智算中心论文专站

AIDC Research Papers

Liquid Cooling AI Data Center Power & Thermal Systems
Current Issue

Volume 2026 · Issue 09-14

按期刊卷期页方式整理本期论文。每条仅使用日报已列出的可追溯公开来源,不新增未经核验事实。

Research Article热管理与液冷

GreenPassport: Request-Level Carbon Accounting for Cross-Border AI Inference

Rui Lu

Published 2026-09-09 · arXiv · Credibility S

AI

Abstract, interpretation and reference

Abstract

AI

中文解读

背景:AI 数据中心负载、功率密度和能源约束同步上升,液冷、热管理和数据中心能效正在成为智算中心设计的关键变量。问题:论文聚焦现有方案在效率、可靠性或工程协同上的瓶颈。方法:摘要显示作者采用文献摘要中的模型、实验或案例分析,把运行负载、冷却/能源系统和基础设施约束放在同一分析框架中。结果:研究重点指向冷却效率、能源利用或运维策略的改进方向。意义:对日报读者而言,它可用于判断液冷方案、热管理路线和高密度部署节奏。仍需结合全文实验条件、样本范围和成本假设核验。

参考文献

Rui Lu. GreenPassport: Request-Level Carbon Accounting for Cross-Border AI Inference[J/OL]. (2026-09-09)[2026-09-14]. https://arxiv.org/abs/2609.06784.

Full text 中文海报
热管理与液冷 论文图示
Research Article算电协同

From Grid to Chip: Power Architecture, Stability, and Flexibility of AI Data Centers

Yubo Song、Rui Kong、Takuro Umihara、Pooya Davari、Frede Blaabjerg、Subham Sahoo

Published 2026-09-10 · arXiv · Credibility S

The rapid growth of artificial intelligence ( AI

Abstract, interpretation and reference

Abstract

The rapid growth of artificial intelligence ( AI

中文解读

背景:AI 数据中心负载、功率密度和能源约束同步上升,算力负载与电网侧资源的协同调度正在成为智算中心设计的关键变量。问题:论文聚焦现有方案在效率、可靠性或工程协同上的瓶颈。方法:摘要显示作者采用文献摘要中的模型、实验或案例分析,把运行负载、冷却/能源系统和基础设施约束放在同一分析框架中。结果:研究重点指向AI 负载波动对电网设备寿命和调频边界的影响。意义:对日报读者而言,它可用于判断智算中心建设是否受电网容量、负载波动和调度机制约束。仍需结合全文实验条件、样本范围和成本假设核验。

参考文献

Yubo Song, Rui Kong, Takuro Umihara, 等. From Grid to Chip: Power Architecture, Stability, and Flexibility of AI Data Centers[J/OL]. (2026-09-10)[2026-09-14]. https://arxiv.org/abs/2609.11649.

Full text 中文海报
算电协同 论文图示
Research Article热管理与液冷

Bio-inspired Learning and Decision-Making with Probabilistic In-Memory Computing Hardware: Part 2

Thomas Dalgaty、Eiji Kawasaki、Miguel de Prado、Tommaso Salvatori、Germain Haugou、Eric Flamand

Published 2026-09-10 · arXiv · Credibility S

This report extends our previous work (Part 1), which introduced an energy

Abstract, interpretation and reference

Abstract

This report extends our previous work (Part 1), which introduced an energy

中文解读

背景:AI 数据中心负载、功率密度和能源约束同步上升,液冷、热管理和数据中心能效正在成为智算中心设计的关键变量。问题:论文聚焦现有方案在效率、可靠性或工程协同上的瓶颈。方法:摘要显示作者采用文献摘要中的模型、实验或案例分析,把运行负载、冷却/能源系统和基础设施约束放在同一分析框架中。结果:研究重点指向冷却效率、能源利用或运维策略的改进方向。意义:对日报读者而言,它可用于判断液冷方案、热管理路线和高密度部署节奏。仍需结合全文实验条件、样本范围和成本假设核验。

参考文献

Thomas Dalgaty, Eiji Kawasaki, Miguel de Prado, 等. Bio-inspired Learning and Decision-Making with Probabilistic In-Memory Computing Hardware: Part 2[J/OL]. (2026-09-10)[2026-09-14]. https://arxiv.org/abs/2609.11288.

Full text 中文海报
热管理与液冷 论文图示
Research Article算电协同

Grid-Mode-Aware Model Predictive Control of Hybrid Energy Storage Systems for AI Data Center Power Smoothing

Xin Chen

Published 2026-09-03 · arXiv · Credibility S

To facilitate the grid-friendly integration of highly variable AI

Abstract, interpretation and reference

Abstract

To facilitate the grid-friendly integration of highly variable AI

中文解读

背景:AI 数据中心负载、功率密度和能源约束同步上升,算力负载与电网侧资源的协同调度正在成为智算中心设计的关键变量。问题:论文聚焦现有方案在效率、可靠性或工程协同上的瓶颈。方法:摘要显示作者采用仿真建模和情景分析,把运行负载、冷却/能源系统和基础设施约束放在同一分析框架中。结果:研究重点指向AI 负载波动对电网设备寿命和调频边界的影响。意义:对日报读者而言,它可用于判断智算中心建设是否受电网容量、负载波动和调度机制约束。仍需结合全文实验条件、样本范围和成本假设核验。

参考文献

Xin Chen. Grid-Mode-Aware Model Predictive Control of Hybrid Energy Storage Systems for AI Data Center Power Smoothing[J/OL]. (2026-09-03)[2026-09-14]. https://arxiv.org/abs/2609.04398.

Full text 中文海报
算电协同 论文图示
Research Article热管理与液冷

THz-SynC: Collective Synthesis with Contextual-Bandit-Assisted Coordination for Reconfigurable Hybrid Optical-THz AI Datacenters

Jingting Jiang、Chong Han

Published 2026-09-03 · arXiv · Credibility S

Terahertz (THz) wireless interconnects offer high-capacity, low-latency, and energy

Abstract, interpretation and reference

Abstract

Terahertz (THz) wireless interconnects offer high-capacity, low-latency, and energy

中文解读

背景:AI 数据中心负载、功率密度和能源约束同步上升,液冷、热管理和数据中心能效正在成为智算中心设计的关键变量。问题:论文聚焦现有方案在效率、可靠性或工程协同上的瓶颈。方法:摘要显示作者采用文献摘要中的模型、实验或案例分析,把运行负载、冷却/能源系统和基础设施约束放在同一分析框架中。结果:研究重点指向冷却效率、能源利用或运维策略的改进方向。意义:对日报读者而言,它可用于判断液冷方案、热管理路线和高密度部署节奏。仍需结合全文实验条件、样本范围和成本假设核验。

参考文献

Jingting Jiang, Chong Han. THz-SynC: Collective Synthesis with Contextual-Bandit-Assisted Coordination for Reconfigurable Hybrid Optical-THz AI Datacenters[J/OL]. (2026-09-03)[2026-09-14]. https://arxiv.org/abs/2609.04025.

Full text 中文海报
热管理与液冷 论文图示
Research Article热管理与液冷

SALTED: a symmetry-adapted machine-learning program for predicting electron-densities in molecules and materials

Zekun Lou、Alan M. Lewis、Théophane Bernhard、Lukas Seifert、Agustin Salcedo、Florian Kleemiss、Mariana Rossi、Andrea Grisafi

Published 2026-09-03 · arXiv · Credibility S

SALTED provides an open-source Python package for machine learning the quantum-mechanical electron density, $n(\mathbf{r})$, in molecular and condensed-phase systems based on input atomic coordinates

Abstract, interpretation and reference

Abstract

SALTED provides an open-source Python package for machine learning the quantum-mechanical electron density, $n(\mathbf{r})$, in molecular and condensed-phase systems based on input atomic coordinates

中文解读

背景:AI 数据中心负载、功率密度和能源约束同步上升,液冷、热管理和数据中心能效正在成为智算中心设计的关键变量。问题:论文聚焦现有方案在效率、可靠性或工程协同上的瓶颈。方法:摘要显示作者采用文献摘要中的模型、实验或案例分析,把运行负载、冷却/能源系统和基础设施约束放在同一分析框架中。结果:研究重点指向冷却效率、能源利用或运维策略的改进方向。意义:对日报读者而言,它可用于判断液冷方案、热管理路线和高密度部署节奏。仍需结合全文实验条件、样本范围和成本假设核验。

参考文献

Zekun Lou, Alan M. Lewis, Théophane Bernhard, 等. SALTED: a symmetry-adapted machine-learning program for predicting electron-densities in molecules and materials[J/OL]. (2026-09-03)[2026-09-14]. https://arxiv.org/abs/2609.03576.

Full text 中文海报
热管理与液冷 论文图示
Research Article热管理与液冷

Artificial Intelligence for Energy Optimization in Data Centers

Mohammed Basharath Ullah、Summaiya Unnisa Begum、Mohammed Nadeem Ullah

Published 2026-09-03 · arXiv · Credibility S

Data

Abstract, interpretation and reference

Abstract

Data

中文解读

背景:AI 数据中心负载、功率密度和能源约束同步上升,液冷、热管理和数据中心能效正在成为智算中心设计的关键变量。问题:论文聚焦现有方案在效率、可靠性或工程协同上的瓶颈。方法:摘要显示作者采用建模优化、调度分析或算法评估,把运行负载、冷却/能源系统和基础设施约束放在同一分析框架中。结果:研究重点指向冷却效率、能源利用或运维策略的改进方向。意义:对日报读者而言,它可用于判断液冷方案、热管理路线和高密度部署节奏。仍需结合全文实验条件、样本范围和成本假设核验。

参考文献

Mohammed Basharath Ullah, Summaiya Unnisa Begum, Mohammed Nadeem Ullah. Artificial Intelligence for Energy Optimization in Data Centers[J/OL]. (2026-09-03)[2026-09-14]. https://arxiv.org/abs/2609.03716.

Full text 中文海报
热管理与液冷 论文图示
Research Article热管理与液冷

A High- and Variable-Dimensional Measurement of the $Z$+jets Differential Cross Section with the ATLAS Experiment and Artificial Intelligence

Kevin Greif

Published 2026-08-28 · arXiv · Credibility S

Proton-proton collisions at the Large Hadron Collider (LHC) offer the opportunity to observe the interactions of fundamental particles at very high energy

Abstract, interpretation and reference

Abstract

Proton-proton collisions at the Large Hadron Collider (LHC) offer the opportunity to observe the interactions of fundamental particles at very high energy

中文解读

背景:AI 数据中心负载、功率密度和能源约束同步上升,液冷、热管理和数据中心能效正在成为智算中心设计的关键变量。问题:论文聚焦现有方案在效率、可靠性或工程协同上的瓶颈。方法:摘要显示作者采用实验验证、原型测试或测量对比,把运行负载、冷却/能源系统和基础设施约束放在同一分析框架中。结果:研究重点指向冷却效率、能源利用或运维策略的改进方向。意义:对日报读者而言,它可用于判断液冷方案、热管理路线和高密度部署节奏。仍需结合全文实验条件、样本范围和成本假设核验。

参考文献

Kevin Greif. A High- and Variable-Dimensional Measurement of the $Z$+jets Differential Cross Section with the ATLAS Experiment and Artificial Intelligence[J/OL]. (2026-08-28)[2026-09-14]. https://arxiv.org/abs/2608.28449.

Full text 中文海报
热管理与液冷 论文图示