arXiv subjects

Guijiang Li

Publications and source records attributed to Guijiang Li.

4 recordsLinked to original sources

arXiv2026

Smart Commander: A Hierarchical Reinforcement Learning Framework for Fleet-Level PHM Decision Optimization

Decision-making in military aviation Prognostics and Health Management (PHM) faces significant challenges due to the "curse of dimensionality" in large-scale fleet operations, combined with sparse feedback and stochastic mission profiles. To address these issues, this paper proposes Smart Commander, a novel Hierarchical Reinforcement Learning (HRL) framework designed to optimize sequential maintenance and logistics decisions. The framework decomposes the complex control problem into a two-tier hierarchy: a strategic General Commander manages fleet-level availability and cost objectives, while tactical Operation Commanders execute specific actions for sortie generation, maintenance scheduling, and resource allocation. The proposed approach is validated within a custom-built, high-fidelity discrete-event simulation environment that captures the dynamics of aircraft configuration and support logistics.By integrating layered reward shaping with planning-enhanced neural networks, the method effectively addresses the difficulty of sparse and delayed rewards. Empirical evaluations demonstrate that Smart Commander significantly outperforms conventional monolithic Deep Reinforcement Learning (DRL) and rule-based baselines. Notably, it achieves a substantial reduction in training time while demonstrating superior scalability and robustness in failure-prone environments. These results highlight the potential of HRL as a reliable paradigm for next-generation intelligent fleet management.

cs.LG↗

arXiv2026

LiteInception: A Lightweight and Interpretable Deep Learning Framework for General Aviation Fault Diagnosis

General aviation fault diagnosis and efficient maintenance are critical to flight safety; however, deploying deep learning models on resource-constrained edge devices poses dual challenges in computational capacity and interpretability. This paper proposes LiteInception--a lightweight interpretable fault diagnosis framework designed for edge deployment. The framework adopts a two-stage cascaded architecture aligned with standard maintenance workflows: Stage 1 performs high-recall fault detection, and Stage 2 conducts fine-grained fault classification on anomalous samples, thereby decoupling optimization objectives and enabling on-demand allocation of computational resources. For model compression, a multi-method fusion strategy based on mutual information, gradient analysis, and SE attention weights is proposed to reduce the input sensor channels from 23 to 15, and a 1+1 branch LiteInception architecture is introduced that compresses InceptionTime parameters by 70%, accelerates CPU inference by over 8x, with less than 3% F1 loss. Furthermore, knowledge distillation is introduced as a precision-recall regulation mechanism, enabling the same lightweight model to adapt to different scenarios--such as safety-critical and auxiliary diagnosis--by switching training strategies. Finally, a dual-layer interpretability framework integrating four attribution methods is constructed, providing traceable evidence chains of "which sensor x which time period." Experiments on the NGAFID dataset demonstrate a fault detection accuracy of 81.92% with 83.24% recall, and a fault identification accuracy of 77.00%, validating the framework's favorable balance among efficiency, accuracy, and interpretability.

cs.AI↗

arXiv2012

Compensation effect and magnetostriction in CoCr2-xFexO4

The magnetic compensation and magnetostriction properties in Fe doped CoCr2O4 samples have been investigated. Structural and magnetic measurements imply that the doped Fe3+ ions initially occupy the B1(Cr) sites when x<0.1, and then mainly take the A(Co) sites. This behaviour results in a role conversion of magnetic contributors and a composition compensation between two competitively magnetic sublattices at x=0.1. Temperature dependence compensation has also been found in the samples with x=0.1~0.22, with the compensation temperature in the range of 40~104 K. The Fe3+ doping also modulates the exchange interaction of the system and prevents the formation of long range conical order of spins. The magnetoelectric transition temperature at 23K in CoCr2O4 is shifted to lower temperature by increasing the dopants. The magnetostriction effect in this system has been observed for the first time. The strain has a maximum value of about 280ppm at x=0.4. The magnetostriction is in consistent with the behaviour of the two magnetic compensations.

cond-mat.mtrl-sci↗

arXiv2012

Stable magnetostructural coupling with tunable magnetoresponsive effects in hexagonal phase-transition ferromagnets

The magnetostructural coupling between the structural and the magnetic transition plays a crucial role in magnetoresponsive effects in a martensitic-transition system. A combination of various magnetoresponsive effects based on this coupling may facilitate the multifunctional applications of a host material. Here, we demonstrate a possibility to obtain a stable magnetostructural coupling in a broad temperature window from 350 to 70 K, showing tunable magnetoresponsive effects, based on simultaneous manipulation of the phase stability and the magnetic structure by suitable chemical substitution of iron in MnNiGe. The resultant MnNiGe:Fe exhibits a magnetic-field-induced martensitic transition from paramagnetic austenite to ferromagnetic martensite, featuring (i) a large volume increase, (ii) a distinct magnetization change, (iii) small thermal hysteresis and (iv) a giant negative magnetocaloric effect. The results indicate that stable magnetostructural coupling is accessible in hexagonal phase-transition systems to attain the magnetoresponsive effects with broad tunability.

cond-mat.mtrl-sci↗