检索结果-内蒙古大学图书馆

Machine Learning-Based Optimization of a Mini-Channel Heatsink Geometry

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING 2023年第9期48卷 12107-12124页

作者： Saeed, Muhammed Kalule, Ramanzani S. S. Berrouk, Abdallah S. S. Alshehhi, Mohamed Almatrafi, Eydhah Khalifa Univ Sci & Technol Mech Engn Dept POB 127788 Abu Dhabi U Arab Emirates Khalifa Univ Sci & Technol Ctr Catalysis & Separat CeCas POB 127788 Abu Dhabi U Arab Emirates Abu Dhabi Maritime Acad POB 54477 Abu Dhabi U Arab Emirates King Abdulaziz Univ Mech Engn Dept POB 80200 Jeddah Saudi Arabia

The research on cooling electronics devices in workstations, servers, and other devices has become one of the rapidly growing technologies today, with the generation of high heat fluxes associated with component compactness and increased power consumption. Hence, the current study aims to enhance the thermohydraulic performance of a compact mini-channel heat removal system based on a machine learning-based optimization technique. The design variables considered for the optimization study are the channel's Reynolds number (Re), fin thickness (t(f)), fin spacing (s(f)), and fin height (h(f)). Initially, the performance of the heat sink geometry with different fin configurations is computed using 3D-RANS simulations. The generated data are then used to train six regression techniques in machine learning to propose the best approach capable of accurately predicting the heat transfer coefficient and pressure drop data. The selected machine learning model is then coupled with the multi-objective genetic algorithm to find the optimal heat sink geometry. It is found that the multilayered perceptron approach re-designed to a deep neural network model efficiently predicts both the heat transfer coefficient and the pressure drop from the available data. The overall performance of the optimized channel geometry is found up to 2.1 times improved than the best available channel configuration in the data pool. Further, the optimized channel geometry's heat transfer coefficient was 14% higher, and the corresponding pressure drop was five times lower.

关键词： Heatsinks Machine learning Neural networks Mini channels multi-objective genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Modeling and design optimization of concentric recoil systems for mortar and tank guns

引用

JOURNAL OF DEFENSE MODELING AND SIMULATION-APPLICATIONS METHODOLOGY TECHNOLOGY-JDMS 2025年

作者： Selim, Mohamed S. Elsaady, Wael A. Abdelsalam, Ossama R. Ibrahim, Ahmed Z. Armed Forces Tech Res Ctr Cairo Egypt Mil Tech Coll Mech Engn Branch POB 4460001 Cairo Egypt

Nowadays, the mounting of guns on wheeled/tracked vehicles is increasingly investigated as it enhances gun mobility. To reduce the force imparted by mortars mounted on light vehicles, this paper presents modeling and design optimization of a new recoiling mortar system that potentially replaces traditional towed, fixed-barrel mortars. A concentric recoil system for the mortar was designed, and the recoil cycle has been mathematically modeled using MATLAB software to determine the time histories of recoil force, displacement, and velocity. The mathematical model has been validated by comparing the calculated and measured firing results of a similar recoil system used with a tank gun. Then, multi-objective genetic algorithm (MOGA) optimization technique was utilized to determine the optimum design parameters of the mortar recoil system. It was found that the newly developed recoil system is capable of reducing the force imparted by the mortar by 63.7%. Further reductions of recoil force, recoil track, impact velocity, and recoil cycle time are obtained by applying the optimization technique.

关键词： Recoiling mortar system tank gun recoil system hydraulic resistance design optimization multi-objective genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

multi-objective OPTIMIZATION OF HOT BLAST STOVE HEAT EXCHANGER

引用

INMATEH-AGRICULTURAL ENGINEERING 2023年第1期69卷 245-259页

作者： Zhang, Ye Jiang, Qing Xu, Liangyuan Huang, Chao Gan, Huasheng Anhui Agr Univ Coll Engn Hefei Anhui Peoples R China Anhui Prov Engn Lab Intelligent Agr Machinery & Eq Hefei Anhui Peoples R China

Shell-and-tube heat exchangers are widely used in many research fields and industrial production processes, but little research has been conducted on the use of heat exchangers for drying crops. This study conducted a numerical simulation of the temperature, velocity, and pressure fields based on the shell-and-tube fluids of a heat exchanger in a biomass particle hot-blast stove. The correctness of the simulation results was verified by test data before simulation, and the mesh was verified to be irrelevant. The application of a multi-objective genetic algorithm in heat exchanger design and optimization was explored, considering five design variables, such as hot tube diameter, transverse pitch, longitudinal pitch, cold flow velocity, and hot flow velocity for optimization. The Nusselt number, friction factor, and comprehensive performance coefficient were used as objective functions for 2D and 3D response surface analysis. The final design variables P1=74.91 mm, P2=104.23 mm, P3=121.37 mm, P4=4.83 m/s, and P5=8.48 m/s were obtained to improve the comprehensive performance coefficient by 16.11%. The heat transfer performance was improved by 9.55% and the resistance performance was reduced by 15%.

关键词： Heat exchanger Hot-blast stove multi-objective genetic algorithm Numerical simulation Response surface method

来源：评论

学校读者我要写书评

暂无评论

Simulation-based optimization and parametric identification of a new skeletal shape-based fins’ heat exchanger

引用

Journal of Thermal Analysis and Calorimetry 2025年 1-31页

作者： Nimbona, Fabrice Jai, Mostapha El Akhrif, Iatimad Rihani, Nadir Fahime, Benaissa El Radouani, Mohammed Mechanic Mechatronic and Command Laboratory ENSAM-Meknes Moulay Ismail University Marjane II Meknes Morocco Euromed University of Fes UEMF Fes Morocco Ecole Normale Supérieure Bujumbura Burundi

This study introduces a new type of fins’ heat sinks, namely the skeleton-based heat exchanger;it combines multi-objective optimization along with the parametric identification of the new design to improve the heat exchange performances. Then, the materials and geometry features were selected according to multi-objective genetic algorithm optimization based on thermal/CFD simulations. By taking advantages of ANSYS Fluent platform, including design of experiments and MOGA, the optimal feasible space of the skeletal geometry features was detected;optimal materials’ performance was obtained near AlSi10Mg aluminum alloy. The second step focused on the system identification of the optimal design;it was mainly based on thermal solicitations namely the Dirac impulse along with the corresponding temperature output;the optimal design was detected to be a 1st-order system, while the validation involved both step and ramp input signals. In sum, results indicate a strong potential to improve heat exchangers using MOGA of the skeletal geometry with the AlSi10Mg;this includes the system effectiveness, the fins’ efficiency, and the overall efficiency that ranged above 90% according to the simulations. In addition, the study was corroborated by a benchmark that allowed positioning the present findings within existing literature;the authors especially pointed out the existing contribution and those of the current work. However, this opens new prospects to further research and practical applications in electronics cooling where future studies are currently under development, focusing on experimenting the proposed design integrating along with the phase change materials integration to enhance this new system. © Akadémiai Kiadó Zrt 2025.

关键词： Energy efficiency Industrial system cooling multi-objective genetic algorithm Simulation-based optimization Skeletal fins’ heat exchanger System identification

来源：评论

学校读者我要写书评

暂无评论

Shape Sensing in Plate Structures through Inverse Finite Element Method Enhanced by multi-objective genetic Optimization of Sensor Placement and Strain Pre-Extrapolation

引用

SENSORS 2024年第2期24卷 608页

作者： Del Priore, Emiliano Lampani, Luca Sapienza Univ Roma Dipartimento Ingn Meccan & Aerosp I-00184 Rome Italy

The real-time reconstruction of the displacement field of a structure from a network of in situ strain sensors is commonly referred to as "shape sensing". The inverse finite element method (iFEM) stands out as a highly effective and promising approach to perform this task. In the current investigation, this technique is employed to monitor different plate structures experiencing flexural and torsional deformation fields. In order to reduce the number of installed sensors and obtain more accurate results, the iFEM is applied in synergy with smoothing element analysis (SEA), which allows the pre-extrapolation of the strain field over the entire structure from a limited number of measurement points. For the SEA extrapolation to be effective for a multitude of load cases, it is necessary to position the strain sensors appropriately. In this study, an innovative sensor placement strategy that relies on a multi-objective genetic algorithm (NSGA-II) is proposed. This approach aims to minimize the root mean square error of the pre-extrapolated strain field across a set of mode shapes for the examined plate structures. The optimized strain reconstruction is subsequently utilized as input for the iFEM technique. Comparisons are drawn between the displacement field reconstructions obtained using the proposed methodology and the conventional iFEM. In order to validate such methodology, two different numerical case studies, one involving a rectangular cantilevered plate and the other encompassing a square plate clamped at the edges, are investigated. For the considered case studies, the results obtained by the proposed approach reveal a significant improvement in the monitoring capabilities over the basic iFEM algorithm with the same number of sensors.

关键词： structural health monitoring shape sensing sensor placement inverse finite element method smoothing element analysis multi-objective genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Enhancing agricultural sustainability: Optimizing crop planting structures and spatial layouts within the water-land-energy-economy-environment-food nexus

引用

GEOGRAPHY AND SUSTAINABILITY 2025年第3期6卷

作者： Wu, Haowei Li, Zhihui Deng, Xiangzheng Zhao, Zhe Chinese Acad Sci Inst Geog Sci & Nat Resources Res Beijing 100101 Peoples R China Chinese Acad Sci Key Lab Land Surface Pattern & Simulat Beijing 100101 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China Liaoning Univ Fac Econ Sch Econ Shenyang 110136 Peoples R China

In the new phase of sustainable development, agriculture is seeking sustainable management of the water-land-energy-economy-environment-food nexus. At present, there are few studies on optimizing crop planting structure and analyzing its spatial layout with consideration of natural and socio-economic factors. Herein, we proposed a framework for addressing this issue. In this framework, the NSGA-II algorithm was used to construct the multi-objective optimization model of crop planting structures with consideration of water and energy consumption, greenhouse gas (GHG) emissions, economic benefits, as well as food, land, and water security constraints, while the model for planting spatial layout optimization was established with consideration of crop suitability using the MaxEnt model and the improved Hungarian algorithm. This framework was further applied in the Black Soil Region of Northeast China (BSRNC) for analyzing optimized crop planting structures and spatial layouts of three main crops (rice, maize, and soybean) under various scenarios. This study showed that the sown area of rice in the BSRNC decreased by up to 40.73 % and 35.30 % in the environmental priority scenario and economic-environmental balance scenario, respectively, whereas that of soybean increased by up to 112.44 % and 63.31 %, respectively. In the economic priority scenario, the sown area of rice increased by up to 93.98 %. Expanding the sown area of soybean was effective in reducing GHG emissions. On the contrary, rice production led to greater environmental costs though it provided higher economic returns. Among the three crops, maize exhibited an advantage in balancing environmental and economic benefits. Hegang-Jixi area in the northeast of the BSRNC was identified as the key area with the most intense crop planting transfer among different scenarios. Overall, this framework provides a new methodology for optimizing crop planting structures and spatial layouts with consideration of the nexus

关键词： Planting structure optimization Crop spatial layout multi-objective genetic algorithm Food security Agricultural sustainability Water-land-energy-economy-environment-food nexus

来源：评论

学校读者我要写书评

暂无评论

A circular approach to foster additive manufacturing early design stages sustainability: a methodological proposal

引用

INTERNATIONAL JOURNAL OF INTERACTIVE DESIGN AND MANUFACTURING - IJIDEM 2023年第2期18卷 815-836页

作者： Faveto, Alberto Lombardi, Franco Chiabert, Paolo Segonds, Frederic Politecn Torino Dept Management & Prod Engn Corso Duca Abruzzi 24 I-10129 Turin Italy HESAM Univ LCPI Arts & Metiers Inst Technol 151 Blvd Hop F-75013 Paris France

The design of new products is now influenced by shifting consumer demands and technological advancements. Products must satisfy high-quality standards and have a low environmental impact. New phenomena such as distributed and urban manufacturing are emerging to cope with this. A new manufacturing era is coming where methods that prevent waste, support small workshops and encourage do-it-yourself are crucial. In the early design stage, the process knowledge is minimal, and the decision taken is vital. For this reason, it is essential to support designers in anticipating the impact of decisions on the final product. This paper establishes the groundwork for decision-support methodologies for sustainable design in One-of-a-Kind additive manufacturing prototyping. Our proposed method is applied to a Fused Filament Fabrication case study, wherein we evaluate the impact of nine variables on factors such as process time, energy and material consumption, environmental footprint, and product quality. The initial step aims to generate fresh insights through Taguchi experimentation, while the subsequent step formulates and resolves a multi-objective optimization problem using the NSGA-II algorithm. The resulting Pareto-optimal solutions serve as the basis for a novel visual-based design support tool. The proposed approach can evaluate the trade-offs between product quality and environmental impact by offering users a visual heatmap based on quantitative data. This heatmap can guide the user in the material and production parameter selection. Integrating the decision support tool into the product design process can empower designers to create environmentally responsible products while fostering innovation.

关键词： Additive manufacturing One-of-a-Kind production Sustainability Sustainable manufacturing multi-objective genetic algorithm Design support tool

来源：评论

学校读者我要写书评

暂无评论

Optimizing Stakeholder Behavior for Groundwater Management in Arid and Semi-arid Climates: A multi-agent System Approach in Karaj Plain, Iran

引用

Iranian Journal of Science and Technology - Transactions of Civil Engineering 2024年第6期48卷 4751-4767页

作者： Khajeem Moghadam, Atena Saghafian, Bahram Najarchi, Mohsen Delavar, Majid Department of Civil Engineering Islamic Azad University Arak Branch Arak Iran Civil Department Science and Research Branch Islamic Azad University Tehran Iran Department of Water Resources Engineering Tarbiat Modares University Tehran Iran

The ever-increasing water demand has not only disrupted the demand-supply balance, causing conflicts among multiple stakeholders, but has also led to environmental impacts due to aquifer depletion. Today, agent-based modeling (ABM) is recommended for studying the effects of stakeholders’ social aspects on water resources, particularly groundwater resources. Aim of this study proposes a model for optimizing stakeholder behavior, which is coupled with numerical groundwater resource simulation and water resource optimization models in Karaj Plain, Iran. The simulation results of the aquifer showed that continuing the current situation will cause the aquifer level to drop by about 5 m. However, implementing the water resource optimization model, with goals of reducing water shortages and improving underground water conditions, will increase the aquifer reserve. This model will also provide 99.7% of the region’s drinking water at a rate of 3.03 MCM per year, raising the aquifer level by 3 m. Additionally, the optimization model of beneficiaries’ behavior and the analysis of proposed scenarios demonstrated that reducing agricultural water rights, combined with policies aimed at changing agents’ behavior in the agent base model, would increase the average groundwater level across the entire aquifer by about 5.5 m. Taking into account the local knowledge of the main water consumers, it is recommended to use multi-agent system models to incorporate the ideas of other agents (decision-makers) and develop the proposed cultivation pattern for effectively managing water resources and reducing environmental risks in arid and semi-arid climates. © The Author(s), under exclusive licence to Shiraz University 2024.

关键词： Agent-based model Environment Groundwater multi-objective genetic algorithm Optimal water allocation Simulation

来源：评论

学校读者我要写书评

暂无评论

multi-classification of Alzheimer's Disease by NSGA-II Slices Optimization and Fusion Deep Learning 17th

Multi-classification of Alzheimer's Disease by NSGA-II Slice...

引用

17th Italian Workshop on Artificial Life and Evolutionary Computation (WIVACE)

作者： Rojas-Valenzuela, Ignacio Rojas, Ignacio Delgado-Marquez, Elvira Valenzuela, Olga Univ Granada ETSIIT Granada Spain Univ Leon Dept Econ & Stat Leon Spain Univ Granada Fac Sci Granada Spain

ISBN: (纸本)9783031574290;9783031574306

This paper presents a two-phase hierarchical classifier for determining the different states in Alzheimer's disease (AD). In the first phase, an evolutionary system is developed to determine the most relevant slices (in both X-axis and Y-axis) of the magnetic resonance imaging (MRI) for the construction of a classifier. To obtain the image features, the biorthogonal wavelet transform 3.3 was used at level 2. Due to the high number of coefficients, a dimensionality reduction is performed using minimum Redundancy - Maximum Relevance algorithm (mRMR) and Principal Component Analysis (PCA). An evolutionary algorithm on a high-performance computer with GPU was used to optimize the slides. Support vector machine (SVM) was used in the fitness function to estimate the features of the classifier in a computationally simple way. In the second phase, using the different solutions of the Pareto front obtained by the evolutionary algorithm, a multiple deep learning system was developed, each of the systems having as input one of the selected slices of the analyzed solution. The solution with three slices (trade-off between complexity and accuracy) was used as the solution. The obtained hierarchical deep learning system fused the information from each system and analyzed the probabilities obtained for each class. As a final result, an accuracy of 92% was obtained for the six classes. A total of 1,200 patients from the Alzheimer's disease neuroimaging initiative (ADNI) database were used, corresponding to six different classes of patients (with varying degrees of dementia).

关键词： Alzheimer's Disease multiclass Classification multi-objective genetic algorithm Hierarchical System Deep Learning Ensemble System

来源：评论

学校读者我要写书评

暂无评论

A design approach of bonding head guiding platform for die to wafer hybrid bonding application using compliant mechanism

引用

INTERNATIONAL JOURNAL OF INTERACTIVE DESIGN AND MANUFACTURING - IJIDEM 2023年第6期17卷 2855-2866页

作者： Duc Nam Nguyen Minh Phung Dang Dixit, Saurav Thanh-Phong Dao Ind Univ Ho Chi Minh City Fac Mech Engn Ho Chi Minh City Vietnam Ho Chi Minh City Univ Technol & Educ Fac Mech Engn Ho Chi Minh City Vietnam Uttaranchal Univ Div Res & Innovat Dehra Dun 248007 Uttarakhand India Ton Duc Thang Univ Inst Computat Sci Div Computat Mechatron Ho Chi Minh City Vietnam Ton Duc Thang Univ Fac Elect & Elect Engn Ho Chi Minh City Vietnam

In semiconductor manufacturing process, the wafer direct bonding, so-called die-to-wafer hybrid bonding, still has unexpected voids on the boning surface which decreases the bonding efficiency. Therefore, this paper proposes a design approach for the bonding head guiding platform (BHGP) based on compliant mechanism for the die-to-wafer hybrid bonding. Firstly, a bridge amplification module (BAM) is designed and simulated. Secondly, a dual parallelogram module (DPM) is designed and analyzed. Then, the BAM is combined with the DPM to create a full design of the BHGP. To investigate the parametric effect of the geometric inputs on the output behaviors of the platform, the parameter sensitivity is conducted. The multi-objective optimization study is performed using multi-objective genetic algorithm. The results determined the displacement can gain up to 191.88 mu m, the parasitic error has a small value of 1.47 mu m, and the von Mises stress is minimal value of 52 MPa. The crosstalk of the platform is about 3.82%, and this ratio ensures a linear characteristic of the platform. A small value of stress allows a high safety factor during operation. The result of this study is expected be utilized for removing voids in the die-to-wafer hybrid bonding process.

关键词： Die-to-wafer hybrid bonding Bonding head guiding platform Bridge amplification module Dual parallelogram module multi-objective genetic algorithm

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

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

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：