optimization of energy consumption and latency for task offloading in cloud-edge collaboration system

引用

PHYSICAL COMMUNICATION 2025年 70卷

作者： Jiang, Xue Dou, Haie Wang, Lei Xia, Zhijie Nanjing Univ Posts & Telecommun Sch Internet Things Nanjing Peoples R China Nanjing Univ Posts & Telecommun Jiangsu Engn Res Ctr Commun & Network Technol Nanjing Peoples R China Nanjing Univ Posts & Telecommun Key Lab Broadband Wireless Commun & Sensor Network Nanjing Peoples R China

Ina cloud-edge collaboration system, ensuring consistent wireless connectivity that meets the quality of service (QoS) for users with varying requirements is challenging. Therefore, the approach to task offloading for all mobile devices (MDs) should be carefully designed to enhance performance based on different requirements. This paper introduces a novel multi-objective genetic algorithm (MOGA) to address the task offloading problem, considering minimal energy consumption and latency simultaneously. Experimental results demonstrate that the proposed multi-objective genetic algorithm achieves rapid convergence and optimizes delay and energy consumption.

关键词： Cloud-edge collaborative Task offloading multi-objective genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Simulation Optimization-Based model for Decision-Making in the stroke clinical pathway

引用

COMPUTERS & INDUSTRIAL ENGINEERING 2025年 204卷

作者： Boareto, Pedro A. Moretti, Leonardo N. Safanelli, Juliana Liberato, Rafaela B. Moro, Carla H. C. Pecora Junior, Jose E. Moro, Claudia M. C. B. Coelho, Leandro dos S. Loures, Eduardo F. R. Deschamps, Fernando Santos, Eduardo A. Portela Pontificia Univ Catolica Parana Ind & Syst Engn Grad Program PPGEPS 1155 Imaculada Conceicao St BR-80215901 Curitiba PR Brazil Univ Fed Parana Elect Engn Grad Program Curitiba PR Brazil Joinville Stroke Registry 238 Getulio Vargas Ave BR-89202000 Joinville SC Brazil Municipal Hosp Sao Jose 238 Getulio Vargas Ave BR-89202000 Joinville SC Brazil Univ Fed Parana 632 Prefeito Lothario Meissner Ave BR-80210170 Curitiba PR Brazil Pontificia Univ Catolica Parana Hlth Technol Grad Program Grad Program Hlth Technol 1155 Imaculada Conceicao St BR-80215901 Curitiba PR Brazil Univ Fed Parana Polytech Ctr Dept Mech Engn 230 Cel Francisco H dos Santos AveBloc 4 BR-81530000 Curitiba PR Brazil

Healthcare is highly complex, sensitive and needs constant improvements. Several works have already been developed to support those processes. However, finding the optimum solution takes much work and time. multi-objective genetic algorithms (MOGA) improve the results by finding the optimal trade-off between multiple conflicting objectives and exploring the problem space more thoroughly. This study presents an enhanced framework that integrates Process Mining (PM), Discrete Event Simulation (DES), and multi-objective genetic algorithms (MOGAs) into an optimized, end-to-end pipeline. This framework builds upon an existing non-optimized approach to enable decision-makers to explore and implement Key Performance Indicator (KPI)-oriented solutions directly from raw log data. The framework navigates vast and complex solution spaces by embedding MOGAs into the KPI-oriented simulation process, delivering optimized scenarios with improved performance, boosting decision-making efficiency. The clinical stroke pathway, covering symptoms' onset to hospital discharge, was utilized as a case. This research demonstrates how optimization techniques with classical techniques into one unified framework can accelerate healthcare improvements, offering scalable applications to other domains beyond stroke care. The results demonstrate that using MOGA leads to improved solutions compared to the non-optimized framework, and this approach can be evaluated in short periods due to its performance. The findings underscore the solutions' sensitivity to changes in simulation parameters, emphasizing the importance of considering multiple objectives when dealing with complex decision-making problems in the healthcare industry. Future studies are suggested to extend the model, compare the effectiveness of different optimization methods within the framework, and test the framework's applicability to other domains.

关键词： multi-objective genetic algorithm Stroke Decision making support Optimization Healthcare Discrete Event Simulation

来源：评论

学校读者我要写书评

暂无评论

Efficient risk-based inspection framework: Balancing safety and budgetary constraints

引用

RELIABILITY ENGINEERING & SYSTEM SAFETY 2025年 253卷

作者： Javid, Y. Kharazmi Univ Engn Fac Ind Engn Dept Tehran Iran

Efficient equipment maintenance is paramount across various industries to mitigate energy wastage and avert potential disasters such as hazardous emissions, fires, and explosions. Within this context, the adoption of risk- based inspection strategies has emerged as a crucial method for assessing equipment integrity. This study integrates the principles of Risk-Based Inspection (RBI) with a novel two-objective mathematical model, resulting in a comprehensive framework for equipment inspection programs. The primary aim of this framework is to reduce overall risk exposure while optimizing inspection expenditures. Unlike conventional approaches, this methodology eliminates the necessity to define threshold risk levels. By integrating inspection costs into the model, the assessment of Failure Consequences, and thereby, the decision-making process has been streamlined. This innovative algorithm effectively balances the reduction of failure likelihood with the minimization of inspection costs, enhancing decision-making capabilities. Importantly, this approach offers significant protection against energy wastage and the occurrence of leaks through robust risk management strategies. The algorithm employs specialized operators to expedite the discovery of optimal solutions. Empirical validation through a case study conducted at a Petrochemical Plant highlights the practicality and effectiveness of the proposed framework.

关键词： Inspection program Risk-based inspection multi-objective genetic algorithm Risk reduction

来源：评论

学校读者我要写书评

暂无评论

A resource pre-allocation method for cognitive analytics-based social media service in edge computing

引用

WIRELESS NETWORKS 2024年第6期30卷 6135-6150页

作者： Wang, Ruizhi Zhang, Difei Shanxi Agr Univ Jinzhong 030801 Peoples R China Qufu Normal Univ Sch Math Sci Qufu 273165 Peoples R China

With the rapid development of 5G mobile communication technology, the growing demand for media services in society is becoming a characteristic of smart cities. Technical study has exhaustively investigated the low-latency resource supply of edge computing systems for web hosting services. However, external variables (such as transmission and network delays) impact the transmission between edge servers and service requests, resulting in service request delays. In addition, resource service requests that are constantly updated and in dynamic distribution may overload some servers while others are idle, resulting in poor load balancing. Consequently, this research presents a Resource Pre-Allocation method (RPA) for cognitive analytics-based social media platforms, which aims to improve the load balancing of edge servers while serving requests with strict latency requirements, so as to obtain the optimal resource allocation strategy. First, the resource requirement prediction algorithm is developed based on temporal-spatial demand history. Then, we propose a multi-objective algorithm combined with the optimal solution selection techniques to obtain the ideal resource allocation decisions. Finally, the performance of RPA is tested and evaluated. The experimental results show that RPA can allocate resources better than other methods.

关键词： 5G edge computing Resource requirement prediction Resource pre-allocation multi-objective genetic algorithm Optimal solution selection

来源：评论

学校读者我要写书评

暂无评论

Optimization of Residential Demand Response Program Cost With Consideration for Occupants Thermal Comfort and Privacy

引用

IEEE ACCESS 2024年 12卷 15194-15207页

作者： Nematirad, Reza Ardehali, M. M. Khorsandi, Amir Mahmoudian, Arash Kansas State Univ Elect & Comp Engn Dept Manhattan KS 66506 USA Amirkabir Univ Technol Tehran Polytech Dept Elect Engn Tehran *** Iran Bellevue Univ Coll Sci & Technol Bellevue NE 68005 USA

Residential consumers can optimize their participation in demand response programs (DRPs) using home energy management systems (HEMS). By automatically adjusting air conditioning (AC) setpoints and shifting certain appliances to off-peak hours, HEMS can lead to significant cost reductions. While HEMS aims to adjust AC temperature setpoints, it is important to consider the occupants thermal comfort. This study aims to develop a multi-objective model for the application of DRPs in a smart residential house. The objectives of the model are to achieve (a) reduction in electrical load demand, (b) adjustment in thermal comfort temperature setpoints, and (c) minimization of consumer costs subject to the related constraints. Determining occupancy status through HEMS provides more economic benefits and thermal comfort for consumers. However, traditional methods such as direct occupancy monitoring are often costly, inaccurate, and can intrusively collect data on residents' activities, locations, and routines, compromising their privacy. To tackle these challenges, this study introduces advanced forecasting algorithms such as random forest, light gradient boosting machine, and multilayer perception artificial neural networks to predict occupancy by utilizing indirect data sources, such as energy consumption patterns. This approach enables the prediction of residential presence without direct monitoring. However, inherent uncertainty associated with predicted parameters can compromise the effectiveness of DRPs, and potentially lead to non-optimal energy savings, jeopardizing consumer comfort, and even system instability. To address these uncertainties, this study integrates robust counterpart optimization techniques, augmented with uncertainty budgets to control uncertainty variation making them less conservative. Simulations show uncertainty increases costs by 36% and reduces AC temperature setpoints. Further, implementing DRPs reduces demand by shifting some appliances to off

关键词： Demand response program thermal comfort occupant privacy occupancy forecasting learning algorithms robust counterpart optimization multi-objective genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Through hole-cutting conic posture optimization for a redundant 3D laser cutting machine

引用

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY 2024年第1-2期132卷 443-461页

作者： Ding, Zhipeng Soccio, Pietro Indri, Marina Rizzo, Alessandro Politecn Torino Dept Elect & Telecommun Corso Duca Abruzzi 24 I-10129 Turin Italy Eft EUROPE Srl Corso Duca Abruzzi 2 I-10128 Turin Italy

Productively reducing the time required to cut numerous through holes in automotive workpieces is crucial for enhancing parts manufacturing in the 3D laser cutting process. However, the conventional cutting strategy, in which the laser beam maintains a stationary posture along the hole path, lacks flexibility and fails to effectively leverage processing tolerances. In this study, we conduct a thorough analysis of the kinematics of a six-axis redundant laser cutting machine and resolve through a decoupling method with singularity management. We propose an innovative conic posture cutting strategy for 3D laser hole-cutting with thin materials. This approach adopts the geometry of a cone as the posture while cutting the hole path. In order to obtain the optimal vertex of the cone while minimizing the taper error generated by the conic posture and kinetic energy consumption of the actuators during motion, we formulate a multi-objective optimization problem and solve it using a genetic algorithm. Furthermore, we enhance the optimization by adopting a time minimization approach. Through the implementation of a B-pillar workpiece cutting experiment, we have successfully validated the credibility of our proposed cutting strategy, thereby demonstrating an enhancement of time on 26 hole-cutting paths.

关键词： 3D laser cutting Through hole-cutting Redundant kinematics Posture optimization multi-objective genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Design and multi-objective optimization of reactive pressure-swing distillation process for separating tetrahydrofuran-methanol-water

引用

SEPARATION AND PURIFICATION TECHNOLOGY 2024年 329卷

作者： Liu, Xiaojing Xu, Qilei Ma, Cuncheng Zhang, Fangkun Cui, Peizhe Wang, Yinglong Shan, Baoming Qingdao Univ Sci & Technol Coll Automat & Elect Engn Qingdao 266061 Peoples R China Qingdao Univ Sci & Technol Coll Chem Engn Qingdao 266042 Peoples R China

The low-energy and low-cost process development and design has always been a hot topic and challenge in separating azeotropic systems in the world. In this paper, a novel, energy-saving reactive pressure-swing distillation (RPSD) was, for the first time, designed for separating the ternary mixture of tetrahydrofuran (THF)-methanol-water. In the reactive distillation column, the process of reacting water with ethylene oxide to produce ethylene glycol for the first time was used to separate THF-methanol-water mixture, then the residual mixtures were separated by two pressure-swing distillation columns. Two feasible RPSD separation processes were investigated in terms of thermodynamic modeling and T-xy phase diagram analysis to achieve optimal separation sequence. The two designed separation processes were optimized by a multi-objective genetic algorithm and heat integration to maximize economic benefit. Results demonstrated that, compared to the conventional pressure-swing distillation process, the total annual cost of the designed two heat-integrated processes based on the RPSD-B process were greatly reduced, i.e., 45.9% and 52.2%, respectively. This paper has important guiding significance for the separation and purification of multicomponent azeotropes.

关键词： Reactive pressure-swing distillation THF-methanol-water multi-objective genetic algorithm Energy savings

来源：评论

学校读者我要写书评

暂无评论

An improved multi-objective genetic algorithm for fuzzy flexible job-shop scheduling problem

引用

INTERNATIONAL JOURNAL OF COMPUTER APPLICATIONS IN TECHNOLOGY 2013年第2-3期47卷 280-288页

作者： Wang, Xiaojuan Li, Wenfeng Zhang, Ying Wuhan Univ Technol Sch Logist Engn Wuhan Peoples R China

In many real-world applications, processing time may vary dynamically due to human factors or operating faults and there are some other uncertain factors in the scheduling problems. In this paper, fuzzy sets are used to model uncertain processing time and due date. In addition, an improved multi-objective genetic algorithm is presented to solve the multi-objective fuzzy Flexible Job-shop Scheduling Problem (FJSP). About this improved multi-objective genetic algorithm, Pareto-optimality is applied, including the non-dominated sorting scheme and an improved elite reservation strategy based on NSGA-II. Meanwhile, the immune and entropy principle is used to preserve the diversity of individuals. Moreover, the advanced crossover and the mutation operators are used to adapt to the special chromosome structure. The computational results demonstrate the effectiveness of the proposed algorithm.

关键词： fuzzy FJSP multi-objective genetic algorithm pareto-optimality immune and entropy principle

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：