arXiv subjects

Thomas Krug

Publications and source records attributed to Thomas Krug.

6 recordsLinked to original sources

arXiv2026

BuilDyn: Excitation-Driven Data Generation for Building Thermal Dynamics Modeling and Control

Machine learning (ML) is increasingly used for data-driven modeling of buildings to enable downstream tasks such as fault detection and diagnosis, and energy-efficient control. While recent work improves generalization across building characteristics, weather, and occupancy, generalization also depends on sufficient exploration of the control-driven system state space. Existing real-world datasets and simulation environments predominantly reflect stationary operation under fixed control policies, resulting in limited excitation and reduced robustness to unseen operating conditions. This paper introduces BuilDyn, a package based on BuilDa that enables customizable excitation strategies for control-oriented data generation. BuilDyn further supports sampling from representative building distributions and provides a Python interface for easy integration into machine learning pipelines. We demonstrate the benefits of BuilDyn by comparing the performance of data-driven ML models trained on non-excited and excited data for one building. With BuilDyn, we hope to advance scalable control-oriented modeling and support future directions such as transfer learning and building-specific foundation models.

eess.SY↗

arXiv2026

Reinforcement Learning with Symbolic Reward Machines

Reward Machines (RMs) are an established mechanism in Reinforcement Learning (RL) to represent and learn sparse, temporally extended tasks with non-Markovian rewards. RMs rely on high-level information in the form of labels that are emitted by the environment alongside the observation. However, this concept requires manual user input for each environment and task. The user has to create a suitable labeling function that computes the labels. These limitations lead to poor applicability in widely adopted RL frameworks. We propose Symbolic Reward Machines (SRMs) together with the learning algorithms QSRM and LSRM to overcome the limitations of RMs. SRMs consume only the standard output of the environment and process the observation directly through guards that are represented by symbolic formulas. In our evaluation, our SRM methods outperform the baseline RL approaches and generate the same results as the existing RM methods. At the same time, our methods adhere to the widely used environment definition and provide interpretable representations of the task to the user.

cs.LG↗

arXiv2025

A Highly Configurable Framework for Large-Scale Thermal Building Data Generation to drive Machine Learning Research

Data-driven modeling of building thermal dynamics is emerging as an increasingly important field of research for large-scale intelligent building control. However, research in data-driven modeling using machine learning (ML) techniques requires massive amounts of thermal building data, which is not easily available. Neither empirical public datasets nor existing data generators meet the needs of ML research in terms of data quality and quantity. Moreover, existing data generation approaches typically require expert knowledge in building simulation. To fill this gap, we present a thermal building data generation framework which we call BuilDa. BuilDa is designed to produce synthetic data of adequate quality and quantity for ML research. The framework does not require profound building simulation knowledge to generate large volumes of data. BuilDa uses a single-zone Modelica model that is exported as a Functional Mock-up Unit (FMU) and simulated in Python. We demonstrate BuilDa by generating data and utilizing it for a transfer learning study involving the fine-tuning of 486 data-driven models.

eess.SY↗

arXiv2025

BUILDA: A Thermal Building Data Generation Framework for Transfer Learning

Transfer learning (TL) can improve data-driven modeling of building thermal dynamics. Therefore, many new TL research areas emerge in the field, such as selecting the right source model for TL. However, these research directions require massive amounts of thermal building data which is lacking presently. Neither public datasets nor existing data generators meet the needs of TL research in terms of data quality and quantity. Moreover, existing data generation approaches typically require expert knowledge in building simulation. We present BuilDa, a thermal building data generation framework for producing synthetic data of adequate quality and quantity for TL research. The framework does not require profound building simulation knowledge to generate large volumes of data. BuilDa uses a single-zone Modelica model that is exported as a Functional Mock-up Unit (FMU) and simulated in Python. We demonstrate BuilDa by generating data and utilizing it for pretraining and fine-tuning TL models.

cs.LG↗

arXiv2025

GenTL: A General Transfer Learning Model for Building Thermal Dynamics

Transfer Learning (TL) is an emerging field in modeling building thermal dynamics. This method reduces the data required for a data-driven model of a target building by leveraging knowledge from a source building. Consequently, it enables the creation of data-efficient models that can be used for advanced control and fault detection & diagnosis. A major limitation of the TL approach is its inconsistent performance across different sources. Although accurate source-building selection for a target is crucial, it remains a persistent challenge. We present GenTL, a general transfer learning model for single-family houses in Central Europe. GenTL can be efficiently fine-tuned to a large variety of target buildings. It is pretrained on a Long Short-Term Memory (LSTM) network with data from 450 different buildings. The general transfer learning model eliminates the need for source-building selection by serving as a universal source for fine-tuning. Comparative analysis with conventional single-source to single-target TL demonstrates the efficacy and reliability of the general pretraining approach. Testing GenTL on 144 target buildings for fine-tuning reveals an average prediction error (RMSE) reduction of 42.1 % compared to fine-tuning single-source models.

eess.SY↗

arXiv2022

Towards Digital Twin-enabled DevOps for CPS providing Architecture-Based Service Adaptation & Verification at Runtime

Industrial Product-Service Systems (IPSS) denote a service-oriented (SO) way of providing access to CPS capabilities. The design of such systems bears high risk due to uncertainty in requirements related to service function and behavior, operation environments, and evolving customer needs. Such risks and uncertainties are well known in the IT sector, where DevOps principles ensure continuous system improvement through reliable and frequent delivery processes. A modular and SO system architecture complements these processes to facilitate IT system adaptation and evolution. This work proposes a method to use and extend the Digital Twins (DTs) of IPSS assets for enabling the continuous optimization of CPS service delivery and the latter's adaptation to changing needs and environments. This reduces uncertainty during design and operations by assuring IPSS integrity and availability, especially for design and service adaptations at CPS runtime. The method builds on transferring IT DevOps principles to DT-enabled CPS IPSS. The chosen design approach integrates, reuses, and aligns the DT processing and communication resources with DevOps requirements derived from literature. We use these requirements to propose a DT-enabled self-adaptive CPS model, which guides the realization of DT-enabled DevOps in CPS IPSS. We further propose detailed design models for operation-critical DTs that integrate CPS closed-loop control and architecture-based CPS adaptation. This integrated approach enables the implementation of A/B testing as a use case and central concept to enable CPS IPSS service adaptation and reconfiguration. The self-adaptive CPS model and DT design concept have been validated in an evaluation environment for operation-critical CPS IPSS. The demonstrator achieved sub-millisecond cycle times during service A/B testing at runtime without causing CPS operation interferences and downtime.

cs.SE↗