检索结果-内蒙古大学图书馆

multiobjective bayesian optimization for a 15-dB back-off high-efficiency load modulated balanced amplifier design

INTERNATIONAL JOURNAL OF NUMERICAL MODELLING-ELECTRONIC NETWORKS DEVICES AND FIELDS 2024年第2期37卷

作者： Chen, Peng Qi, Lin Zhao, Yinshuang Yu, Luqi Yu, Chao Southeast Univ Sch Informat Sci & Engn Nanjing 210096 Peoples R China

In this article, multiobjecive bayesian optimization (MBO) with a non-penalization strategy is proposed to design a 15-dB back-off high-efficiency load modulated balanced amplifier (LMBA). Applying the proposed method, the output matching networks of the LMBA are optimized to achieve proper load modulation behaviors that providing good performance. To verify the proposed method, a 2.0-GHz LMBA is designed and measured. Experimental results show that LMBA achieves an output power of 44.5 dBm with a gain higher than 6.6 dB at saturation. The measured drain efficiency (DE)/power-added efficiency (PAE) are 67%/52% at saturation, 54%/43% at 9-dB back-off, and 52%/46% at 15-dB back-off, respectively. Tested with 20-MHz signal with 10-dB peak-to-average power ratio (PAPR) and a 256-quadrature amplitude modulation (QAM) scheme, the LMBA achieves adjacent channel leakage ratio (ACLR) levels of -46.7/-47.5 dBc and an error vector magnitude (EVM) of 1.2% after digital predistortion.

关键词： high-efficiency load modulated balanced amplifier multiobjective bayesian optimization

来源：评论

学校读者我要写书评

暂无评论

multiobjective bayesian optimization for a 1.7-3.7GHz GaN Power Amplifier Design 9

Multiobjective Bayesian Optimization for a 1.7-3.7GHz GaN Po...

引用

2022 International Wireless Symposium

作者： Zhao, Yinshuang Chen, Peng Zhu, Xiaowei Yu, Chao Southeast Univ State Key Lab Millimeter Waves Nanjing Peoples R China Purple Mt Lab Nanjing Peoples R China

ISBN: (数字)9781665481977

ISBN: (纸本)9781665481977

This work designs a broadband power amplifier (PA) using multiobjective bayesian optimization (MOBO) algorithm. Based on a 10-W gallium nitride (GaN) device, the work firstly uses Chebyshev impedance transformer to design preliminary matching networks, and then applies MOBO to build surrogate models and search for the optimum parameters of the matching networks. The design approach is implemented automatically on a co-simulation platform consisting of a programming environment and a commercial software. Measured results verify that the optimized PA using MOBO algorithm achieves a power-added efficiency higher than 46.7% with output power more than 10W from 1.7 to 3.7GHz, demonstrating a broadband high-efficiency design.

关键词： Broadband multiobjective bayesian optimization power amplifier

来源：评论

学校读者我要写书评

暂无评论

Fabrication of SERS Substrates Using Silver-Coated Gold Nanostars for Chemical Sensing: A multiobjective bayesian optimization Approach

引用

ACS APPLIED NANO MATERIALS 2025年第23期8卷 11930-11939页

作者： Giordano, Andrea N. Franqui-Rios, Samuel Quarin, Steven M. Vang, Der Austin, Drake R. Doyle, Abigail G. Baldwin, Luke A. Strobbia, Pietro Rao, Rahul Air Force Res Lab Mat & Mfg Directorate Wright Patterson AFB OH 45433 USA Univ Cincinnati Dept Chem Cincinnati OH 45221 USA Univ Calif Los Angeles Dept Chem & Biochem Los Angeles CA 90095 USA CNR Washington DC 20001 USA

Field detection of trace analytes remains a challenge in many sectors, including national security, public health, and environmental monitoring. This challenge arises from the scarcity of field-deployable technologies capable of identifying trace analytes in the complex matrices of real-world samples. Surface-enhanced Raman scattering (SERS) is a promising technique for chemical sensing because it offers sensitive and specific analyte detection with commercial portable Raman systems. Despite the immense potential of SERS, commercially available SERS technologies that are accurate, reliable, cost-effective, and compatible with portable spectrometers remain scarce. This scarcity is due to the challenges associated with the complex design and resource-intensive optimization of the manufactured substrates that can achieve both high SERS intensity and signal uniformity. Machine learning (ML) tools are ideal for addressing complex optimization problems, but their application in SERS substrate fabrication remains largely unexplored. We present a pioneering example of applying ML to optimize SERS substrate fabrication using silver-coated gold nanostars for chemical sensing. In this study, we used multiobjective bayesian optimization to achieve both high SERS intensity and signal uniformity across the SERS substrate. Optimal fabrication parameters were identified for spin and flow coating of silver-coated gold nanostars. The optimized spin coated substrate achieved an experimental limit of detection of 100 nM with an application relevant compound (ammonium nitrate), suggesting its potential for trace-level sensing. This work underscores the potential for ML to accelerate the development of field-deployable SERS technologies for trace analyte detection. Furthermore, we provide a general framework for optimized fabrication of nanoparticles on surfaces for other applications such as catalysis, electronics, and medicine.

关键词： SERS silver-coated gold nanostars machinelearning multiobjective bayesian optimization ammonium nitrate chemical sensor flow coating spin coating Raman spectroscopy

来源：评论

学校读者我要写书评

暂无评论

On lower confidence bound improvement matrix-based approaches for multiobjective bayesian optimization and its applications to thin-walled structures

引用

THIN-WALLED STRUCTURES 2021年 161卷

作者： Sun, Guangyong Li, Linsong Fang, Jianguang Li, Qing Hunan Univ State Key Lab Adv Design & Mfg Vehicle Body Changsha 410082 Hunan Peoples R China Univ Sydney Sch Aerosp Mech & Mechatron Engn Sydney NSW 2006 Australia Univ Technol Sydney Sch Civil & Environm Engn Sydney NSW 2007 Australia

In engineering practice, most design criteria require time-consuming functional evaluation. To tackle such design problems, multiobjective bayesian optimization has been widely applied to generation of optimal Pareto solutions. However, improvement function-based expected improvement (EI) and the hypervolume improvementbased lower confidence bound (LCB) infill-criteria are frequently criticized for their high computational cost. To address this issue, this study proposes a novel approach for developing multiobjective LCB criteria on the basis of LCB improvement matrix. Specifically, three cheap yet efficient infill-criteria are suggested by introducing three different improvement functions (namely, hypervolume improvement, Euclidean distance and maximin distance) that assemble the improvement matrix to a scalar value, which is then maximized for adding solution points sequentially. All these criteria have closed-form expressions and can maintain the anticipated properties, thereby largely reducing computational efforts without either integration or expensive evaluation of hypervolume indicator. The efficiency of the proposed criteria is demonstrated through the ZDT and DTLZ tests with different numbers of design variables and different complexities of objectives. The testing results exhibit that the proposed criteria have faster convergence, and enable to generate satisfactory Pareto front with fairly low computational cost compared with other conventional criteria. Finally, the best performing criterion is further applied to real-life design problems of tailor rolled blank (TRB) thin-walled structures under impact loads, which demonstrates a strong search capability for with good distribution of Pareto points, potentially providing an effective means to engineering design with strong nonlinearity and sophistication.

关键词： multiobjective bayesian optimization Lower confidence bound (LCB) Infill-criteria Crashworthiness design

来源：评论

学校读者我要写书评

暂无评论

Simultaneous Improvement of Multiple Electrical Characteristics in Ultrathin a-IGZO TFTs Using Machine Learning optimization

引用

ACS APPLIED MATERIALS & INTERFACES 2025年第7期17卷 10865-10875页

作者： Yang, Hyunkyu Lee, Jiho Park, Chaeyoung Lee, Chan Jin, Minho Kim, Jiyeon Shin, Jong Chan Hwang, Suhwan Lee, Eungkyu Kim, Youn Sang Seoul Natl Univ Dept Chem & Biol Engn Seoul 08826 South Korea Seoul Natl Univ Inst Chem Proc Coll Engn Seoul 08826 South Korea Samsung Elect Co Suwon 16677 Gyeonggi Do South Korea Kyung Hee Univ Dept Elect Engn Yongin 17104 Gyeonggi Do South Korea Seoul Natl Univ Grad Sch Convergence Sci & Technol Program Nano Sci & Technol Seoul 08826 South Korea Seoul Natl Univ Grad Sch Convergence Sci & Technol Dept Appl Bioengn Seoul 08826 South Korea Adv Inst Convergence Technol Suwon 16229 South Korea

While IGZO is emerging as a promising channel material to address the scaling limitations of conventional silicon-based DRAM, its application in next-generation 3D DRAM requires further advancements in achieving ultrathin structures and excellent performance tailored to DRAM characteristics. Specifically, optimizing process variables is essential for enhancing mobility in ultrathin structures, where mobility tends to degrade significantly, while maintaining a constant threshold voltage, a task that is both experimentally intensive and resource-demanding. This study employed multi-objective bayesian optimization (MOBO) machine learning (ML) to simultaneously optimize multiple electrical objectives, aiming to achieve high mobility and a near-zero threshold voltage for ultrathin IGZO thin-film transistors (TFTs) under complex sputtering conditions, involving a wide range of possible combinations of Ar gas flow, sputtering power, and working pressure. Integrating empirical insights and expert knowledge into feature extraction, the MOBO approach leveraged human-driven expertise to optimize field-effect mobility and threshold voltage within the solution space. With ML assistance, a Pareto-optimal front was constructed to visualize trade-offs, achieving high field-effect mobility of 33.1 cm2/Vs and near-zero threshold voltage of -0.05 V at a 7.47 nm channel thickness. This approach is expected advance next-generation semiconductor technologies, offering exceptional gains in both efficiency and performance.

关键词： DRAM a-IGZO thin-film transistor machine learning multiobjective bayesian optimization

来源：评论

学校读者我要写书评

暂无评论

A Multi-Fidelity bayesian optimization Approach for Constrained Multi-Objective optimization Problems

引用

JOURNAL OF MECHANICAL DESIGN 2024年第7期146卷 071702-071702页

作者： Lin, Quan Hu, Jiexiang Zhou, Qi Shu, Leshi Zhang, Anfu Huazhong Univ Sci & Technol Sch Aerosp Engn Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Mech Sci & Engn State Key Lab Digital Mfg Equipment & Technol Wuhan 430074 Peoples R China Wuhan Second Ship Design & Res Inst Wuhan 430064 Peoples R China

In this paper, a multi-fidelity bayesian optimization approach is presented to tackle computationally expensive constrained multiobjective optimization problems (MOPs). The proposed approach consists of a three-stage optimization framework designed to search for promising candidate points. In the first stage, an acquisition function is proposed to identify a feasible solution if none is available in the current set of sampling points. Subsequently, a new multi-fidelity weighted expected hypervolume improvement function is developed to find better solutions. In the third stage, a constrained weighted lower confidence bound acquisition function is presented to enhance the constraint predictions and refine the solutions near the constraint boundary. Additionally, a filter strategy is suggested to determine whether constraint updating is necessary, aiming to save computational resources and improve optimization efficiency. Moreover, to expedite the optimization process, a parallel optimization approach is further developed based on the suggested three-stage optimization framework. To achieve this, a multi-fidelity influence function is introduced, allowing the proposed approach to determine a desired number of candidate points within a single iteration. Lastly, the proposed approach is demonstrated through six numerical benchmark examples, which verifies its significant advantages in addressing expensive constrained MOPs. Besides, the proposed approach is applied to the multiobjective optimization of a metamaterial vibration isolator, resulting in the attainment of satisfactory solutions.

关键词： multiobjective bayesian optimization multi-fidelity surrogate expensive constrained optimization parallel computing simulation-based design

来源：评论

学校读者我要写书评

暂无评论

Obtaining high H2-rich syngas yield and carbon conversion efficiency from biomass gasification: From characterization to process optimization using machine learning with experimental validation

引用

FUEL 2024年 378卷

作者： Kundu, Kaushik Kumar, Avan Kodamana, Hariprasad Pant, Kamal K. Indian Inst Technol Delhi Dept Chem Engn Catalyt React Engn Lab New Delhi 110016 India Indian Inst Technol Delhi Dept Chem Engn Control & Analyt Proc Syst Lab New Delhi 110016 India Indian Inst Technol Delhi Yardi Sch Artificial Intelligence New Delhi 110016 India Arizona State Univ Sch Sustainabil Tempe AZ 85287 USA

Gasification, among the other thermochemical processes, stands as a promising method for sustainable waste management while generating valuable H-2-rich syngas. In the last few years, there has been a significant surge in the adoption of machine learning (ML) techniques for modelling biomass as well as gasification and pyrolysis of waste. The present study unveiled the supremacy of Gaussian Process Regression (GPR) algorithm as compared to other ML models, in predicting syngas yield [Y-syn (Nm(3)/kg of biomass)] and carbon conversion efficiency [eta Carbon(%)] respectively, with a high prediction accuracy of R-2 > 0.9 in both the cases. The necessary datasets were collected from experimental trials, at reaction conditions of temperature (T): 550-1000 degrees C, air flow rate (AFR): 0.02-0.04 L/min, and reaction time of 2 h, in a two-stage fixed bed reactor, which were identified on the Basis of thermodynamic as well as stoichiometric analysis. Additionally, the models were employed in bayesian optimization within a weighted multi-objective framework to obtain the optimal operational parameters, focusing on maximizing both yield and conversion simultaneously. Finally, this result was validated experimentally to check the sanctity of the data-driven optimization.

关键词： H-2 -rich syngas Biomass gasification Gaussian process regression multiobjective bayesian optimization Experimental validation

来源：评论

学校读者我要写书评

暂无评论

Framework for discovering porous materials: Structural hybridization and bayesian optimization of conditional generative adversarial network

引用

DIGITAL CHEMICAL ENGINEERING 2022年 5卷

作者： Matsuda, Yosuke Ookawara, Shinichi Yasuda, Tomoki Yoshikawa, Shiro Matsumoto, Hideyuki Tokyo Inst Technol Dept Chem Sci & Engn 2-12-1 S1-26 OokayamaMeguro Ku Tokyo 1528552 Japan Egypt Japan Univ Sci & Technol Dept Energy Resources Engn POB 179-21934 New Borg El Arab City 21934 Alexandria Egypt

Although deep-learning-based materials discovery has attracted considerable research attention, the application of deep learning has been limited to discovery of materials within single material types such that the discovered materials are similar to existing ones. Thus, we developed an alternative approach for discovering porous materials. A materials discovery design space was configured using several key porous materials and a conditional generative adversarial network (CGAN), which structurally hybridizes them. Hybridization was controlled using a vector design variable input as a condition to the CGAN, and each design variable component represented a key porous material intensity, which was defined as a conceptual quantity for expressing materialness. Furthermore, by varying the vector latent variable input to the CGAN, any number of similar hybrid porous materials could be generated using the same design variable. The multiobjective bayesian optimizer efficiently discovered hybrid porous materials constituting Pareto solutions in the objective function space of the material properties, which were mapped from the design space by the CGAN structural hybridization and computational fluid dynamics property evaluation. Tradeoff properties such as the pressure drop and filtration efficiency were multiobjectively optimized for key porous materials exhibiting random, packed-sphere, and sponge structures. By flexibly adopting their structural parameters depending on the required pressure drop and filtration efficiency, the discovered optimized hybrid porous materials outperformed the key porous ones. The results demonstrated the effectiveness of the proposed framework for discovering porous materials.

关键词： Porous material Permeability Filtration efficiency Conditional generative adversarial network Computational fluid dynamics simulation multiobjective bayesian 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：