News on China's scientific and technological development.

Japan: Let put export control in Crycoolers sales to China.
Chinese Cryocoolers companies: OH NO that's horrible, please don't leave the Chinese market.
Chinese Cryocoolers companies: But we can still sell in your country?
Japan: Yes I guest.
Chinese Cryocoolers companies: YES...OH NO that's horrible.

Boreas Cryogenics Cryocoolers Technology

Chinese researchers found a way to boost the performance of Nvidia GPU 800-fold for some computing tasks.

Chinese algorithm boosts Nvidia GPU performance 800-fold in science computing​

A breakthrough by Chinese researchers could help solve complex problems in industries ranging from aerospace to bridge design​

Click to expand...

TikTok investing $3.7B in Thailand data center project. It seems like TikTok has been investing in quite a few data centers across ASEAN region.

CAS invented a new model for studying "combustion hysteresis phenomenon" of scramjet engines, increasing the total numbers of elements in computer aided simulations by 2 orders of magnitude, from millions in the traditional models to hundreds of millions.


超燃冲压发动机中存在一种特殊的现象，即相同流动控制参数，由于历史路径的不同，可能造成发动机存在非唯一的运行模态和工作裕度。超燃冲压发动机在模态转换过程中存在燃烧迟滞效应在近年来的国内外地面实验中得到广泛证实。作为高超声速飞行的首选动力，超燃冲压发动机中的迟滞效应给发动机的主动控制带来巨大困难。

目前关于迟滞效应的研究多为实验现象观察。通过数值手段准确复现燃烧迟滞现象并揭示其机理是超声速燃烧研究的挑战性难题之一。这是因为燃烧迟滞现象涉及诸多的多物理耦合因素，过度简化的物理模型和燃烧化学反应机理会“遗漏”这一现象。另一个原因是燃烧迟滞现象是一个动态演化过程，其数值复现对计算资源的需求极大。近年来，中国科学院力学研究所空天飞行器数值模拟课题组提出了以动态分区火焰面模型（DZFM）为核心的“六位一体”超声速燃烧模型体系，在高保真的同时实现了计算效率的量级式提升（相比FLUENT提升60倍）。相关工作以“Combustion Hysteresis Phenomenon in a Dual-Mode Scramjet”为题发表于航空航天领域顶级期刊AIAA Journal。

基于传统方法的超声速燃烧工程模拟多基于百万级网格，而动态分区火焰面模型可以在不显著增加计算资源的前提下实现亿级网格的超声速燃烧大涡模拟。动态分区概念的提出有效降低了大涡模拟等高解析度计算方法的工程应用门槛，为数值复现燃烧迟滞现象进而揭示其内在机理奠定了方法基础。


The combustion hysteresis of a hydrogen-fueled ramp-based dual-mode scramjet under a flight Mach number of 5 was examined by using IDDES based on up to 115.75 million cells and a detailed 13s/33r H2/O2 mechanism. As is evident by the wall pressure measurements, the scramjet and ramjet modes were successfully reproduced for four global fuel equivalence ratios (Φ). However, a hysteresis in the mode transition occurs under the Φ-increasing and Φ-decreasing paths. A smooth ram-to-scramjet mode transition with gradual pressure drop occurs when Φ decreases to 0.08 under the Φ-decreasing path, whereas a jumped scram-to-ram mode transition with abrupt pressure rise occurs at Φ=0.17 under the Φ-increasing path. The combustion efficiency under the Φ-decreasing path is always higher than those under the Φ-increasing path within the hysteresis loop. This higher efficiency can be attributed to mixing enhancement and kinetic strengthening. For the former, longer residence time and richer vortexes promote the mixing. Meanwhile, the latter kinetic strengthening is contributed by the higher pressure, higher temperature, and richer radicals inherited from the former combustion state. A dynamic regulation mechanism realized through a feedback loop with a characteristic response time of 0.89 flush through time resides the pseudo-shock wave in the isolator and sustains the ramjet mode. Hysteresis causes more total pressure loss due to stronger pseudo-shock structure and more heat addition. Within the hysteresis loop, higher combustion efficiency dominates the higher thrust under Φ-decreasing path.

Tsinghua developed a new model in predicting battery degradation trajectory. It is based on "physics-informed learning" and 25 times faster than the "traditional capacity calibration test".

Non-destructive degradation pattern decoupling for early battery trajectory prediction via physics-informed learning​

Manufacturing complexities and uncertainties have impeded the transition from material prototypes to commercial batteries, making their verification a critical quality assessment link. A fundamental challenge is to decouple electrochemical interactions for establishing a quantitative mapping from electrochemical parameters to macro battery performance. Here, we show that the proposed physics-informed learning model can quantify and visualize temporally resolved thermodynamic and kinetic parameters from field accessible electric signals, facilitating a non-destructive degradation pattern decoupling. The lifetime trajectory prediction is 25 times faster than the traditional capacity calibration test while retaining a 95.1% average accuracy across temperatures, underpinned by projected electrochemical data from early cycle observations which have not yet been established. We rationalize this predictability to the interpretation of statistical insights from material-agnostic featurization, excited by a multistep charging scheme with different current intensities and their switching conditions. The waste management of defective prototypes is enabled by statistically and non-destructively interpreting internal electrochemical states, demonstrating a 19.76 billion USD defective material recycling market by 2060. This paper highlights the potential of revisiting electrochemical degradation behaviors using physics-informed learning and dynamic current excitations, facilitating next-generation battery manufacturing, reuse, and recycling sustainability.


随着全球对可再生能源需求的日益增长，电池作为储能系统关键技术的地位愈发凸显。然而，电池研发过程中面临着从材料原型到商业产品转化的重重挑战，电池原型验证效率低、研发成本高以及生产废料管理不善等问题制约着电池行业的可持续发展。

在电池制造领域，传统的容量校准方法在原型验证时需耗费大量时间。同时，制造的不一致性和电池老化的多样性，使得电池原型的性能评估变得极为复杂。为此，清华大学深圳国际研究生院张璇、周光敏、李阳副教授团队与合作者提出了一种基于物理信息学习的电池衰减轨迹早期预测方法。该方法通过计算热力学和动力学参数，并将其关联至未来状态变化，从而实现对电池整个衰减轨迹的早期预测。与传统方法相比，该方法仅利用电池原型的早期循环数据（50次循环，占总寿命4%），即可达成95.1%全寿命平均预测准确率，将原型验证速度提升了至少25倍。