http://bura.brunel.ac.uk/handle/2438/13035 2026-04-18T04:45:00Z 2026-04-18T04:45:00Z Mao, C Wen, C Wang, Z Liu, W Yang, J http://bura.brunel.ac.uk/handle/2438/33156 2026-04-16T02:00:37Z 2026-03-21T00:00:00Z

Title: Tldiag: a federated learning-based DB-mixer for privacy-preserving and fault diagnosis in transmission lines Authors: Mao, C; Wen, C; Wang, Z; Liu, W; Yang, J Abstract: As a critical component of smart grids where faults occur frequently, the rapid and accurate diagnosis of transmission line faults is essential for enhancing system reliability and response efficiency. However, existing centralized diagnostic methods face significant challenges, including data privacy concerns, limited feature representation capabilities, and class imbalance issues in real-world power grids. To address these issues, this paper proposes an adaptive diagnostic framework based on federated learning and a dual-branch mixer (TLDiag) to address the privacy-sensitive cross-domain fault diagnosis problem in dynamic grid environments. Specifically, a distributed federated learning architecture is designed to ensure data privacy by enabling localized model training through collaborative client–server interactions. Furthermore, a dual-branch mixer, i.e., DB-Mixer, guided by a dual-branch attention mechanism (DBAM), is developed to enhance feature representation by jointly modeling spatial and channel-wise information. Additionally, an adaptive dual-field loss (ADF loss) is introduced, incorporating dynamic task weighting and physical constraints to effectively mitigate class imbalance and improve diagnostic robustness. Extensive experiments conducted on the IEEE 5-bus system demonstrate that TLDiag achieves superior performance in both fault type classification (97.53%) and fault location identification (98.34%), while exhibiting stable convergence across varying client scales. Compared to baseline methods such as MDCNN and CNN-LSTM, TLDiag significantly outperforms in accuracy and robustness. By deeply integrating federated learning with the physical characteristics of power systems, this approach offers a high-accuracy, privacy-preserving solution for real-time fault diagnosis in smart grid scenarios. Description: Data availability: The data supporting the findings of this study are publicly available from the IEEE Data Port (https://ieee-dataport.org/documents/transmission-line-fault-using-line-voltages-and-currents-features).

2026-03-21T00:00:00Z Caputo, S Sacchi, L Tucker, A http://bura.brunel.ac.uk/handle/2438/32970 2026-03-13T03:00:29Z 2026-01-01T00:00:00Z

Title: Integrating Generative Models with Pseudo-Time to Build Realistic Image Trajectories of Eye Disease Authors: Caputo, S; Sacchi, L; Tucker, A Abstract: ... Description: ...

2026-01-01T00:00:00Z Zhu, Z Chen, Z Zhu, C Si, W Wang, F http://bura.brunel.ac.uk/handle/2438/32841 2026-02-24T03:00:34Z 2026-02-09T00:00:00Z

Title: Optimizing potential-based reward automata in partially observable reinforcement learning using genetic local search Authors: Zhu, Z; Chen, Z; Zhu, C; Si, W; Wang, F Abstract: Partially observable reinforcement learning extends the reinforcement learning framework to environments in which agents have limited visibility of the state space, making it particularly relevant for applications in robotics and autonomous vehicle navigation. However, a primary challenge in partially observable reinforcement learning is defining effective reward functions that can guide the learning process despite partial observability. To address this challenge, this paper introduces a novel approach for constructing potential-based reward automata by employing genetic local search methods. Specifically, our method constructs these automata from compressed representations of exploration trajectories, which succinctly capture critical decision points and essential state transitions while eliminating redundant steps. By optimizing trajectory samples and shortening agent trajectories to their crucial transitions, our technique significantly reduces computational overhead. Formally, we define the learning objective as an optimization problem aimed at maximizing the log-likelihood of future observations while simultaneously minimizing the structural complexity of the learned reward automata. Furthermore, by incorporating value-based strategies to estimate potential values within the reward automata, our approach improves learning efficiency and facilitates the identification of optimal reward structures. We empirically evaluate our proposed method on seven partially observable grid-world benchmarks. Experimental results demonstrate that our method achieves superior performance relative to state-of-the-art reward automata-based techniques, exhibiting both accelerated learning speeds and higher accumulated rewards. Additionally, our genetic local search algorithm consistently outperforms comparative heuristic methods in terms of learning curves and reward accumulation. Description: Highlights: • Introduce a new RA learning mechanism with reward constraints for better strategies. • Develop evolutionary algorithms to optimize RA and policies in various environments. • Conduct experiments showing superior performance in six partially observable domains. • Analyze exploration–exploitation balance and environmental randomness effects. • Demonstrate the stability and efficiency of our genetic local search method.; Data availability: No data was used for the research described in the article.

2026-02-09T00:00:00Z Tolani, S Mitra, D Dantu, SC Kumar, A http://bura.brunel.ac.uk/handle/2438/32693 2026-01-23T03:00:36Z 2025-12-09T00:00:00Z

Title: Navigating the labyrinth of drugging the disordered Authors: Tolani, S; Mitra, D; Dantu, SC; Kumar, A Abstract: Intrinsically disordered proteins (IDPs) and intrinsically disordered regions (IDRs) are central players in many cellular processes, which make their involvement across cellular dysregulation and diseases evident. Unlike structured proteins, they carry out their functions mainly through interactions with a variety of partners. Recent progress, from large-scale proteome studies to detailed atomic-level investigations, has shed light on how their dynamic nature shapes complex structure-function relationships. Despite this flexibility, IDPs and IDRs are also being explored as promising drug targets. However, successful therapeutic efforts first require a careful understanding of their dynamic behavior and interactions. In this review, we outline how protein disorder influences both health and disease, and highlight relentless approaches aimed at turning this flexibility into opportunities for drug discovery. Description: Data availability: No datasets were generated or analysed during the current study.

2025-12-09T00:00:00Z