YDYi Dong

Yi Dong

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Yi Dong has a total of 51 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Multimodal LLM Framework with Time-Series-Guided Visual Routing for Ultra-Short-Term Solar Forecasting

    by Yiheng Li, Maizhen Ning, Xiaoyang Chen, Guoping Ruan, Junwei Wu, Fangyu Wu, Yi Dong, Eng Gee Lim, Yang Du

    Accurate ultra-short-term solar irradiance forecasting is important for photovoltaic power management, storage scheduling, and the secure operation of power systems with increasing renewable penetration. However, historical irradiance measurements primarily reflect atmospheric variations after they have affected the observation site, providing limited information about approaching clouds that may induce rapid irradiance ramps. Although ground-based sky images provide complementary spatial information, using complete visual representations can introduce irrelevant regions and weaken effective multimodal interaction. This study proposes a time-series-guided multimodal forecasting framework that integrates historical irradiance, sky images, statistical prompts, and a frozen LLM. Historical irradiance sequences are divided into temporal patches and reprogrammed into the LLM representation space, while a hierarchical visual encoder extracts spatial tokens from sky images. A time-series-conditioned visual router evaluates the relevance of individual image patches according to the current temporal representation and retains task-relevant regions through hard Top-K selection with a differentiable straight-through estimator. The selected visual tokens are then integrated with the temporal representation through TS-guided cross-modal attention, in which temporal tokens serve as queries and visual tokens provide keys and values. A residual connection preserves the original temporal information, and statistical prompt embeddings are subsequently concatenated with the fused temporal tokens before processing by the frozen LLM. Experiments on the Folsom dataset achieve nRMSE values of 14.89%, 17.19%, and 19.91% at 10-, 30-, and 60-min forecasting horizons, respectively, outperforming all evaluated baselines. Ablation and sensitivity analyses further confirm the contributions of time-series-conditioned visual selection and explicit cross-modal fusion, while an intermediate number of selected patches provides a favorable balance between information retention and visual redundancy. Cross-dataset validation on the Stanford SKIPP'D dataset further demonstrates the applicability of the proposed framework to PV power forecasting under a different site and prediction target.

  • Beyond shortcuts: Mitigating spurious correlations in radiological diagnosis with causal intervention

    by Xinyi Zeng, Jia Wang, Jian Zhang, Yi Dong, Wei Wang, Yanji Jiang, Yonggang Zhang, Haiyang Zhang

  • Hierarchical Testing With Rabbit Optimization for Industrial Cyber-Physical Systems

    by Jinwei Hu, Zezhi Tang, Xin Jin, Benyuan Zhang, Yi Dong, Xiaowei Huang

    This paper presents HERO (Hierarchical Testing with Rabbit Optimization), a novel black-box adversarial testing framework for evaluating the robustness of deep learning-based Prognostics and Health Management systems in Industrial Cyber-Physical Systems. Leveraging Artificial Rabbit Optimization, HERO generates physically constrained adversarial examples that align with real-world data distributions via global and local perspective. Its generalizability ensures applicability across diverse ICPS scenarios. This study specifically focuses on the Proton Exchange Membrane Fuel Cell system, chosen for its highly dynamic operational conditions, complex degradation mechanisms, and increasing integration into ICPS as a sustainable and efficient energy solution. Experimental results highlight HERO's ability to uncover vulnerabilities in even state-of-the-art PHM models, underscoring the critical need for enhanced robustness in real-world applications. By addressing these challenges, HERO demonstrates its potential to advance more resilient PHM systems across a wide range of ICPS domains.


  • Open Access

    A Temporal Scale Transformer Framework for Precise Remaining Useful Life Prediction in Fuel Cells

    by Zezhi Tang, Xiaoyu Chen, Xin Jin, Benyuan Zhang, Wenyu Liang, Tengyang Gong, Yi Dong, Zepeng Liu

    In the task of exploring Predictive Health Management (PHM) strategies for Proton Exchange Membrane Fuel Cells (PEMFC), the Transformer model, due to its outstanding performance in numerous fields, has been widely applied within data-driven models. However, for time series analysis tasks, using the self-attention mechanism in the Transformer model results in a complexity of the input sequence squared, often leading to low computational efficiency. Additionally, the model faces the challenge of effectively extracting both global long-term dependencies and local detail features simultaneously. To address these issues, this paper proposes a novel solution in the form of an improved version of the inverted Transformer (iTransformer) - the Temporal Scale Transformer (TSTransformer). Specifically, unlike traditional Transformers, which model each timestep as an input token, the new embedding method maps sequences of different lengths into tokens at different stages for inter-sequence modelling, allowing multivariate correlations to be described through the attention mechanism and sequence representations to be encoded through feed forward networks (FFN). Notably, incorporating a one-dimensional convolutional layer into the multivariate attention at different stages for multi-level scaling of the K and V matrices not only enhances the ability to extract local features and capture characteristics at different temporal scales but also reduces the number of tokens and computational costs. This enables the model to effectively capture the ability to extract local features in time series data. Through rigorous experimental research on the model and comparative analysis with established models (such as Long Short-Term Memory (LSTM), iTransformer, and Transformer). The results collectively highlight the potential of the TSTransformer model as a powerful tool for advancing PHM in the renewable energy sector and effectively solving the problems associated with pure Transformer backbone models in data-driven time series tasks.

  • Open Access

    Nash equilibrium seeking in non-cooperative heterogeneous multi-robot systems via output regulation

    by Yi Dong, Zhongguo Li, Shoaib Ehsan, Zhengtao Ding, Xiaowei Huang


  • Reachability Verification Based Reliability Assessment for Deep Reinforcement Learning Controlled Robotics and Autonomous Systems

    by Yi Dong, Xingyu Zhao, Sen Wang, Xiaowei Huang

    Deep Reinforcement Learning (DRL) has achieved impressive performance in robotics and autonomous systems (RAS). A key challenge to its deployment in real-life operations is the presence of spuriously unsafe DRL policies. Unexplored states may lead the agent to make wrong decisions that could result in hazards, especially in applications where DRL-trained end-to-end controllers govern the behaviour of RAS. This letter proposes a novel quantitative reliability assessment framework for DRL-controlled RAS, leveraging verification evidence generated from formal reliability analysis of neural networks. A two-level verification framework is introduced to check the safety property with respect to inaccurate observations that are due to, e.g., environmental noise and state changes. Reachability verification tools are leveraged locally to generate safety evidence of trajectories. In contrast, at the global level, we quantify the overall reliability as an aggregated metric of local safety evidence, corresponding to a set of distinct tasks and their occurrence probabilities. The effectiveness of the proposed verification framework is demonstrated and validated via experiments on real RAS.


  • Privacy-Preserving Distributed Learning for Residential Short-Term Load Forecasting

    by Yi Dong, Yingjie Wang, Mariana Gama, Mustafa A. Mustafa, Geert Deconinck, Xiaowei Huang

    In the realm of power systems, the increasing involvement of residential users in load forecasting applications has heightened concerns about data privacy. Specifically, the load data can inadvertently reveal the daily routines of residential users, thereby posing a risk to their property security. While federated learning (FL) has been employed to safeguard user privacy by enabling model training without the exchange of raw data, these FL models have shown vulnerabilities to emerging attack techniques, such as deep leakage from gradients and poisoning attacks. To counteract these, we initially employ a secure-aggregation (SecAgg) algorithm that leverages multiparty computation cryptographic techniques to mitigate the risk of gradient leakage. However, the introduction of SecAgg necessitates the deployment of additional subcenter servers for executing the MPC protocol, thereby escalating computational complexity and reducing system robustness, especially in scenarios where one or more subcenters are unavailable. To address these challenges, we introduce a Markovian switching-based distributed training framework, the convergence of which is substantiated through rigorous theoretical analysis. The distributed Markovian switching (DMS) topology shows strong robustness toward the poisoning attacks as well. Case studies employing real-world power system load data validate the efficacy of our proposed algorithm. It not only significantly minimizes communication complexity but also maintains accuracy levels comparable to traditional FL methods, thereby enhancing the scalability of our load forecasting algorithm.


  • Open Access

    Reliability Assessment and Safety Arguments for Machine Learning Components in System Assurance

    by Yi Dong, Wei Huang, Vibhav Bharti, Victoria Cox, Alec Banks, Sen Wang, Xingyu Zhao, Sven Schewe, Xiaowei Huang

    The increasing use of Machine Learning (ML) components embedded in autonomous systems—so-called Learning-Enabled Systems (LESs)—has resulted in the pressing need to assure their functional safety. As for traditional functional safety, the emerging consensus within both, industry and academia, is to use assurance cases for this purpose. Typically assurance cases support claims of reliability in support of safety, and can be viewed as a structured way of organising arguments and evidence generated from safety analysis and reliability modelling activities. While such assurance activities are traditionally guided by consensus-based standards developed from vast engineering experience, LESs pose new challenges in safety-critical application due to the characteristics and design of ML models. In this article, we first present an overall assurance framework for LESs with an emphasis on quantitative aspects, e.g., breaking down system-level safety targets to component-level requirements and supporting claims stated in reliability metrics. We then introduce a novel model-agnostic Reliability Assessment Model (RAM) for ML classifiers that utilises the operational profile and robustness verification evidence. We discuss the model assumptions and the inherent challenges of assessing ML reliability uncovered by our RAM and propose solutions to practical use. Probabilistic safety argument templates at the lower ML component-level are also developed based on the RAM. Finally, to evaluate and demonstrate our methods, we not only conduct experiments on synthetic/benchmark datasets but also scope our methods with case studies on simulated Autonomous Underwater Vehicles and physical Unmanned Ground Vehicles.