arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2012.00642physics.flu-dyn
收藏

基于贝叶斯神经网络的粒子图像测速分析及其同时的不确定性量化

Particle image velocimetry analysis with simultaneous uncertainty quantification using Bayesian neural networks

Mia Morrell, Kyle Hickmann, Brandon Wilson

摘要:

粒子图像测速技术(PIV)是实验流体力学中从图像中提取流场的有效工具。最近,卷积神经网络(CNN)已被用于执行PIV分析,其精度与经典方法相当。在这里,我们扩展了CNN的使用,分析PIV数据,同时提供推断流场的同时不确定性量化。本文采用的方法是贝叶斯卷积神经网络(BCNN),它通过变分贝叶斯学习CNN权重的分布。我们比较了三种不同BCNN模型的性能。第一网络仅从图像询问区域估计流速。我们的第二个模型学习从图像询问区域和询问区域互相关图中推断速度。最后,我们表现最好的网络仅从询问区域互相关图中获得速度。我们发现,使用询问区域互相关图作为输入的BCNN比仅使用询问窗口作为输入的BCNN表现更好,并讨论了为什么会出现这种情况。此外,我们在完整的测试图像对上测试了性能最好的BCNN,结果表明可以在95%的置信区间内捕获100%的真实粒子位移。最后,我们表明,BCNN可以推广到多通道PIV算法中使用,具有适度的准确性损失,这可以通过未来的微调和训练计划来克服。据我们所知,这是第一次尝试使用贝叶斯神经网络来执行粒子图像测速。

英文摘要:

Particle image velocimetry (PIV) is an effective tool in experimental fluid mechanics to extract flow fields from images. Recently, convolutional neural networks (CNNs) have been used to perform PIV analysis with accuracy on par with classical methods. Here we extend the use of CNNs to analyze PIV data while providing simultaneous uncertainty quantification on the inferred flow field. The method we apply in this paper is a Bayesian convolutional neural network (BCNN) which learns distributions of the CNN weights through variational Bayes. We compare the performance of three different BCNN models. The first network estimates flow velocity from image interrogation regions only. Our second model learns to infer velocity from both the image interrogation regions and interrogation region cross-correlation maps. Finally, our best performing network derives velocities from interrogation region cross-correlation maps only. We find that BCNNs using interrogation region cross-correlation maps as inputs perform better than those using interrogation windows only as inputs and discuss reasons why this may be the case. Additionally, we test the best performing BCNN on a full test image pair, showing that 100% of true particle displacements can be captured within its 95% confidence interval. Finally, we show that BCNNs can be generalized to be used with multi-pass PIV algorithms with a moderate loss in accuracy, which may be overcome by future work on finetuning and training schemes. To our knowledge, this is the first effort to use Bayesian neural networks to perform particle image velocimetry.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑