A large-scale microgrid cluster (MGC) using centralized energy management faces computational and data processing burdens, making it rather hard to obtain real-time optimal energy dispatch strategies. Network partitioning method can reduce such complexity by dividing the MGC into self-adequate partitions. However, existing dynamic network partitioning methods normally update the network partitioning results at fixed time intervals, which limits the ability to adapt to spatiotemporal variations in output of renewable energy sources (RESs), load demand, and grid topology. To address this, a dynamic network partitioning method based on an improved genetic algorithm (IGA) combined with depth-first search (DFS) is proposed. IGA is employed to optimize the partitioning index, and DFS is subsequently applied to obtain the network partitioning result. Furthermore, a dynamic update mechanism is designed to adaptively update the partitioning results according to spatiotemporal variations in system conditions. Simulation results on the modified IEEE 123-bus system show that the proposed method achieves better partitioning quality than the genetic algorithm (GA), spectral clustering, and Louvain algorithms while maintaining computational efficiency.

Although the Internet of Robotic Things (IoRT) has enhanced the productivity of robotic systems in conjunction with the Internet of Things (IoT), it does not inherently support seamless human–robot collaboration. This article presents HuBotVerse, a unified framework designed to foster the evolution of the Internet of Human and Intelligent Robotic Things (IoHIRT). HuBotVerse is advantageous due to its unique features, including security, user-friendliness, manageability, and its open source nature. Moreover, this framework can seamlessly integrate various human–robot interaction (HRI) interfaces to facilitate collaborative control between humans and robots. Here, we emphasize a digital twin (DT)-based mixed reality (MR) interface, which enhances teleoperation efficiency by offering users an intuitive and immersive way to interact with. To evaluate the effectiveness of HuBotVerse, we conducted user studies based on a pick-and-place task. Feedback was gathered through questionnaires, complemented by a quantitative analysis of key performance metrics, user experience, and the National Aeronautics and Space Administration Task Load Index (NASA-TLX). Results indicate that the fusion of MR and HuBotVerse within a comprehensive framework significantly improves the efficiency and user experience of teleoperation. Moreover, the follow-up questionnaires reflect the advantages of the HuBotVerse framework in terms of evident user-friendliness, manageability, and usability in home-care or health-care applications. For codes, project videos, tutorials, technical details, case studies, and Q&A, please check our website ( https://sites.google.com/view/iohirtplusmr/home ).

The expansion of transnational higher education creates opportunities for global programme delivery, but also introduces challenges in maintaining academic standards, pedagogical integrity, and a consistent student experience across international branch campuses. These issues are particularly acute in interdisciplinary STEM fields such as financial technology (fintech), where expertise is distributed and academic staff are in high demand. This paper presents a case study of the design and planned delivery of the MSc Financial Technology with Data Science at the University of Bristol's Mumbai Enterprise Campus (MEC), launching in August 2026. Originally developed in the UK within a Faculty of Science and Engineering, the programme was awarded Best Academic Programme of the Year Supporting Fintech Companies at the Fintech Awards South West. It integrates multiple disciplines, including data science, artificial intelligence, cloud-based data engineering, and blockchain development, within an industry-aligned curriculum focused on real-world financial applications. To address the challenges of transnational STEM delivery, the paper proposes and operationalises an integrated pedagogical-operational model that combines block teaching, flying faculty, and co-teaching within a single coordinated framework. The model is designed to address three interrelated challenges: maintaining academic consistency across locations, accessing and coordinating interdisciplinary expertise under staffing constraints, and integrating teaching across geographically distributed academic teams. The paper outlines the design and implementation of this model and introduces a longitudinal comparative study to evaluate the pedagogical and operational implications of delivery format during the initial years of programme delivery, providing a basis for assessing its effectiveness and informing subsequent phases of empirical evaluation. The proposed approach offers a potentially transferable framework for transnational STEM delivery in contexts characterised by constrained local academic capacity and distributed interdisciplinary expertise.

Temporal Graph Learning (TGL) is crucial for capturing the evolving nature of stock markets. Traditional methods often ignore the interplay between dynamic temporal changes and static relational structures between stocks. To address this issue, we propose the Dynamic Graph Representation with Contrastive Learning (DGRCL) framework, which integrates dynamic and static graph relations to improve the accuracy of stock trend prediction. Our framework introduces two key components: the Embedding Enhancement (EE) module and the Contrastive Constrained Training (CCT) module. The EE module focuses on dynamically capturing the temporal evolution of stock data, while the CCT module enforces static constraints based on stock relations, refined within contrastive learning. This dual-relation approach allows for a more comprehensive understanding of stock market dynamics. Our experiments on two major U.S. stock market datasets, NASDAQ and NYSE, demonstrate that DGRCL significantly outperforms state-of-the-art TGL baselines. Ablation studies indicate the importance of both modules. Overall, DGRCL not only enhances prediction ability but also provides a robust framework for integrating temporal and relational data in dynamic graphs. Code and data are available for public access.