Free Workshops
Registration
Registration for the Conference workshops is now open through the Registration Form. Please register latest by October 1st since we need to be ablse to chose properly sized rooms for all tutorials, and provide credentials for cloud-based tutorials.
1. Introduction to Modelica and Thermo-fluid Modeling with Applications from the Buildings Library
Instructors: David Blum, Michael Wetter, and Hongxiang (Casper) Fu, Lawrence Berkeley National Laboratory
Time: October 14th, 2:00 pm to 5:00 pm
Abstract:
This workshop will give an introduction to the modeling of thermo-fluid systems and their control using the Modelica Buildings Library with Modelon Impact or with OpenModelica. We will be using applications from the Modelica Buildings Library, but the course will be structured to also appeal to modelers from other application domains that are concerned with thermo-fluid systems. After an introductory presentation that discusses the scope of the Modelica Buildings Library, we discuss best practices in setting up thermo-fluid flow models and how to avoid potential problems. In hands-on exercises, participants will build models of simple heating and air conditioning systems, link them to a thermal load, and add feedback control. The models will be built using components from the Modelica Buildings Library. Simulations will be run on Modelon Impact.
Duration: 3 hours
Suggested Experience and Software Requirements
- Have prior knowledge in Modelica and either have gone through the online Impact tutorial for basic usage (https://help.modelon.com/latest/application_overview/getting_started_overview/) or have knowledge of using OpenModelica’s OMEdit, or
- Have participated in the workshop provided by Modelon on introduction to Modelica modeling and Impact.
Laptop with recent version of either Google Chrome or Opera installed. A minimum of 8GB of memory is recommended.
2. Introduction to Modeling, Simulation, Debugging, and Interoperability with Modelica and OpenModelica
Instructor: Lena Buffoni, Linköping University
Time: October 14th, 10:00 am to 1:00 pm
Abstract:
This tutorial gives an introduction to the Modelica language, the OpenModelica environment, and an overview of modeling and simulation in a number of application areas. Some advanced features of OpenModelica will be presented. A number of hands-on exercises will be done during the tutorial, both graphical modeling using the Modelica standard library and textual modeling. Bring your laptop for exercises.
Duration: 3 hours
Suggested Experience and Software Requirements
No prior experience necessary
Please bring a laptop with Windows, Linux or Mac installed. It is best to have Openmodelica installed prior to the workshop, see https://openmodelica.org/download/download-windows/.
3. Deployment of models as Web-Apps with Modelon Impact
Instructors: Hubertus Tummescheit and Lixiang Li, Modelon
Time: October 14th, 2:00 pm to 5:00 pm
Abstract
In this session you will learn more about a deployment approach that allows custom interfaces to be created for parameterization and post-processing. The course will make use of the Modelon Impact REST-API, documented on https://modelon-impact-client.readthedocs.io/en/latest/ , Jupyter notebooks as the app-development environment, and a few utility-routines that will speed up building a web-app. The notebook can be deployed as a web-app through the Voilà framework (https://github.com/voila-dashboards/voila). Participants who know the concepts of Modelica tools should be able to run a simple, self-developed web app on top of Modelon Impact, and understand the possibilities to automate workflows and deploy them as app in Modelon Impact.
Course take-aways and learnings:
- Introduction to the Modelon Impact REST-API
- Understanding how to build a simple web-app using jupyter-notebooks and Voilà
Recommended pre-reading:
Tutorials at https://help.modelon.com/latest/
Duration: 3.0 hours
Suggested Experience and Software Requirements
Please bring a laptop with a recent version of Chrome, Edge or Opera installed, with at least 8 GB of memory. No installation needs, registered participants will have a login to Modelon Imapct on the cloud.
Familiarity with any Modelica tool required, familiarity with Modelon Impact beneficial. Some familiarity with Python programming. Participants are incouraged to familiarize themselves with Modelon Impact before the course at https://help.modelon.com/latest/
4. Using Python as Interface for the Modelica and FMI Standards
Instructor: Behnam Afsharpoya, PhD
Time: October 14th, 10:00 am to 1:00 pm
Abstract
Python is a very well-known coding language with valuable libraries that provide toolboxes for Artificial intelligence and Machine learning methods. Integration of Python with modelica and FMI standard can provide useful toolboxes for physical and numerical analyzes.
Duration: 4 hours
Suggested Experience and Software Requirements
Basic level for Modelica, Python and FMI
Python, Modelica and FMI tools will be provided. Compiler needs to be installed
5. Integrating Declarative and Imperative Acausal Modeling with ModelingToolkit
Instructor: Chris Rakaukas, Juliahub and MIT
Time: October 14th, 10:00 am to 1:00 pm
Abstract
Modelica and Remodel are designed to work as declarative acausal modeling systems. ModelingToolkit hits an interesting middle ground in the design space of modeling systems by having declarative semantics which are interacted with imperatively. In this workshop we will explore this idea by demonstrating some of the ergonomic advantages gained by such an approach. Computer generation of model variations for more easily comparing between models, data-driven generation of models via symbolic regression and evolutionary approaches, and non-standard model transformations (moment closure approximations) will be used to showcase how the interactivity of the ModelingToolkit compiler can be used to improve modeling workflows. The disadvantages will also be discussed, in particular the difficulties of certain analyses and GUI integration. Altogether, participants will get a better view of the engineering trade-offs being explored in the modeling and simulation space and leave with a better perspective of alternative modeling workflows.
Duration: 4 hours
Suggested Experience and Software Requirements
Only Modelica experience is required.
Software requirements: A computer that can install Julia.
The workshop will start with a quick introduction to ModelingToolkit itself (https://docs.sciml.ai/ModelingToolkit/stable/). Then, some of the examples from the ModelingToolkit Course (https://sciml.github.io/ModelingToolkitCourse/dev/lectures/lecture2/) will be used in order to highlight some of the computational generation of models. Automated model generation (https://docs.sciml.ai/Overview/stable/showcase/missing_physics/) will also be highlighted.