We present a neural-network-based framework for the solution of three-dimensional boundary value problems where the solution is expressible in terms of harmonic potentials. The approach leverages the Whittaker integral formula, which allows representing the solution through functions that are holomorphic with respect to a suitable complex variable. These functions are subsequently approximated using holomorphic neural networks, which guaranty fulfillment of the holomorphicity requirement. A key feature of the proposed formulation is that the governing partial differential equations (PDEs) are satisfied exactly by construction. Therefore, in contrast to standard physics-informed neural networks, no residual minimization of PDEs is required in the interior of the domain, and training is based exclusively on boundary collocation points. The method is validated against three-dimensional Laplace and linear elasticity problems, where, in the latter case, displacement and stress fields are expressed via the Papkovich-Neuber potentials. The numerical results show an accurate approximation of both scalar and vector fields, with errors remaining controlled throughout the domain. Overall, the work demonstrates that the incorporation of analytical structures into neural network architectures provides a natural and effective framework for the meshless approximation of three-dimensional boundary value problems while preserving the underlying properties of the governing equations.
Modern power systems require fast and accurate dynamic simulations for stability assessment, digital twins, and real-time control, but classical ODE solvers are often too slow for large-scale or online applications. We propose a neural-operator framework for surrogate modeling of power system components, using Deep Operator Networks (DeepONets) to learn mappings from system states and time-varying inputs to full trajectories without step-by-step integration. To enhance generalization and data efficiency, we introduce Physics-Informed DeepONets (PI-DeepONets), which embed the residuals of governing equations into the training loss. Our results show that DeepONets, and especially PI-DeepONets, achieve accurate predictions under diverse scenarios, providing over 30 times speedup compared to high-order ODE solvers. Benchmarking against Physics-Informed Neural Networks (PINNs) highlights superior stability and scalability. Our results demonstrate neural operators as a promising path toward real-time, physics-aware simulation of power system dynamics.
Electric-assisted bicycles (e-bikes) represent a key example of how emerging technologies can influence travel patterns and broader sustainability transitions. While existing studies have examined motivations, substitution patterns, and behavioural impacts of e-bike adoption, limited evidence is available on the processes through which individuals adapt their travel behaviour after adoption. Moreover, the predominant reliance on either quantitative or qualitative approaches has constrained the ability to capture both statistical generalisation and context-specific behavioural mechanisms. This study addresses these gaps by employing a mixed-method approach to identifying the individual behaviour change process of e-bike adopters in Denmark. The analysis combines four waves of an online panel survey of 967 individuals intending to adopt an e-bike with before-and-after semi-structured interviews conducted with a subsample of 9 participants. Survey participants either adopted or did not adopt e-bikes, and their travel behaviour measures, motivational-attitudinal profiles, and interview narratives before and after potential adoption are used to understand how and why e‑bikes can support modal shifts and what conditions prevent behaviour change.
Bycatch of marine mammals in commercial fisheries presents conservation, operational, and regulatory challenges, particularly in the Greater North Sea where trawl fisheries targeting forage fish overlap with seal foraging habitats. This study evaluates the effectiveness of two seal excluder devices—the DPPO’s excluder tube and the PFA’s excluder grid—designed to reduce incidental seal captures in sandeel and herring fisheries, respectively. Using spatially explicit, industry-led self-reporting bycatch data, we identified bycatch hotspots and quantified reductions in seal captures: 85% in the sandeel fishery and an estimated 89% in the herring fishery, acknowledging broader uncertainty around the latter estimate. These findings demonstrate the potential of gear-based mitigation to address marine mammal bycatch in high-risk areas. The datasets used are unique in their origin, representing voluntary, operational-level reporting by fishers, with exceptionally high coverage of the respective fisheries. This collaboration between researchers and industry not only enhances data quality and quantity but also fosters stakeholder ownership, which is critical for the policy acceptance and take-up of conservation measures. Our results support the continued and potentially mandatory use of excluder devices in areas of elevated bycatch risk, contributing to ecosystem-based fisheries management and marine mammal protection.
Materials discovery is the linchpin to advance sustainable energy technologies, which mitigate resource scarcity and climate change. Establishing a carbon-neutral society relies heavily on the introduction of highly active and stable electrocatalysts for sustainable reaction engineering. In the current expression of high-throughput materials science, both high-throughput synthesis (HTS) and high-throughput characterization (HTC) are required to effectively cover an ever-increasing search space. HTS relies on combinatorial design and multimodal growth to overcome the constraints of traditional electrocatalyst fabrication. Concomitantly, HTC rapidly elucidates relationships between the composition, structure, and processing of electrocatalysts and their (multi-)functional properties, such as, activity, selectivity, and stability. In this review, we discuss emerging trends in high-throughput experimentation (HTE) to discover, design, and optimize electrocatalytically active materials for sustainable energy sources. We highlight the potential of individual HTS and HTC building blocks to be integrated into a holistic HTE workflow in conjunction with AI-driven feedback loops to truly accelerate the discovery process. Finally, we discuss future challenges and opportunities to improve the rational design strategy of electrocatalysts for sustainable energy sources based on integrated HTE workflows.
