According to Hypersphere World-Universe Model, dark matter particles DIRACs are magnetic dipoles consisting of two Dirac’s monopoles. We conclude that DIRACs are the subject of Maxwell’s equations. So-called “auxil...According to Hypersphere World-Universe Model, dark matter particles DIRACs are magnetic dipoles consisting of two Dirac’s monopoles. We conclude that DIRACs are the subject of Maxwell’s equations. So-called “auxiliary” magnetic field intensity H is indeed current density of magnetic dipoles. The developed approach to magnetic field can explain a wealth of discovered phenomena in Cosmic Magnetism: a dark magnetic field, the large-scale structure of the Milky Way’s magnetic field, and other magnetic phenomena which are only partly related to objects visible in other spectral ranges.展开更多
With the help of skew-symmetric differential forms the hidden properties of the mathematical physics equations are revealed. It is shown that the equations of mathematical physics can describe the emergence of various...With the help of skew-symmetric differential forms the hidden properties of the mathematical physics equations are revealed. It is shown that the equations of mathematical physics can describe the emergence of various structures and formations such as waves, vortices, turbulent pulsations and others. Such properties of the mathematical physics equations, which are hidden (they appear only in the process of solving these equations), depend on the consistency of derivatives in partial differential equations and on the consistency of equations, if the equations of mathematical physics are a set of equations. This is due to the integrability of mathematical physics equations. It is shown that the equations of mathematical physics can have double solutions, namely, the solutions on the original coordinate space and the solutions on integrable structures that are realized discretely (due to any degrees of freedom). The transition from the solutions of the first type to one of the second type describes discrete transitions and the processes of origin of various structures and observable formations. Only mathematical physics equations, on what no additional conditions such as the integrability conditions are imposed, can possess such properties. The results of the present paper were obtained with the help of skew-symmetric differential forms.展开更多
This paper deals with Mckean-Vlasov backward stochastic differential equations with weak monotonicity coefficients.We first establish the existence and uniqueness of solutions to Mckean-Vlasov backward stochastic diff...This paper deals with Mckean-Vlasov backward stochastic differential equations with weak monotonicity coefficients.We first establish the existence and uniqueness of solutions to Mckean-Vlasov backward stochastic differential equations.Then we obtain a comparison theorem in one-dimensional situation.展开更多
This paper studies high order compact finite volume methods on non-uniform meshes for one-dimensional elliptic and parabolic differential equations with the Robin boundary conditions.An explicit scheme and an implicit...This paper studies high order compact finite volume methods on non-uniform meshes for one-dimensional elliptic and parabolic differential equations with the Robin boundary conditions.An explicit scheme and an implicit scheme are obtained by discretizing the equivalent integral form of the equation.For the explicit scheme with nodal values,the algebraic system can be solved by the Thomas method.For the implicit scheme with both nodal values and their derivatives,the system can be implemented by a prediction-correction procedure,where in the correction stage,an implicit formula for recovering the nodal derivatives is introduced.Taking two point boundary value problem as an example,we prove that both the explicit and implicit schemes are convergent with fourth order accuracy with respect to some standard discrete norms using the energy method.Two numerical examples demonstrate the correctness and effectiveness of the schemes,as well as the indispensability of using non-uniform meshes.展开更多
For the complex nonlinear elliptic partial differential equations,it is not easy to prove the Pogorelov type estimates.In this paper,we study a class of complex Hessian quotient type equations in convex cone Γ_(k),wh...For the complex nonlinear elliptic partial differential equations,it is not easy to prove the Pogorelov type estimates.In this paper,we study a class of complex Hessian quotient type equations in convex cone Γ_(k),which is of mixed Hessian type.For the Γ_(k)-admissible solutions,we establish the corresponding Pogorelov estimates with 0≤l<k<n and Pogorelov type estimates with l+1<k=n and l≥0.These extend the previous related results on the real Hessian quotient type equations.展开更多
A new quadrilateral edge element method is proposed and analyzed for Maxwell equations.This proposed method is based on Duan-Liang quadrilateral element(Math.Comp.73(2004),pp.1–18).When applied to the eigenvalue prob...A new quadrilateral edge element method is proposed and analyzed for Maxwell equations.This proposed method is based on Duan-Liang quadrilateral element(Math.Comp.73(2004),pp.1–18).When applied to the eigenvalue problem,the method is spectral-correct and spurious-free.Stability and error estimates are obtained,including the interpolation error estimates and the error estimates between the finite element solution and the exact solution.The method is suitable for singular solution as well as smooth solution,and consequently,the method is valid for nonconvex domains which may have a number of reentrant corners.Of course,the method is suitable for arbitrary quadrilaterals(under the usual shape-regular condition).展开更多
Deep learning methods have achieved significant progress in solving partial differential equations.However,when applied to the widely used anisotropic scattering neutron transport equations in reactor engineering,thes...Deep learning methods have achieved significant progress in solving partial differential equations.However,when applied to the widely used anisotropic scattering neutron transport equations in reactor engineering,these encounter significant challenges.To address this issue,this study introduces a multi-antiderivative transformation alternating iterative deep learning method(M-AIM).This method transforms the integral terms of the scattering and fission sources in the transport equation into multiple antiderivative functions corresponding to the integrand,converts the differential-integral form of the transport equation into an exact differential equation,and establishes the necessary constraints for a unique solution.The M-AIM uses multiple deep neural networks to map the unknown angular flux density of transport equations and represents various forms of antiderivative functions.It constructs the corresponding weighted loss functions.By alternating iterative training with deep learning methods applied to these neural networks,the loss is reduced gradually.When the loss decreases to a preset minimum,the neural network approaches a numerical solution for both angular flux density and antiderivative functions.This paper presents a numerical verification of geometries such as flat plates and spheres.It verifies the validity of the theoretical framework and associated methods.The study contributes to the development of novel technical approaches for applying deep learning to solve anisotropic scattering neutron transport equations in reactor engineering.展开更多
This work is devoted to the study of initial boundary value problem for k-component system of semilinear wave equations with several fundamental boundary conditions(namely,the Dirichlet,Neumann,and Robin boundary cond...This work is devoted to the study of initial boundary value problem for k-component system of semilinear wave equations with several fundamental boundary conditions(namely,the Dirichlet,Neumann,and Robin boundary conditions).Blow-up results and lifespan estimates of solutions to the problem with two different types of weak damping terms and power nonlinearities in the sub-critical and critical cases on exterior domain are obtained.The test function technique is performed in the proofs.It is worth observing that our results in Theorem 1.1 in this article contain the results in[6]as a special case whenθ=0.To the best of our knowledge,the results in Theorems 1.1-1.2 are new.展开更多
The existence of a global attractor is established for generalized Navier-Stokes equations incorporating damping term within the periodic domainΩ=[−π,π]^(n).Initially,we show the existence and uniqueness of strong ...The existence of a global attractor is established for generalized Navier-Stokes equations incorporating damping term within the periodic domainΩ=[−π,π]^(n).Initially,we show the existence and uniqueness of strong solutions.Subsequently,we verify the continuity of the associated semigroup when max{2n+1/n-1,5n+2/3n-2} < β <3n+2/n-2.Finally,we establish the existence of both H^(α)-global attractor and H^(2α)-global attractor.展开更多
We investigate the three-dimensional(3D)generalized magnetohydrodynamic(MHD)-Boussinesq system with fractional dissipation and damping terms,aiming to establish the well-posedness theory for the 3D incompressible temp...We investigate the three-dimensional(3D)generalized magnetohydrodynamic(MHD)-Boussinesq system with fractional dissipation and damping terms,aiming to establish the well-posedness theory for the 3D incompressible temperature-dependent MHD-Boussinesq equations with damping.By exploiting the structural properties of the system and performing refined a priori estimates,we address the global well-posedness of solutions under the weakest possible dissipation conditions.展开更多
We investigate the constrained fractional Choquard equation■where m>0,N>2s with s∈(0,1)being the order of the fractional Laplacian operator and I_(α)forα∈(0,N)denotes the Riesz potential.The parameterμ∈ℝa...We investigate the constrained fractional Choquard equation■where m>0,N>2s with s∈(0,1)being the order of the fractional Laplacian operator and I_(α)forα∈(0,N)denotes the Riesz potential.The parameterμ∈ℝappears as a Lagrange multiplier.By imposing general mass-supercritical conditions on F,we confirm the existence of normalized solutions that characterize the global minimizer on the Pohozaev manifold.Our proof does not depend on the assumption that all weak solutions satisfy the Pohozaev identity,a challenge that remains unsolved for this doubly nonlocal equation.展开更多
In this paper,the growth characteristic of meromorphic solutions for the following difference equation An(z)f(z+n)+…+A1(z)f(z+1)+A0(z)f(z)=0 with no dominating coefficient is studied.By imposing certain restriction o...In this paper,the growth characteristic of meromorphic solutions for the following difference equation An(z)f(z+n)+…+A1(z)f(z+1)+A0(z)f(z)=0 with no dominating coefficient is studied.By imposing certain restriction on the entire coefficients associated with Petrenko's deviation of the above equation,we obtain some results and partially address a question posed byⅠ.Laine and C.C.Yang.Furthermore,for the entire solutions f(z)of the difference equation An(z)f(z+n)+…+A1(z)f(z+1)+A0(z)f(z)=F(z),where Aj(z)(j=0,…,n),F(z)are entire functions,we discover a close relationship between the measure of common transcendental directions associated with classical difference operators of f(z)and Petrenko's deviations of the coefficients.展开更多
The neutron diffusion equation plays a pivotal role in nuclear reactor analysis.Nevertheless,employing the physics-informed neural network(PINN)method for its solution entails certain limitations.Conventional PINN app...The neutron diffusion equation plays a pivotal role in nuclear reactor analysis.Nevertheless,employing the physics-informed neural network(PINN)method for its solution entails certain limitations.Conventional PINN approaches generally utilize a fully connected network(FCN)architecture that is susceptible to overfitting,training instability,and gradient vanishing as the network depth increases.These challenges result in accuracy bottlenecks in the solution.In response to these issues,the residual-based resample physics-informed neural network(R2-PINN)is proposed.It is an improved PINN architecture that replaces the FCN with a convolutional neural network with a shortcut(S-CNN).It incorporates skip connections to facilitate gradient propagation between network layers.Additionally,the incorporation of the residual adaptive resampling(RAR)mechanism dynamically increases the number of sampling points.This,in turn,enhances the spatial representation capabilities and overall predictive accuracy of the model.The experimental results illustrate that our approach significantly improves the convergence capability of the model and achieves high-precision predictions of the physical fields.Compared with conventional FCN-based PINN methods,R 2-PINN effectively overcomes the limitations inherent in current methods.Thus,it provides more accurate and robust solutions for neutron diffusion equations.展开更多
Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay di...Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay differential equations(DDEs)and delay integrodifferential equations(DIDEs)with constant delays,primarily due to their low regularity at delayinduced breaking points.In this paper,a DNN method that combines multi-task learning(MTL)which is proposed to solve both the forward and inverse problems of DIDEs.The core idea of this approach is to divide the original equation into multiple tasks based on the delay,using auxiliary outputs to represent the integral terms,followed by the use of MTL to seamlessly incorporate the properties at the breaking points into the loss function.Furthermore,given the increased training dificulty associated with multiple tasks and outputs,we employ a sequential training scheme to reduce training complexity and provide reference solutions for subsequent tasks.This approach significantly enhances the approximation accuracy of solving DIDEs with DNNs,as demonstrated by comparisons with traditional DNN methods.We validate the effectiveness of this method through several numerical experiments,test various parameter sharing structures in MTL and compare the testing results of these structures.Finally,this method is implemented to solve the inverse problem of nonlinear DIDE and the results show that the unknown parameters of DIDE can be discovered with sparse or noisy data.展开更多
In order to solve the problem of the variable coefficient ordinary differen-tial equation on the bounded domain,the Lagrange interpolation method is used to approximate the exact solution of the equation,and the error...In order to solve the problem of the variable coefficient ordinary differen-tial equation on the bounded domain,the Lagrange interpolation method is used to approximate the exact solution of the equation,and the error between the numerical solution and the exact solution is obtained,and then compared with the error formed by the difference method,it is concluded that the Lagrange interpolation method is more effective in solving the variable coefficient ordinary differential equation.展开更多
In this paper,we construct two fully decoupled,second-order semi-discrete numerical schemes for the Boussinesq equations based on the scalar auxiliary variable(SAV)approach.By introducing a scalar auxiliary variable,t...In this paper,we construct two fully decoupled,second-order semi-discrete numerical schemes for the Boussinesq equations based on the scalar auxiliary variable(SAV)approach.By introducing a scalar auxiliary variable,the original Boussinesq system is transformed into an equivalent one.Then we discretize it using the second-order backward di erentiation formula(BDF2)and Crank-Nicolson(CN)to obtain two second-order time-advanced schemes.In both numerical schemes,a pressure-correction method is employed to decouple the velocity and pressure.These two schemes possess the desired property that they can be fully decoupled with satisfying unconditional stability.We rigorously prove both the unconditional stability and unique solvability of the discrete schemes.Furthermore,we provide detailed implementations of the decoupling procedures.Finally,various 2D numerical simulations are performed to verify the accuracy and energy stability of the proposed schemes.展开更多
In this paper,we mainly focus on a type of nonlinear Choquard equations with nonconstant potential.Under appropriate hypotheses on potential function and nonlinear terms,we prove that the above Choquard equation with ...In this paper,we mainly focus on a type of nonlinear Choquard equations with nonconstant potential.Under appropriate hypotheses on potential function and nonlinear terms,we prove that the above Choquard equation with prescribed 2-norm has some normalized solutions by introducing variational methods.展开更多
Mesophase pitch carbon fibers have an ultra-high modulus and thermal conductivity that are unmatched by other carbon fibers,making it irreplaceable in many fields.However,due to the high temperature dependence of the ...Mesophase pitch carbon fibers have an ultra-high modulus and thermal conductivity that are unmatched by other carbon fibers,making it irreplaceable in many fields.However,due to the high temperature dependence of the viscosity of the melted pitch and the poor mechanical properties of pitch fibers,it is difficult to reduce the fiber diameter when using continuous spinning.We used the Mathworks Matlab software to optimize the mesophase pitch melt spinning model and to simulate the effects of spinning temperature,mass flow rate,winder speed,and quenching air temperature near the spinneret on the maximum shear rate during drawing.Simulation results demonstrate that applying gradient cooling to the melt upon exiting the spinneret significantly reduces the maximum shear rate and extends the drawing zone,thereby promoting the spinning stability and helping reduce the fiber diameter.In the experiment,instead of quenching in air,we applied gradient cooling to the melt,whose temperature decreased according to the equation Ta=298+278exp(−11.4z),where Ta is the final air temperature in Kelvin,and z is the distance from the spinneret in meters.It was found the gradient cooling greatly improved the draw-down ratio,reducing the average diameter of the pitch fibers from 20.8 to 13.1μm,along with improved process stability.The experimental results are in excellent agreement with the predictions.At the same time,the tensile strength of the 1150℃ carbonized fibers increased from 0.6 to 1.1 GPa.Although the degree of orientation of the fibers decreased slightly,the tight bonding between microcrystals,the suppression of splitting,and the smaller diameter improved the mechanical properties of carbon fibers.This study provides an effective method for reducing the fiber diameter while improving continuity.展开更多
In this work,we investigate the existence and asymptotic stability in mean square of mild solutions for non-linear impulsive neutral stochastic evolution equations with infinite delays in distribution in a real separa...In this work,we investigate the existence and asymptotic stability in mean square of mild solutions for non-linear impulsive neutral stochastic evolution equations with infinite delays in distribution in a real separable Hilbert space.By using the Banach fixed point principle,some sufficient conditions are derived to ensure the asymptotic stability of mild solutions.Moreover,we investigate the Hyers-Ulam stability for such stochastic system.Finally,an illustrative example is given to demonstrate the effectiveness of the obtained results.展开更多
This paper studies the Smoluchowski–Kramers approximation for a discrete-time dynamical system modeled as the motion of a particle in a force field.We show that the approximation holds for the drift-implicit Euler–M...This paper studies the Smoluchowski–Kramers approximation for a discrete-time dynamical system modeled as the motion of a particle in a force field.We show that the approximation holds for the drift-implicit Euler–Maruyama discretization and derive its convergence rate.In particular,the solution of the discretized system converges to the solution of the first-order limit equation in the mean-square sense,and this convergence is independent of the order in which the mass parameterμand the step size h tend to zero.展开更多
文摘According to Hypersphere World-Universe Model, dark matter particles DIRACs are magnetic dipoles consisting of two Dirac’s monopoles. We conclude that DIRACs are the subject of Maxwell’s equations. So-called “auxiliary” magnetic field intensity H is indeed current density of magnetic dipoles. The developed approach to magnetic field can explain a wealth of discovered phenomena in Cosmic Magnetism: a dark magnetic field, the large-scale structure of the Milky Way’s magnetic field, and other magnetic phenomena which are only partly related to objects visible in other spectral ranges.
文摘With the help of skew-symmetric differential forms the hidden properties of the mathematical physics equations are revealed. It is shown that the equations of mathematical physics can describe the emergence of various structures and formations such as waves, vortices, turbulent pulsations and others. Such properties of the mathematical physics equations, which are hidden (they appear only in the process of solving these equations), depend on the consistency of derivatives in partial differential equations and on the consistency of equations, if the equations of mathematical physics are a set of equations. This is due to the integrability of mathematical physics equations. It is shown that the equations of mathematical physics can have double solutions, namely, the solutions on the original coordinate space and the solutions on integrable structures that are realized discretely (due to any degrees of freedom). The transition from the solutions of the first type to one of the second type describes discrete transitions and the processes of origin of various structures and observable formations. Only mathematical physics equations, on what no additional conditions such as the integrability conditions are imposed, can possess such properties. The results of the present paper were obtained with the help of skew-symmetric differential forms.
基金Supported by the National Natural Science Foundation of China(12001074)the Research Innovation Program of Graduate Students in Hunan Province(CX20220258)+1 种基金the Research Innovation Program of Graduate Students of Central South University(1053320214147)the Key Scientific Research Project of Higher Education Institutions in Henan Province(25B110025)。
文摘This paper deals with Mckean-Vlasov backward stochastic differential equations with weak monotonicity coefficients.We first establish the existence and uniqueness of solutions to Mckean-Vlasov backward stochastic differential equations.Then we obtain a comparison theorem in one-dimensional situation.
文摘This paper studies high order compact finite volume methods on non-uniform meshes for one-dimensional elliptic and parabolic differential equations with the Robin boundary conditions.An explicit scheme and an implicit scheme are obtained by discretizing the equivalent integral form of the equation.For the explicit scheme with nodal values,the algebraic system can be solved by the Thomas method.For the implicit scheme with both nodal values and their derivatives,the system can be implemented by a prediction-correction procedure,where in the correction stage,an implicit formula for recovering the nodal derivatives is introduced.Taking two point boundary value problem as an example,we prove that both the explicit and implicit schemes are convergent with fourth order accuracy with respect to some standard discrete norms using the energy method.Two numerical examples demonstrate the correctness and effectiveness of the schemes,as well as the indispensability of using non-uniform meshes.
文摘For the complex nonlinear elliptic partial differential equations,it is not easy to prove the Pogorelov type estimates.In this paper,we study a class of complex Hessian quotient type equations in convex cone Γ_(k),which is of mixed Hessian type.For the Γ_(k)-admissible solutions,we establish the corresponding Pogorelov estimates with 0≤l<k<n and Pogorelov type estimates with l+1<k=n and l≥0.These extend the previous related results on the real Hessian quotient type equations.
基金supported by the National Natural Science Foundation of China(12401482)the second author was supported by the National Natural Science Foundation of China(12371371,12261160361,11971366)supported by the Open Research Fund of Hubei Key Laboratory of Computational Science,Wuhan University.
文摘A new quadrilateral edge element method is proposed and analyzed for Maxwell equations.This proposed method is based on Duan-Liang quadrilateral element(Math.Comp.73(2004),pp.1–18).When applied to the eigenvalue problem,the method is spectral-correct and spurious-free.Stability and error estimates are obtained,including the interpolation error estimates and the error estimates between the finite element solution and the exact solution.The method is suitable for singular solution as well as smooth solution,and consequently,the method is valid for nonconvex domains which may have a number of reentrant corners.Of course,the method is suitable for arbitrary quadrilaterals(under the usual shape-regular condition).
基金supported by the National Natural Science Foundation of China(No.12575189)。
文摘Deep learning methods have achieved significant progress in solving partial differential equations.However,when applied to the widely used anisotropic scattering neutron transport equations in reactor engineering,these encounter significant challenges.To address this issue,this study introduces a multi-antiderivative transformation alternating iterative deep learning method(M-AIM).This method transforms the integral terms of the scattering and fission sources in the transport equation into multiple antiderivative functions corresponding to the integrand,converts the differential-integral form of the transport equation into an exact differential equation,and establishes the necessary constraints for a unique solution.The M-AIM uses multiple deep neural networks to map the unknown angular flux density of transport equations and represents various forms of antiderivative functions.It constructs the corresponding weighted loss functions.By alternating iterative training with deep learning methods applied to these neural networks,the loss is reduced gradually.When the loss decreases to a preset minimum,the neural network approaches a numerical solution for both angular flux density and antiderivative functions.This paper presents a numerical verification of geometries such as flat plates and spheres.It verifies the validity of the theoretical framework and associated methods.The study contributes to the development of novel technical approaches for applying deep learning to solve anisotropic scattering neutron transport equations in reactor engineering.
基金Supported by Fundamental Research Program of Shanxi Province(20210302123045,20210302123182)National Natural Science Foundation of China(11601446)。
文摘This work is devoted to the study of initial boundary value problem for k-component system of semilinear wave equations with several fundamental boundary conditions(namely,the Dirichlet,Neumann,and Robin boundary conditions).Blow-up results and lifespan estimates of solutions to the problem with two different types of weak damping terms and power nonlinearities in the sub-critical and critical cases on exterior domain are obtained.The test function technique is performed in the proofs.It is worth observing that our results in Theorem 1.1 in this article contain the results in[6]as a special case whenθ=0.To the best of our knowledge,the results in Theorems 1.1-1.2 are new.
文摘The existence of a global attractor is established for generalized Navier-Stokes equations incorporating damping term within the periodic domainΩ=[−π,π]^(n).Initially,we show the existence and uniqueness of strong solutions.Subsequently,we verify the continuity of the associated semigroup when max{2n+1/n-1,5n+2/3n-2} < β <3n+2/n-2.Finally,we establish the existence of both H^(α)-global attractor and H^(2α)-global attractor.
文摘We investigate the three-dimensional(3D)generalized magnetohydrodynamic(MHD)-Boussinesq system with fractional dissipation and damping terms,aiming to establish the well-posedness theory for the 3D incompressible temperature-dependent MHD-Boussinesq equations with damping.By exploiting the structural properties of the system and performing refined a priori estimates,we address the global well-posedness of solutions under the weakest possible dissipation conditions.
基金supported by the Guangdong Basic and Applied Basic Research Foundation(2022A1515012138)the NSFC(12271436,12371119)supported by the Natural Science Basic Research Program of Shaanxi(2022JC-04).
文摘We investigate the constrained fractional Choquard equation■where m>0,N>2s with s∈(0,1)being the order of the fractional Laplacian operator and I_(α)forα∈(0,N)denotes the Riesz potential.The parameterμ∈ℝappears as a Lagrange multiplier.By imposing general mass-supercritical conditions on F,we confirm the existence of normalized solutions that characterize the global minimizer on the Pohozaev manifold.Our proof does not depend on the assumption that all weak solutions satisfy the Pohozaev identity,a challenge that remains unsolved for this doubly nonlocal equation.
基金Supported by the National Natural Science Foundation of China(Grant No.11661043)and the ScienceTechnology Research Project of Jiangxi Provincial Department of Education(Grant No.GJJ2200320).
文摘In this paper,the growth characteristic of meromorphic solutions for the following difference equation An(z)f(z+n)+…+A1(z)f(z+1)+A0(z)f(z)=0 with no dominating coefficient is studied.By imposing certain restriction on the entire coefficients associated with Petrenko's deviation of the above equation,we obtain some results and partially address a question posed byⅠ.Laine and C.C.Yang.Furthermore,for the entire solutions f(z)of the difference equation An(z)f(z+n)+…+A1(z)f(z+1)+A0(z)f(z)=F(z),where Aj(z)(j=0,…,n),F(z)are entire functions,we discover a close relationship between the measure of common transcendental directions associated with classical difference operators of f(z)and Petrenko's deviations of the coefficients.
基金supported by the Science and Technology on Reactor System Design Technology Laboratory(No.LRSDT12023108)supported in part by the Chongqing Postdoctoral Science Foundation(No.cstc2021jcyj-bsh0252)+2 种基金the National Natural Science Foundation of China(No.12005030)Sichuan Province to unveil the list of marshal industry common technology research projects(No.23jBGOV0001)Special Program for Stabilizing Support to Basic Research of National Basic Research Institutes(No.WDZC-2023-05-03-05).
文摘The neutron diffusion equation plays a pivotal role in nuclear reactor analysis.Nevertheless,employing the physics-informed neural network(PINN)method for its solution entails certain limitations.Conventional PINN approaches generally utilize a fully connected network(FCN)architecture that is susceptible to overfitting,training instability,and gradient vanishing as the network depth increases.These challenges result in accuracy bottlenecks in the solution.In response to these issues,the residual-based resample physics-informed neural network(R2-PINN)is proposed.It is an improved PINN architecture that replaces the FCN with a convolutional neural network with a shortcut(S-CNN).It incorporates skip connections to facilitate gradient propagation between network layers.Additionally,the incorporation of the residual adaptive resampling(RAR)mechanism dynamically increases the number of sampling points.This,in turn,enhances the spatial representation capabilities and overall predictive accuracy of the model.The experimental results illustrate that our approach significantly improves the convergence capability of the model and achieves high-precision predictions of the physical fields.Compared with conventional FCN-based PINN methods,R 2-PINN effectively overcomes the limitations inherent in current methods.Thus,it provides more accurate and robust solutions for neutron diffusion equations.
文摘Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay differential equations(DDEs)and delay integrodifferential equations(DIDEs)with constant delays,primarily due to their low regularity at delayinduced breaking points.In this paper,a DNN method that combines multi-task learning(MTL)which is proposed to solve both the forward and inverse problems of DIDEs.The core idea of this approach is to divide the original equation into multiple tasks based on the delay,using auxiliary outputs to represent the integral terms,followed by the use of MTL to seamlessly incorporate the properties at the breaking points into the loss function.Furthermore,given the increased training dificulty associated with multiple tasks and outputs,we employ a sequential training scheme to reduce training complexity and provide reference solutions for subsequent tasks.This approach significantly enhances the approximation accuracy of solving DIDEs with DNNs,as demonstrated by comparisons with traditional DNN methods.We validate the effectiveness of this method through several numerical experiments,test various parameter sharing structures in MTL and compare the testing results of these structures.Finally,this method is implemented to solve the inverse problem of nonlinear DIDE and the results show that the unknown parameters of DIDE can be discovered with sparse or noisy data.
文摘In order to solve the problem of the variable coefficient ordinary differen-tial equation on the bounded domain,the Lagrange interpolation method is used to approximate the exact solution of the equation,and the error between the numerical solution and the exact solution is obtained,and then compared with the error formed by the difference method,it is concluded that the Lagrange interpolation method is more effective in solving the variable coefficient ordinary differential equation.
基金Supported by Research Project Supported by Shanxi Scholarship Council of China(2021-029)International Cooperation Base and Platform Project of Shanxi Province(202104041101019)+2 种基金Basic Research Plan of Shanxi Province(202203021211129)Shanxi Province Natural Science Research(202203021212249)Special/Youth Foundation of Taiyuan University of Technology(2022QN101)。
文摘In this paper,we construct two fully decoupled,second-order semi-discrete numerical schemes for the Boussinesq equations based on the scalar auxiliary variable(SAV)approach.By introducing a scalar auxiliary variable,the original Boussinesq system is transformed into an equivalent one.Then we discretize it using the second-order backward di erentiation formula(BDF2)and Crank-Nicolson(CN)to obtain two second-order time-advanced schemes.In both numerical schemes,a pressure-correction method is employed to decouple the velocity and pressure.These two schemes possess the desired property that they can be fully decoupled with satisfying unconditional stability.We rigorously prove both the unconditional stability and unique solvability of the discrete schemes.Furthermore,we provide detailed implementations of the decoupling procedures.Finally,various 2D numerical simulations are performed to verify the accuracy and energy stability of the proposed schemes.
基金Supported by the National Natural Science Foundation of China(11671403,11671236,12101192)Henan Provincial General Natural Science Foundation Project(232300420113)。
文摘In this paper,we mainly focus on a type of nonlinear Choquard equations with nonconstant potential.Under appropriate hypotheses on potential function and nonlinear terms,we prove that the above Choquard equation with prescribed 2-norm has some normalized solutions by introducing variational methods.
文摘Mesophase pitch carbon fibers have an ultra-high modulus and thermal conductivity that are unmatched by other carbon fibers,making it irreplaceable in many fields.However,due to the high temperature dependence of the viscosity of the melted pitch and the poor mechanical properties of pitch fibers,it is difficult to reduce the fiber diameter when using continuous spinning.We used the Mathworks Matlab software to optimize the mesophase pitch melt spinning model and to simulate the effects of spinning temperature,mass flow rate,winder speed,and quenching air temperature near the spinneret on the maximum shear rate during drawing.Simulation results demonstrate that applying gradient cooling to the melt upon exiting the spinneret significantly reduces the maximum shear rate and extends the drawing zone,thereby promoting the spinning stability and helping reduce the fiber diameter.In the experiment,instead of quenching in air,we applied gradient cooling to the melt,whose temperature decreased according to the equation Ta=298+278exp(−11.4z),where Ta is the final air temperature in Kelvin,and z is the distance from the spinneret in meters.It was found the gradient cooling greatly improved the draw-down ratio,reducing the average diameter of the pitch fibers from 20.8 to 13.1μm,along with improved process stability.The experimental results are in excellent agreement with the predictions.At the same time,the tensile strength of the 1150℃ carbonized fibers increased from 0.6 to 1.1 GPa.Although the degree of orientation of the fibers decreased slightly,the tight bonding between microcrystals,the suppression of splitting,and the smaller diameter improved the mechanical properties of carbon fibers.This study provides an effective method for reducing the fiber diameter while improving continuity.
基金supported by the Excellent Talents Support Program for Colleges and Universities in Anhui Province(Grant No.gxyqZD2020003)。
文摘In this work,we investigate the existence and asymptotic stability in mean square of mild solutions for non-linear impulsive neutral stochastic evolution equations with infinite delays in distribution in a real separable Hilbert space.By using the Banach fixed point principle,some sufficient conditions are derived to ensure the asymptotic stability of mild solutions.Moreover,we investigate the Hyers-Ulam stability for such stochastic system.Finally,an illustrative example is given to demonstrate the effectiveness of the obtained results.
基金supported by the PhD Research Startup Foundation of Hubei University of Economics(Grand No.XJ23BS42).
文摘This paper studies the Smoluchowski–Kramers approximation for a discrete-time dynamical system modeled as the motion of a particle in a force field.We show that the approximation holds for the drift-implicit Euler–Maruyama discretization and derive its convergence rate.In particular,the solution of the discretized system converges to the solution of the first-order limit equation in the mean-square sense,and this convergence is independent of the order in which the mass parameterμand the step size h tend to zero.
