Deep Learning for Flow Sculpting: Insights into Efficient Learning using Scientific Simulation Data

Stoecklein, Daniel; Lore, Kin Gwn; Davies, M. C. R.; Sarkar, Soumik; Ganapathysubramanian, Baskar

doi:10.1038/srep46368

Cited by 57 publications

(40 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As has been well studied in machine learning field [5], a further regularization can be introduced to the NNLS problem. In this study, a commonly used ridge regression [71], or called Tikhonov regularization, is applied to address the ill-posed issues, and the NNLS problem (29) where the regularized coefficient is…”

Section: Solving Non-negative Least Squaresmentioning

confidence: 99%

“…In conjunction with machine learning techniques such as manifold learning [21] or neural networks [22], the recent studies [23][24][25] offer a new paradigm for data-driven computing for various applications such as design of materials [26]. There is a vast body of literature devoted to these subjects, including the recent developments based on nonlinear dimensionality reduction [24], nonlinear regression, deep learning [27][28][29], among others.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A physics-constrained data-driven approach based on locally convex reconstruction for noisy database

Chen

2020

Computer Methods in Applied Mechanics and Engineering

View full text Add to dashboard Cite

Physics-constrained data-driven computing is a hybrid approach that integrates universal physical laws with data-based models of experimental data to enhance the scientific computing. A new data-driven simulation approach enriched with a locally convex reconstruction, termed the local convexity data-driven (LCDD) computing, is proposed to enhance accuracy and robustness against noise and outliers in data sets in the data-driven computing. In this approach, for a given state obtained by the physical simulation, the corresponding optimum experimental solution is sought by projecting the state onto the associated local convex manifold reconstructed based on the nearest experimental data. This learning process of local data structure is less sensitivity to noisy data and consequently yields better accuracy. A penalty relaxation is also introduced to recast the local learning solver in the context of non-negative least squares that can be solved effectively. The reproducing kernel approximation with stabilized nodal integration are employed for the solution of the physical manifold to allow reduced stress-strain data at the discrete points for enhanced effectiveness in the LCDD learning solver. Due to the inherent manifold learning properties, LCDD performs well for high-dimensional data sets that are relatively sparse in real-world engineering applications. Numerical tests demonstrated that LCDD enhances nearly one order of accuracy compared to the standard distance-minimization data-driven scheme when dealing with noisy database, and a linear exactness is achieved when local stress-strain relation is linear.

show abstract

Section: Solving Non-negative Least Squaresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A physics-constrained data-driven approach based on locally convex reconstruction for noisy database

Chen

2020

Computer Methods in Applied Mechanics and Engineering

View full text Add to dashboard Cite

show abstract

“…Due to this special ability of ML algorithms to be input agnostic, i.e., the ability to automatically evaluate features from input data, they have found utility in a wide variety of applications including recommendation systems 16 and self-driving cars. 17 These approaches are slowly gaining popularity in physics and engineered systems, [18][19][20] where modern sensor and computational developments have paved the way for structured data generation. 21,22 Here, we utilize the versatility of CNNs to map the active layer morphology of thin film OPVs to a performance metric, which is the short-circuit current J sc .…”

Section: Introductionmentioning

confidence: 99%

Interpretable deep learning for guided microstructure-property explorations in photovoltaics

et al. 2019

Self Cite

View full text Add to dashboard Cite

The microstructure determines the photovoltaic performance of a thin film organic semiconductor film. The relationship between microstructure and performance is usually highly non-linear and expensive to evaluate, thus making microstructure optimization challenging. Here, we show a data-driven approach for mapping the microstructure to photovoltaic performance using deep convolutional neural networks. We characterize this approach in terms of two critical metrics, its generalizability (has it learnt a reasonable map?), and its intepretability (can it produce meaningful microstructure characteristics that influence its prediction?). A surrogate model that exhibits these two features of generalizability and intepretability is particularly useful for subsequent design exploration. We illustrate this by using the surrogate model for both manual exploration (that verifies known domain insight) as well as automated microstructure optimization. We envision such approaches to be widely applicable to a wide variety of microstructure-sensitive design problems.

show abstract

“…In fluid mechanics, neural networks have been reported recently [3][4][5] as tools that can help computational fluid mechanics simulations by mapping the estimates of low-resolution simulations to those with higher fidelity. In microfluidics, neural networks have been used to estimate various quantities for different applications [6][7][8][9] . Mahdi and Daoud 10 used neural networks to predict the size of the droplets in an emulsion while Khor et al 11 .…”

mentioning

confidence: 99%

Learning from droplet flows in microfluidic channels using deep neural networks

Hadikhani

Borhani

Hashemi

2019

Sci Rep

View full text Add to dashboard Cite

A non-intrusive method is presented for measuring different fluidic properties in a microfluidic chip by optically monitoring the flow of droplets. A neural network is used to extract the desired information from the images of the droplets. We demonstrate the method in two applications: measurement of the concentration of each component of a water/alcohol mixture, and measurement of the flow rate of the same mixture. A large number of droplet images are recorded and used to train deep neural networks (DNN) to predict the flow rate or the concentration. It is shown that this method can be used to quantify the concentrations of each component with a 0.5% accuracy and the flow rate with a resolution of 0.05 ml/h. The proposed method can in principle be used to measure other properties of the fluid such as surface tension and viscosity.Machine learning is a framework that learns from the data without being programmed. Deep leaning is a specific approach of machine learning 1 which has emerged in recent years as a powerful technique for a broad range of applications. Deep learning consists of several representation layers where each layer is obtained by the non-linear transformation of the previous layer 2 . Deep neural networks use the combination of these transformations to learn complex functions. In fluid mechanics, neural networks have been reported recently 3-5 as tools that can help computational fluid mechanics simulations by mapping the estimates of low-resolution simulations to those with higher fidelity. In microfluidics, neural networks have been used to estimate various quantities for different applications 6-9 . Mahdi and Daoud 10 used neural networks to predict the size of the droplets in an emulsion while Khor et al. 11 . used them to predict the stability of droplets in an emulsion. In our work, we employ neural networks to estimate fluid and flow parameters by observing the droplet formation process in a passive microfluidic chip. We extract this information by monitoring the flow of droplets with an optical microscope and training a neural network to obtain useful information from the recorded images. In one experiment, we trained a network to accurately measure the flow velocity at the inlet of the channel from droplet images. In a separate demonstration, a network was trained to identify the concentration of isopropanol in water in the droplet-forming solution. Our experiments demonstrate that DNNs can capture the complex nonlinear phenomena that result in the flow patterns and droplet shapes.The flow rate is a parameter that affects droplet formation 12 and can change the size, generation frequency, and pattern of droplets 13 . Similarly, the dilution ratio of isopropanol (IPA) in water affects the generation frequency and flow pattern of the droplets. Such effects can also be recognized by the DNN to measure the concentration. There are alternative methods for measuring the flow rate 14-17 or the dilution ratio 18-21 in a microfluidic chip. The two prominent methods for on-chip flow measurement...

show abstract

Deep Learning for Flow Sculpting: Insights into Efficient Learning using Scientific Simulation Data

Cited by 57 publications

References 36 publications

A physics-constrained data-driven approach based on locally convex reconstruction for noisy database

A physics-constrained data-driven approach based on locally convex reconstruction for noisy database

Interpretable deep learning for guided microstructure-property explorations in photovoltaics

Learning from droplet flows in microfluidic channels using deep neural networks

Contact Info

Product

Resources

About