arXiv subjects

Janina Zittel

Publications and source records attributed to Janina Zittel.

10 recordsLinked to original sources

arXiv2026

Robust Unit Commitment in District Heating Networks: Chance-Constrained and CVaR Optimization Under Demand Uncertainty

While district heating networks are a key component of the energy transition, their operational planning is challenging due to substantial uncertainty in heat demand. We address this volatility without compromising system reliability. By adapting chance-constrained programming (CC) and conditional value-at-risk (CVaR) optimization to the mixed-integer unit-commitment problem for district-heating networks, we can compute optimal unit-commitment schedules under demand uncertainty. We generate heat demand time series using a Bayesian model, which explicitly quantifies forecasting uncertainty and yields predictive distributions as input to the network optimization. To handle the inherent uncertainty in heat demand, we apply both robust optimization approaches: chance-constrained programming limits the probability of unmet demand, while CVaR optimization penalizes severe shortfalls in the tail of the distribution. We evaluate the approaches both on real-world data from the Berlin district heating network and on a benchmark set of synthetic, realistically parameterized instances of varying sizes. The results compare the two uncertainty-handling methods with respect to solution quality, risk exposure, and computational effort.

math.OC↗

arXiv2026

Computational acceleration strategies for large-scale energy system optimization: a comparative study of GPU-accelerated and distributed-memory solvers

Energy system optimization models are increasing in scope and resolution, yielding large and challenging linear programs. For a long time, the standard way to address such problems has relied on shared-memory interior-point methods (IPM), which combine robustness and accuracy but face scalability limits as model instance size grows. Recently, two promising directions for specialized solver architectures have emerged: (i) GPU-accelerated first-order methods (FOM); and (ii) distributed-memory IPM, which can exploit block structure that arises in many energy system models. This paper presents a computational study comparing these solver classes on a diverse test set of large-scale linear programs arising from energy system analysis, including scenario-based formulations derived from stochastic programming. The results illustrate that distributed-memory IPM can leverage problem structure to deliver substantial speed-ups on specific problems with block-angular structures. GPU-accelerated FOMs demonstrate strong scalability but may yield solutions with higher relative infeasibilities, which, depending on the use case and model uncertainty, can still be acceptable. Overall, our findings indicate that recent algorithmic and hardware advances substantially broaden the computational toolbox available to the energy system optimization community. Each solver class exhibits distinct advantages: shared-memory IPMs remain a powerful tool for reliably obtaining high-accuracy solutions; distributed-memory IPMs can extend scalability to hundreds of cores for certain structured models, enabling faster time-to-solution; and GPU-based FOM can deliver fast solutions when such lower accuracy levels are appropriate. Together, they help make high-resolution, multi-scenario energy system optimization models tractable across a broader range of problem sizes and computing environments.

math.OC↗

arXiv2026

Benchmarking Realistic Synthetic Instances Against a Large-Scale District Heating Network: A Multi-Objective Optimization Study for Berlin

Decarbonizing urban energy systems requires optimization approaches capable of handling the operational complexity of large-scale district heating networks. However, existing studies typically focus on a single real-world network, limiting comparability and the transferability of insights. To address this, realistic synthetic instances provide controlled, reproducible environments for testing optimization algorithms independent of specific case studies while preserving key structural and temporal characteristics of real systems. Such instances enable systematic benchmarking, methodological development, and comparative studies across algorithms and modeling choices. In this work, we generate a suite of large-scale synthetic instances for multi-objective optimization of district heating systems. The instances are openly available as network topologies in JSON format and as mixed-integer programs (MPS files) for benchmarking. They are constructed via a transparent procedure that allows reproduction, extension, and transfer to other network-based problems. We apply the method to Berlins district heating network, the most complex in Western Europe, formulating a tri-objective mixed-integer model for unit commitment over up to 25 years with 4-hour temporal resolution. A computational study provides a detailed comparison between the synthetic instances and the real-world Berlin data, showing under which conditions the generated instances reproduce realistic optimization characteristics. Furthermore, we investigate which features make the resulting models computationally challenging. The findings highlight how well-designed synthetic instances can support robust benchmarking practices and enable meaningful assessment of (multi-objective) optimization methods for large-scale district heating systems.

math.OC↗

arXiv2026

Reconnecting Fragmented Citation Networks with Semantic Augmentation

Citation graphs are fundamental tools for modeling scientific structure, but are often fragmented due to missing citations of scientifically connected articles. To address this issue, we propose a computationally efficient hybrid framework integrating citation topology with large language model (LLM)-based text similarity. Using 662,369 Web of Science publications in Mathematics and Operations Research & Management Science, we augment the original graph by adding semantic edges from small, disconnected components and weighting existing citations according to textual similarity. Semantic augmentation substantially reduces fragmentation while preserving disciplinary homogeneity. Compared to embedding-only clustering, cluster detection on augmented graphs using the Leiden algorithm retains structural interpretability while offering multi-scale organization. The method scales efficiently to large datasets and offers a practical strategy for strengthening citation-based indicators without collapsing disciplinary boundaries.

cs.DL↗

arXiv2025

Long-Term Multi-Objective Optimization for Integrated Unit Commitment and Investment Planning for District Heating Networks

The need to decarbonize the energy system has intensified the focus on district heating networks in urban and suburban areas. Therefore, exploring transformation pathways with reasonable trade-offs between economic viability and environmental goals became necessary. We introduce a network-flow-based model class integrating unit commitment and long-term investment planning for multi-energy systems. While the integration of unit commitment and investment planning has been applied to multi-energy systems, a formal introduction and suitability for the application of long-term portfolio planning of an energy provider on an urban scale has yet to be met. Based on mixed integer linear programming, the model bridges the gap between overly detailed industrial modeling tools not designed for computational efficiency at scale and rather abstract academic models. The formulation is tested on Berlin's district heating network. Hence, the challenge lies in a large number of variables and constraints and the coupling of time steps, for example, through investment decisions. A case study explores different solutions on the Pareto front defined by optimal trade-offs between minimizing costs and CO2 emissions through a lexicographic optimization approach. The resulting solution catalog can provide decision-makers valuable insights into feasible transformation pathways, highlighting distinctions between robust and target-dependent investments.

math.OC↗

arXiv2025

Warm-starting Strategies in Scalarization Methods for Multi-Objective Optimization

We explore how warm-starting strategies can be integrated into scalarization-based approaches for multi-objective optimization in (mixed) integer linear programming. Scalarization methods remain widely used classical techniques to compute Pareto-optimal solutions in applied settings. They are favored due to their algorithmic simplicity and broad applicability across continuous and integer programs with an arbitrary number of objectives. While warm-starting has been applied in this context before, a systematic methodology and analysis remain lacking. We address this gap by providing a theoretical characterization of warm-starting within scalarization methods, focusing on the sequencing of subproblems. However, optimizing the order of subproblems to maximize warm-start efficiency may conflict with alternative criteria, such as early identification of infeasible regions. We quantify these trade-offs through an extensive computational study.

math.OC↗

arXiv2025

Leveraging Transfer Learning to Overcome Data Limitations in Czochralski Crystal Growth

The Czochralski (Cz) method is a widely used process for growing high-quality single crystals, critical for applications in semiconductors, optics, and advanced materials. Achieving optimal growth conditions requires precise control of process and furnace design parameters. Still, data scarcity -- especially for new materials -- limits the application of machine learning (ML) in predictive modeling and optimization. This study proposes a transfer learning approach to overcome this limitation by adapting ML models trained on a higher data volume of one source material (Si) to a lower data volume of another target material (Ge and GaAs). The materials were deliberately selected to assess the robustness of the transfer learning approach in handling varying data similarity, with Cz-Ge being similar to Cz-Si, and GaAs grown via the liquid encapsulated Czochralski method (LEC), which differs from Cz-Si. We explore various transfer learning strategies, including Warm Start, Merged Training, and Hyperparameters Transfer, and evaluate multiple ML architectures across two different materials. Our results demonstrate that transfer learning significantly enhances predictive accuracy with minimal data, providing a practical framework for optimizing Cz growth parameters across diverse materials.

cond-mat.mtrl-sci↗

arXiv2025

Enhancing Multi-Energy Modeling: The Role of Mixed-Integer Optimization Decisions

The goal to decarbonize the energy sector has led to increased research in modeling and optimizing multi-energy systems. One of the most promising techniques for modeling (multi-)energy optimization problems is mixed-integer programming (MIP), valued for its ability to represent the complexities of integrated energy systems. While the literature often focuses on deriving mathematical formulations and parameter settings, less attention is given to critical post-formulation decisions. Modeling multi-energy systems as a MIP demands decisions across multiple degrees of freedom. Key steps include reducing a real-world multi-energy network into an abstract topology, defining variables, formulating the relevant (in-)equalities to represent technical requirements, setting objectives, and integrating these elements into a MIP. However, with these elements fixed, the specific transformation of the abstract topology into a graph structure and the construction of the MIP remain non-uniquely. These choices can significantly impact user-friendliness, problem size, and computational efficiency, thus affecting the feasibility and efficiency of modeling efforts. We identify and analyze the additional degrees of freedom and describe two distinct approaches to address them. The approaches are compared regarding mathematical equivalence, suitability for solution algorithms, and clarity of the underlying topology. A case study on a realistic subarea of Berlin's district heating network involving tri-objective optimization for a unit commitment problem demonstrates the practical significance of these decisions. By highlighting these critical yet often overlooked aspects, our work equips energy system modelers with insights to improve computational efficiency, scalability, and interpretability in their optimization efforts.

math.OC↗

arXiv2022

Optimal discrete pipe sizing for tree-shaped CO2 networks

Many energy-intensive industries, like the steel industry, plan to switch to renewable energy sources. Other industries, such as the cement industry, have to rely on carbon capture utilization and storage (CCUS) technologies to reduce their production processes' inevitable carbon dioxide (CO2) emissions. However, a new transport infrastructure needs to be established to connect the point of capture and the point of storage or utilization. Given a tree-shaped network transporting captured CO2 from multiple sources to a single sink, we investigate how to select optimal pipeline diameters from a discrete set of diameters. The general problem of optimizing arc capacities in potential-based fluid networks is already a challenging mixed-integer nonlinear optimization problem. The problem becomes even more complex when adding the highly sensitive nonlinear behavior of CO2 regarding temperature and pressure changes. We propose an iterative algorithm that splits the problem into two parts: a) the pipe-sizing problem under a fixed supply scenario and temperature distribution and b) the thermophysical modeling, including mixing effects, the Joule-Thomson effect, and the heat exchange with the surrounding environment. We show the effectiveness of our approach by applying our algorithm to a real-world network planning problem for a CO2 network in Germany.

math.OC↗

arXiv2022

Optimizing transient gas network control for challenging real-world instances using MIP-based heuristics

Optimizing the transient control of gas networks is a highly challenging task. The corresponding model incorporates the combinatorial complexity of determining the settings for the many active elements as well as the non-linear and non-convex nature of the physical and technical principles of gas transport. In this paper, we present the latest improvements of our ongoing work to solve this problem for real-world, large-scale problem instances: By adjusting our mixed-integer non-linear programming model regarding the gas compression capabilities in the network, we reflect the technical limits of the underlying units more accurately while maintaining a similar overall model size. In addition, we introduce a new algorithmic approach that is based on splitting the complexity of the problem by first finding assignments for discrete variables and then determining the continuous variables as locally optimal solution of the corresponding non-linear program. For the first task, we design multiple different heuristics based on concepts for general time-expanded optimization problems that find solutions by solving a sequence of sub-problems defined on reduced time horizons. To demonstrate the competitiveness of our approach, we test our algorithm on particularly challenging historic demand scenarios. The results show that high-quality solutions are obtained reliably within short solving times, making the algorithm well-suited to be applied at the core of time-critical industrial applications.

math.OC↗