Communications engineering最新文献_第5页

EdgeVolution: democratizing multi-objective neural architecture search and end-to-end deployment on microcontrollers. EdgeVolution：在微控制器上实现多目标神经架构搜索和端到端部署。

IF 7

Communications engineering Pub Date : 2026-06-23 DOI: 10.1038/s44172-026-00708-2

René Groh, Stefan Dendorfer, Mateo Ávila Pava, Fabio Egle, Sebastian Zimmermann, Andreas M Kist

{"title":"EdgeVolution: democratizing multi-objective neural architecture search and end-to-end deployment on microcontrollers.","authors":"René Groh, Stefan Dendorfer, Mateo Ávila Pava, Fabio Egle, Sebastian Zimmermann, Andreas M Kist","doi":"10.1038/s44172-026-00708-2","DOIUrl":"10.1038/s44172-026-00708-2","url":null,"abstract":"Edge AI holds great potential for extending the use of artificial neural networks to resource-constrained edge devices, such as microcontrollers. Despite this potential, optimizing and deploying neural networks on these platforms remains challenging due to a lack of tools for hardware-specific adaptation, leading to reproducibility issues and suboptimal performance. To address these challenges, we present EdgeVolution, an end-to-end hardware-in-the-loop platform that facilitates multi-objective optimization, neural architecture selection, and direct deployment onto target hardware. We demonstrate the versatility of EdgeVolution through four application use cases, showcasing its wide-ranging applicability. By offering a generic and adaptable pipeline, EdgeVolution enables the creation and deployment of neural network models tailored to specific datasets, classification tasks, and hardware constraints, thereby improving accessibility, performance, and reproducibility for AI applications on edge devices.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":"5 1","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13291272/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148310644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Ghost noise in single-fiber bidirectional transmission links and its suppression approaches. 单光纤双向传输链路中的鬼噪及其抑制方法。

IF 7

Communications engineering Pub Date : 2026-06-22 DOI: 10.1038/s44172-026-00712-6

Wujie Wang, Qihao Hu, Lina Ma, Jiali Gao, Hongze Leng, Junqiang Song

{"title":"Ghost noise in single-fiber bidirectional transmission links and its suppression approaches.","authors":"Wujie Wang, Qihao Hu, Lina Ma, Jiali Gao, Hongze Leng, Junqiang Song","doi":"10.1038/s44172-026-00712-6","DOIUrl":"10.1038/s44172-026-00712-6","url":null,"abstract":"Single-fiber bidirectional transmission (Bidi transmission) enables efficient reuse of existing global communication cables for sensing, yet environmental disturbances induce link noise that fundamentally limits long-distance sensing performance. Despite extensive research efforts dedicated to Bidi transmission link noise suppression, existing methods fail to eliminate link noise, forcing the abandonment of the Bidi transmission scheme in large-scale deployments. Here, we uncover that the origin of this residual noise is the \"ghost noise,\" which can be decomposed into polarization cross-coupling noise (PCN) and calculation cross-coupling noise (CCN). To address this ghost noise, we develop a polarization-traversal interrogation (PTI) technique integrated with pseudo-random binary sequence (PRBS) phase modulation. This hybrid approach achieves zero noise increase over a 7 km Bidi transmission link and realizes 94.8% background noise reduction in a 10 km Bidi transmission link. This breakthrough extends the low-noise Bidi analog transmission distance by six times, holding significant implications for the construction of sensing networks through global communication infrastructure.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":" ","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13646289/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148304210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

LightPro: a linear photonic processor with full programmability. LightPro：具有完全可编程性的线性光子处理器。

IF 7

Communications engineering Pub Date : 2026-06-17 DOI: 10.1038/s44172-026-00707-3

Amin Shafiee, Zahra Ghanaatian, Benoit Charbonnier, Mahdi Nikdast

{"title":"LightPro: a linear photonic processor with full programmability.","authors":"Amin Shafiee, Zahra Ghanaatian, Benoit Charbonnier, Mahdi Nikdast","doi":"10.1038/s44172-026-00707-3","DOIUrl":"10.1038/s44172-026-00707-3","url":null,"abstract":"The physical scaling of photonic matrix-vector multiplication hardware for deep neural network acceleration is fundamentally limited by accumulated optical losses, crosstalk noise, and the prohibitive footprint of conventional devices such as Mach-Zehnder interferometers. Here we present LightPro, a fully programmable linear photonic processor designed to optimize scalability, power efficiency, and area footprint. At its core, our architecture integrates a neural architecture search and pruning framework with tunable phase-change material directional couplers. By thermally modulating the phase-change material state, we dynamically adjust coupling coefficients to achieve precise splitting ratios, facilitating highly optimized topologies for matrix-vector multiplication operations. The underlying phase-change material-based devices are evaluated using numerical multiphysics simulations and compact models, which are validated against reported experimental data from prior work. System-level evaluations demonstrate that the neural architecture search-optimized LightPro architectures achieve up to an 85% footprint reduction and a greater than 50% decrease in power consumption. Network scaling evaluations using handwritten digit and Gaussian datasets yield an inference accuracy degradation of less than 5%. Experimental prototyping on a commercial photonic processor validates the computational accuracy of LightPro, establishing a scalable and efficient pathway for next-generation photonic artificial intelligence accelerators.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":" ","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13650130/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148273965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Test-based equivalent-material method for collapse qualification of helically wound and layered cylindrical structures. 螺旋缠绕和层状圆柱结构倒塌鉴定的试验等效材料法。

IF 7

Communications engineering Pub Date : 2026-06-16 DOI: 10.1038/s44172-026-00699-0

Yuteng Zhang, Mohsen Saneian, Yong Bai, Pan Fang, Peng Cheng, Liang Zhao

{"title":"Test-based equivalent-material method for collapse qualification of helically wound and layered cylindrical structures.","authors":"Yuteng Zhang, Mohsen Saneian, Yong Bai, Pan Fang, Peng Cheng, Liang Zhao","doi":"10.1038/s44172-026-00699-0","DOIUrl":"10.1038/s44172-026-00699-0","url":null,"abstract":"Reliable qualification of collapse resistance is critical across pressurised cylindrical shells ranging from offshore risers and subsea cables to aerospace tanks and even biological ducts. Existing approaches either rely on costly hyperbaric chambers or oversimplified geometric models that neglect the forming-induced residual stresses and imperfections. Here, we propose a back-inferred equivalent-material method (EMM), which transforms a simple benchtop flat-plate compression test result into a nonlinear constitutive law that implicitly embeds cold work, residual stress, friction, and small-scale compliance. Embedding this equivalent law into a homogeneous finite-element model reproduces hydrostatic collapse without the need for full-scale testing. We demonstrate the method on flexible-pipe carcass layers and steel-strip reinforced thermoplastic pipes, showing close agreement with explicit-geometry simulations and hyperbaric measurements. More broadly, the geometry-agnostic formulation enables rapid, low-cost, and physically faithful qualification across diverse layered shells, offering a general route to replace hyperbaric testing in industries spanning energy, aerospace, and biomedical engineering.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":" ","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13631229/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148267181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

End-to-end robust system discovery in electrical dynamical systems using scientific machine learning. 利用科学机器学习在电动力系统中发现端到端鲁棒系统。

IF 7

Communications engineering Pub Date : 2026-06-16 DOI: 10.1038/s44172-026-00689-2

Jhelum Chakravorty, Nicolò Ripamonti, Tor Laneryd

{"title":"End-to-end robust system discovery in electrical dynamical systems using scientific machine learning.","authors":"Jhelum Chakravorty, Nicolò Ripamonti, Tor Laneryd","doi":"10.1038/s44172-026-00689-2","DOIUrl":"10.1038/s44172-026-00689-2","url":null,"abstract":"System discovery is an important part of the power systems asset management process. In this paper, we introduce an end-to-end approach for robust system discovery for a class of electrical dynamical systems with polynomial dynamics. To that end, we provide a theoretical analysis of the problem setting and the solution approach using a particular Scientific Machine Learning method called Physics-Informed Machine Learning. We introduce model architecture and training and validation methods for deterministic as well as probabilistic approaches to predict the solution to pertinent inverse problems and propose a sampling method to make the predictions robust to data sparsity. With empirical evidence drawn from a case study of thermal modeling of electrical induction machines, we establish the merits of the proposed method and discuss the implications of this research on possible future directions. In particular, we benchmark our proposed end-to-end method with multiple deterministic and probabilistic approaches and show that it outperforms all but one baseline in terms of performance in a range of 6% to 78% reduction of validation error. The drop in performance (increase in validation error) in our end-to-end probabilistic approach compared to a Bayesian approach is compensated by a large reduction (86%) in the computation time, making our method more balanced in terms of performance and cost. The experimental results establish that our end-to-end approach outperforms the baseline methods by demonstrating consistent balancing among performance and computation cost.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":" ","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13645738/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148267172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unlocking the Nickel Value of Ultramafic Resources Through Innovative Thermal Treatment Process. 创新热处理工艺释放超镁铁资源镍的价值。

IF 7

Communications engineering Pub Date : 2026-06-15 DOI: 10.1038/s44172-026-00704-6

Wei Lv, Fanmao Wang, Brian Makuza, Lucy Liu, Frederick D Ford, Manqiu Xu, Sam Marcuson, Mansoor Barati

{"title":"Unlocking the Nickel Value of Ultramafic Resources Through Innovative Thermal Treatment Process.","authors":"Wei Lv, Fanmao Wang, Brian Makuza, Lucy Liu, Frederick D Ford, Manqiu Xu, Sam Marcuson, Mansoor Barati","doi":"10.1038/s44172-026-00704-6","DOIUrl":"10.1038/s44172-026-00704-6","url":null,"abstract":"Nickel is a critical metal with broad applications in stainless steel and nickel-based alloys. In recent years, its rapidly growing use in secondary batteries has become particularly important, underscoring its strategic role in enabling low-carbon, green, and sustainable societal development. As high-grade nickel sulfide resources become increasingly scarce and global demand for nickel accelerates-fueled by the expansion of clean energy technologies-there is a pressing need to develop environmentally benign extraction strategies for low-grade nickel sulfide ores to safeguard the stability and sustainability of the global nickel supply chain. Here we report a low-temperature predominantly solid-state process for extracting nickel from unconventional, low-grade ultramafic ores, estimated to contain approximately 45 million tonnes of untapped nickel. The method leverages cheap metallic iron as a nickel getter offering a sustainable pathway for high-value nickel recovery aligned with decarbonized metal production. The process produces ferronickel alloys (16-24% nickel) with several key advantages, including rapid processing time ( ~ 3 hours), low operational temperatures ( < 950 °C), and the elimination of SO2 emissions. Temperature, atmosphere, and iron addition were tailored to create favorable thermodynamic conditions within the reactor, enabling selective partitioning of nickel into metallic alloys while effectively sequestering sulfur as stable solid sulfide phases. Controlled tuning of alloy particle size and morphology facilitates efficient separation from gangue, yielding a ferronickel which can be converted to battery-grade nickel through conventional refining. This method, verified to mini-plant scale broadens the technological landscape of nickel extraction and contributes to a more equitable and resilient global nickel supply chain.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":" ","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13635238/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148260688","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A machine learning perspective on three decades of methanol synthesis: research framework and experimental operation insights. 甲醇合成三十年的机器学习视角：研究框架和实验操作见解。

IF 7

Communications engineering Pub Date : 2026-06-10 DOI: 10.1038/s44172-026-00706-4

Mo Yan, Chengxi Yao, Shihao Wu, Muhammad Asif, Nuning Anugrah Putri Namari, Joohwan Ha, Ziyang Wang, Jaewon Lee, Dayoung Yu, Seokchan Lee, Seungjae Lee, Hao Yu, Xiaoming Duan, Taesung Kim

{"title":"A machine learning perspective on three decades of methanol synthesis: research framework and experimental operation insights.","authors":"Mo Yan, Chengxi Yao, Shihao Wu, Muhammad Asif, Nuning Anugrah Putri Namari, Joohwan Ha, Ziyang Wang, Jaewon Lee, Dayoung Yu, Seokchan Lee, Seungjae Lee, Hao Yu, Xiaoming Duan, Taesung Kim","doi":"10.1038/s44172-026-00706-4","DOIUrl":"10.1038/s44172-026-00706-4","url":null,"abstract":"Heterogeneous catalysis optimization relies on the coordinated selection of catalyst composition, synthesis procedure, and reaction conditions, yet the empirical links between these variables and catalytic performance remain scattered across individual studies. Here, we show that an interpretable machine learning (ML) framework can decode this complexity by integrating disparate literature data. We curate, to our knowledge, the first unified dataset from 200 papers spanning three decades of CO2-to-methanol synthesis over Cu/ZnO-based catalysts, and derive a subset of apparent activation energy (Eapp). Using the dataset, we optimize six ML models via Bayesian hyperparameter tuning and apply regularization strategies to mitigate overfitting. We then use SHAP to quantify the non-linear impact of catalyst properties, synthesis procedures, and reaction conditions on methanol selectivity, CO2 conversion, and methanol space-time yield. Crucially, this data-driven approach uncovers previously obscured operating windows for achieving high methanol productivity at milder pressures than conventionally assumed. Extending the ML framework to the Eapp kinetic subset further identifies regime-sensitive signatures. A controlled experiment over commercial catalysts corroborates the positive space velocity-Eapp relationship inferred from ML models. These results show how interpretable ML built on curated legacy experiments can support operating-window selection, hypothesis-driven kinetic experiments, and practical optimization of mature catalytic technologies.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":" ","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13594224/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148213216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Real-time 3D ultrasound in augmented reality accelerates training and narrows novice-expert performance gaps. 增强现实中的实时3D超声加速了培训并缩小了新手与专家的性能差距。

IF 7

Communications engineering Pub Date : 2026-06-10 DOI: 10.1038/s44172-026-00692-7

Jason F Hou, Shrihari Viswanath, Cinay Dilibal, Bowen Wu, Tanisha Shende, Canan Dagdeviren

{"title":"Real-time 3D ultrasound in augmented reality accelerates training and narrows novice-expert performance gaps.","authors":"Jason F Hou, Shrihari Viswanath, Cinay Dilibal, Bowen Wu, Tanisha Shende, Canan Dagdeviren","doi":"10.1038/s44172-026-00692-7","DOIUrl":"10.1038/s44172-026-00692-7","url":null,"abstract":"Ultrasound imaging requires users to infer three-dimensional anatomy from two-dimensional slices, imposing steep training demands that limit broader adoption. Here we present AR-VIU, a mixed-reality platform that streams real-time volumetric ultrasound as point-cloud renderings into an augmented-reality headset with true-scale spatial registration. To isolate the contributions of volumetric imaging and immersive display, we tested four conditions-two-dimensional imaging on a screen, two-dimensional imaging in augmented reality, three-dimensional imaging on a screen, and three-dimensional imaging in augmented reality-in a controlled study with 18 participants (9 novices, 9 experts). Participants performed object recognition and localization tasks. The augmented-reality volumetric system was associated with the highest accuracy, lowest variability, and near-elimination of the novice-expert performance gap. These results demonstrate technical feasibility for real-time three-dimensional ultrasound in mixed reality and establish an evaluation framework for perceptual and cognitive performance in clinically relevant scenarios, with near-term applications in training and education.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":"5 1","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13254270/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148220893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Thermal-mechanical coupling in shale geothermal reservoirs: a framework for temperature-dependent seismic characterization. 页岩地热储层的热-力学耦合：温度相关地震表征的框架。

IF 7

Communications engineering Pub Date : 2026-06-10 DOI: 10.1038/s44172-026-00702-8

Qamar Yasin, Ghulam Mohyuddin Sohail, Eshimiakhe Daniel, Atif Ismail, Zuorong Chen, David A Wood, Mengdi Sun

{"title":"Thermal-mechanical coupling in shale geothermal reservoirs: a framework for temperature-dependent seismic characterization.","authors":"Qamar Yasin, Ghulam Mohyuddin Sohail, Eshimiakhe Daniel, Atif Ismail, Zuorong Chen, David A Wood, Mengdi Sun","doi":"10.1038/s44172-026-00702-8","DOIUrl":"10.1038/s44172-026-00702-8","url":null,"abstract":"Reliable seismic characterization of geothermal reservoirs requires quantifying the effects of temperature on rock elastic properties. We investigate temperature-dependent seismic velocity evolution in shale through integrated laboratory ultrasonic measurements and finite-element modeling. The P- and S-wave velocities were measured on shale samples from China and the United States with diverse mineralogies over a temperature range from 20 °C to 200 °C, with numerical simulations extending the predictions to 300 °C. Results show a systematic decrease in P-wave velocity with increasing temperature. In contrast, S-wave velocity exhibits enhanced sensitivity to thermally induced micro-fracturing and pore-structure evolution, displaying a biphasic temperature response associated with the competing effects of thermoelastic softening and progressive microcrack development. Temperature-dependent rock-physics models incorporating evolving elastic moduli accurately reproduce the observed velocity trends. These findings establish a quantitative thermo-mechanical framework for interpreting seismic responses in shale-dominated geothermal systems and improve the reliability of seismic monitoring under high-temperature conditions.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":" ","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13594094/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148213297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A shape optimization method for resistance reduction of local piping components with multiscale validation. 基于多尺度验证的局部管道构件减阻形状优化方法。

IF 7

Communications engineering Pub Date : 2026-06-09 DOI: 10.1038/s44172-026-00705-5

Ao Tian, Ran Gao, Angui Li, Junkai Ren, Yingying Wang, Yibu Gao, Ruoyin Jing, Yi Wang, Yan Tian, Zijing Fan, Shu He

{"title":"A shape optimization method for resistance reduction of local piping components with multiscale validation.","authors":"Ao Tian, Ran Gao, Angui Li, Junkai Ren, Yingying Wang, Yibu Gao, Ruoyin Jing, Yi Wang, Yan Tian, Zijing Fan, Shu He","doi":"10.1038/s44172-026-00705-5","DOIUrl":"10.1038/s44172-026-00705-5","url":null,"abstract":"Reducing the flow resistance of local components in building transmission and distribution systems is a key pathway to building energy conservation. Here we propose a novel low-resistance optimization method applied to U-bend shape design, with the minor axis a, major axis b, and offset c defined as shape features. The sample size is 150, and the parameter ranges are as follows: [50 mm, 150 mm] for a, [50 mm, 150 mm] for b, and [-30 mm, 30 mm] for c. On this basis, five representative machine learning regression modeling paradigms, including ridge regression, support vector regression, random forest, multilayer perceptron, and Gaussian process regression, are systematically compared, and the model with the best predictive performance is selected as the surrogate model for U-bend optimization. The proposed method is validated through full-scale experiments, numerical simulations, and turbulent energy dissipation. The results show that within a Reynolds number range of 1.0 × 105 to 2.4 × 105, the optimized U-bend achieves a resistance reduction rate of 13-24% relative to the traditional circular U-bend. This study provides a reference for the low-resistance design and energy-saving optimization of building transmission and distribution systems.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":" ","pages":""},"PeriodicalIF":7.0,"publicationDate":"2026-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13590639/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148213237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0