采用混合粗糙数据物理信息神经网络的扑翼气动性能预测方法

胡付佳; 周逸伦; 刘小民

doi:10.7652/xjtuxb202311019

您当前的位置：

首页 >

文章列表页 >

采用混合粗糙数据物理信息神经网络的扑翼气动性能预测方法

—————————————————————— | 更新时间：2025-12-08

- 采用混合粗糙数据物理信息神经网络的扑翼气动性能预测方法
- A Predicting Method on Aerodynamic Performance of Flapping Wing Using Hybrid Coarse Data Driven Physical Informed Neural Network
- 西安交通大学学报 2023年57卷第11期页码：194-205
- 作者机构：
  
  1. 西安交通大学能源与动力工程学院,西安,710049
  2. 西安陕鼓动力股份有限公司,西安,710075
- 作者简介：
  
  : 2023-04-04。作者简介: 胡付佳(1994—),女,博士生
  刘小民(通信作者),男,教授,博士生导师。基金项目: 陕西省重点研发计划资助项目(2023-YBGY-280)。
- 基金信息：
  
  : 2023-04-04。作者简介: 胡付佳(1994—),女,博士生;刘小民(通信作者),男,教授,博士生导师。基金项目: 陕西省重点研发计划资助项目(2023-YBGY-280)。
- DOI：10.7652/xjtuxb202311019
  中图分类号： TP18;V211
- 网络首发：2023-11-10，
  
  纸质出版：2023
- 稿件说明：
移动端阅览
胡付佳, 周逸伦, 刘小民. 采用混合粗糙数据物理信息神经网络的扑翼气动性能预测方法[J]. 西安交通大学学报, 2023,57(11):194-205.

HU Fujia, ZHOU Yilun, LIU Xiaomin. A Predicting Method on Aerodynamic Performance of Flapping Wing Using Hybrid Coarse Data Driven Physical Informed Neural Network[J]. 2023, 57(11): 194-205.
胡付佳, 周逸伦, 刘小民. 采用混合粗糙数据物理信息神经网络的扑翼气动性能预测方法[J]. 西安交通大学学报, 2023,57(11):194-205. DOI： 10.7652/xjtuxb202311019.

HU Fujia, ZHOU Yilun, LIU Xiaomin. A Predicting Method on Aerodynamic Performance of Flapping Wing Using Hybrid Coarse Data Driven Physical Informed Neural Network[J]. 2023, 57(11): 194-205. DOI： 10.7652/xjtuxb202311019.

摘要

为解决时空尺度上扑翼流动控制方程求解需要花费大量时间和计算资源的问题

基于强非线性曲线拟合能力的物理信息神经网络(PINN)深度学习方法

提出了一种混合粗糙数据驱动物理信息神经网络模型(HCDD-PINN)

研究了模型对涉及非定常流动特征和动边界二维俯仰扑翼问题的训练和预测性能。通过使用相较于传统计算流体动力学方法(CFD)更为粗糙的数据驱动模型训练

将扑翼流动控制方程嵌入神经网络损失中

并施加初始条件和边界条件约束

采用一阶自适应矩优化算法(ADAM)和二阶拟牛顿法优化算法(L-BFGS-B)

以前馈-反向传播方式最小化模型损失函数

从而提高模型预测控制方程数值解的准确性和可靠性。结果表明:与原始PINN模型相比

HCDD-PINN模型显著降低了流场的预测误差

能够准确地预测扑翼瞬时气动力和瞬时速度及压力场

训练时间缩短了75%。此外

训练完成的HCDD-PINN模型可以快速获得流场任意时刻的物理信息

而传统CFD方法则需要重新对流场进行计算。该研究为求解扑翼流动控制方程乃至流体非线性偏微分方程组(PDEs)提供了一种有效的替代方案。

Abstract

The solution of the governing equations of the flapping wing problem on a spatiotemporal scale takes a great amount of time and computational resources. To solve this challenge

this paper proposes a hybrid coarse data-driven with physics-informed neural network model(HCDD-PINN). The model leverages its strong nonlinear curve fitting ability to investigate the training and predicting performance of the model for two-dimensional pitching flapping wing problems

which involves unsteady flow characteristics and motion boundaries. The model is trained using an order-of-magnitude grid

coarser than the traditional computational fluid dynamic(CFD)required. The architecture is devised to enforce the initial and boundary conditions and incorporate the governing equations into the loss of the neural network. The optimization process is conducted by ADAM and L-BFGS-B methods with a feed forward-back propagation manner

to improve the accuracy and reliability of predicting the time evolution of nonlinear PDEs solutions. The results show that the proposed HCDD-PINN framework exhibits improved stability and accuracy compared to the PINN models that lack internal data. It effectively reduces the prediction error of the flow field

and accurately predicts the instantaneous aerodynamic forces

velocity and pressure fields of flapping wings. Moreover

the training time has accelerated by approximately four times. Additionally

the interested variables of the flow field at any instant can be rapidly obtained by the trained HCDD-PINN model

which is superior to the traditional CFD method that usually needs to be re-run. This study provides an effective alternative for solving flapping wing governing equations and even nonlinear partial differential equations(PDEs).

关键词

Keywords

references

GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning [M]. Cambridge, MA, USA: The MIT Press, 2016.

XIONG Huiy, ALIPANAHI B, LEE L J, et al. The human splicing code reveals new insights into the genetic determinants of disease [J]. Science, 2015, 347(6218): 1254806.

周逸伦, 孙永瑞, 胡付佳, 等. 仿鸮翼离心压缩机叶轮的参数优化及性能研究 [J]. 西安交通大学学报, 2023, 57(7): 151-159.

ZHOU Yilun, SUN Yongrui, HU Fujia, et al. Parameters optimization and performance investigation of centrifugal compressor with bionic impeller inspired by owl wing [J]. Journal of Xi'an Jiaotong University, 2023, 57(7): 151-159.

RAISSI M, PERDIKARIS P, KARNIADAKIS G E. Physics informed deep learning: part I data-driven solutions of nonlinear partial differential equations [EB/OL].(2017-11-28)[2023-03-12]. https://arxiv. org/abs/1711.10561.

RAISSI M, PERDIKARIS P, KARNIADAKIS G E. Physics informed deep learning: part II data-driven discovery of nonlinear partial differential equations [J].(2017-11-28)[2023-03-12]. https://arxiv.org/abs/1711.10566.

RAISSI M, PERDIKARIS P, KARNIADAKIS G E. Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations [J]. Journal of Computational Physics, 2019, 378: 686-707.

RAO Chengping, SUN Hao, LIU Yang. Physics-informed deep learning for incompressible laminar flows [J]. Theoretical and Applied Mechanics Letters, 2020, 10(3): 207-212.

CHOI S, JUNG I, KIM H, et al. Physics-informed deep learning for data-driven solutions of computational fluid dynamics [J]. Korean Journal of Chemical Engineering, 2022, 39(3): 515-528.

田十方, 李彪. 梯度优化物理信息神经网络(GOPINNs):求解复杂非线性问题的深度学习方法 [J/OL]. 物理学报: 1-20[2023-03-12]. http://kns.cnki. net/kcms/detail/11.1958. O4.20230201.2155.004.html

TIAN Shifang, LI Biao. Gradient-optimized physics-informed neural networks(GOPINNs): a deep learning method for solving complex nonlinear problems [J/OL]. Acta Physica Sinica: 1-20[2023-03-12]. http://kns.cnki.net/kcms/detail/11.1958.O4.20230201.2155.004.html

CHENG Chen, ZHANG Guangtao. Deep learning method based on physics informed neural network with Resnet block for solving fluid flow problems [J]. Water, 2021, 13(4): 423.

RAISSI M, WANG Zhicheng, TRIANTAFYLLOU M S, et al. Deep learning of vortex-induced vibrations [J]. Journal of Fluid Mechanics, 2019, 861: 119-137.

崔永赫, 张文耀, 闫慧龙, 等. “硬”边界低阶导数型物理信息神经网络: 一种流动求解器 [J]. 西安交通大学学报, 2022, 56(9): 123-133.

CUI Yonghe, ZHANG Wenyao, YAN Huilong, et al. “Hard” boundary low-order derivative physics informed neural network: a fluid flow solver [J]. Journal of Xi'an Jiaotong University, 2022, 56(9): 123-133.

王意存, 邢江宽, 罗坤, 等. 基于物理信息神经网络的燃烧化学微分方程求解 [J]. 浙江大学学报(工学版), 2022, 56(10): 2084-2092.

WANG Yicun, XING Jiangkuan, LUO Kun, et al. Solving combustion chemical differential equations via physics-informed neural network [J]. Journal of Zhejiang University(Engineering Science), 2022, 56(10): 2084-2092.

吴国正, 王发杰, 程隋福, 等. 基于物理信息神经网络的内部声场正反问题数值计算 [J]. 计算物理, 2022, 39(6): 687-698.

WU Guozheng, WANG Fajie, CHENG Suifu, et al. Numerical simulation of forward and inverse problems of internal sound field based on physics-informed neural network [J]. Chinese Journal of Computational Physics, 2022, 39(6): 687-698.

WOOTTON R. From insects to microvehicles [J]. Nature, 2000, 403(6766): 144-145.

SRYGLEY R B, THOMAS A L R. Unconventional lift-generating mechanisms in free-flying butterflies [J]. Nature, 2002, 420(6916): 660-664.

HU Fujia, LIU Xiaomin. Effects of stroke deviation on hovering aerodynamic performance of flapping wings [J]. Physics of Fluids, 2019, 31(11): 111901.

HU Fujia, WANG Yuanying, LI Dian, et al. Effects of asymmetric stroke deviation on the aerodynamic performance of flapping wing [J]. Proceedings of the Institution of Mechanical Engineers: Part G Journal of Aerospace Engineering, 2023, 237(2): 480-499.

SUN Luning, GAO Han, PAN Shaowu, et al. Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data [J]. Computer Methods in Applied Mechanics and Engineering, 2020, 361: 112732.

FUKS O, TCHELEPI H A. Limitations of physics informed machine learning for nonlinear two-phase transport in porous media [J]. Journal of Machine Learning for Modeling and Computing, 2020, 1(1): 19-37.

RAISSI M. Deep hidden physics models: deep learning of nonlinear partial differential equations [J]. Journal of Machine Learning Research, 2018, 19(1): 932-955.

RAISSI M, YAZDANI A, KARNIADAKIS G E. Hidden fluid mechanics: learning velocity and pressure fields from flow visualizations [J]. Science, 2020, 367(6481): 1026-1030.

KOCHKOV D, SMITH J A, ALIEVA A, et al. Machine learning-accelerated computational fluid dynamics [J]. Proceedings of the National Academy of Sciences of the United States of America, 2021, 118(21): e2101784118.

BAYDIN A G, PEARLMUTTER B A, RADUL A A, et al. Automatic differentiation in machine learning: a survey [J]. Journal of Machine Learning Research, 2017, 18(1): 5595-5637.

ELLINGTON C P, VAN DEN BERG C, WILLMOTT A P, et al. Leading-edge vortices in insect flight [J]. Nature, 1996, 384(6610): 626-630.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

深度学习驱动的碳纤维管壁缺陷镜面红外检测无损定量评估方法

航空发动机高温结构损伤检测研究与应用进展

面向模型预测控制的风电机组深度Koopman全局线性建模方法

跨声速压气机转子通用k-ω湍流模型的机器学习辅助优化

仿鸮翼前网罩对环状叶轮循环扇性能的影响