FAQFAQ   SuchenSuchen     BenutzergruppenBenutzergruppen   RegistrierenRegistrieren 
 ProfilProfil   Einloggen, um private Nachrichten zu lesenEinloggen, um private Nachrichten zu lesen   LoginLogin 

Sense and Attack to Turbulence with Deep Learning

 
Neues Thema eröffnen   Neue Antwort erstellen     Foren-Übersicht -> Tipps zum Bauen und Werfen
Vorheriges Thema anzeigen :: Nächstes Thema anzeigen  
Autor Nachricht
Mustafa Umut Sarac
Wurfkreis-Choleriker


Anmeldungsdatum: 19.09.2018
Beiträge: 131
Wohnort: Istanbul - Turkey

BeitragVerfasst am: 14.04.2019, 18:02    Titel: Sense and Attack to Turbulence with Deep Learning Antworten mit Zitat

Piezo Sensor Films are flexible and ultralight. They can be used to cover a boomerang without adding too much weight.

I will write an interesting subject and I dont know how to handle in real world but idea is good.

Think, you covered the boomerang wings bottom and top with piezo electric sensor film.

Film will sense the fluid flow pressure difference and pressure difference indicates where the fluid flow flows.

And there is the turbulence.

At back side of the camber , there is turbulence and complex pressure .
differences on the surface.

Piezo sensor produces the electric when the force changes.

And the piezo produces vibration when it is electrified.

We need a algorithm which manages the double nature of input and output of the film.

When the turbulence occurs , it depends to elementary electronics knowledge to sense turbulance on surface nature and map it and than counter attack it with surface vibration.

I think we need to investigate a newer way called deep learning where albatross flight with air drones been imitated with the result of digital signal processor and neural networks training.

The purpose of machine learning is to enter sensor input and surface vibration input and the gyro and gps input to train a new computer program output which will be used at new products.

This can be done.

PS. Additionally , above sensing and counter attack surface vibration might be needed not to use entire surface piezo cover but it might be done with pixels.

Pixels and their circuit can be difficult but it can be simplified , resolution can be lowered and circuit can be printed to under the sensor, on to wing with inkjet printer or laser printed on to carrier and can be transferred on to wing with release agent.

Fourier transform can be important to understand the flow map and counter attack it.

All computations can be done on boomerang , or boomerang rc or rc cell phone computer.

Experiments can be done in a water tank or can be done with covering the surface with a wet powder where it goes dried in few minutes and reveals the surface flow.

Deep learning subject had been told at albatross flight thread and can be read.


Mustafa Umut Sarac
Istanbul
Nach oben
Benutzer-Profile anzeigen Private Nachricht senden AIM-Name
Mustafa Umut Sarac
Wurfkreis-Choleriker


Anmeldungsdatum: 19.09.2018
Beiträge: 131
Wohnort: Istanbul - Turkey

BeitragVerfasst am: 15.04.2019, 01:18    Titel: Antworten mit Zitat

Deep Dynamical Modeling and Control of Unsteady Fluid Flows

Jeremy Morton∗
jmorton2@stanford.edu

Freddie D. Witherden∗†
fdw@stanford.edu

Antony Jameson †
antony.jameson@tamu.edu

Mykel J. Kochenderfer∗
mykel@stanford.edu

Abstract

The design of flow control systems remains a challenge due to the nonlinear nature of the equations that govern fluid flow. However, recent advances in computational fluid dynamics (CFD) have enabled the simulation of complex fluid flows with high accuracy, opening the possibility of using learning-based approaches to facilitate controller design. We present a method for learning the forced and unforced dy-namics of airflow over a cylinder directly from CFD data. The proposed approach,grounded in Koopman theory, is shown to produce stable dynamical models that can predict the time evolution of the cylinder system over extended time horizons.
Finally, by performing model predictive control with the learned dynamical models,we are able to find a straightforward, interpretable control law for suppressing vortex shedding in the wake of the cylinder.


PAPER IS HEREBELOW :

https://drive.google.com/file/d/189RggWjPiLVTZyejaQTIia09p983CfCj/view?usp=sharing
Nach oben
Benutzer-Profile anzeigen Private Nachricht senden AIM-Name
Beiträge der letzten Zeit anzeigen:   
Neues Thema eröffnen   Neue Antwort erstellen     Foren-Übersicht -> Tipps zum Bauen und Werfen Alle Zeiten sind GMT + 1 Stunde
Seite 1 von 1

 
Gehe zu:  
Du kannst keine Beiträge in dieses Forum schreiben.
Du kannst auf Beiträge in diesem Forum nicht antworten.
Du kannst deine Beiträge in diesem Forum nicht bearbeiten.
Du kannst deine Beiträge in diesem Forum nicht löschen.
Du kannst an Umfragen in diesem Forum nicht mitmachen.


Powered by phpBB © 2001, 2002 phpBB Group