84 citations · 277 across the 26 of their papers we have counts for
28 papers
Data-Driven Conditional Flexibility Index
Moritz Wedemeyer, Eike Cramer, Alexander Mitsos +1
With the increasing flexibilization of processes, determining robust scheduling decisions has become an important goal. Traditionally, the flexibility index has been used to identi…
Estimating Dense-Packed Zone Height in Liquid-Liquid Separation: A Physics-Informed Neural Network Approach
Mehmet Velioglu, Song Zhai, Alexander Mitsos +3
Separating liquid-liquid dispersions in gravity settlers is critical in chemical, pharmaceutical, and recycling processes. The dense-packed zone height is an important performance…
End-to-End Reinforcement Learning of Koopman Models for eNMPC of an Air Separation Unit
Daniel Mayfrank, Kayra Dernek, Laura Lang +2
With our recently proposed method based on reinforcement learning (Mayfrank et al. (2024), Comput. Chem. Eng. 190), Koopman surrogate models can be trained for optimal performance…
DeepEOSNet: Capturing the dependency on thermodynamic state in property prediction tasks
Jan Pavšek, Alexander Mitsos, Manuel Dahmen +2
We propose a machine learning (ML) architecture to better capture the dependency of thermodynamic properties on the independent states. When predicting state-dependent thermodynami…
Molecular Machine Learning in Chemical Process Design
Jan G. Rittig, Manuel Dahmen, Martin Grohe +2
We present a perspective on molecular machine learning (ML) in the field of chemical process engineering. Recently, molecular ML has demonstrated great potential in (i) providing h…
Accelerating Deterministic Global Optimization via GPU-parallel Interval Arithmetic
Hongzhen Zhang, Tim Kerkenhoff, Neil Kichler +4
Spatial Branch and Bound (B&B) algorithms are widely used for solving nonconvex problems to global optimality, yet they remain computationally expensive. Though some works have bee…
Sample-Efficient Reinforcement Learning of Koopman eNMPC
Daniel Mayfrank, Mehmet Velioglu, Alexander Mitsos +1
Reinforcement learning (RL) can be used to tune data-driven (economic) nonlinear model predictive controllers ((e)NMPCs) for optimal performance in a specific control task by optim…
Robust Energy System Design via Semi-infinite Programming
Moritz Wedemeyer, Eike Cramer, Alexander Mitsos +1
Time-series information needs to be incorporated into energy system optimization to account for the uncertainty of renewable energy sources. Typically, time-series aggregation meth…
Physics-Informed Neural Networks for Dynamic Process Operations with Limited Physical Knowledge and Data
Mehmet Velioglu, Song Zhai, Sophia Rupprecht +3
In chemical engineering, process data are expensive to acquire, and complex phenomena are difficult to fully model. We explore the use of physics-informed neural networks (PINNs) f…
Task-optimal data-driven surrogate models for eNMPC via differentiable simulation and optimization
Daniel Mayfrank, Na Young Ahn, Alexander Mitsos +1
Mechanistic dynamic process models may be too computationally expensive to be usable as part of a real-time capable predictive controller. We present a method for end-to-end learni…
Optimal design of a local renewable electricity supply system for power-intensive production processes with demand response
Sonja H. M. Germscheid, Benedikt Nilges, Niklas von der Assen +2
This work studies synergies arising from combining industrial demand response and local renewable electricity supply. To this end, we optimize the design of a local electricity gen…
Multivariate Scenario Generation of Day-Ahead Electricity Prices using Normalizing Flows
Hannes Hilger, Dirk Witthaut, Manuel Dahmen +3
Trading on the day-ahead electricity markets requires accurate information about the realization of electricity prices and the uncertainty attached to the predictions. Deriving acc…
End-to-End Reinforcement Learning of Koopman Models for Economic Nonlinear Model Predictive Control
Daniel Mayfrank, Alexander Mitsos, Manuel Dahmen
(Economic) nonlinear model predictive control ((e)NMPC) requires dynamic models that are sufficiently accurate and computationally tractable. Data-driven surrogate models for mecha…
A Recursively Recurrent Neural Network (R2N2) Architecture for Learning Iterative Algorithms
Danimir T. Doncevic, Alexander Mitsos, Yue Guo +4
Meta-learning of numerical algorithms for a given task consists of the data-driven identification and adaptation of an algorithmic structure and the associated hyperparameters. To…
Physical Pooling Functions in Graph Neural Networks for Molecular Property Prediction
Artur M. Schweidtmann, Jan G. Rittig, Jana M. Weber +4
Graph neural networks (GNNs) are emerging in chemical engineering for the end-to-end learning of physicochemical properties based on molecular graphs. A key element of GNNs is the…
Graph neural networks for the prediction of molecular structure-property relationships
Jan G. Rittig, Qinghe Gao, Manuel Dahmen +2
Molecular property prediction is of crucial importance in many disciplines such as drug discovery, molecular biology, or material and process design. The frequently employed quanti…
Graph Neural Networks for Temperature-Dependent Activity Coefficient Prediction of Solutes in Ionic Liquids
Jan G. Rittig, Karim Ben Hicham, Artur M. Schweidtmann +2
Ionic liquids (ILs) are important solvents for sustainable processes and predicting activity coefficients (ACs) of solutes in ILs is needed. Recently, matrix completion methods (MC…
Graph Machine Learning for Design of High-Octane Fuels
Jan G. Rittig, Martin Ritzert, Artur M. Schweidtmann +7
Fuels with high-knock resistance enable modern spark-ignition engines to achieve high efficiency and thus low CO2 emissions. Identification of molecules with desired autoignition p…
Demand Response for Flat Nonlinear MIMO Processes using Dynamic Ramping Constraints
Florian Joseph Baader, Philipp Althaus, André Bardow +1
Volatile electricity prices make demand response (DR) attractive for processes that can modulate their production rate. However, if nonlinear dynamic processes must be scheduled si…
Multivariate Probabilistic Forecasting of Intraday Electricity Prices using Normalizing Flows
Eike Cramer, Dirk Witthaut, Alexander Mitsos +1
Electricity is traded on various markets with different time horizons and regulations. Short-term intraday trading becomes increasingly important due to the higher penetration of r…
Normalizing Flow-based Day-Ahead Wind Power Scenario Generation for Profitable and Reliable Delivery Commitments by Wind Farm Operators
Eike Cramer, Leonard Paeleke, Alexander Mitsos +1
We present a specialized scenario generation method that utilizes forecast information to generate scenarios for day-ahead scheduling problems. In particular, we use normalizing fl…
Nonlinear Isometric Manifold Learning for Injective Normalizing Flows
Eike Cramer, Felix Rauh, Alexander Mitsos +2
To model manifold data using normalizing flows, we employ isometric autoencoders to design embeddings with explicit inverses that do not distort the probability distribution. Using…
Simultaneous mixed-integer dynamic scheduling of processes and their energy systems
Florian Joseph Baader, André Bardow, Manuel Dahmen
Increasingly volatile electricity prices make simultaneous scheduling optimization desirable for production processes and their energy systems. Simultaneous scheduling needs to acc…
Validation Methods for Energy Time Series Scenarios from Deep Generative Models
Eike Cramer, Leonardo Rydin Gorjão, Alexander Mitsos +3
The design and operation of modern energy systems are heavily influenced by time-dependent and uncertain parameters, e.g., renewable electricity generation, load-demand, and electr…