Siemens (Austria)

Offshore wind generators are expected to exceed 15 MVA in the coming years, necessitating compact and light-weight step-up transformers. While high-temperature superconducting (HTS) transformers offer high efficiency and compact size, their complex AC loss behavior in large-scale devices is still not sufficiently understood. In this work, we present a comprehensive investigation on practical strategies for reducing AC loss in a 15 MVA HTS transformer based on the primary industrial criteria of efficiency, weight, and cost. We employed an efficient numerical method to model up to thousands of HTS tapes in low-voltage (LV) and high-voltage (HV) windings in detail. Using this framework, we evaluated the influence of key design and operating parameters on the AC loss, including winding height, axial gap between the cables, radial distance between HV and LV windings, winding height difference, number of parallel conductors, voltage per turn, magnetic flux diverters, and temperature. These analyses enable the development of four improved transformer designs at operating temperatures of 20 K, 70 K, and two configurations at 77 K that balance AC loss reduction with conductor cost and core weight. The results provide validated design principles and practical guidelines for developing next-generation HTS transformers that are compact, light-weight, cost-effective, and suitable for large-scale offshore wind energy systems. • Numerical framework developed for AC loss evaluation in HTS transformers • Impact of key operating and design parameters on AC loss quantified • Reduction of AC loss at winding ends is identified as key design strategy • Design trade-offs between AC loss, conductor length, and cost are analyzed • Optimized HTS transformer designs proposed for 20 K, 70 K, and 77 K

Building operations still rely heavily on manual review of unstructured maintenance records, which is slow, error-prone, and difficult to standardize across systems and teams. We present a knowledge-graph-centric middleware framework that mediates between heterogeneous IT/OT sources and analytics applications to support reliable, low-latency root-cause analysis. The framework integrates large language models (LLMs) with a domain knowledge graph (KG) under explicit quality criteria to ensure reliable construction and evolution. It begins with LLM-assisted schema generation, guided by Knowledge Graph Schema (KGS) metrics that ensure consistency, completeness, scalability, and minimal redundancy. Building on this foundation, the graph structure is refined using Graph Convolutional Network (GCN) embeddings combined with recursive KMeans clustering, which groups similar types and lowers query overhead. Next, a two-stage extraction pipeline converts unstructured text into a structured knowledge data frame and enriches relations by leveraging contextual proximity through weighted aggregation. To accommodate evolving data, the system supports incremental updates for both seen and unseen entities, preserving chronological consistency while avoiding disruptive schema revisions. Together, these components allow applications to remain decoupled from heterogeneous source formats while ensuring robust governance of both schema and data. The design further reduces query latency, improves retrieval precision, and reduces manual effort, thus supporting more timely and consistent maintenance decisions.

Abstract The scheduling of complex job shop manufacturing environments is a difficult and NP-hard optimization problem. It is tackled increasingly with AI-based methods, often by reinforcement learning approaches. So far, this is mostly an academic endeavor as the trust in such systems is limited. The neural networks representing the policy are often seen as a black box by domain experts. By proposing a holistic approach to make such policies more interpretable and explainable, we hope to bridge the gap between academia and industry. With increasing trust in the solution, it becomes closer to being deployed in a real manufacturing environment. We propose a combination of different statistical and well-established ML-based methodologies to analyze single state-action explanations as well as the overall strategy and apply them to pretrained agents on open-source benchmark simulation models. This allows us to deliver humanly understandable explanations for the policies of the analyzed agents.

This paper demonstrates the application and validation of a dynamic moisture model designed to predict the moisture behavior in transformers. The study includes a discussion of both laboratory measurements and field data, and it outlines necessary steps to obtain reliable simulation results and how to use them effectively.

Laser irradiation can induce local modulations of functional material properties, such as a decreased resistivity or a variation in reflectance. Recent studies investigated the laser-induced phase separation of 4 H-SiC into carbon and silicon on top of regrown SiC to customize its electrical conductivity for the application in electronic devices. To understand the physical processes leading to the laser-induced phase separation, time-resolved pump-probe measurements represent a suitable tool. This study advances the state of the art by characterizing the transient reflectance changes in 4 H-SiC upon irradiation by spatially resolved pump-probe reflectometry. Since the laser heating alters the reflectance of the sample, the spatially resolved measurement enables to observe the heat conduction from the irradiated to the non-irradiated areas, which sustains for several milliseconds. Numerical simulations of the temperature evolution reveal a restricted one-dimensional heat conduction into depth due to the broad lateral extent of the irradiated area. The associated sustained increased temperature within the irradiated area most certainly abets the feasibility of the phase separation. These findings offer practical insights for optimizing the applied laser parameters to tailor the material properties via phase separation.

Domain-specific heuristics are a crucial technique for the efficient solving of problems that are large or computationally hard. Answer Set Programming (ASP) systems support declarative specifications of domain-specific heuristics to improve solving performance. However, such heuristics must be invented manually so far. Inventing domain-specific heuristics for answer-set programs requires expertise with the domain under consideration and familiarity with ASP syntax, semantics, and solving technology. The process of inventing useful heuristics would highly profit from automatic support. This paper presents a novel approach to the automatic learning of such heuristics. We use Inductive Logic Programming (ILP) to learn declarative domain-specific heuristics from examples stemming from (near-)optimal answer sets of small but representative problem instances. Our experimental results indicate that the learned heuristics can improve solving performance and solution quality when solving larger, harder instances of the same problem.