v0.3.0

PaddleCFD 0.3.0 Release Notes

中文版：参见 release_notes_cn.md.

PaddleCFD 0.3.0 Overview

PaddleCFD is a deep learning toolkit based on the PaddlePaddle deep learning framework, focused on Computational Fluid Dynamics (CFD) tasks. It is used for fluid mechanics equation discovery, numerical simulation acceleration, flow shape optimization, and flow control strategy discovery. The PaddleCFD 0.3.0 release mainly focuses on accelerating CFD numerical simulations, constructing fluid computation surrogate models based on the PaddlePaddle framework. It has achieved breakthroughs in model accuracy and computational efficiency, providing high baseline models for researchers in the field of fluid mechanics and turnkey tool libraries for enterprise users. The main features of the PaddleCFD 0.3.0 release are as follows:

• Coverage of mainstream and cutting-edge surrogate models: The PaddleCFD 0.3.0 release includes specific models for scientific computing, such as Fourier Neural Operators (FNO), DeepONet, and cutting-edge deep learning models like Transformer, diffusion models, and KAN. These models possess powerful operator learning, prediction, and generation capabilities and represent the forefront of AI + fluid mechanics research (Data source: Clarivate Analytics Essential Science Indicators).
• Leading precision/computational efficiency: PaddleCFD 0.3.0 has improved and adapted models from publicly available papers, achieving a dual increase in model accuracy and computational efficiency. The PPFNO model, with integral correction, has an average relative error of less than 3% for drag coefficient in the test dataset (baseline model ~8%); via operator fusion, training speed is increased by 188%, and inference speed by 159%, achieving second-level inference for tens of millions of grids. The PPTransfomer model, through PaddlePaddle's dynamic-to-static conversion and neural network compiler, boosted training speed by 29.4%, with dual-card parallel efficiency reaching 90.2%, supporting parallel inference for models with tens of millions of grids. The PPKAN model, compared to traditional MLP neural networks, has a 30% improvement in accuracy with an equivalent number of parameters. The PPDifusion model, through data parallelism, achieves scalable acceleration, with single-machine multi-card parallel efficiency of over 99.4%. The PPDeepONet model, based on the MultiONet network structure and through the second-order optimizer SOAP, achieves an approximate 10% improvement in model accuracy.
• Full-scenario support for industry implementation: PaddleCFD 0.3.0 places greater emphasis on industrial application implementation, enhancing model accuracy and computational efficiency through real industrial business scenarios and improving model functional modules. For example, the PPFNO model developed a complete set of functional modules for the drag coefficient prediction task, including training/inference data (volume mesh & surface mesh) preprocessing, distributed training, offline inference, online inference, etc., enabling the model to be containerized for training, inference, and deployment, with applications in leading high-speed train industry enterprises. Additionally, PaddleCFD 0.3.0 provides data parsing modules for various CFD data formats, enabling seamless integration with multiple traditional CFD simulation software.
• Single-folder strategy to enhance usability: PaddleCFD 0.3.0, referencing the successful single-file strategy of the HuggingFace AI suite, places all modules related to the model in the same folder, and avoids over-wrapping of deep learning framework APIs, reducing the learning cost and maintenance cost for users.

PaddleCFD 0.3.0 Updates

New Features

• The PPDeepONet model has added the PirateNets network structure, improving the stability and efficiency of model training.
• The PPFNO model has open-sourced the custom operator fused-segment-csr, which provides higher model training and inference throughput, decouples memory usage from sample size, and supports training and inference for models with tens of millions of grids.

Bug Fixes

• The PPTransformer model has fixed a model import error issue.
• By installing custom operators from source code, the issue of the toolkit environment lacking a specific dynamic library is fixed.
• Updated the PaddleCFD installation documentation, fixing potential errors encountered during installation.
• Fixed an issue where the PPFNO model inference-data preprocessing module imported an unnecessary package causing an error.

Contributors

guhaohao0991, HydrogenSulfate, KaiCHEN-HT, liaoxin2, lijialin03, wangguan1995, XiaoguangHu01

v0.2.0

PaddleCFD 0.2.0 Release Notes

中文版：参见 release_notes_cn.md.

Overview of PaddleCFD 0.2.0 Version

PaddleCFD is a deep learning toolkit based on the PaddlePaddle deep learning framework, focused on Computational Fluid Dynamics (CFD) tasks. It's used for fluid mechanics equations discovery, numerical simulation acceleration, flow shape optimization, and flow control strategy discovery. The PaddleCFD 0.2.0 version mainly focuses on accelerating CFD numerical simulations, creating surrogate models of fluid computation based on the PaddlePaddle framework. It has achieved breakthroughs in model accuracy and computational efficiency, providing high baseline models for researchers in the field of fluid mechanics and turnkey tool libraries for enterprise users. The main features of the PaddleCFD 0.2.0 version are as follows:

• Coverage of Mainstream Cutting-edge Surrogate Models: The PaddleCFD 0.2.0 version includes specific scientific computing models like the Fourier Neural Operator (FNO) and DeepONet, as well as cutting-edge deep learning models such as Transformer, Diffusion models, and KAN. These models possess strong operator learning, prediction, and generation capabilities, and are at the forefront of research in the AI+ fluid mechanics field (data source: Clarivate Analytics Essential Science Indicators).
• Leading Precision/Computational Efficiency: The PaddleCFD 0.2.0 version has improved and adapted models from publicly available versions in papers, achieving dual enhancements in model accuracy and computational efficiency. The ppfno model, through integral learning, has an average relative error of less than 3% in drag coefficient predictions in the test dataset (baseline model ~8%). Through operator fusion, training speed has increased by 75%, and inference speed has tripled, achieving second-level inference for tens of millions of grids. The pptransfomer model, through PaddlePaddle's dynamic-to-static conversion and Compiler Infrastructure for Neural Networks (CINN), has improved training speed by 29.4%, with dual-GPU parallel efficiency reaching 90.2%, supporting parallel inference for tens of millions of grids. The ppkan model, compared to traditional MLP neural networks, has a 30% accuracy improvement with a similar number of parameters. The ppdifusion model, through data parallelism, achieves scalable acceleration of the model, with single-machine multi-GPU parallel efficiency reaching over 99.4%. The ppdeeponet model, based on the MultiONet network structure and through the second-order optimizer SOAP, achieves approximately a 10% improvement in model accuracy.
• Full Scenario Support for Industrial Implementation: The PaddleCFD 0.2.0 version pays more attention to the industrial implementation, polishing and enhancing model accuracy and computational efficiency through real industrial business scenarios, and perfecting model function modules. For example, for the ppfno model's task of predicting aerodynamic drag coefficients, a complete set of functional modules has been developed, including training/inference data (volumetric & surface grid) preprocessing, distributed training, offline inference, online inference, etc., enabling containerized deployment of model training and inference, with applications in leading enterprises in the high-speed train industry. Additionally, the PaddleCFD 0.2.0 version offers data parsing modules for various CFD data formats, enabling seamless integration with multiple traditional CFD simulation software.
• Single Folder Strategy to Enhance Usability: Drawing from the successful suite strategy of HuggingFace in the AI field, the PaddleCFD 0.2.0 version places all modules involved in the model into the same folder. This avoids excessive encapsulation of deep learning framework APIs, reducing the learning curve and maintenance costs for users.

List of Contributors

guhaohao0991, HydrogenSulfate, KaiCHEN-HT, liaoxin2, lijialin03, wangguan1995, XiaoguangHu01