NWP | Bias Correction
應用「模式輸出統計」校正數碼氣溫預報 | HKO | 2011 多元線性回歸及卡爾曼濾波 出現次數最多的最佳因子，首三位為2米溫度、925hPa經向風及850hPa溫度 應用「模式輸出統計」校正相對濕度及風的數碼預報 | HKO | 2012 把2010年9月至2011年8月的數據分為四季，分別找出不同統計模型在四季及全年(即2010年9月至2011年8月)的最佳因子組合。 不同季節的比較
ML | CorrDiff (NVIDIA)
News 生成式AI與傳統數值天氣預報的結合NVIDIA CorrDiff天氣降尺度模型 | 科學月刊 | 2024-07-15 輝達（NVIDIA）執行長黃仁勳今（2024）年3月在美國舉辦的GTC（GPU technology conference）大會及6月在臺灣舉辦的臺北國際電腦展（COMPUTEX）上，皆展示了一項稱為「CorrDiff」（ Corrector Diffusion ）
CFD | DNS
DNS A direct numerical simulation (DNS) is a simulation in computational fluid dynamics (CFD) in which the Navier–Stokes equations are numerically solved without any turbulence model. This means that
ML | Genesis
What is Genesis? https://github.com/Genesis-Embodied-AI/Genesis Genesis is a physics platform designed for general purpose Robotics/Embodied AI/Physical AI applications. It is simultaneously multip
CFD | Wall Function
Why using wall functions approach? Nowadays turbulence models are heavily used when doing CFD simulations. However, some mature turbulence models such as $ k-$ are only valid in the area of turbulence
CFD | RANS vs. URANS vs. LES
Characteristics of turbulence One characteristic of turbulent flows is their irregularity (or randomness). A fully deterministic approach to characterize turbulent flows is very difficult. Turbulent f
ML | Diffusion Models (Denoising diffusion probabilistic models)
Diffusion Models Diffusion Models are generative models, meaning that they are used to generate data similar to the data on which they are trained. Fundamentally, Diffusion Models work by destroying t
ML | U-Net
U-Net The U-Net architecture follows an encoder-decoder cascade structure, where the encoder gradually compresses information into a lower-dimensional representation. Then the decoder decodes this inf
