direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Dr.-Ing. Erik Esche

Habilitand; Research Assistant with Teaching Obligations and Continuous Tasks

KWT-A 108 A

Tel. +49 30 314 21634

Fax +49 30 314 26915

S/MIME (X509v3): CA der TU Berlin

Fingerprint (SHA1): 8E:37:62:A1:86:09:B9:DB:06:DF:6C:82:6D:C0:4C:BC:BF:8B:D5:96

Curriculum Vitae

Erik Esche
Geboren 1987 in Berlin

2010 - B.Sc. in Energy- and Process Engineering (TU Berlin); 2011 - Master's Thesis at Carnegie Mellon University, Pittsburgh, PA with Lorenz T. Biegler; 2011 - M.Sc. in Energy- and Process Engineering (TU Berlin); 2015 - PhD (cum laude) in Process Engineering on "MINLP Optimization under Uncertainty of a Mini-plant for the Oxidative Coupling of Methane"; since:Habilitand & PostDoc at Process Dynamics and Operations Group of Prof. Repke

Research

Areas of Research:

  • Optimal operation of process plants
  • Hydroformyltion of long-chained olefines
  • Optimale synthesis of chemical processes under uncertainty
  • Chance Constraints (probabilistic contraints)

Research Projects:

Lupe
Lupe

Contact Person for Research Area: Methods for Process Design and Operations

Deputy Contact Person for Research Area: Model Development

Bachelor's and Master's Theses

 Topics on Offer:

  • Knowledge Graphs to Model Chemical Engineering Knowledge and Data
  • Semi-supervised Regression for Control of Dynamic Systems
  • Unscented Transformation for Solution of Joint Chance Constraints
  • Set-based Formulation of Models in Chemical Engineering Using MOSAICmodeling
  • Distributed Database Architectures for Model Development in MOSAICmodeling

Further topics on request.

Publications

Customized code generation based on user specifications for simulation and optimization
Zitatschlüssel Tolksdorf2019
Autor Gregor Tolksdorf and Erik Esche and Günter Wozny and Jens-Uwe Repke
Seiten 670–684
Jahr 2019
ISSN 0098-1354
DOI 10.1016/j.compchemeng.2018.12.006
Journal Computers & Chemical Engineering
Jahrgang 121
Monat feb
Verlag Elsevier BV
Zusammenfassung Model-driven software engineering is a well-known concept in computer science, but scarcely applied in chemical engineering. This contribution presents a first implementation of model-driven development of customized code for simulation and optimization based on equation-oriented models in process science. Transferring the model-driven approach to chemical engineering allows for transformations or discretizations before automated code generation. The customization of this concept poses additional challenges regarding flexibility and tailoring for the user’s needs. We propose the so-called “User-defined Language Specificators (UDLS)”, based on free standards (MathML, XML), to combine the benefits of automated code generation with the flexibility of customization, therefore still avoiding error-prone manual model implementation. The case studies show the successful application of this approach for equation-based flowsheet simulation (using CAPE-OPEN interfaces) as well as optimization.
Link zur Publikation Download Bibtex Eintrag

Contact Details

Dr.-Ing. Erik Esche
Technische Universität Berlin
Sekr. KWT-9 - Fachgebiet Dynamik und Betrieb technischer Anlagen
Str. des 17. Juni 135
D-10623 Berlin
Germany

Tel. +49 (0) 30 314 - 21 634
Fax. +49 (0) 30 314 - 26 915

Zusatzinformationen / Extras

Direktzugang

Schnellnavigation zur Seite über Nummerneingabe

Office Hours:

Dr.-Ing. Erik Esche

All in-person office hours are postponed for now. Separate arrangements can be made by .

KWT-A 110