检索结果-内蒙古大学图书馆

Numerical simulation of underexpanded supersonic jets impingement on an inclined flat plate

THERMOPHYSICS AND AEROMECHANICS 2021年第4期28卷 479-486页

作者： Epikhin, A. S. Elizarova, T. G. Ivannikov Inst Syst Programming RAS Moscow Russia Bauman Moscow State Tech Univ Moscow Russia Keldysh Inst Appl Math RAS Moscow Russia

The paper presents the results of numerical simulation of complex shock-wave structures arising from an under-expanded jets impingement on an inclined flat plate. The plate deflection angles equal to 45 degrees, 60 degrees, and 90 degrees are investigated, which corresponds to different types of shock waves interference. The gas-dynamic characteristics are calculated using the OpenFOAM software package with the QGDFoam solver. This solver based on a system of regularized (quasi-gas dynamic) equations. The numerical simulation results of the flowfields and pressure distributions at the plate are compared against the results of the Kurganov-Tadmor scheme and with experimental data. The features of the applied numerical approach for simulation of complex shock-wave structures with triplet points, contact discontinuities, and low-entropy flows are identified.

关键词： numerical simulation underexpanded jet interaction shock-wave structures regularized (quasi-gas dynamic) equations QGDFoam OpenFOAM

来源：评论

学校读者我要写书评

暂无评论

New aspects of black box conditional gradient: Variance reduction and one point feedback

引用

CHAOS SOLITONS & FRACTALS 2024年 189卷

作者： Veprikov, Andrey Bogdanov, Alexander Minashkin, Vladislav Beznosikov, Aleksandr Moscow Inst Phys & Technol Moscow Russia RAS Ivannikov Inst Syst Programming Moscow Russia Sber AI Lab Moscow Russia Innopolis Univ Innopolis Russia

This paper deals with the black-box optimization problem. In this setup, we do not have access to the gradient of the objective function, therefore, we need to estimate it somehow. We propose a new type of approximation JAGUAR, that memorizes information from previous iterations and requires (9 (1) oracle calls. We implement this approximation in the Frank-Wolfe and Gradient Descent algorithms and prove the convergence of these methods with different types of zero-order oracle. Our theoretical analysis covers scenarios of non-convex, convex and PL-condition cases. Also in this paper, we consider the stochastic minimization problem on the set Q with noise in the zero-order oracle;this setup is quite unpopular in the literature, but we prove that the JAGUAR approximation is robust not only in deterministic minimization problems, but also in the stochastic case. We perform experiments to compare our gradient estimator with those already known in the literature and confirm the dominance of our methods.

关键词： Optimization Zero-order optimization Frank-Wolfe methods

来源：评论

学校读者我要写书评

暂无评论

Numerical investigation of flow around two cylinders in tandem above a scoured bed

引用

PHYSICS OF FLUIDS 2022年第8期34卷 1-27页

作者： Yang, Bo Liang, Bingchen Zhang, Qin Meng, Xun Matvey, Kraposhin Epikhin, Andrey Ocean Univ China Coll Engn Qingdao Peoples R China Ocean Univ China Shandong Prov Key Lab Ocean Engn Qingdao Peoples R China RAS Ivannikov Inst Syst Programming Moscow Russia

Flow mechanisms around two cylinders in tandem arrangement above a scoured bed have been investigated using the three-dimensional unsteady Navier-Stokes equations with the Spalart-Allmaras improved delayed detached-eddy simulation model. A turbulent inlet boundary layer generation method is adopted to obtain more realistic inlet boundary conditions. First, uniform flow over a single cylinder at Re = 3900 and flow over a single cylinder above scoured beds at Re = 6000 were simulated to validate the numerical model, boundary layer generation method, and mesh density effect. Second, two cylinders in the tandem arrangement above scoured beds with six different pitch ratios L/D are investigated numerically in terms of instantaneous vortex characteristics, the hydrodynamic force, and time-averaged flow fields. The simulation results in scoured beds are compared with simulations under the near flat wall and wall-free conditions. The major findings can be summarized as follows. (1) When L / D & LE;2.0, the wake of two tandem cylinders is dominated by the intermittent shedding, and the downstream sand dune in the scoured bed hinders the Karman vortex formation at the rear of the downstream cylinder. Lift force fluctuations of the two cylinders have small amplitudes, and their spectra show a multi-peak distribution and no dominant peak frequency in the spectrum of the upstream cylinder. A squarish cavity-like recirculation zone is formed between two cylinders at L / D <= 2.0. (2) When L / D >= 3.0, the periodic vortex shedding is evident in the wake of the upstream cylinder, and the small sand berm between two cylinders has an impact on the bottom shear layer of the upstream cylinder. The downstream cylinder is periodically impacted by the vortices shed from the upstream cylinder. Lift force spectra of the upstream and downstream cylinders have the same peak frequency. (3) Due to the influence of scoured bed and the inlet boundary layer, the time-averaged lift coefficient of

关键词： Wakes

来源：评论

学校读者我要写书评

暂无评论

Homing Sequence Derivation With Quantified Boolean Satisfiability

引用

IEEE TRANSACTIONS ON COMPUTERS 2022年第3期71卷 696-711页

作者： Tu, Kuan-Hua Wang, Hung-En Jiang, Jie-Hong R. Kushik, Natalia Yevtushenko, Nina Natl Taiwan Univ Grad Inst Elect Engn Taipei 10617 Taiwan Inst Polytech Paris Serv Repartis Architectures MOdelisat Validat Adm CNRS Telicom SudParis F-91000 Evry France RAS Higher Sch Econ Ivannikov Inst Syst Programming Moscow 109004 Russia

Homing sequence derivation for nondeterministic finite state machines (NFSMs) has important applications in software/hardware system testing and verification. Unlike prior methods based on explicit tree-based search, in this article we formulate the derivation of a preset/adaptive homing sequence in terms of quantified Boolean formula (QBF) solving. This formulation exploits compact circuit representation of NFSMs and QBF encoding of the existence condition of homing sequence for effective computation. The implicit circuit representation effectively avoids explicit state enumeration, and can be more scalable. Different encoding schemes and QBF solvers are evaluated for their suitability for the homing sequence derivation. Experiments on various computation methods and benchmarks show the generality and feasibility of a proposed approach.

关键词： Testing Vegetation Encoding Monitoring Complexity theory Benchmark testing Labeling Homing sequence nondeterministic finite state machine quantified Boolean formula

来源：评论

学校读者我要写书评

暂无评论

A novel numerical optimality technique to find the optimal results of Volterra integral equation of the second kind with discontinuous kernel

引用

APPLIED NUMERICAL MATHEMATICS 2023年第1期186卷 202-212页

作者： Noeiaghdam, Samad Sidorov, Denis Dreglea, Aliona Irkutsk Natl Res Tech Univ Baikal Sch BRICS Ind Math Lab Irkutsk 664074 Russia South Ural State Univ Dept Appl Math & Programming Lenin Prospect 76 Chelyabinsk 454080 Russia Russian Acad Sci Melentiev Energy Syst Inst Siberian Branch Irkutsk 664033 Russia

In this study we consider linear and nonlinear Volterra integral equations (VIEs) of the second kind with discontinuous kernel. A novel iterative method using floating point arithmetic (FPA) is presented to solve the problem. Also a convergence theorem and error analysis of the method are presented. The main novelty of this study is to validate the results using the CESTAC (Controle et Estimation Stochastique des Arrondis de Calculs) method which is based on stochastic arithmetic (SA). Moreover instead of applying usual mathematical softwares we use the CADNA (Control of Accuracy and Debugging for Numerical Applications) library which must be done on Linux operating system using C, C++ or Fortran codes. Applying the method and the library we will be able to find the optimal results such as optimal error, optimal iteration of the method and optimal approximation. Proving the main theorem of the CESTAC method we show the equality between the number of common significant digits (NCSDs) of two successive iterations with the exact and approximate solutions. Thus it can help us to apply a new termination criterion instead of absolute error to show the accuracy and efficiency of the method.(c) 2023 IMACS. Published by Elsevier B.V. All rights reserved.

关键词： Volterra integral equation Discontinuous kernel Iterative method Convergence analysis CESTAC method CADNA library

来源：评论

学校读者我要写书评

暂无评论

Deep neural networks generalization and fine-tuning for 12-lead ECG classification

引用

BIOMEDICAL SIGNAL PROCESSING AND CONTROL 2024年 93卷

作者： Avetisyan, Aram Tigranyan, Shahane Asatryan, Ariana Mashkova, Olga Skorik, Sergey Ananev, Vladislav Markin, Yury Russian Acad Sci Ivannikov Inst Syst Programming Moscow Russia Russian Armenian Univ Yerevan Armenia Novgorod State Univ Veliky Novgorod Russia

Numerous studies focus on diagnosing heart diseases using deep learning methods applied to 12-lead electrocardiographic (ECG) records. However, these studies often face limitations due to reliance on specific datasets, which vary in size and parameters such as patient metadata, the number of doctors annotating ECGs, types of devices used for ECG recording, data preprocessing techniques, etc. It is well-known that high-quality deep neural networks trained on one ECG dataset may not necessarily perform well on other datasets or in different clinical settings. In this paper, we propose a methodology designed to enhance the quality of heart disease prediction regardless of the dataset. We achieve this by first training neural networks on a variety of labeled datasets, then fine-tuning for specific datasets, significantly improving prediction accuracy. We demonstrate its applicability by training various neural networks on a large private dataset TIS, which contains a wide range of ECG records from multiple hospitals, and on a relatively smaller public dataset, PTB-XL. Our results show that networks trained on a large dataset improves classification performance. Furthermore, these networks fine-tuned on PTB-XL outperform those trained exclusively on smaller datasets. Additionally, we made the weights of our pre-trained models publicly available, enabling researchers and clinicians to adapt these models to their specific datasets.

关键词： Electrocardiography Time series analysis Deep learning ECG classification Fine-tuning Pre -trained neural networks

来源：评论

学校读者我要写书评

暂无评论

Numerical simulation of the unsteady aerodynamic loads on the tail fin in the vortex breakdown flow

引用

PROGRESS IN COMPUTATIONAL FLUID DYNAMICS 2021年第5期21卷 274-282页

作者： Epikhin, Andrey RAS Ivannikov Inst Syst Programming Moscow 109004 Russia Bauman Moscow State Tech Univ Dept Mech Engn Moscow 105005 Russia

This work presents numerical simulation results of the flow around an airbrake-tail fin system of an aircraft. In this research, airbrake location and deflection angles effects on the unsteady aerodynamic loads acting on the tail fin that leads to the buffeting have been studied. The main flow field structures and their transformation features are investigated. The area of the maximal pressure applied to the tail fin has been determined. Analysing the obtained data during a series of numerical simulations allows to conclude that aerodynamic loads on the tail fin increase by five times. This conclusion has been made for the considered configuration of the airbrake-tail fin system when the airbrake deflection angle grows bigger. Change in the distance between the airbrake and the tail fin does not significantly affect the force loads. The efficiency of using the airbrake under different operating conditions has been estimated.

关键词： vortex flow aerodynamics airbrake tail fin buffeting unsteady loads OpenFOAM

来源：评论

学校读者我要写书评

暂无评论

About One Property of Number Rank in RNS

About One Property of Number Rank in RNS

引用

IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (ElConRus)

作者： Babenko, Mikhail Golimblevskaia, Elena North Caucasus Fed Univ Stavropol Russia Ivannikov Inst Syst Programming Moscow Russia

ISBN: (纸本)9781665404761

an approximate value of the number rank is used to increase the speed of homomorphic encryption algorithms. The use of an approximate rank of a number allows reducing the computational complexity of the data encryption algorithm, but at the same time leads to a loss of accuracy. In this paper, we investigate the properties of the number rank. We have shown that it is possible to reduce the accuracy of calculating the approximate rank of a number, but the homomorphic encryption scheme will have the same properties.

关键词： Residue Number system number rank homomorphic encryption Chinese remainder theorem modular arithmetic

来源：评论

学校读者我要写书评

暂无评论

Applying language models to algebraic topology: generating simplicial cycles using multi-labeling in Wu's formula 41

Applying language models to algebraic topology: generating s...

引用

41st International Conference on Machine Learning (ICML)

作者： Brilliantov, Kirill Pavutnitskiy, Fedor Pasechnyuk, Dmitry Magai, German ETH Zurich Switzerland Beijing Inst Math Sci & Applicat Beijing Peoples R China Mohamed Bin Zayed Univ Artificial Intelligence Abu Dhabi U Arab Emirates Russian Acad Sci Ivannikov Inst Syst Programming Moscow Russia HSE Univ Moscow Russia Noeon Res Tokyo Japan

Computing homotopy groups of spheres has long been a fundamental objective in algebraic topology. Various theoretical and algorithmic approaches have been developed to tackle this problem. In this paper we take a step towards the goal of comprehending the group-theoretic structure of the generators of these homotopy groups by leveraging the power of machine learning. Specifically, in the simplicial group setting of Wu's formula, we reformulate the problem of generating simplicial cycles as a problem of sampling from the intersection of algorithmic datasets related to Dyck languages. We present and evaluate language modelling approaches that employ multi-label information for input sequences, along with the necessary group-theoretic toolkit and non-neural baselines.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Adaptive Methods for Variational Inequalities with Relatively Smooth and Reletively Strongly Monotone Operators

引用

programming AND COMPUTER SOFTWARE 2023年第6期49卷 485-492页

作者： Ablaev, S. S. Stonyakin, F. S. Alkousa, M. S. Pasechnyk, D. A. Moscow Inst Phys & Technol Inst Skii Per 9 Dolgoprudnyi 141701 Russia Vernadsky Crimean Fed Univ Pr Akad Vernadskogo 4 Simferopol 295007 Russia Natl Res Univ Higher Sch Econ Ul Myasnitskaya 20 Moscow 101000 Russia Russian Acad Sci Ivannikov Inst Syst Programming Trusted Artificial Intelligence Res Ctr Ul Solzhenitsyna 25 Moscow 109004 Russia

This paper is devoted to some adaptive methods for variational inequalities with relatively smooth and relatively strongly monotone operators. Based on the recently proposed proximal version of the extragradient method for this class of problems, we study in detail the method with adaptively selected parameter values. The rate of convergence of this method is estimated. The result is generalized to the class of variational inequalities with relatively strongly monotone delta-generalized smooth operators. For the ridge regression problem and variational inequality associated with box-simplex games, numerical experiments are carried out to demonstrate the effectiveness of the proposed technique for adaptive parameter selection during the execution of the algorithm.

关键词： Variational techniques

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：