arXiv subjects

Feiyang Zhang

Publications and source records attributed to Feiyang Zhang.

18 recordsLinked to original sources

arXiv2026

The Radioactive Background of the JUNO Calibration System

The Jiangmen Underground Neutrino Observatory (JUNO) experiment is a reactor antineutrino detector employing 20 kton of ultra-pure liquid scintillator to determine the neutrino mass ordering and to precisely measure oscillation parameters. The total singles background rate from radioactivity is required to be below 10 Hz in the energy range of 0.7-12 MeV within the fiducial volume for reactor neutrino analysis. The calibration system is designed to characterize the detector energy and position responses, while several of its components are located close to the target and may contribute to the background budget. Therefore, extensive material screening and selection are required to construct a low-background calibration system and to ensure that its contribution remains within the design requirements. In this work, a comprehensive study of the radioactive background induced by the calibration system is presented, including material radioactivity measurements using high-purity germanium detectors and neutron activation analysis techniques, detailed Monte Carlo simulations to evaluate the background, and comparisons with in-situ detector data to validate the predictions. In this data analysis, dedicated spatial selection methods are developed to isolate calibration-related contributions and to suppress the liquid scintillator background. The total radioactivity contribution from the calibration system is estimated to be less than 76 mHz, which satisfies the requirement of 200 mHz (2% of the total background budget). The results from in-situ data are found to be consistent with the expectations based on material assay and simulation within uncertainties. These results demonstrate that the calibration-induced background is well understood, in agreement between data and simulation, and negligible for reactor antineutrino measurements in JUNO.

physics.ins-det↗

arXiv2026

DBPnet: Damper Characteristics-Based Bayesian Physics-Informed Neural Network for Wheel Load Estimation

Advanced driver assistance systems (ADAS) play an important role in modern automotive intelligence, significantly enhancing vehicle safety and stability. The performance of ADAS critically relies on accurate and reliable vehicle state estimation, particularly from vehicle dynamic sensors. Among these signals, wheel load is a key variable for chassis control and safety-critical functions, yet it remains difficult to estimate robustly due to complex suspension geometry, nonlinear dynamics, and measurement noise. To address this issue, we propose DBPnet, a Bayesian physics-informed neural network (PINN) with a physics-aware embedding module inspired by damper characteristics. First, this paper presents a suspension linkage-level modeling (SLLM) approach that constructs a nonlinear instantaneous dynamic model by explicitly considering the complex geometric structure of the suspension. Building upon SLLM, Bayesian inference is integrated into the PINN to effectively cope with noise and uncertainty in the vehicle chassis system, thereby improving the model's robustness. Then, a physics-informed loss function is employed to ensure consistency with fundamental physical principles, while the damper characteristics-inspired embedding module extracts temporal variation features of input signals and incorporates them into each layer of the PINN, ensuring that physical observations guide the neural network without being constrained by fixed physical models. Extensive evaluations on high-fidelity simulations and real-world experiments demonstrate that our DBPnet consistently achieves lower RMSE and MaxError than baseline methods. These results highlight the potential of our DBPnet to advance wheel load estimation and contribute to the development of more reliable ADAS actuator functions.

eess.SY↗

arXiv2026

PILD: Physics-Informed Learning via Diffusion

Diffusion models have emerged as powerful generative tools for modeling complex data distributions, yet their purely data-driven nature limits applicability in engineering and scientific problems where physical laws must be respected. This paper proposes Physics-Informed Learning via Diffusion (PILD), a framework that unifies diffusion modeling and physical constraints through a probabilistic residual formulation with a virtual residual observation sampled from a Laplace distribution. To make this formulation practical under noisy diffusion states, we introduce a Jensen-gap-aware adaptive residual scale, which reduces the bias induced by residual likelihood marginalization. Additionally, we develop a physics-conditional alignment mechanism for conditional tasks that encourages intermediate latent representations to remain consistent with the observation conditions during denoising. The proposed framework is concise, modular, and broadly applicable to problems governed by ordinary differential equations, partial differential equations, as well as algebraic equations or inequality constraints. Extensive experiments across engineering and scientific tasks show that PILD improves physical fidelity and predictive accuracy over representative physics-informed and diffusion-based baselines.

cs.LG↗

arXiv2026

Damper-B-PINN: Damper Characteristics-Based Bayesian Physics-Informed Neural Network for Vehicle State Estimation

Accurate state estimation is fundamental to intelligent vehicles. Wheel load, one of the most important chassis states, serves as an essential input for advanced driver assistance systems (ADAS) and exerts a direct influence on vehicle stability and safety. However, wheel load estimation remains challenging due to the complexity of chassis modeling and the susceptibility of nonlinear systems to noise. To address these issues, this paper first introduces a refined suspension linkage-level modeling approach that constructs a nonlinear instantaneous dynamic model by explicitly considering the complex geometric structure of the suspension. Building upon this, we propose a damper characteristics-based Bayesian physics-informed neural network (Damper-B-PINN) framework to estimate dynamic wheel load, which leverages the suspension dynamics as physical guidance of PINN while employing Bayesian inference to mitigate the effects of system noise and uncertainty. Moreover, a damper-characteristic physics conditioning (DPC) module is designed for embedding physical prior. The proposed Damper-B-PINN is evaluated using both high-fidelity simulation datasets generated by CarSim software and real-world datasets collected from a Formula Student race car. Experimental results demonstrate that our Damper-B-PINN consistently outperforms existing methods across various test conditions, particularly extreme ones. These findings highlight the potential of the proposed Damper-B-PINN framework to enhance the accuracy and robustness of dynamic wheel load estimation, thereby improving the reliability and safety of ADAS applications.

cs.AI↗

arXiv2025

Monolithic Expandable-FOV Metalens Enabled by Radially Gradient-Tilted Meta-Atoms

Metalens, as the most promising and applicable emerging optical device, has long been constrained by the limited field of view (FOV). Recent studies employing phase engineering or multi-layer strategies have made some progress, but they all rely on upright meta-atoms. This leads us to consider whether tilted meta-atoms could represent a promising yet underexplored approach for enhancing the field of view? In this work, we introduce the control of the tilt angle of meta-atoms as a new degree of freedom into the design of metalenses and propose a wide-field-of-view (WFOV) metalens design framework utilizing radially gradient-tilted meta-atoms. Based on the proposed method, we designed two WFOV metalenses with distinct tilt configurations to meet specific application requirements of efficiency or precision. Simulation results demonstrate that both designs, exhibiting distinct performance characteristics, achieve diffraction-limited focusing within a 120 degree FOV. Additionally, the FOV can be expanded with this method by tuning the tilt angle configurations of meta-atoms and the metalens diameter. These findings will establish a promising pathway towards compact optical systems capable of combining ultra-wide angular coverage with high-resolution imaging.

physics.optics↗

arXiv2025

Direct Observation of d-Wave Superconducting Gap Symmetry in Pressurized La3Ni2O7-delta Single Crystals

The recent discovery of superconductivity in pressure-stabilized bulk La3Ni2O7-delta, with a critical temperature (Tc) exceeding 77 K, has opened a new frontier in high-temperature superconductivity research beyond cuprates. Yet, the superconducting gap amplitude and symmetry, the key parameters to characterize a superconductor, remain elusive due to the overwhelming challenges of gap studies under high pressure. Here, we introduce in situ directional point-contact spectroscopy conducted under truly hydrostatic pressure, enabling the direct mapping of the superconducting gap in pressurized La3Ni2O7-delta single crystals. Depending on the junction orientation, differential conductance (dI/dV) spectra exhibit distinct V-shaped quasiparticle features and a sharp zero-bias peak, indicating a predominant d-wave-like pairing symmetry. Measurement of the c-axis gap amplitude Delta yields a gap-to-Tc ratio of 2Delta/kBTc = 4.2(5), positioning La3Ni2O7-delta firmly among unconventional, nodal high-Tc superconductors. These findings set stringent constraints on theoretical models for nickelate superconductors and establish a robust spectroscopic approach for understanding superconductors under extreme pressures.

cond-mat.supr-con↗

arXiv2025

Skyrmions with customized intensity distribution and trajectory

Optical skyrmions, which are topological protection quasi-particles with nontrivial textures, hold a pivotal focus in current structured light research for their potential in diverse applications. In this work, the angular spectrum theory is first introduced into the generation of optical skyrmions and modulation of the intensity and trajectory of skyrmions at will. We propose a novel theoretical approach for the generation of skyrmions, including Neel-type, Bloch-type, anti-type and 2nd order. The simultaneous and independent modulation of intensity distribution and trajectory of isolated skyrmions is first achieved with the combination of phase-shifting theory with angular spectrum theory. By controlling the displacement phase factor (DPF), the customized shape of skyrmions array with controllable intensity distribution and trajectory is also generated. Our findings in this work allow a greater exploration of skyrmions, which promote applications in particle manipulation and high-density storage.

physics.optics↗

arXiv2025

Fast Fourier Correlation is a Highly Efficient and Accurate Feature Attribution Algorithm from the Perspective of Control Theory and Game Theory

The study of neural networks from the perspective of Fourier features has garnered significant attention. While existing analytical research suggests that neural networks tend to learn low-frequency features, a clear attribution method for identifying the specific learned Fourier features has remained elusive. To bridge this gap, we propose a novel Fourier feature attribution method grounded in signal decomposition theory. Additionally, we analyze the differences between game-theoretic attribution metrics for Fourier and spatial domain features, demonstrating that game-theoretic evaluation metrics are better suited for Fourier-based feature attribution. Our experiments show that Fourier feature attribution exhibits superior feature selection capabilities compared to spatial domain attribution methods. For instance, in the case of Vision Transformers (ViTs) on the ImageNet dataset, only $8\%$ of the Fourier features are required to maintain the original predictions for $80\%$ of the samples. Furthermore, we compare the specificity of features identified by our method against traditional spatial domain attribution methods. Results reveal that Fourier features exhibit greater intra-class concentration and inter-class distinctiveness, indicating their potential for more efficient classification and explainable AI algorithms.

cs.LG↗

arXiv2024

Customized calibration sources in the JUNO experiment

We customized a laser calibration system and four radioactive $γ$-ray calibration sources for the Jiangmen Underground Neutrino Observatory (JUNO), a 20-kton liquid scintillator-based neutrino detector. The laser source system was updated to realize the isotropic light emission timing within $\pm0.25$~nsec level and to allow the tuning of the laser intensity covering more than four orders of magnitude. In addition, methods to prepare four different radioactive sources ($^{18}{\rm F}$, $^{40}{\rm K}$, $^{226}{\rm Ra}$, and $^{241}{\rm Am}$), covering energies from O(10)~keV to O(1)~MeV, for the JUNO detector were established in this study. The radioactivity of each source and the risk of impurities leaking into the detector from the source were confirmed to meet the experimental requirements.

physics.ins-det↗

arXiv2024

Comprehensive and Reliable Feature Attribution for Diverse Modalities and Models via Frequency-Domain Insights

Personalized Federal learning(PFL) allows clients to cooperatively train a personalized model without disclosing their private dataset. However, PFL suffers from Non-IID, heterogeneous devices, lack of fairness, and unclear contribution which urgently need the interpretability of deep learning model to overcome these challenges. These challenges proposed new demands for interpretability. Low cost, privacy, and detailed information. There is no current interpretability method satisfying them. In this paper, we propose a novel interpretability method \emph{FreqX} by introducing Signal Processing and Information Theory. Our experiments show that the explanation results of FreqX contain both attribution information and concept information. FreqX runs at least 10 times faster than the baselines which contain concept information.

cs.LG↗

arXiv2023

Simulation Software of the JUNO Experiment

The Jiangmen Underground Neutrino Observatory (JUNO) is a multi-purpose experiment, under construction in southeast China, that is designed to determine the neutrino mass ordering and precisely measure neutrino oscillation parameters. Monte Carlo simulation plays an important role for JUNO detector design, detector commissioning, offline data processing, and physics processing. The JUNO experiment has the world's largest liquid scintillator detector instrumented with many thousands of PMTs. The broad energy range of interest, long lifetime, and the large scale present data processing challenges across all areas. This paper describes the JUNO simulation software, highlighting the challenges of JUNO simulation and solutions to meet these challenges, including such issues as support for time-correlated analysis, event mixing, event correlation and handling the simulation of many millions of optical photons.

hep-ex↗

arXiv2021

Unambiguously Resolving the Potential Neutrino Magnetic Moment Signal at Large Liquid Scintillator Detectors

Non-vanishing electromagnetic properties of the neutrinos have been predicted by many theories beyond the Standard Model, and an enhanced neutrino magnetic moment can have profound implications for fundamental physics. The XENON1T experiment recently detected an excess of electron recoil events in the 1-7 keV energy range, which can be compatible with solar neutrino magnetic moment interaction at a most probable value of $μ_ν = 2.1 \times 10^{-11} μ_{\text{B}}$. However, tritium backgrounds and solar axion interaction in this energy window are equally plausible causes. Upcoming multi-tonne noble liquid detectors will test these scenarios more in depth, but will continue to face similar ambiguity. We report a unique capability of future large liquid scintillator detectors to help resolve the potential neutrino magnetic moment scenario. With a liquid scintillator exposure of $O$(100) kton$\cdot$year, a sensitivity of $μ_ν < 10^{-11} μ_{\text{B}}$ can be reached at an energy threshold greater than 40 keV, where no tritium or solar axion events but only neutrino magnetic moment signal is still present.

hep-ex↗

arXiv2021

Construction and Simulation Bias Study of The Guide Tube Calibration System for JUNO

A Guide Tube Calibration System (GTCS) has been designed for the Jiangmen Underground Neutrino Observatory (JUNO), in order to measure the detector energy response near the outer radius of the active volume. Recently, a prototype system has been constructed and tested, and the calibration algorithm has also been studied to evaluate the risk when the simulation tuning and the error control fail. In this paper, we first report its construction and the performance tests in the lab. Then the influence on the global energy measurement caused by the simulation bias of GTCS is discussed, in order to make sure the algorithm is qualified.

physics.ins-det↗

arXiv2021

The Automatic Calibration Unit in JUNO

This paper describes the design and construction of the automatic calibration unit (ACU) for the JUNO experiment. The ACU is a fully automated mechanical system. It is capable of deploying multiple radioactive sources, an ultraviolet (UV) laser source, or an auxiliary sensor such as a temperature sensor, one at a time, into the central detector of JUNO along the central axis. It is designed as a primary tool to precisely calibrate the energy scale of detector, aligning timing for the photosensors, and partially monitoring the position-dependent energy scale variations.

physics.ins-det↗

arXiv2021

A Precise Method to Determine the Energy Scale and Resolution using Gamma Calibration Sources in a Liquid Scintillator Detector

Gamma sources are routinely used to calibrate the energy scale and resolution of liquid scintillator detectors. However, non-scintillating material surrounding the source introduces energy losses, which may bias the determination of the centroid and width of the full absorption peak. In this paper, we present a general method to determine the true gamma centroid and width to a relative precision of 0.03\% and 0.50\%, respectively, using energy losses predicted by the Monte Carlo simulation. In particular, the accuracy of the assumed source geometry is readily obtained from the fit. The method performs well with experimental data in the Daya Bay detector.

physics.ins-det↗

arXiv2020

Cable Loop Calibration System for Jiangmen Underground Neutrino Observatory

A cable loop source calibration system is developed for the Jiangmen Underground Neutrino Observatory, a 20 kton spherical liquid scintillator neutrino experiment. This system is capable of deploying radioactive sources into different positions of the detector in a vertical plane with a few-cm position accuracy. The design and the performance of the prototype are reported in this paper.

physics.ins-det↗

arXiv2019

Design of Guide Tube Calibration System for JUNO Experiment

Jiangmen Underground Neutrino Observatory (JUNO) is designed to determine the neutrino mass hierarchy using a 20 kton liquid scintillator detector. To calibrate detector boundary effect, the Guide Tube Calibration System (GTCS) has been designed to deploy a radioactive source along a given longitude on the outer surface of the detector. In this paper, we studied the physics case of this system via simulation, which leads to a mechanical design.

physics.ins-det↗

arXiv2018

Laser Calibration System in JUNO

A UV laser calibration system is developed for the Jiangmen Underground Neutrino Observatory (JUNO), a 20-kton liquid scintillator-based, long baseline reactor anti-neutrino experiment. This system is capable of delivering fast laser pulses (<1 ns) into the detector with a photon intensity ranging from hundreds of keV to a few TeV equivalent energy. An independent intensity monitoring system is developed for this laser, which has achieved a < 0.3% relative uncertainty. Such a system will be used to calibrate the response of the JUNO photosensors and electronics in situ to a sub-percent precision.

physics.ins-det↗