为助力师生追踪力学、数学与应用物理学领域前沿动态，提升科研创新能力，拓宽学术交流视野，理学院和科技处特邀请国内外知名专家学者莅临我校开展系列学术讲座。现将具体安排通知如下，欢迎各位师生踊跃参加。
报告一
报告时间：2025 年12月13日 上午8:50
报告地点：新能源楼 317
报告专家：Bobomurat AHMEDOV 教授/院士（乌兹别克斯坦科学院）
报告题目：Study of Ionospheric Perturbations In D- And F-Layers Using VIF Receiver and TashkentKitab GPS Stations
摘要：Abstract: Tashkent International Heliophysical Year (IHY) station is a member of the Atmospheric Weather Electromagnetic System for Observation, Modelling and Education (AWESOME) network being operated globally to study the ionosphere and the magnetosphere with the help of electromagnetic waves in the Very Low Frequency (VLF) band. Regular monitoring of the D- and Flayers of ionosphere over Central Asia territory is being performed on a permanent basis starting year 2008 when one VLF receiver and two SuperSID receivers were provided to the Uzbekistan IHY cite by Stanford University. The results obtained at Tashkent IHY station are applied to earthquake electromagnetic precursors, lightning, and Solar flares and to ionospheric disturbances originating from gamma ray flares of Soft Gamma-Ray Repeaters connected with evolution of strongly magnetized neutron stars believed as magnetars. Regular monitoring of the D-layer of ionosphere over Central Asia territory has been performed on the permanent basis. Several Solar events are observed and the analysis has shown that there is simultaneous correlation between the times of change of amplitude of the waves and the Solar flares. Features of the lightning discharge generated by radio atmospherics are studied and its effectiveness in D-region ionosphere diagnostics is examined. We have mainly analysed GPS derived TEC disturbances from two GPS stations located in Tashkent and Kitab, for possible earthquake ionospheric precursors. The ionospheric anomalies were observed during strong local earthquakes (M greater than 5.0) which occurred mostly in and around Uzbekistan in seismically active zone within 1000 km from the observing GPS stations located in Tashkent and Kitab. The solar and geomagnetic conditions were quiet during occurrence of the selected earthquakes. We produced TEC time series over both sites and apply them to detect anomalous TEC signals preceding or accompanying the earthquakes. The results show anomalous enhancements which are examined in the earthquakes. In general the anomalies occurred 1-6 days before the earthquakes as ionospheric electromagnetic precursors. To identify the anomalous values of TEC we calculated differential TEC (dTEC). dTEC is obtained by subtracting 15 days backward running mean of vTEC from the values of observed vTEC at ach epoch.
报告二
报告时间：2025 年12月13日上午9:20
报告地点：新能源楼 317
报告专家：Anatoly Kudryavtsev 教授（圣彼得堡国立大学/哈尔滨工业大学）
报告题目：Towards to the Creation Electromagnetic Wave Amplifier in Plasma Media with Inverse Electron Distribution Function
摘要：Back in 1961, it was predicted that a plasma with an inverse distribution, not for bound states as in a laser, but for an ensemble of isotropic free electrons (EDF) also has unique properties, such as absolute negative conductivity and negative absorption of radiation, and thus amplifies electromagnetic waves over a wide frequency range. A logical question arises: why were bound-state inversion and lasers quickly created and developed into a giant industry, while the inverse EDF of unbounded free electrons has not yet to be realized? In our opinion, the reason for the current uncertain situation is related with a significant simplification of the problem of solving the Boltzmann kinetic equation by neglecting all terms with spatial gradients (including the ambipolar field).In this report, we will review the current state of affairs in solving this problem and outline solutions for a practical demonstration of the inverse EDF. It has been shown that since real laboratory plasma is always spatially inhomogeneous, finding the EDF and analyzing its formation requires solving the full kinetic equation, which depends on both the velocity and the spatial variables (x, v, t).A criterion for the formation of an inverse EDF is obtained. It was shown that in order to obtain inversion in the field accelerating the electron to the anode, EDF in the low-energy region (and, correspondingly, the electron density) must decrease toward the anode. Kinetic simulations and analysis have shown that one of the simplest to implement inverse EDFs is a glow discharge in a hollow cathode (HC), when the electron density decreases in an axial electric field accelerating electron toward the anode. Another potential object with an inverse EDF is a positive column of glow discharge with an expanding cross-section.
报告三
报告时间：2025 年12月13日上午 10:10
报告地点：新能源楼317
报告专家：边伟教授（哈尔滨工业大学）
报告题目：Sparse Optimization Problems with Cardinality Penalties: Continuous Relaxation Method and Proximal-based Algorithms
摘要：Cardinality penalties represent a foundational approach for enforcing sparsity in optimization models. In this work, we propose a unified framework of continuous relaxation methods and design different proximal-friendly algorithms to four critical classes of sparse optimization problems with cardinality penalties: 1) element-wise sparse models, 2) group-structured sparse models, 3) rank-regularized models, and 4) sparse minmax models. These problems are inherently challenged by discontinuous objective functions, which complicate both analysis and computation. To bridge this gap, we first develop the exact continuous relaxation models that share the same global solutions as the original problems. A rigorous theoretical analysis is then conducted to characterize the relationships between global solutions, local solutions, and stationary points across all problem classes. Notably, we provide a more rigorous characterization to the global solutions, leading to two key definitions: (i) the strong local solutions for cardinality problems, and (ii) the strong stationary points of the continuous relaxation models, which exhibit a one-to-one correspondence. Next, leveraging this theoretical foundation, we design the proximal-based algorithms for the first three problem classes. These algorithms are provably guaranteed to converge to the strong local solutions, with some convergence rates explicitly analyzed. Some comprehensive numerical experiments further validate the efficiency of the proposed algorithms, demonstrating superior performance in both solution quality and computational scalability compared to baseline methods.
报告四
报告时间：2025 年12月13日上午10:40
报告地点：新能源楼317
报告专家：朱全新教授（湖南师范大学）
报告题目：Stabilization of Stochastic Nonlinear Delay Systems Driven by Lévy Processes
摘要：In this talk, we address the stabilization problem for a class of stochastic nonlinear delay systems driven by Lévy processes. By introducing a novel event-triggered strategy and employing techniques from stochastic analysis, we establish sufficient conditions for the practical pth moment exponential stability of the closed-loop system. Unlike many existing works, our approach does not rely on the global Lipschitz condition or linear matrix inequalities (LMIs). Moreover, in contrast to results focusing either on discrete-time stochastic systems or on continuous-time systems with Brownian motion alone, our study specifically targets the event-triggered sampling control of continuous-time systems subject to both time delays and Lévy noise, which encompasses both diffusion and jump components. This leads to a more general framework with broader applicability.
报告五
报告时间：2025 年12月13日下午14:00
报告地点：新能源楼 317
报告专家：袁承勋教授（哈尔滨工业大学）
报告题目：Research on Gas Component Detection Based on Plasma Technology
摘要：With the continuous development of China’s economy and the increasing demands for the quality of living environments, the need for online analysis of process gases has been steadily rising across various fields, such as industrial process control, atmospheric environment monitoring, exhaust gas diagnostics, and biomedical applications. Consequently, the application scenarios for gas component detection have become more diverse. This study focuses on the design, numerical modeling, and experimental investigation of a micro hollow cathode discharge device, aiming to realize parameter diagnostics and material composition detection based on micro hollow cathode discharge plasma. In terms of numerical modeling, a hybrid model combining electron kinetic descriptions with heavy particle fluid descriptions was developed to simulate plasma parameters and fast electron dynamics in the negative glow region. The model accurately predicted the fast electron energy distribution function characterized by distinctive electron peaks formed through Penning ionization reactions and superelastic collisions. Its reliability was verified via probe diagnostics. Experimentally, a micro hollow cathode discharge device was designed to generate a stable non-local negative glow plasma under atmospheric pressure conditions. By covering the outer surface of the cathode, the current-voltage characteristics of the discharge were enhanced and the negative glow plasma region was expanded, thereby improving the resolution of the fast electron energy spectrum. This enabled the analysis and detection of gas components based on plasma electron energy spectroscopy technology. Additionally, plasma diagnostics including measurements of electron density, electron temperature, and electron energy distribution function were conducted using wall probes.This research deepens the understanding of plasma dynamics and provides a powerful tool for gas phase component analysis in non-equilibrium plasmas.
报告六
报告时间：2025 年12月13日下午 14:30
报告地点：新能源楼 317
报告专家：张强教授（南京大学）
报告题目：Optimal Error Estimate of Discontinuous Galerkin Method for One-Dimensional Linear Hyperbolic Equations with Degenerate Points Moving Along Space-Time Curves
摘要：In this talk, we shall consider the optimal error estimate on a semi-discrete discontinuous Galerkin method with purely upwind numerical flux to solve a one-dimensional linear variable coefficient hyperbolic equation, where the flow speed has different signs when passing through the degenerate points. The purpose of this talk is to present a rigorous proof for a special case that the degenerate points move along smooth curves depending on both space and time variables. To address the analysis difficulties due to the change of the flow speed’s sign regarding time, we propose a novel analysis framework with the help of a time-dependent projection based on the hybrid application of Gauss-Radau projections and a carefully designed time-space subdivision associated with the spatial mesh.
报告七
报告时间：2025 年12月13日下午 15:20
报告地点：新能源楼 317
报告专家：李小舟 副研究院（电子科技大学）
报告题目：Spectral Symbol Analysis of Space–Time FE–DG Discretizations: Foundations for Parallel-inTime Solvers
摘要：In this talk, we study the space–time discretization of the multidimensional heat equation using a discontinuous Galerkin (DG) method in time and finite element (FE) approximations in space. The DG time discretization induces a characteristic block structure in the fully discrete system, motivating a detailed spectral analysis. Within the framework of Generalized Locally Toeplitz (GLT) theory, we show that the resulting space–time matrices admit an asymptotic spectral symbol as the spatial and temporal mesh parameters tend to zero. This symbol compactly captures the interaction between the temporal DG operator and the spatial FE discretization and elucidates key structural properties of the scheme. Numerical experiments confirm the theoretical predictions. The analysis provides structural insight into FE–DG space–time systems and forms a theoretical basis for the future development of advanced solvers, including time-parallel approaches.
报告八
报告时间：2025 年12月13日下午 15:50
报告地点：新能源楼 317
报告专家：李佳副教授（哈尔滨工业大学）
报告题目：Optimal Error Estimates for the Discontinuous Finite Element Method with Generalized Numerical Flux
摘要：The discontinuous finite element method (DG method) is a high-order accurate numerical approach for solving partial differential equations. Its numerical scheme is characterized by the use of an independent set of polynomial bases on each computational element, allowing the basis functions to be discontinuous across element boundaries. This feature makes the DG method particularly effective for problems with discontinuous solutions. Numerical fluxes are an essential component of the DG method’s numerical scheme. This report studies the DG method employing generalized numerical fluxes. The generalized flux uses information from the numerical solutions on both sides of an element boundary, providing greater flexibility to the numerical scheme. However, this also couples the information across elements in the error analysis, making the use of traditional analytical tools insufficient to prove error estimates. By constructing a special global projection, this work ultimately establishes the optimal error estimate properties of the discontinuous finite element method with generalized numerical flux.
科技处、理学院
2025.12.12