检索结果-内蒙古大学图书馆

Low Capillary Elastic Flow model optimization Using the Lattice Boltzmann Method and Non-Dominated Sorting Genetic Algorithm

引用

MICROMACHINES 2025年第3期16卷 298-298页

作者： Hou, Yaqi Zhang, Wei Hu, Jiahua Gao, Feiyu Zong, Xuexue Taiyuan Univ Technol Coll Chem & Chem Engn Shanxi Key Lab Chem Prod Engn Taiyuan 030024 Peoples R China Lanzhou Univ Technol Coll Energy & Power Engn Lanzhou 730050 Peoples R China

In simulations of elastic flow using the lattice Boltzmann method (LBM), the steady-state behavior of the flow at low capillary numbers is typically poor and prone to the formation of bubbles with inhomogeneous lengths. This phenomenon undermines the precise control of heat transfer, mass transfer, and reactions within microchannels and microreactors. This paper establishes an LBM multiphase flow model enhanced by machine learning. The hyperparameters of the machine learning model are optimized using the particle swarm algorithm. In contrast, the non-dominated sorting genetic algorithm (NSGA-II) is incorporated to optimize bubble lengths and stability. This results in a coupled multiphase flow numerical simulation model that integrates LBM, machine learning, and the particle swarm algorithm. Using this model, we investigate the influence of elastic flow parameters on bubble length and stability in a T-shaped microchannel. The simulation results demonstrate that the proposed LBM multiphase flow model can effectively predict bubble elongation rates under complex conditions. Furthermore, multi-objective optimization determines the optimal gas-liquid two-phase inlet flow rate relationship, significantly mitigating elastic flow instability at low capillary numbers. This approach enhances the controllability of the elastic flow process and improves the efficiency of mass and heat transfer.

关键词： lattice Boltzmann method microchannel elastic flow machine learning NSGA-II model optimization

来源：评论

学校读者我要写书评

暂无评论

Automated deep-learning model optimization framework for microcontrollers

引用

ETRI JOURNAL 2025年第2期47卷 179-192页

作者： Hong, Seungtae Park, Gunju Kim, Jeong-Si Elect & Telecommun Res Inst Artificial Intelligence Comp Res Lab Daejeon South Korea Univ Sci & Technol Daejeon South Korea

This paper introduces a framework for optimizing deep-learning models on microcontrollers (MCUs) that is crucial in today's expanding embedded device market. We focus on model optimization techniques, particularly pruning and quantization, to enhance the performance of neural networks within the limited resources of MCUs. Our approach combines automatic iterative optimization and code generation, simplifying MCU model deployment without requiring extensive hardware knowledge. Based on experiments with architectures, such as ResNet-8 and MobileNet v2, our framework substantially reduces the model size and enhances inference speed that are crucial for MCU efficiency. Compared with TensorFlow Lite for MCUs, our optimizations for MobileNet v2 reduce static random-access memory use by 51%-57% and flash use by 17%-62%, while increasing inference speed by approximately 1.55 times. These advancements highlight the impact of our method on performance and memory efficiency, demonstrating its value in embedded artificial intelligence and broad applicability in MCU-based neural network optimization.

关键词： automated framework deep learning memory efficiency microcontrollers model optimization

来源：评论

学校读者我要写书评

暂无评论

model optimization method and connected-pipe experiment of a liquid fuel ramjet engine

引用

航空动力学报 2013年第6期28卷 1277-1285页

作者： MA Qian-rong GUO Xin WU Hu CHOU Qian School of Power and Energy Northwest Polytechnical University Aero-Engine Altitude Simulated Key Laboratory China Gas Turbine EstablishmentAviation Industry Corporation of China

The optimization method of a mathematical model and connected-pipe experimental technique for a test in altitude test facility (ATF) of a liquid fuel ramjet engine was *** optimization of the simple mathematical model was divided into two ***,using the test engine's geometry configuration size data,a preliminary adjustment was ***,using experimental test data,the components' experiential coefficients were modified *** was laid on the simulation technique of flight condition and parameters measurement *** experimental technique was applied to a ramjet ATF test *** comparison results show that the optimized-model has higher precision and the nozzle gross thrust difference drops from 12% to about 4%.

关键词： ramjet engine model optimization altitude test facility(ATF) connected-pipe experiment simulation technique

来源：评论

学校读者我要写书评

暂无评论

model optimization of cadmium and accumulation in switchgrass (Panicum virgatum L.): potential use for ecological phytoremediation in Cd-contaminated soils

引用

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH 2015年第21期22卷 16758-16771页

作者： Wang, Quanzhen Gu, Muyu Ma, Xiaomin Zhang, Hongjuan Wang, Yafang Cui, Jian Gao, Wei Gui, Jing Northwest A&F Univ Coll Anim Sci & Technol Dept Grassland Sci Yangling 712100 Shaanxi Peoples R China Northwest A&F Univ Coll Life Sci Inst Plant Sci Yangling 712100 Shaanxi Peoples R China

Soil pollution with heavy metals is an increasingly serious threat to the environment, food security, and human health. Therefore, it is urgent to develop economic and highly efficient soil restoration technology for environmental improvement;phytoremediation is an option that is safe, has low cost, and is environmentally friendly. However, in selecting hyperaccumulators or tolerant plants, theories and operation technologies for optimal restoration should be satisfied. In this study, the switchgrass growth response and performance of phytoextraction under the coupling effect of Cd and pH were investigated by evaluating seed germination, seedling growth, and the Cd content in the plant to evaluate the potential use of switchgrass as a phytoremediation plant in cadmium contaminated soil. This study conducted three sets of independent experiments with five levels of Cd concentrations, including two orthogonal matrix designs of combining Cd with pH values. The results showed that switchgrass was germinated well under all treatments (Cd concentration of 0-500 mu M), but the seedling growth was significantly affected by Cd and pH, as shown by multivariate regression analyses. Hormesis was found during the growth of switchgrass plants exposed to low Cd concentrations under hydroponic conditions, and switchgrass plants were capable of developing with a Cd concentration of 100-175 mu M and pH of 4.1-5.9. Mild acidic conditions can enhance the ability of Cd to accumulate in switchgrass. Switchgrass was moderately tolerant to Cd and may be used as a phytoremediation plant for Cd-contaminated soils in the future. Our results also suggest that hormetic effects should be taken into consideration in the phytoremediation of Cd-contaminated soils. We discuss the physiological and biochemical mechanisms contributing to the effective application of the plant for the phytoremediation of Cd-contaminated soils.

关键词： model optimization Cadmium accumulation Switchgrass Phytoremediation Cd-contaminated soils

来源：评论

学校读者我要写书评

暂无评论

model optimization of Pollution Deposition Criterion of Insulators Based on Energy

引用

IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION 2017年第5期24卷 2920-2929页

作者： Lv, Yukun Li, Jingang Zhao, Weiping North China Elect Power Univ Sch Energy Power & Mech Engn Baoding 071003 Peoples R China Jinneng Clean Energy Ltd Co Taiyuan 030002 Peoples R China

In order to study the contamination characteristics of insulators under different conditions and to ensure the safe operation of electrical systems, the mechanical properties and motor characteristics/behavior of particles near the insulators were analyzed. A wind tunnel contamination test of a porcelain insulator was also conducted. The deposition criteria model based on energy of particles on the insulator surface was progressively optimized from three aspects: particle positive and negative charge ratio, insulator and particle surface energy, and particle size distribution. The effect of the contamination characteristics of the porcelain insulators in the wind tunnel was investigated, and the results showed that three optimization models were obtained. Compared with the Wind-tunnel contamination test results, the average relative error of the simulation results decreased by 6.6%, 21.5%, and 6.4%, respectively, and their surface deposit amount all display a consistent changing pattern with DC voltage variation and a V-shaped distribution, which verifies the rationality of the three optimization models. Thus, it can be used as a reference for numerical simulation of porcelain insulators.

关键词： Porcelain insulator contamination characteristics deposition criteria model optimization numerical simulation

来源：评论

学校读者我要写书评

暂无评论

model optimization and validation with experimental data using the case study of a building equipped with photovoltaic panel on roof: Coupling of the building thermal simulation code ISOLAB with the generic optimization program GenOpt

引用

ENERGY AND BUILDINGS 2013年 58卷 333-347页

作者： Bigot, Dimitri Miranville, Frederic Boyer, Harry Bojic, Milorad Guichard, Stephane Jean, Aurelien Univ La Runion St Denis France Univ Kragujevac Kragujevac Serbia

This article deals with the optimization and validation of a thermal model of photovoltaic panels integrated to buildings. The PV thermal model is implemented in the building simulation code ISOLAB and have already been the subject of previous works. The optimization is made by using the generic optimization program GenOpt. First, these two programs are presented. Secondly, the case study is presented both experimental and modeling way, and main results of previous work are given. Thirdly, the validation methodology is described. Fourthly, the coupling between ISOIAB and GenOpt is presented, and optimization and validation criteria are given. Finally, results of the experimental optimization are presented and the model is validated. (C) 2012 Elsevier B.V. All rights reserved.

关键词： Thermal modeling model optimization model validation GenOpt ISOLAB PV modeling Building simulation

来源：评论

学校读者我要写书评

暂无评论

model optimization for Very-short-term Wind Power Forecasting Using Hilbert-Huang Transform

Model Optimization for Very-short-term Wind Power Forecastin...

引用

6th International Conference on Smart Grid and Clean Energy Technologies (ICSGCE)

作者： Shi, Jie Gong, Yulei Liu, Xiaofei Zhu, Xiaoxun Univ Jinan Sch Phys & Technol Jinan Peoples R China Univ Jinan Sch Elect Engn Jinan Peoples R China Jinan Urban Planning Advisory Serv Ctr Jinan Peoples R China North China Elect Power Univ Sch Energy Power & Mech Engn Baoding Peoples R China

ISBN: (纸本)9781467389044

Wind power forecasting is one of the most effective solutions of large scale wind power integration. However, the data which contains wind speed and wind power output is the non-stationary stochastic signal to be taken as a time-energy series, and the data varies largely in every frequency component. It creates forecasting errors when treating wind speed or wind power. In addition, the very short-term wind power forecasting, which the time frame is several minutes ahead, can help balance the supply and need of grid, facilitating the steady and healthy operation of power market. Facing the wind data feature and forecasting importance, this paper proposes a novel approach using autoregressive model (AR) and Hilbert-Huang Transform (HHT) to improve the accuracy of wind power forecasting. The input data is decomposed into several components, in every of which, the data satisfies the linear characteristic better than the one before transformation. Case study from a wind farm in Texas, U.S. A shows that the proposed approach performs better than the traditional linear forecasting method.

关键词： wind power very short-term forecasting Autoregressive model (AR) Hilbert-Huang Transform (HHT) model optimization

来源：评论

学校读者我要写书评

暂无评论

model optimization in Imbalanced Regression 25th

Model Optimization in Imbalanced Regression

引用

25th International Conference on Discovery Science (DS)

作者： Silva, Anibal Ribeiro, Rita P. Moniz, Nuno Univ Porto Fac Sci Porto Portugal INESC TEC Porto Portugal

ISBN: (纸本)9783031188398;9783031188404

Imbalanced domain learning aims to produce accurate models in predicting instances that, though underrepresented, are of utmost importance for the domain. Research in this field has been mainly focused on classification tasks. Comparatively, the number of studies carried out in the context of regression tasks is negligible. One of the main reasons for this is the lack of loss functions capable of focusing on minimizing the errors of extreme (rare) values. Recently, an evaluation metric was introduced: Squared Error Relevance Area (SERA). This metric posits a bigger emphasis on the errors committed at extreme values while also accounting for the performance in the overall target variable domain, thus preventing severe bias. However, its effectiveness as an optimization metric is unknown. In this paper, our goal is to study the impacts of using SERA as an optimization criterion in imbalanced regression tasks. Using gradient boosting algorithms as proof of concept, we perform an experimental study with 36 data sets of different domains and sizes. Results show that models that used SERA as an objective function are practically better than the models produced by their respective standard boosting algorithms at the prediction of extreme values. This confirms that SERA can be embedded as a loss function into optimization-based learning algorithms for imbalanced regression scenarios.

关键词： Imbalanced regression Asymmetric loss functions model optimization Boosting

来源：评论

学校读者我要写书评

暂无评论

model optimization for model-Based Compression of Real World Video Data 4

Model Optimization for Model-Based Compression of Real World...

引用

4th IEEE International Conference on Consumer Electronics Berlin (ICCE-Berlin)

作者： Feller, Christian Wuenschmann, Juergen Wagner, Raimar Rothermel, Albrecht Univ Ulm Inst Microelect Albert Einstein Allee 43 D-89081 Ulm Germany

ISBN: (纸本)9781479961658

The availability of affordable cameras with additional depth sensor boosted the advances in the field of Simultaneous Localization and Mapping (SLAM). This area of research addresses creating a 3D representation of unknown environment by fusing sensor information. Whereas a coarse representation suffices for navigational purposes, the resulting 3D models are not accurate enough to allow rendering a visually pleasing video stream based on the gathered information, when applying a model based video compression standard such as MPEG-4 Part 25. This publication focuses on increasing the accuracy of the created 3D models in order to improve the perceived quality when using model-based video coding and aims for performance comparable to hybrid video encoding.

关键词： cameras data compression image fusion video coding 3D models 3D representation MPEG-4 Part 25 SLAM coarse representation depth sensor model based video compression standard model optimization model-based compression model-based video coding navigational purposes perceived quality improvement real world video data sensor information fusion simultaneous localization-and-mapping unknown environment visually pleasing video stream rendering Accuracy Cameras Data models Equations Mathematical model Solid modeling Three-dimensional displays

来源：评论

学校读者我要写书评

暂无评论

Task-Aware Dynamic model optimization for Multi-Task Learning

引用

IEEE ACCESS 2023年 11卷 137709-137717页

作者： Choi, Sujin Jin, Hyundong Kim, Eunwoo Chung Ang Univ Dept Artificial Intelligence Seoul 06974 South Korea Chung Ang Univ Sch Comp Sci & Engn Seoul 06974 South Korea

Multi-task learning (MTL) is a field in which a deep neural network simultaneously learns knowledge from multiple tasks. However, achieving resource-efficient MTL remains challenging due to entangled network parameters across tasks and varying task-specific complexity. Existing methods employ network compression techniques while maintaining comparable performance, but they often compress uniformly across all tasks without considering individual complexity. This can lead to suboptimal solutions due to entangled network parameters and memory inefficiency, as the parameters for each task may be insufficient or excessive. To address these challenges, we propose a framework called Dynamic model optimization (DMO) that dynamically allocates network parameters to groups based on task-specific complexity. This framework consists of three key steps: measuring task similarity and task difficulty, grouping tasks, and allocating parameters. This process involves the calculation of both weight and loss similarities across tasks and employs sample-wise loss as a measure of task difficulty. Tasks are grouped based on their similarities, and parameters are allocated with dynamic pruning according to task difficulty within their respective groups. We apply the proposed framework to MTL with various classification datasets. Experimental results demonstrate that the proposed approach achieves high performance while taking fewer network parameters than other MTL methods.

关键词： Multi-task learning resource-efficient learning model optimization

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：