• Research Overview (German) - Here you will find a brief overview of our research in German.

  forschungsubersicht_deutsch.pdf

   

• Current Research (English) - Here you can view the research topics organized by researcher.

Research Overview

Expertise

Experimental Modeling (Identification)

• Building models from data, not from first principles (physical, chemical, ... laws)
• Improving models derived from first principles using measurement data (gray-box models)
• Metamodeling: Fast models that approximate complex, time-consuming calculations. Suitable for dynamic simulation, optimization, and real-time applications
• Nonlinear and/or dynamic models

Design of Experiments

• Where to measure? (Not the location, but which operating points?)
• How to describe the boundaries of the operating regime?

Input Selection

• How relevant are the various inputs for modeling?
• Which inputs have strongly nonlinear effects?

Classification, novelty detection, extrapolation detection, ...

Models: What Are They For?

• Models form the basis for most advanced techniques in many disciplines.
• Model-based techniques can be divided into a (sometimes iterative) two-step procedure:
  1. Building a model.
  2. Applying a technique based on this model.

Applications of models:

Models: What Are They Based On?

Topics

1. Experimental Modeling (Identification)

• Local model networks.
• Projects:
  LOLIMOT Lookup Tables

2. Design of Experiments (DoE)

• Optimal maximin Latin hypercubes (unsupervised).
• Optimization of the next measurement point to gather the most information using HILOMOT-DoE (supervised).
• Projects:
  Preprocessing for Data-Driven Modeling with Probability Density Estimation
  Density-Based Data Weighting
  Growing DoE with 1- and 2-Class Classifiers 

3. Automatic Input Selection
 

4. Metamodeling

• For CFD, FEM, and Look-up Tables.

5. Nonlinear Dynamic Models

• Local model networks with OBF and FIR.
• Projects:
  Nonlinear Regularized FIR Models
  Nonlinear Laguerre and Kautz Models
  Nonlinear State-Space Identification
  Gray-Box State-Space Modeling

6. Classification

• Least-squares support vector machine.
• Projects:
  One-Class Classifier for Data Assessment
  Robust One- and Two-Class Classifiers via Zero Leave-One-Out Cross-Validation Errors

7. Design of Excitation Signals for Identification

• Projects:
  Excitation Signal Design for Nonlinear Dynamic Systems

8. Applications 

• Combustion engines
• Driveability Calibration
• Automatic transmission modeling
• Structural health monitoring
• Metamodels for CFD simulations
• Metamodels for FEM simulations
• Location estimation for inductive charging
• Projects:
  Model Predictive Control with Local Model Networks
  Structural Health Monitoring (SHM) using Probability Density Estimation
  Adaptive Local Model Networks (ALMN)
  Process Optimization of a Combine Harvester
  Data-Driven Optimization of Multi-Stage Forming Processes