Modelica: Component Oriented Modeling of Physical Systems
summary
Modelica compiles component models into algebraic and differential equations, then applies equation sorting into BLT form to drop solve complexity from cubic to linear. A demo robot model has about 4,000 variables, 1,342 parameters and ~500 components and compilers do index reduction, state selection and tearing before numeric integration. There is an npm-style experiment that packages a Modelica model as code and npm demo runs and returns trajectories, so the simulation pipeline actually works.
AI-written summary · about
chapters (15)
- 0:00Intro
- 0:29Background
- 2:32The First Computer
- 3:15Artillery
- 3:51Math
- 4:51Programming
- 7:57Modelica
- 12:05Connectors
- 15:59Inheritance
- 18:55Industrial Robot
- 20:34Component to Equations
- 22:02Matrix
- 23:59Equation Sorting
- 26:20Tools
- 33:37Questions
related talks
26mCOLib - A Collaborative Open-source Library for Power System C...
35mModeling and more - Experiences in building an efficient model infrastructure
30mCreate Detailed Traffic Models and Simulations
46mMake more money by modeling and optimizing your business processes with Kotlin
05mRelational Programming
06mML for Systems and Systems for ML
01h 03mQuery Optimization and Evaluation 3: Variable Elimination and Tensor Decomposition
34mConstraint Logic Propagation Conflict Spreadsheets