A dynamic masking technique for particle image velocimetry using convolutional autoencoders

Vennemann, Bernhard; Rösgen, Thomas

doi:10.1007/s00348-020-02984-w

Cited by 21 publications

(16 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, listing semantic units requires consideration of text and images based on learning independent context vectors from text encoders [9] and image encoders [10], respectively; secondly, in the process of learning spaces for text and images, sentences are constrained to be closer to the meaning of the sentence in the spaces, that is, to be semantically identical. Finally, the concepts of semantic unit and semantic unit representation are set up, specifically as follows:…”

Section: Enumeration Semantic Unitmentioning

confidence: 99%

Automatic Calibration Algorithm for English Text Translation Based on Semantic Features

Liu

2021

Security and Communication Networks

View full text Add to dashboard Cite

At present, the existing methods of English article flip calibration neglect to extract English semantic features, which leads to errors in English flip results and has a great impact on the accuracy and time consumption of translation sentence calibration. Therefore, a semantic feature-based automatic text flipping calibration algorithm is proposed. According to the features of semantic information in machine translation, a semantic grammar tree is constructed to complete the machine turning of English articles. The CART decision tree attribute is obtained, and the random forest method is introduced to extract the input matrix and output matrix of the corpus feature as samples to determine the spatial attribute feature of the mistranslated sentences. Choose 10000 English sentences about human body parts as the experimental object and design the simulation experiment. The experimental results show that the minimum and maximum accuracy rates are 95.4% and 100.0%, respectively. The proposed algorithm is time-consuming, and the KSMR value is lower than that of the traditional method. It is proved that the error rate of English article flipping is significantly reduced.

show abstract

Section: Enumeration Semantic Unitmentioning

confidence: 99%

Automatic Calibration Algorithm for English Text Translation Based on Semantic Features

Liu

2021

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…Autoencoders have recently become popular for the nonlinear dimensionality reduction of datasets extracted from several high dimensional systems. These have been motivated by the extraction of coherent structures that parameterize low-dimensional embeddings in manifolds [23,24,25], and the utilization of these embeddings for efficient surrogate models of nonlinear dynamical systems [26,27,28,29,30]. In this work, we utilize convolutional autoencoders to identify low-dimensional representations of experimentally collected data for building parameter-observation maps where the former are obtained through meteorological and wind turbine data and the latter are LiDAR measurements collected in the wake generated by wind turbines.…”

Section: Convolutional Autoencodermentioning

confidence: 99%

Data-Driven Wind Turbine Wake Modeling via Probabilistic Machine Learning

Renganathan¹,

Maulik²,

Letizia³

et al. 2021

Preprint

View full text Add to dashboard Cite

Wind farm design primarily depends on the variability of the wind turbine wake flows to the atmospheric wind conditions, and the interaction between wakes. Physics-based models that capture the wake flow-field with high-fidelity are computationally very expensive to perform layout optimization of wind farms, and, thus, data-driven reduced order models can represent an efficient alternative for simulating wind farms. In this work, we use real-world light detection and ranging (LiDAR) measurements of wind-turbine wakes to construct predictive surrogate models using machine learning. Specifically, we first demonstrate the use of deep autoencoders to find a low-dimensional latent space that gives a computationally tractable approximation of the wake LiDAR measurements. Then, we learn the mapping between the parameter space and the (latent space) wake flow-fields using a deep neural network. Additionally, we also demonstrate the use of a probabilistic machine learning technique, namely, Gaussian process modeling, to learn the parameter-space-latent-space mapping in addition to the epistemic and aleatoric uncertainty in the data. Finally, to cope with training large datasets, we demonstrate the use of variational Gaussian process models that provide a tractable alternative to the conventional Gaussian process models for large datasets. Furthermore, we introduce the use of active learning to adaptively build and improve a conventional Gaussian process model predictive capability. Overall, we find that our approach provides accurate approximations of the wind-turbine wake flow field that can be queried at an orders-of-magnitude cheaper cost than those generated with high-fidelity physics-based simulations.Keywords Wind turbine • renewable energy • wake • deep learning • Gaussian processes • lidar

show abstract

“…The accuracy of the employed PIV algorithm has been previously exstensively assessed in literature (e.g., Guérin et al (2020), Vennemann and Rösgen (2020), Mohammadshahi et al (2020), Narayan et al (2020)). However, in order to ensure the physical consistency of our PIV measurements, two benchmark cases are here presented.…”

Section: Assessment Of Piv Chain Analysismentioning

confidence: 99%

“…DM techniques are included in few commercially available PIV analysis software packages (TSI Instruments 2014;DantecDynamics 2018). Recent developments (Vennemann and Rösgen 2020) envisage the application of neural-network automatic masking techniques, which however require synthetic datasets to be generated in order to train the network.…”

Section: Introductionmentioning

confidence: 99%

A new dynamic masking technique for time resolved PIV analysis

Lombardi

Rocca

Prestininzi

2021

J Vis

View full text Add to dashboard Cite

Time resolved PIV encompassing moving and/or deformable objects interfering with the light source requires the employment of dynamic masking (DM). A few DM techniques have been recently developed, mainly in microfluidics and multiphase flows fields. Most of them require ad-hoc design of the experimental setup, and may spoil the accuracy of the resulting PIV analysis. A new DM technique is here presented which envisages, along with a dedicated masking algorithm, the employment of fluorescent coating to allow for accurate tracking of the object. We show results from measurements obtained through a validated PIV setup demonstrating the need to include a DM step even for objects featuring limited displacements. We compare the proposed algorithm with both a no-masking and a static masking solution. In the framework of developing low cost, flexible and accurate PIV setups, the proposed algorithm is made available through a freeware application able to generate masks to be used by an existing, freeware PIV analysis package. Graphic abstract

show abstract

A dynamic masking technique for particle image velocimetry using convolutional autoencoders

Cited by 21 publications

References 8 publications

Automatic Calibration Algorithm for English Text Translation Based on Semantic Features

Automatic Calibration Algorithm for English Text Translation Based on Semantic Features

Data-Driven Wind Turbine Wake Modeling via Probabilistic Machine Learning

A new dynamic masking technique for time resolved PIV analysis

Contact Info

Product

Resources

About