In this paper we introduce JuliaSim, a high-performance programming environment designed to blend traditional modeling and simulation with machine learning. JuliaSim can build surrogates from component-based models, including Functional Mockup Units, using continuous-time echo state networks (CTESN). The foundation of this environment, Modeling-Toolkit.jl, is an acausal-modeling language which can compose the trained surrogates as components. We present the JuliaSim model library, consisting of differential-algebraic equations and pre-trained surrogates, which can be composed using the modeling system. We demonstrate a surrogate-accelerated approach on HVAC dynamics by showing that the CTESN surrogates capture dynamics at less than 4% error with an acceleration of 340x, and speed up design optimization by two orders of magnitude. We showcase the surrogate deployed in a co …
%0 Journal Article
%1 rackauckas2022composing
%A Rackauckas, Chris
%A Gwozdz, Maja
%A Jain, Anand
%A Ma, Yingbo
%A Martinuzzi, Francesco
%A Rajput, Utkarsh
%A Saba, Elliot
%A Shah, Viral B
%A Anantharaman, Ranjan
%A Edelman, Alan
%A Gowda, Shashi
%A Pal, Avik
%A Laughman, Chris
%D 2022
%I IEEE
%K imported topic_earthenvironment
%P 1-17
%T Composing modeling and simulation with machine learning in Julia
%X In this paper we introduce JuliaSim, a high-performance programming environment designed to blend traditional modeling and simulation with machine learning. JuliaSim can build surrogates from component-based models, including Functional Mockup Units, using continuous-time echo state networks (CTESN). The foundation of this environment, Modeling-Toolkit.jl, is an acausal-modeling language which can compose the trained surrogates as components. We present the JuliaSim model library, consisting of differential-algebraic equations and pre-trained surrogates, which can be composed using the modeling system. We demonstrate a surrogate-accelerated approach on HVAC dynamics by showing that the CTESN surrogates capture dynamics at less than 4% error with an acceleration of 340x, and speed up design optimization by two orders of magnitude. We showcase the surrogate deployed in a co …
@article{rackauckas2022composing,
abstract = {In this paper we introduce JuliaSim, a high-performance programming environment designed to blend traditional modeling and simulation with machine learning. JuliaSim can build surrogates from component-based models, including Functional Mockup Units, using continuous-time echo state networks (CTESN). The foundation of this environment, Modeling-Toolkit.jl, is an acausal-modeling language which can compose the trained surrogates as components. We present the JuliaSim model library, consisting of differential-algebraic equations and pre-trained surrogates, which can be composed using the modeling system. We demonstrate a surrogate-accelerated approach on HVAC dynamics by showing that the CTESN surrogates capture dynamics at less than 4% error with an acceleration of 340x, and speed up design optimization by two orders of magnitude. We showcase the surrogate deployed in a co …},
added-at = {2024-11-29T11:53:34.000+0100},
author = {Rackauckas, Chris and Gwozdz, Maja and Jain, Anand and Ma, Yingbo and Martinuzzi, Francesco and Rajput, Utkarsh and Saba, Elliot and Shah, Viral B and Anantharaman, Ranjan and Edelman, Alan and Gowda, Shashi and Pal, Avik and Laughman, Chris},
biburl = {https://puma.scadsai.uni-leipzig.de/bibtex/294fae6978b690b275182e3700ce0ca9b/joum576e},
citation = {2022 Annual Modeling and Simulation Conference (ANNSIM), 1-17, 2022},
conference = {2022 Annual Modeling and Simulation Conference (ANNSIM)},
interhash = {3ff2d3301193a6e0ba801a504986975a},
intrahash = {94fae6978b690b275182e3700ce0ca9b},
keywords = {imported topic_earthenvironment},
pages = {1-17},
publisher = {IEEE},
timestamp = {2024-11-29T11:53:34.000+0100},
title = {Composing modeling and simulation with machine learning in Julia},
year = 2022
}
