1,319 Hits in 4.2 sec

Identifying Grey-box Thermal Models with Bayesian Neural Networks [article]

Md Monir Hossain, Tianyu Zhang, Omid Ardakanian
2020 arXiv   pre-print
We show that Bayesian neural networks can be used to estimate parameters of a grey-box thermal model if sufficient training data is available, and this model outperforms several black-box models in terms  ...  Leveraging real data from 8,884 homes equipped with smart thermostats, we discuss how the prior knowledge about the model parameters can be utilized to quickly build an accurate thermal model for another  ...  CONCLUSION This paper studies the problem of identifying grey-box thermal models (RC-network models) with Bayesian neural networks leveraging time series data generated by smart thermostats and metadata  ... 
arXiv:2009.05889v1 fatcat:urh6qrhcvjf3pi52n2drsfvzue

Thermal error model of linear motor feed system based on Bayesian neural network (June 2021)

Shengsen Liu, Zeqing Yang, Qiang Wei, Yingshu Chen, Libing Liu
2021 IEEE Access  
by combining the NARMAX time series model with the neural network model.  ...  Bayesian neural network algorithm steps. FIGURE 3 . 3 Bayesian neural network Structure. FIGURE 4 . 4 Grey Relation Analysis process. FIGURE 5 . 5 Location of temperature measuring point.  ... 
doi:10.1109/access.2021.3103972 fatcat:vbm7jnf27nhbtl7fxjmdvqawfa

Embedding Physics Domain Knowledge into a Bayesian Network Enables Layer-by-Layer Process Innovation for Photovoltaics [article]

Zekun Ren, Felipe Oviedo, Muang Thway, Siyu I.P. Tian, Yue Wang, Hansong Xue, Jose Dario Perea, Mariya Layurova, Thomas Heumueller, Erik Birgersson, Armin Aberle, Christoph J. Brabec, Rolf Stangl (+4 others)
2019 arXiv   pre-print
Herein, we demonstrate that embedding physics domain knowledge into a Bayesian network enables an optimization approach that identifies the root cause(s) of underperformance with layer by-layer resolution  ...  (e.g., cell efficiency), using a Bayesian inference framework with an autoencoder-based surrogate device-physics model that is 100x faster than numerical solvers.  ...  with black-box regression and the Bayesian network with the Arrhenius parametrization.  ... 
arXiv:1907.10995v2 fatcat:ogx3mm2pubagbhomh6koeb43s4

Mechanistic Grey-Box Modeling of a Packed-Bed Regenerator for Industrial Applications

Verena Halmschlager, Stefan Müllner, René Hofmann
2021 Energies  
This work examines the development of a mechanistic grey-box model for a sensible thermal energy storage, a packed-bed regenerator.  ...  While the existing data-driven model lacks robustness and the purely physical model lacks accuracy, the hybrid grey-box models do not show significant disadvantages.  ...  Additionally, the presented grey-box model is a major addition to the authors' previous publication [18] , where the PBR was modeled with a primarily data-driven modeling approach using Neural Networks  ... 
doi:10.3390/en14113174 fatcat:5diiwsobsrfjjhydx5g2mff7vy

A Review on Battery Modelling Techniques

S. Tamilselvi, S. Gunasundari, N. Karuppiah, Abdul Razak RK, S. Madhusudan, Vikas Madhav Nagarajan, T. Sathish, Mohammed Zubair M. Shamim, C. Ahamed Saleel, Asif Afzal
2021 Sustainability  
The approaches, advantages and disadvantages of black box and grey box type battery modelling are analysed.  ...  This work details the charging and discharging characteristics using the black box and grey box techniques for modelling the lithium-ion battery.  ...  black box model (e.g., artificial neural network (ANN) model) [11] .  ... 
doi:10.3390/su131810042 fatcat:bhnzlfivuvgcvkdh3zixihsroe

Global approach test improvement using a neural network model identification to characterise solar combisystem performances

Antoine Leconte, Gilbert Achard, Philippe Papillon
2012 Solar Energy  
The proposed model to identify is a "grey box" model, mixing a "White Box" model composed of known physical equations and a "Black Box" model, which is an Artificial Neural Network (ANN).  ...  Compared to those annual results, "Grey Box" SCS models trained from a twelve days sequence are able to predict energy consumption with a good accuracy for 27 different environments.  ...  He has tried two kind of model: a grey box model (based on known physical model and parameters identification) and a Dynamic Adaptative Neural Network.  ... 
doi:10.1016/j.solener.2012.04.003 fatcat:s4ytn6wlwzf4pkeq5ymbftceke

Past, present and future mathematical models for buildings

Xiaoshu Lu, Derek Clements-Croome, Martti Viljanen
2009 Intelligent Buildings International  
In contrast with the early modelling techniques, model approaches adopted in neural networks, expert systems, fuzzy logic and genetic models provide a promising method to accommodate these complications  ...  The advantages and limitations of the applied mathematical models are identified and the models are classified in terms of their application range and goal.  ...  Box 4 illustrates a simple example of neural networks with three inputs, one hidden layer of neurons containing four nodes and one output.  ... 
doi:10.3763/inbi.2009.0009 fatcat:xgmboh5uazcrpcnxwczxrdog3a

Leveraging artificial intelligence for pandemic preparedness and response: a scoping review to identify key use cases

Ania Syrowatka, Masha Kuznetsova, Ava Alsubai, Adam L. Beckman, Paul A. Bain, Kelly Jean Thomas Craig, Jianying Hu, Gretchen Purcell Jackson, Kyu Rhee, David W. Bates
2021 npj Digital Medicine  
The objective of this scoping review was to identify the key use cases for involving AI for pandemic preparedness and response from the peer-reviewed, preprint, and grey literature.  ...  The search identified 1167 articles that reported on traditional modeling approaches, which highlighted additional areas where ML could be leveraged for improving the accuracy of estimations or projections  ...  Zoe Co for assistance with data abstraction. A.S. is supported by a Fellowship Award from the Canadian Institutes of Health Research.  ... 
doi:10.1038/s41746-021-00459-8 pmid:34112939 fatcat:itib6lu3cbdwvlwbkejpm3fusu

Suitability Analysis Of Modeling And Assessment Approaches In Energy Efficiency In Buildings

Christos Koulamas, Athanasios Kalogeras, Rosalia Pacheco-Torres, Jorge Casillas, Luca Ferrarini
2018 Zenodo  
This work presents a comprehensive analysis of the mostimportant results today, along with their various classification and assessment approaches for modelingenergy building consumption.  ...  Grey box models represent a balance between the good generalization capability of white box models and high accuracy of black box models.  ...  to model predictive control, although most of them are better classified as grey models, as in [23] and [24] . 15 As mentioned before, white box models require a thermal engineering expert to model  ... 
doi:10.5281/zenodo.1211106 fatcat:f54qzt5tsrc47cnhqfcb7evz6i

Thermal Infrared Face Recognition

Vincent A Weidlich
2021 Cureus  
Previous research regarding temperature variations, mathematical formulas, wave types, and methods in thermal infrared face recognition is reviewed.  ...  Thermal infrared face recognition helps recognize faces that are not able to be recognized in visible light and can additionally recognize facial blood vessel structure.  ...  Constructive Auto-associative Neural Network (CANet), which is a novel neural network inspired by neural biology, has been proposed.  ... 
doi:10.7759/cureus.13736 pmid:33842114 pmcid:PMC8021210 fatcat:rorchuvponeltneoznkymdjjte

Analysing uncertainty in parameter estimation and prediction for grey-box building thermal behaviour models

O.M. Brastein, A. Ghaderi, C.F. Pfeiffer, N.-O. Skeie
2020 Energy and Buildings  
Simplified thermal network models, often expressed as grey-box Resistor-Capacitor circuit analogue models, have been successfully applied in the prediction setting.  ...  If a Bayesian framework is used, treating the parameters as random variables with a probability distribution in the parameter space, projections of the posterior distribution can be studied by using the  ...  A reasonable compromise between the physics-based whitebox and the data-driven black-box models is the use of grey-box thermal network models [3, 4, 7, 9, 5] .  ... 
doi:10.1016/j.enbuild.2020.110236 fatcat:qxtqtlsmqvcfdnm2tptjgb2eka

Review of data-driven energy modelling techniques for building retrofit

C. Deb, A. Schlueter
2021 Renewable & Sustainable Energy Reviews  
Differentiating between 1) bottom-up approaches that can be divided into white-, grey-and black-box modelling, and 2) top-down approaches that utilize analytical methods of clustering and regression, this  ...  This paper reviews current retrofit methodologies with a focus on the contrast between data-driven approaches that utilize measured building data, acquired through either 1) on-site sensor deployment or  ...  Grey-box models (resistance-capacitance analogue models) A grey-box model combines a partial theoretical structure with measured data to complete the model.  ... 
doi:10.1016/j.rser.2021.110990 fatcat:3k7qpvbhlvevrljsvpo7aby7te

Numerical and experimental results of a novel and generic methodology for energy performance evaluation of thermal systems using renewable energies

Amine Lazrak, Antoine Leconte, David Chèze, Gilles Fraisse, Philippe Papillon, Bernard Souyri
2015 Applied Energy  
the "grey box model", combines simplified physical equations (collectors, building, storage, 69 auxiliary = "white box") and an artificial neural network ANN ("black box").  ...  Perspectives and methodology limitations are discussed as well. 19 KEY-WORDS 20 Thermal systems, Renewable energy, Performance estimation, Dynamic modelling, Artificial neural networks, System testing.  ...  of a neural network MLP with one hidden layer (right)236 3.3.2.  ... 
doi:10.1016/j.apenergy.2015.08.049 fatcat:yfjtisq25bg3vivwel2esupg5e

Forecasting Thermoacoustic Instabilities in Liquid Propellant Rocket Engines Using Multimodal Bayesian Deep Learning [article]

Ushnish Sengupta, Günther Waxenegger-Wilfing, Jan Martin, Justin Hardi, Matthew P. Juniper
2021 arXiv   pre-print
We train an autoregressive Bayesian neural network model to forecast the amplitude of the dynamic pressure time series, inputting multiple sensor measurements (injector pressure/ temperature measurements  ...  The Bayesian nature of our algorithms allows us to work with a dataset whose size is restricted by the expense of each experimental run, without making overconfident extrapolations.  ...  In this study, on the other hand, we use nonlinear autoregressive time series modeling with a Bayesian Neural Network to forecast, with uncertainties, the future amplitude of pressure fluctuations given  ... 
arXiv:2107.06396v2 fatcat:sc3silupvbc7hi4w267w5yo6vi

Forecasting Energy Use in Buildings Using Artificial Neural Networks: A Review

Jason Runge, Radu Zmeureanu
2019 Energies  
Amongst the many different types of models, artificial neural networks rank among the most popular data-driven approaches applied to date.  ...  This paper offers a review of the studies published since the year 2000 which have applied artificial neural networks for forecasting building energy use and demand, with a particular focus on reviewing  ...  mathematical approach calibrated with measurements; and (ii) grey-box models, which couple a physical model of the HVAC system or building, with a black-box model applied at key parameters within the physical  ... 
doi:10.3390/en12173254 fatcat:zbqk5s3igjdadko6h5pyiqkcwa
« Previous Showing results 1 — 15 out of 1,319 results