scholarly journals Reduced order models and machine learning in analysis and optimum design of structures

2019 ◽  
Author(s):  
Νικόλαος Καλλιώρας
Keyword(s):  
Machine Learning ◽  
Optimum Design ◽  
Isotropic Material ◽  
Harmony Search ◽  
Reduced Order Models ◽  
Generative Design ◽  
Reduced Order ◽  
Boltzmann Machines ◽  
Deep Boltzmann Machines

Ο βέλτιστος σχεδιασμός κατασκευών απασχολεί τον άνθρωπο από την εποχή της πρώτης κατασκευής. Το ενδιαφέρον αυτό αυξήθηκε με την αύξηση του μεγέθους και της πολυπλοκότητας των κατασκευών. Η ανάλυση των κατασκευών και ο υπολογιστικός της φόρτος εξαρτάται από το μέγεθος και την πολυπλοκότητα των κατασκευών. Η χρήση προσεγγιστικών μεθόδων αυξήθηκε λόγω της αύξησης του υπολογιστικού φόρτου που απαιτούν οι ακριβείς μέθοδοι. Στην παρούσα διδακτορική διατριβή παρουσιάζεται συνεισφορά στους μεταευρετικούς αλγόριθμους, τα μοντέλα μειωμένης τάξης, τους αλγόριθμους μηχανικής μάθησης, την βελτιστοποίησης τοπολογίας και το generative design. Συγκεκριμένα, παρουσιάζεται ένας νέος μεταευρετικός αλγόριθμος που δημιουργήθηκε στα πλαίσια της διδακτορικής διατριβής αλλά και μια βελτιωμένη έκδοση του αλγόριθμου Harmony Search που αρχικά έχει προταθεί από τον Καθηγητή κ. Zong Woo Geem. Επίσης παρουσιάζονται τέσσερις διαφορετικές μεθοδολογίες συνδυασμού αλγορίθμων βαθιών νευρωνικών δικτύων και του αλγόριθμου βελτιστοποίησης τοπολογίας Solid Isotropic Material with Penalization (SIMP). Η πρώτη μεθοδολογία, DL-TOP, χρησιμοποιεί Deep Boltzmann Machines για να προβλέψει την τελική πυκνότητα των πεπερασμένων στοιχείων στη διαδικασία της βελτιστοποίησης τοπολογίας μελετώντας πλήθος αρχικών τιμών τους. Η μεθοδολογία DL-SCALE χρησιμοποιεί Deep Boltzmann Machines σε μια λογική Model Upgrading για να επιταχύνει την βελτιστοποίηση τοπολογίας μέσω μοντέλων μειωμένης τάξης και πύκνωσης του πλέγματος των πεπερασμένων στοιχείων. Η Τρίτη μεθοδολογία, DLRM-TOP, χρησιμοποιεί βαθιά νευρωνικά δίκτυα για να προβλέψει την τελική πυκνότητα κάθε πεπερασμένου στοιχείου βάση πληροφορίας από την τελική κατάσταση των μοντέλων μειωμένης τάξης. Η τέταρτη μεθοδολογία, CN-TOP, χρησιμοποιεί βαθιά συνελικτικά νευρωνικά δίκτυα που βελτιώνουν την ποιότητα εικόνας για την επιτάχυνση της βελτιστοποίησης τοπολογίας. Τέλος παρουσιάζεται μια λογική συνδυασμού βαθιών νευρωνικών δικτύων και SIMP για την αυτόματη παραγωγή πληθώρας σχεδιασμών χωρίς την παρέμβαση του χρήστη σε μια λογική generative design. Το μόνο που απαιτείται από τον χρήστη είναι ο ορισμός του προβλήματος. Τα αποτελέσματα των παραπάνω μεθοδολογιών (επιτάχυνση διαδικασιών και παραγωγή σχεδιασμών) που παρουσιάζονται στην διδακτορική διατριβή κάνουν φανερό πως η μηχανική μάθηση και οι σύγχρονες τεχνικές της μπορούν να αποτελέσουν σημαντικά εργαλεία στην επιστήμη του πολιτικού μηχανικού.

scholarly journals Non-intrusive nonlinear model reduction via machine learning approximations to low-dimensional operators

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Zhe Bai ◽  
Liqian Peng
Keyword(s):  
Machine Learning ◽  
Nonlinear Dynamical Systems ◽  
Reduced Order Models ◽  
Simulation Code ◽  
Nonlinear Dynamical ◽  
Reduced Order ◽  
Nonlinear Model Reduction ◽  
Regression Techniques ◽  
Low Dimensional ◽  
Modern Machine

AbstractAlthough projection-based reduced-order models (ROMs) for parameterized nonlinear dynamical systems have demonstrated exciting results across a range of applications, their broad adoption has been limited by their intrusivity: implementing such a reduced-order model typically requires significant modifications to the underlying simulation code. To address this, we propose a method that enables traditionally intrusive reduced-order models to be accurately approximated in a non-intrusive manner. Specifically, the approach approximates the low-dimensional operators associated with projection-based reduced-order models (ROMs) using modern machine-learning regression techniques. The only requirement of the simulation code is the ability to export the velocity given the state and parameters; this functionality is used to train the approximated low-dimensional operators. In addition to enabling nonintrusivity, we demonstrate that the approach also leads to very low computational complexity, achieving up to $$10^3{\times }$$ 10 3 × in run time. We demonstrate the effectiveness of the proposed technique on two types of PDEs. The domain of applications include both parabolic and hyperbolic PDEs, regardless of the dimension of full-order models (FOMs).

scholarly journals Machine learning augmented reduced-order models for FFR-prediction

2021 ◽  
Vol 384 ◽  
pp. 113892
Author(s):  
Fredrik E. Fossan ◽  
Lucas O. Müller ◽  
Jacob Sturdy ◽  
Anders T. Bråten ◽  
Arve Jørgensen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Reduced Order Models ◽  
Reduced Order

scholarly journals A machine learning framework for rapid forecasting and history matching in unconventional reservoirs

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Shriram Srinivasan ◽  
Daniel O’Malley ◽  
Maruti K. Mudunuru ◽  
Matthew R. Sweeney ◽  
Jeffrey D. Hyman ◽  
...  
Keyword(s):  
Machine Learning ◽  
Real Time ◽  
Transfer Learning ◽  
History Matching ◽  
Time History ◽  
Unconventional Reservoirs ◽  
Production Data ◽  
High Fidelity ◽  
Reduced Order Models ◽  
Reduced Order

AbstractWe present a novel workflow for forecasting production in unconventional reservoirs using reduced-order models and machine-learning. Our physics-informed machine-learning workflow addresses the challenges to real-time reservoir management in unconventionals, namely the lack of data (i.e., the time-frame for which the wells have been producing), and the significant computational expense of high-fidelity modeling. We do this by applying the machine-learning paradigm of transfer learning, where we combine fast, but less accurate reduced-order models with slow, but accurate high-fidelity models. We use the Patzek model (Proc Natl Acad Sci 11:19731–19736, 10.1073/pnas.1313380110, 2013) as the reduced-order model to generate synthetic production data and supplement this data with synthetic production data obtained from high-fidelity discrete fracture network simulations of the site of interest. Our results demonstrate that training with low-fidelity models is not sufficient for accurate forecasting, but transfer learning is able to augment the knowledge and perform well once trained with the small set of results from the high-fidelity model. Such a physics-informed machine-learning (PIML) workflow, grounded in physics, is a viable candidate for real-time history matching and production forecasting in a fractured shale gas reservoir.

scholarly journals Li-ion Battery Aging with Hybrid Physics-Informed Neural Networks and Fleet-wide Data

2021 ◽  
Vol 13 (1) ◽  
Author(s):  
Renato G. Nascimento ◽  
Matteo Corbetta ◽  
Chetan S. Kulkarni ◽  
Felipe A. C. Viana
Keyword(s):  
Machine Learning ◽  
Computational Cost ◽  
Lithium Ion ◽  
Unmanned Aircraft ◽  
Model Parameters ◽  
Reduced Order Models ◽  
Reduced Order ◽  
Model Form ◽  
Battery Capacity ◽  
Battery Discharge

Lithium-ion batteries are commonly used to power electric unmanned aircraft vehicles (UAVs).Therefore, the ability to model both the state of charge as well as battery health is very important for reliable and affordable operation of UAV fleets.Even though models based on first principles are accurate and trustworthy, the complex electro-chemistry that governs battery discharge and aging makes it hard to build and use such models for in-time monitoring of battery conditions.Moreover, the careful tuning or estimation of high-fidelity model parameters hampers the straightforward deployment in the field.Alternatively, reduced order models have the advantage of capturing the overall behavior of battery discharge. Reduced-order principle-based models are built by carefully simplifying the physics/chemistry such that computational cost is dramatically reduced while the overall behavior of the system is still captured.These simplifications also lead to a number of parameters to be estimated based on data as well as residual discrepancy (model-form uncertainty).This approach can lead to a number of parameters to be estimated based on data as well as residual model-form uncertainty; a property shared with machine learning models. The latter are solely built on the basis of data, and can still capture unexpected nonlinearities.The drawback is that traditional machine learning tends to require large number of data points hard to retrieve in many scientific and engineering fields like, for example, the field of battery discharge and degradation prediction. In this paper, we will present a hybrid modeling approach for tracking and forecasting battery aging based on ``as-used'' conditions.Our approach directly implements a reduced-order model based on Nerst and Butler-Volmer equations within a deep neural network framework.While most of the input-output relationship is captured by reduced-order models, the data-driven kernels reduce the gap between predictions and observations.The hybrid model estimates the overall battery discharge, and a multilayer perceptron models the battery internal voltage.Battery aging is characterized by time-dependent internal resistance and the amount of available Li-ions.We address the difficult issue of building and updating the aging model by reducing the need for reference discharge cycles.This is beneficial to operators, since it reduces the need of taking the batteries out of commission.We compensate for lack of reference discharge cycles by using a probabilistic model that leverages previously available fleet-wide information. We validate our approach using data publicly available through the NASA Prognostics Center of Excellence website.Results showed that our hybrid battery prognosis model can be successfully calibrated, even with a limited number of observations.Moreover, the model can help optimizing battery operation by offering long-term forecast of battery capacity.

scholarly journals Non-intrusive Nonlinear Model Reduction via Machine Learning Approximations to Low-dimensional Operators

2021 ◽  
Author(s):  
Zhe Bai ◽  
Liqian Peng
Keyword(s):  
Machine Learning ◽  
Nonlinear Dynamical Systems ◽  
Reduced Order Models ◽  
Simulation Code ◽  
Nonlinear Dynamical ◽  
Reduced Order ◽  
Nonlinear Model Reduction ◽  
Regression Techniques ◽  
Low Dimensional ◽  
Modern Machine

Abstract Although projection-based reduced-order models (ROMs) for parameterized nonlinear dynamical systems have demonstrated exciting results across a range of applications, their broad adoption has been limited by their intrusivity: implementing such a reduced-order model typically requires significant modifications to the underlying simulation code. To address this, we propose a method that enables traditionally intrusive reduced-order models to be accurately approximated in a non-intrusive manner. Specifically, the approach approximates the low-dimensional operators associated with projection-based reduced-order models (ROMs) using modern machine-learning regression techniques. The only requirement of the simulation code is the ability to export the velocity given the state and parameters; this functionality is used to train the approximated low-dimensional operators. In addition to enabling nonintrusivity, we demonstrate that the approach also leads to very low computational complexity, achieving up to $10^3\times$ in run time. We demonstrate the effectiveness of the proposed technique on two types of PDEs.

Machine learning for predictive coal combustion CFD simulations—From detailed kinetics to HDMR Reduced-Order models

Fuel ◽  
2020 ◽  
Vol 274 ◽  
pp. 117720
Author(s):  
P. Debiagi ◽  
H. Nicolai ◽  
W. Han ◽  
J. Janicka ◽  
C. Hasse
Keyword(s):  
Machine Learning ◽  
Coal Combustion ◽  
Reduced Order Models ◽  
Cfd Simulations ◽  
Reduced Order ◽  
Detailed Kinetics
Sign in / Sign up

Export Citation Format

Share Document