Searcharxiv⌕ Search

arXiv subjects

Junyong Liu

Publications and source records attributed to Junyong Liu.

5 recordsLinked to original sources

RaGCD: Reliability-Aware 2D-3D Multimodal Generalized Category Discovery

Generalized Category Discovery (GCD) aims to recognize known categories while discovering novel ones from partially labeled data. Existing GCD studies primarily focus on unimodal observations or image-text interactions, leaving the complementarity between visual appearance and explicit 3D geometry underexplored. To exploit complementary visual appearance and spatial geometry, we investigate GCD with paired images and point clouds. However, the relative reliability of these modalities varies across samples, so unreliable evidence can compromise both multimodal fusion and cross-modal knowledge transfer. This challenge is particularly pronounced for novel categories, where labeled supervision is unavailable. To address it, we propose RaGCD, a reliability-aware framework for 2D-3D multimodal GCD. Specifically, Reliability-aware Modality Allocation (RMA) combines prediction-derived reliability cues with modality-specific features to adaptively allocate image and point-cloud contributions at the channel level. Reliability-conditioned Alignment (RCA) further reweights feature-level and prediction-level alignment and selects reliable observations for prototype-level alignment, regulating cross-modal interaction according to sample reliability. Extensive experiments on ModelNet40, Washington RGB-D, and ShapeNet-ViPC demonstrate that RaGCD outperforms the compared methods in overall and novel-category clustering accuracy. Ablation studies support the complementary roles of RMA and RCA, while controlled corruption experiments show that modality allocation shifts toward the less corrupted modality. These results support the effectiveness of reliability-aware interaction for category discovery from paired appearance and geometry.

eess.SP↗

The Role of Insurance in Power-System Risk Management: A Critical Review

Modern power systems are becoming more renewable-rich, digitalized, and interconnected, making risk propagation and economic consequences increasingly complex. Physical mitigation remains the primary means of risk control, but technical and economic limits leave residual risk. Insurance can complement these measures by transferring associated economic losses and influencing power-system decisions through risk-responsive terms. Despite growing interest, the literature remains fragmented across power-system engineering, actuarial science, and risk finance. This review develops an integrated physical-financial perspective for reviewing insurance in power systems. It examines how power system risks generate residual economic losses and the conditions under which these losses can be insured and transferred. It then develops an actuarial framework for insured-loss assessment and pricing. Insurance-aware decision-making is synthesized through four coupling levels describing how financial protection interacts with power-system decisions. Further progress is constrained by weak physical-financial loss validation, incomplete representation of systemic and evolving portfolio risk, and limited empirical and actuarial grounding of insurance-decision feedback. Research priorities include stronger physical-financial data linkage, forward-looking portfolio assessment, multi-timescale decision feedback, and empirical validation. Addressing these priorities would help move power-system insurance from fragmented applications toward a more coherent approach to residual-risk management in future power systems.

math.OC↗

Renewable energy exporting consumption-oriented transfer limit switching control: A unsupervised learning-based method

A method for generating unsupervised conditional mapping rules for multi-inter-corridor transfer limits and their integration into unit commitment through banding-switching is proposed in this paper. The method starts by using Ant colony clustering(ACC) to identify different operating modes with renewable energy penetration. For each sub-pattern, coupling inter-corridors are determined using correlation coefficients. An algorithm for constructing coupled inter-corridors' limits boundaries, employing grid partitioning, is proposed to establish conditional mappings from sub-patterns to multi-inter-corridor limits. Additionally, a banding matching model is proposed, incorporating distance criteria and the Big-M method. It also includes a limit-switching method based on Lagrange multipliers. Case studies on the IEEE 39-node system illustrate the effectiveness of this method in increasing consumption of renewable energy and reducing operational costs while adhering to stability verification requirements.

eess.SY↗

Physics-Guided Graph Neural Networks for Real-time AC/DC Power Flow Analysis

The increasing scale of alternating current and direct current (AC/DC) hybrid systems necessitates a faster power flow analysis tool than ever. This letter thus proposes a specific physics-guided graph neural network (PG-GNN). The tailored graph modelling of AC and DC grids is firstly advanced to enhance the topology adaptability of the PG-GNN. To eschew unreliable experience emulation from data, AC/DC physics are embedded in the PG-GNN using duality. Augmented Lagrangian method-based learning scheme is then presented to help the PG-GNN better learn nonconvex patterns in an unsupervised label-free manner. Multi-PG-GNN is finally conducted to master varied DC control modes. Case study shows that, relative to the other 7 data-driven rivals, only the proposed method matches the performance of the model-based benchmark, also beats it in computational efficiency beyond 10 times.

eess.SY↗

Analytic Deep Learning-based Surrogate Model for Operational Planning with Dynamic TTC Constraints

The increased penetration of wind power introduces more operational changes of critical corridors and the traditional time-consuming transient stability constrained total transfer capability (TTC) operational planning is unable to meet the real-time monitoring need. This paper develops a more computationally efficient approach to address that challenge via the analytical deep learning-based surrogate model. The key idea is to resort to the deep learning for developing a computationally cheap surrogate model to replace the original time-consuming differential-algebraic constraints related to TTC. However, the deep learning-based surrogate model introduces implicit rules that are difficult to handle in the optimization process. To this end, we derive the Jacobian and Hessian matrices of the implicit surrogate models and finally transfer them into an analytical formulation that can be easily solved by the interior point method. Surrogate modeling and problem reformulation allow us to achieve significantly improved computational efficiency and the yielded solutions can be used for operational planning. Numerical results carried out on the modified IEEE 39-bus system demonstrate the effectiveness of the proposed method in dealing with com-plicated TTC constraints while balancing the computational efficiency and accuracy.

eess.SY↗