東華大學圖書館 |

Data-Driven Identification and Control of Turbulent Structures Using Deep Neural Networks.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Data-Driven Identification and Control of Turbulent Structures Using Deep Neural Networks./
作者:	Jagodinski, Eric.
面頁冊數:	1 online resource (135 pages)
附註:	Source: Dissertations Abstracts International, Volume: 84-07, Section: B.
Contained By:	Dissertations Abstracts International84-07B.
標題:	Fluid mechanics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30244657click for full text (PQDT)
ISBN:	9798368413372

Data-Driven Identification and Control of Turbulent Structures Using Deep Neural Networks.
Jagodinski, Eric.

Data-Driven Identification and Control of Turbulent Structures Using Deep Neural Networks. - 1 online resource (135 pages)

Source: Dissertations Abstracts International, Volume: 84-07, Section: B.

Thesis (Ph.D.)--Florida Atlantic University, 2022.

Includes bibliographical references

Wall-bounded turbulent flows are pervasive in numerous physics and engineering applications. Such flows tend to have a strong impact on the design of ships, airplanes and rockets, industrial chemical mixing, wind and hydrokinetic energy, utility infrastructure and innumerable other fields. Understanding and controlling wall-bounded turbulence has been a long-pursued endeavor yielding plentiful scientific and engineering discoveries, but there is much that remains unexplained from a fundamental viewpoint. One unexplained phenomenon is the formation and impact of coherent structures like the ejections of slow near-wall fluid into faster moving flow which have been shown to correlate with increases in friction drag. This thesis focuses on recognizing and regulating organized structures within wall-bounded turbulent flows using a variety of machine learning techniques to overcome the nonlinear nature of this phenomenon. Deep Learning has provided new avenues of analyzing large amounts of data by applying techniques modeled after biological neurons. These techniques allow for the discovery of nonlinear relationships in massive, complex systems like the data found frequently in fluid dynamics simulation. Using a neural network architecture called Convolutional Neural Networks that specializes in uncovering spatial relationships, a network was trained to estimate the relative intensity of ejection structures within turbulent flow simulation without any a priori knowledge of the underlying flow dynamics. To explore the underlying physics that the trained network might reveal, an interpretation technique called Gradient-based Class Activation Mapping was modified to identify salient regions in the flow field which most influenced the trained network to make an accurate estimation of these organized structures. Using various statistical techniques, these salient regions were found to have a high correlation to ejection structures, and to high positive kinetic energy production, low negative production, and low energy dissipation regions within the flow. Additionally, these techniques present a general framework for identifying nonlinear causal structures in general three-dimensional data in any scientific domain where the underlying physics may be unknown. While identification is an important goal, control and regulation of these coherent structures has countless scientific and practical applications, like modulating drag on ships or efficiency in utility infrastructure. However, control of turbulent flows has been a challenging problem because of the inherent nonlinear evolution of the coherent structures. Deep Reinforcement Learning may help overcome these obstacles by leveraging artificial neural networks to devise an effective control scheme, even without a priori knowledge of the underlying dynamics. The proposed approach is to utilize Deep Reinforcement Learning for control of blowing-suction actuators within a wall-bounded turbulent simulation with the goal of friction drag reduction. Preliminary results aiming to reduce the intensity of ejection structures with a simple representation of the flow field achieved a notable reduction in the intensity. A framework to expand upon this control scheme has been developed in which two-dimensional Convolutional Neural Networks were used to reveal spatial patterns, and Long Short-Term Memory was used to reveal temporal relationships in the flow history. This framework can be used to uncover causal relationships between the autonomously determined actuation policy and the subsequent influence on the turbulent flow state.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798368413372Subjects--Topical Terms:

528155
Fluid mechanics.
Subjects--Index Terms:

Coherent structuresIndex Terms--Genre/Form:

542853
Electronic books.

Data-Driven Identification and Control of Turbulent Structures Using Deep Neural Networks.
LDR:05036nmm a2200397K 4500 001 2354982
005 20230505090519.5
006 m o d
007 cr mn ---uuuuu
008 241011s2022 xx obm 000 0 eng d
020 $a 9798368413372
035 $a (MiAaPQ)AAI30244657
035 $a AAI30244657
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Jagodinski, Eric. $3 3695369
245 1 0 $a Data-Driven Identification and Control of Turbulent Structures Using Deep Neural Networks.
264 0 $c 2022
300 $a 1 online resource (135 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 84-07, Section: B.
500 $a Advisor: Verma, Siddhartha.
502 $a Thesis (Ph.D.)--Florida Atlantic University, 2022.
504 $a Includes bibliographical references
520 $a Wall-bounded turbulent flows are pervasive in numerous physics and engineering applications. Such flows tend to have a strong impact on the design of ships, airplanes and rockets, industrial chemical mixing, wind and hydrokinetic energy, utility infrastructure and innumerable other fields. Understanding and controlling wall-bounded turbulence has been a long-pursued endeavor yielding plentiful scientific and engineering discoveries, but there is much that remains unexplained from a fundamental viewpoint. One unexplained phenomenon is the formation and impact of coherent structures like the ejections of slow near-wall fluid into faster moving flow which have been shown to correlate with increases in friction drag. This thesis focuses on recognizing and regulating organized structures within wall-bounded turbulent flows using a variety of machine learning techniques to overcome the nonlinear nature of this phenomenon. Deep Learning has provided new avenues of analyzing large amounts of data by applying techniques modeled after biological neurons. These techniques allow for the discovery of nonlinear relationships in massive, complex systems like the data found frequently in fluid dynamics simulation. Using a neural network architecture called Convolutional Neural Networks that specializes in uncovering spatial relationships, a network was trained to estimate the relative intensity of ejection structures within turbulent flow simulation without any a priori knowledge of the underlying flow dynamics. To explore the underlying physics that the trained network might reveal, an interpretation technique called Gradient-based Class Activation Mapping was modified to identify salient regions in the flow field which most influenced the trained network to make an accurate estimation of these organized structures. Using various statistical techniques, these salient regions were found to have a high correlation to ejection structures, and to high positive kinetic energy production, low negative production, and low energy dissipation regions within the flow. Additionally, these techniques present a general framework for identifying nonlinear causal structures in general three-dimensional data in any scientific domain where the underlying physics may be unknown. While identification is an important goal, control and regulation of these coherent structures has countless scientific and practical applications, like modulating drag on ships or efficiency in utility infrastructure. However, control of turbulent flows has been a challenging problem because of the inherent nonlinear evolution of the coherent structures. Deep Reinforcement Learning may help overcome these obstacles by leveraging artificial neural networks to devise an effective control scheme, even without a priori knowledge of the underlying dynamics. The proposed approach is to utilize Deep Reinforcement Learning for control of blowing-suction actuators within a wall-bounded turbulent simulation with the goal of friction drag reduction. Preliminary results aiming to reduce the intensity of ejection structures with a simple representation of the flow field achieved a notable reduction in the intensity. A framework to expand upon this control scheme has been developed in which two-dimensional Convolutional Neural Networks were used to reveal spatial patterns, and Long Short-Term Memory was used to reveal temporal relationships in the flow history. This framework can be used to uncover causal relationships between the autonomously determined actuation policy and the subsequent influence on the turbulent flow state.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Fluid mechanics. $3 528155
650 4 $a Ocean engineering. $3 660731
653 $a Coherent structures
653 $a Convolutional neural networks
653 $a Reinforcement learning
653 $a Turbulence
653 $a Wall-bounded flow
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0204
690 $a 0800
690 $a 0547
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a Florida Atlantic University. $b Ocean Engineering. $3 3173823
773 0 $t Dissertations Abstracts International $g 84-07B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30244657 $z click for full text (PQDT)