SLIDE 1
Automated Configuration
- f Co-simulation with
Co-simulation on the rise
- Keyword analysis
2
Automated Configuration of Co-simulation with Domain Specific Hints - - PowerPoint PPT Presentation
Automated Configuration of Co-simulation with Domain Specific Hints - - PowerPoint PPT Presentation
Automated Configuration of Co-simulation with Domain Specific Hints Co-simulation on the rise Keyword analysis 2 Co-simulation on the rise Research Projects 3 Co-simulation on the rise Applications 4 Co-simulation on the rise
SLIDE 2
SLIDE 3
Co-simulation on the rise
- Research Projects
3
SLIDE 4
Co-simulation on the rise
- Applications
4
SLIDE 5
Co-simulation on the rise
- Surveys
5
SLIDE 6
Co-simulation on the rise
6
SLIDE 7
Industry Pains
Lack of tools that sufficiently support FMI Not enough cooperation between theorists and practitioners. Difficult to implement FMUs Insufficient documentation
7
SLIDE 8
Everyday Challenges
* Not published: under review Judging the validity of the co-sim Insufficient communication between theorists and practitioners Defining macro-step size. Defining tolerances Algebraic loops.
8
SLIDE 9
Automated Configuration
- Why is this so difficult?
- Why are adaptive step size algorithms not enough?
9
SLIDE 10
Motivating Example
10
SLIDE 11
Example Behavior(s)
11
SLIDE 12
Example Behavior(s)
Solution: Engineers have intuition (or past experience) on the correct behavior Example: Controller is software running at 10000000Hz Plant and Load are connected by a power connection (v*f)
12
SLIDE 13
Example Behavior(s)
Fixed-step Jacobi, step size 1st-order input interpolations, step size Combination of: 1st-order input interpolations, causality preservation, and energy conservation, in selected signals Why is this so difficult? Configuration Effort
13
SLIDE 14
Grand Challenge
Combination of: 1st-order input interpolations, causality preservation, and energy conservation, in selected signals Controller is software running at 10000000Hz Plant and Load are connected by a power connection (v*f) HintCO
14
SLIDE 15
Grand Challenge - Detail
Controller is software running at 10000000Hz Plant and Load are connected by a power connection (v*f) FMUs (interpolations/extrapolations), and connections Step size Set/Get/DoStep invocation sequence Real behavior of the coupled model Co-sim behavior
15
SLIDE 16
- Contrib. A – Hint Language
16
Controller is software running at 1e6 Hz Plant and Load are connected by a power connection (v*f)
SLIDE 17
- Contrib. B – Exploration
- Search Space Encoding:
- Set of all communication step sizes
- Set of all operation sequences
- Set of all adaptations (e.g., interpolation, energy
conservation, etc…) applied to FMUs
17
SLIDE 18
- Contrib. B – Exploration
- Priority Variant Generation:
- Compact representation of variants in a diagram, and
- Prioritizes walk in that diagram
18
SLIDE 19
- Contrib. B – Exploration
- Variant Execution:
- Determine feasible operation ordering.
19
SLIDE 20
- Contrib. B – Exploration
- Translating Hints to Adaptations:
20
SLIDE 21
Results
21
SLIDE 22
Summary
- Practitioners need more support for configuration
- f co-simulations
- Existing master algorithms are not sufficient without
extensive fine tuning.
- There is no general way of obtaining the real
behavior of a coupled system,
- So we propose to leverage engineer’s knowledge and
past experience.
- We provide a tool to tackle this problem.
22
SLIDE 23
Questions
- Practitioners need more support for configuration of co-simulations
- Existing master algorithms are not sufficient without extensive fine tuning.
- There is no general way of obtaining the real behavior of a coupled
system,
- So we propose to leverage engineer’s knowledge and past experience.
- We provide a tool to tackle this problem.
23
SLIDE 24
Demo
24