In this paper,weak optimal inverse problems of interval linear programming(IvLP)are studied based on KKT conditions.Firstly,the problem is precisely defined.Specifically,by adjusting the minimum change of the current ...In this paper,weak optimal inverse problems of interval linear programming(IvLP)are studied based on KKT conditions.Firstly,the problem is precisely defined.Specifically,by adjusting the minimum change of the current cost coefficient,a given weak solution can become optimal.Then,an equivalent characterization of weak optimal inverse IvLP problems is obtained.Finally,the problem is simplified without adjusting the cost coefficient of null variable.展开更多
A class of matrix inverse problems minimizing ‖A-‖ F on the linear manifold l A=｛A∈R n×m ｜‖AX-B‖ F=min｝ is considered. The perturbation analysis of the solution to these problems is carried out. Th...A class of matrix inverse problems minimizing ‖A-‖ F on the linear manifold l A=｛A∈R n×m ｜‖AX-B‖ F=min｝ is considered. The perturbation analysis of the solution to these problems is carried out. The perturbation upper bounds of the solution are given for both the consistent and inconsistent cases. The obtained preturbation upper bounds are with respect to the distance from the perturbed solution to the unperturbed manifold.展开更多
An inverse problem for identification of the coefficient in heat-conduction equation is considered. After reducing the problem to a nonlinear ill-posed operator equation, Newton type iterative methods are considered. ...An inverse problem for identification of the coefficient in heat-conduction equation is considered. After reducing the problem to a nonlinear ill-posed operator equation, Newton type iterative methods are considered. The implicit iterative method is applied to the linearized Newton equation, and the key step in the process is that a new reasonable a posteriori stopping rule for the inner iteration is presented. Numerical experiments for the new method as well as for Tikhonov method and Bakushikskii method are given, and these results show the obvious advantages of the new method over the other ones.展开更多
For the 2-D wave inverse problems introduced from geophysical exploration, in this paper, the author presents integration-characteristic method to solve the velocity parameter, and then applies it to common shotpoint ...For the 2-D wave inverse problems introduced from geophysical exploration, in this paper, the author presents integration-characteristic method to solve the velocity parameter, and then applies it to common shotpoint model data, in noise-free case. The accuracy is quite good.展开更多
This paper studies the influence of a finite container on an ideal gas.The trace of the heat kernel (t) =exp, where are the eigenvalues of the negative Laplacian -in Rn(n = 2 or 3), is studied for a general multi-conn...This paper studies the influence of a finite container on an ideal gas.The trace of the heat kernel (t) =exp, where are the eigenvalues of the negative Laplacian -in Rn(n = 2 or 3), is studied for a general multi-connected bounded drum ft which is surrounded by simply connected bounded domains Ωi with smooth boundaries Ωi(i = 1,… ,m) where the Dirichlet, Neumann and Robin boundary conditions on Ωi(i = 1,…,m) are considered. Some geometrical properties of Ω are determined. The thermodynamic quantities for an ideal gas enclosed in Ω are examined by using the asymptotic expansions of (t) for short-time t. It is shown that the ideal gas can not feel the shape of its container Ω, although it can feel some geometrical properties of it.展开更多
To determine a variation of pipe's inner geometric shape as due to etch, the three-layered feedforward artificial neural network is used in the inverse analysis through observing the elastoplastic strains of the o...To determine a variation of pipe's inner geometric shape as due to etch, the three-layered feedforward artificial neural network is used in the inverse analysis through observing the elastoplastic strains of the outer wall under the working inner pressure. Because of different kinds of inner wail radii and eccentricity. several groups of strains calculated with computational mechanics are used for the network to do learning. Numerical calculation demonstrates that this method is effective and the estimated inner wall geometric parameters have high precision.展开更多
We introduce a multi-cost-functional method for solving inverse problems of wave equations. This method has its simplicity, efficiency and good physical interpretation. It has the advantage of being programmed for two...We introduce a multi-cost-functional method for solving inverse problems of wave equations. This method has its simplicity, efficiency and good physical interpretation. It has the advantage of being programmed for two- or three- (space) dimensional problems as well as for one-dimensional problems.展开更多
The following inverse problem is solved—given the eigenvalues and the potential b(n) for a difference boundary value problem with quadratic dependence on the eigenparameter, λ, the weights c(n) can be uniquely ...The following inverse problem is solved—given the eigenvalues and the potential b(n) for a difference boundary value problem with quadratic dependence on the eigenparameter, λ, the weights c(n) can be uniquely reconstructed. The investi-gation is inductive on m where represents the number of unit intervals and the results obtained depend on the specific form of the given boundary conditions. This paper is a sequel to [1] which provided an algorithm for the solution of an analogous inverse problem, where the eigenvalues and weights were given and the potential was uniquely reconstructed. Since the inverse problem considered in this paper contains more unknowns than the inverse problem considered in [1], an additional spectrum is required more often than was the case in [1].展开更多
In this paper,the physics informed neural network(PINN)deep learning method is applied to solve two-dimensional nonlocal equations,including the partial reverse space y-nonlocal Mel'nikov equation,the partial reve...In this paper,the physics informed neural network(PINN)deep learning method is applied to solve two-dimensional nonlocal equations,including the partial reverse space y-nonlocal Mel'nikov equation,the partial reverse space-time nonlocal Mel'nikov equation and the nonlocal twodimensional nonlinear Schr?dinger(NLS)equation.By the PINN method,we successfully derive a data-driven two soliton solution,lump solution and rogue wave solution.Numerical simulation results indicate that the error range between the data-driven solution and the exact solution is relatively small,which verifies the effectiveness of the PINN deep learning method for solving high dimensional nonlocal equations.Moreover,the parameter discovery of the partial reverse space-time nonlocal Mel'nikov equation is analysed in terms of its soliton solution for the first time.展开更多
In the past decade,notable progress has been achieved in the development of the generalized finite difference method(GFDM).The underlying principle of GFDM involves dividing the domain into multiple sub-domains.Within...In the past decade,notable progress has been achieved in the development of the generalized finite difference method(GFDM).The underlying principle of GFDM involves dividing the domain into multiple sub-domains.Within each sub-domain,explicit formulas for the necessary partial derivatives of the partial differential equations(PDEs)can be obtained through the application of Taylor series expansion and moving-least square approximation methods.Consequently,the method generates a sparse coefficient matrix,exhibiting a banded structure,making it highly advantageous for large-scale engineering computations.In this study,we present the application of the GFDM to numerically solve inverse Cauchy problems in two-and three-dimensional piezoelectric structures.Through our preliminary numerical experiments,we demonstrate that the proposed GFDMapproach shows great promise for accurately simulating coupled electroelastic equations in inverse problems,even with 3%errors added to the input data.展开更多
We present an efficient numerical strategy for the Bayesian solution of inverse problems.Stochastic collocation methods,based on generalized polynomial chaos(gPC),are used to construct a polynomial approximation of th...We present an efficient numerical strategy for the Bayesian solution of inverse problems.Stochastic collocation methods,based on generalized polynomial chaos(gPC),are used to construct a polynomial approximation of the forward solution over the support of the prior distribution.This approximation then defines a surrogate posterior probability density that can be evaluated repeatedly at minimal computational cost.The ability to simulate a large number of samples from the posterior distribution results in very accurate estimates of the inverse solution and its associated uncertainty.Combined with high accuracy of the gPC-based forward solver,the new algorithm can provide great efficiency in practical applications.A rigorous error analysis of the algorithm is conducted,where we establish convergence of the approximate posterior to the true posterior and obtain an estimate of the convergence rate.It is proved that fast(exponential)convergence of the gPC forward solution yields similarly fast(exponential)convergence of the posterior.The numerical strategy and the predicted convergence rates are then demonstrated on nonlinear inverse problems of varying smoothness and dimension.展开更多
This paper proposes a semi‐analytical and local meshless collocation method,the loca-lized method of fundamental solutions(LMFS),to address three‐dimensional(3D)acoustic inverse problems in complex domains.The propo...This paper proposes a semi‐analytical and local meshless collocation method,the loca-lized method of fundamental solutions(LMFS),to address three‐dimensional(3D)acoustic inverse problems in complex domains.The proposed approach is a recently developed numerical scheme with the potential of being mathematically simple,nu-merically accurate,and requiring less computational time and storage.In LMFS,an overdetermined sparse linear system is constructed by using the known data at the nodes on the accessible boundary and by making the remaining nodes satisfy the governing equation.In the numerical procedure,the pseudoinverse of a matrix is solved via the truncated singular value decomposition,and thus the regularization techniques are not needed in solving the resulting linear system with a well‐conditioned matrix.Numerical experiments,involving complicated geometry and the high noise level,confirm the ef-fectiveness and performance of the LMFS for solving 3D acoustic inverse problems.展开更多
This paper considers the estimation of an unknown function h that can be characterized as a solution to a nonlinear operator equation mapping between two infinite dimensional Hilbert spaces. The nonlinear operator is ...This paper considers the estimation of an unknown function h that can be characterized as a solution to a nonlinear operator equation mapping between two infinite dimensional Hilbert spaces. The nonlinear operator is unknown but can be consistently estimated, and its inverse is discontinuous, rendering the problem ill-posed. We establish the consistency for the class of estimators that are regularized using general lower semicompact penalty functions. We derive the optimal convergence rates of the estimators under the Hilbert scale norms. We apply our results to two important problems in economics and finance: (1) estimating the parameters of the pricing kernel of defaultable bonds; (2) recovering the volatility surface implied by option prices allowing for measurement error in the option prices and numerical error in the computation of the operator.展开更多
Let G be a finite abelian group and S be a sequence with elements of G.We say that S is a regular sequence over G if|SH|≤|H|-1 holds for every proper subgroup H of G,where SH denotes the subsequence of S consisting o...Let G be a finite abelian group and S be a sequence with elements of G.We say that S is a regular sequence over G if|SH|≤|H|-1 holds for every proper subgroup H of G,where SH denotes the subsequence of S consisting of all terms of S contained in H.We say that S is a zero-sum free sequence over G if 0■Σ(S),where Σ(S)■G denotes the set of group elements which can be expressed as a sum of a nonempty subsequence of S.In this paper,we study the inverse problems associated with Σ(S)when S is a regular sequence or a zero-sum free sequence over G=Cp■Cp,where p is a prime.展开更多
A general stochastic algorithm for solving mixed linear and nonlinear problems was introduced in[11].We show in this paper how it can be used to solve the fault inverse problem,where a planar fault in elastic half-spa...A general stochastic algorithm for solving mixed linear and nonlinear problems was introduced in[11].We show in this paper how it can be used to solve the fault inverse problem,where a planar fault in elastic half-space and a slip on that fault have to be reconstructed from noisy surface displacement measurements.With the parameter giving the plane containing the fault denoted by m and the regularization parameter for the linear part of the inverse problem denoted by C,both modeled as random variables,we derive a formula for the posterior marginal of m.Modeling C as a random variable allows to sweep through a wide range of possible values which was shown to be superior to selecting a fixed value[11].We prove that this posterior marginal of m is convergent as the number of measurement points and the dimension of the space for discretizing slips increase.Simply put,our proof only assumes that the regularized discrete error functional for processing measurements relates to an order 1 quadrature rule and that the union of the finite-dimensional spaces for discretizing slips is dense.Our proof relies on trace class operator theory to show that an adequate sequence of determinants is uniformly bounded.We also explain how our proof can be extended to a whole class of inverse problems,as long as some basic requirements are met.Finally,we show numerical simulations that illustrate the numerical convergence of our algorithm.展开更多
This paper describes a new method which has been developed for the solution of direct and inverse problems of 3-D compressible flows in turbomachinery.Two types of streamfunctions are proposed in the paper and the str...This paper describes a new method which has been developed for the solution of direct and inverse problems of 3-D compressible flows in turbomachinery.Two types of streamfunctions are proposed in the paper and the streamfunction-coordinate system is applied in numerical computations.The algorithm is applied to stator blades and the results are compared with experimental data,It is shown that the comparisons are very satis- factory.展开更多
Various works from the literature aimed at accelerating Bayesian inference in inverse problems.Stochastic spectral methods have been recently proposed as surrogate approximations of the forward uncertainty propagation...Various works from the literature aimed at accelerating Bayesian inference in inverse problems.Stochastic spectral methods have been recently proposed as surrogate approximations of the forward uncertainty propagation model over the support of the prior distribution.These representations are efficient because they allow affordable simulation of a large number of samples from the posterior distribution.Unfortunately,they do not perform well when the forward model exhibits strong nonlinear behavior with respect to its input.In this work,we first relate the fast(exponential)L2-convergence of the forward approximation to the fast(exponential)convergence(in terms of Kullback-Leibler divergence)of the approximate posterior.In particular,we prove that in case the prior distribution is uniform,the posterior is at least twice as fast as the convergence rate of the forward model in those norms.The Bayesian inference strategy is developed in the framework of a stochastic spectral projection method.The predicted convergence rates are then demonstrated for simple nonlinear inverse problems of varying smoothness.We then propose an efficient numerical approach for the Bayesian solution of inverse problems presenting strongly nonlinear or discontinuous system responses.This comes with the improvement of the forward model that is adaptively approximated by an iterative generalized Polynomial Chaos-based representation.The numerical approximations and predicted convergence rates of the former approach are compared to the new iterative numerical method for nonlinear time-dependent test cases of varying dimension and complexity,which are relevant regarding our hydrodynamics motivations and therefore regarding hyperbolic conservation laws and the apparition of discontinuities in finite time.展开更多
To study the feasibility of using machine learning technology to solve the forward problem(prediction of aerodynamic parameters)and the inverse problem(prediction of geometric parameters)of turbine blades,this paper b...To study the feasibility of using machine learning technology to solve the forward problem(prediction of aerodynamic parameters)and the inverse problem(prediction of geometric parameters)of turbine blades,this paper built a forward problem model based on backpropagation artificial neural networks(BP-ANNs)and an inverse problem model based on radial basis function artificial neural networks(RBF-ANNs).The S2(a stream surface obtained by extending a radial curve in turbo blades)calculation program was used to generate the dataset for single-stage turbo blades,and the back propagation algorithm was used to train the model.The parameters of five blade sections in a single-stage turbine were selected as inputs of the forward problem model,including stagger angle,inlet geometric angle,outlet geometric angle,wedge angle of leading edge pressure side,wedge angle of leading edge suction side,wedge angle of trailing edge,rear bending angle,and leading edge diameter.The outputs are efficiency,power,mass flow,relative exit Mach number,absolute exit Mach number,relative exit flow angle,absolute exit flow angle and reaction degree,which are eight aerodynamic parameters.The inputs and outputs of the inverse problem model are the opposite of that of the forward problem model.The models can accurately predict the aerodynamic parameters and geometric parameters,and the mean square errors(MSEs)of the forward problem test set and the inverse problem test set are 0.001 and 0.00035,respectively.This study shows that machine learning technology based on neural networks can be flexibly applied to the design of forward and inverse problems of turbine blades,and the models built by this method have practical application value in regression prediction problems.展开更多
In this article we consider a sequence of hierarchical space model of inverse problems.The underlying function is estimated from indirect observations over a variety of error distributions including those that are hea...In this article we consider a sequence of hierarchical space model of inverse problems.The underlying function is estimated from indirect observations over a variety of error distributions including those that are heavy-tailed and may not even possess variances or means.The main contribution of this paper is that we establish some oracle inequalities for the inverse problems by using quantile coupling technique that gives a tight bound for the quantile coupling between an arbitrary sample p-quantile and a normal variable,and an automatic selection principle for the nonrandom filters.This leads to the data-driven choice of weights.We also give an algorithm for its implementation.The quantile coupling inequality developed in this paper is of independent interest,because it includes the median coupling inequality in literature as a special case.展开更多
A fractal approximation algorithm is developed to obtain approximate solutions to an inverse initial-value problem IVP(inverse IVP) for the differential equation. Numerical computational results are presented to demon...A fractal approximation algorithm is developed to obtain approximate solutions to an inverse initial-value problem IVP(inverse IVP) for the differential equation. Numerical computational results are presented to demonstrate the effectiveness of this algorithm for solving inverse IVP for a class of specific differential equations.展开更多
基金Supported by the National Natural Science Foundation of China(11971433)First Class Discipline of Zhe-jiang-A(Zhejiang Gongshang University-Statistics,1020JYN4120004G-091),Graduate Scientic Research and Innovation Foundation of Zhejiang Gongshang University.
文摘In this paper,weak optimal inverse problems of interval linear programming(IvLP)are studied based on KKT conditions.Firstly,the problem is precisely defined.Specifically,by adjusting the minimum change of the current cost coefficient,a given weak solution can become optimal.Then,an equivalent characterization of weak optimal inverse IvLP problems is obtained.Finally,the problem is simplified without adjusting the cost coefficient of null variable.
文摘A class of matrix inverse problems minimizing ‖A-‖ F on the linear manifold l A=｛A∈R n×m ｜‖AX-B‖ F=min｝ is considered. The perturbation analysis of the solution to these problems is carried out. The perturbation upper bounds of the solution are given for both the consistent and inconsistent cases. The obtained preturbation upper bounds are with respect to the distance from the perturbed solution to the unperturbed manifold.
文摘An inverse problem for identification of the coefficient in heat-conduction equation is considered. After reducing the problem to a nonlinear ill-posed operator equation, Newton type iterative methods are considered. The implicit iterative method is applied to the linearized Newton equation, and the key step in the process is that a new reasonable a posteriori stopping rule for the inner iteration is presented. Numerical experiments for the new method as well as for Tikhonov method and Bakushikskii method are given, and these results show the obvious advantages of the new method over the other ones.
文摘For the 2-D wave inverse problems introduced from geophysical exploration, in this paper, the author presents integration-characteristic method to solve the velocity parameter, and then applies it to common shotpoint model data, in noise-free case. The accuracy is quite good.
文摘This paper studies the influence of a finite container on an ideal gas.The trace of the heat kernel (t) =exp, where are the eigenvalues of the negative Laplacian -in Rn(n = 2 or 3), is studied for a general multi-connected bounded drum ft which is surrounded by simply connected bounded domains Ωi with smooth boundaries Ωi(i = 1,… ,m) where the Dirichlet, Neumann and Robin boundary conditions on Ωi(i = 1,…,m) are considered. Some geometrical properties of Ω are determined. The thermodynamic quantities for an ideal gas enclosed in Ω are examined by using the asymptotic expansions of (t) for short-time t. It is shown that the ideal gas can not feel the shape of its container Ω, although it can feel some geometrical properties of it.
文摘To determine a variation of pipe's inner geometric shape as due to etch, the three-layered feedforward artificial neural network is used in the inverse analysis through observing the elastoplastic strains of the outer wall under the working inner pressure. Because of different kinds of inner wail radii and eccentricity. several groups of strains calculated with computational mechanics are used for the network to do learning. Numerical calculation demonstrates that this method is effective and the estimated inner wall geometric parameters have high precision.
文摘We introduce a multi-cost-functional method for solving inverse problems of wave equations. This method has its simplicity, efficiency and good physical interpretation. It has the advantage of being programmed for two- or three- (space) dimensional problems as well as for one-dimensional problems.
文摘The following inverse problem is solved—given the eigenvalues and the potential b(n) for a difference boundary value problem with quadratic dependence on the eigenparameter, λ, the weights c(n) can be uniquely reconstructed. The investi-gation is inductive on m where represents the number of unit intervals and the results obtained depend on the specific form of the given boundary conditions. This paper is a sequel to [1] which provided an algorithm for the solution of an analogous inverse problem, where the eigenvalues and weights were given and the potential was uniquely reconstructed. Since the inverse problem considered in this paper contains more unknowns than the inverse problem considered in [1], an additional spectrum is required more often than was the case in [1].
文摘In this paper,the physics informed neural network(PINN)deep learning method is applied to solve two-dimensional nonlocal equations,including the partial reverse space y-nonlocal Mel'nikov equation,the partial reverse space-time nonlocal Mel'nikov equation and the nonlocal twodimensional nonlinear Schr?dinger(NLS)equation.By the PINN method,we successfully derive a data-driven two soliton solution,lump solution and rogue wave solution.Numerical simulation results indicate that the error range between the data-driven solution and the exact solution is relatively small,which verifies the effectiveness of the PINN deep learning method for solving high dimensional nonlocal equations.Moreover,the parameter discovery of the partial reverse space-time nonlocal Mel'nikov equation is analysed in terms of its soliton solution for the first time.
基金the Natural Science Foundation of Shandong Province of China(Grant No.ZR2022YQ06)the Development Plan of Youth Innovation Team in Colleges and Universities of Shandong Province(Grant No.2022KJ140)the Key Laboratory ofRoad Construction Technology and Equipment(Chang’an University,No.300102253502).
文摘In the past decade,notable progress has been achieved in the development of the generalized finite difference method(GFDM).The underlying principle of GFDM involves dividing the domain into multiple sub-domains.Within each sub-domain,explicit formulas for the necessary partial derivatives of the partial differential equations(PDEs)can be obtained through the application of Taylor series expansion and moving-least square approximation methods.Consequently,the method generates a sparse coefficient matrix,exhibiting a banded structure,making it highly advantageous for large-scale engineering computations.In this study,we present the application of the GFDM to numerically solve inverse Cauchy problems in two-and three-dimensional piezoelectric structures.Through our preliminary numerical experiments,we demonstrate that the proposed GFDMapproach shows great promise for accurately simulating coupled electroelastic equations in inverse problems,even with 3%errors added to the input data.
基金The work of Y.Marzouk is supported in part by the DOE Office of Advanced Scientific Computing Research(ASCR)by Sandia Corporation(a wholly owned subsidiary of Lockheed Martin Corporation)as operator of Sandia National Laboratories under US Department of Energy contract number DE-AC04-94AL85000+1 种基金The work of D.Xiu is supported in part by AFOSR FA9550-08-1-0353,NSF CAREER Award DMS-0645035the DOE/NNSA PSAAP center at Purdue(PRISM)under contract number DE-FC52-08NA28617.
文摘We present an efficient numerical strategy for the Bayesian solution of inverse problems.Stochastic collocation methods,based on generalized polynomial chaos(gPC),are used to construct a polynomial approximation of the forward solution over the support of the prior distribution.This approximation then defines a surrogate posterior probability density that can be evaluated repeatedly at minimal computational cost.The ability to simulate a large number of samples from the posterior distribution results in very accurate estimates of the inverse solution and its associated uncertainty.Combined with high accuracy of the gPC-based forward solver,the new algorithm can provide great efficiency in practical applications.A rigorous error analysis of the algorithm is conducted,where we establish convergence of the approximate posterior to the true posterior and obtain an estimate of the convergence rate.It is proved that fast(exponential)convergence of the gPC forward solution yields similarly fast(exponential)convergence of the posterior.The numerical strategy and the predicted convergence rates are then demonstrated on nonlinear inverse problems of varying smoothness and dimension.
基金National Natural Science Foundation of China,Grant/Award Number:11802151Natural Science Foundation of Shandong Province of China,Grant/Award Number:ZR2019BA008+1 种基金supported by the National Natural Science Foundation of China(No.11802151)the Natural Science Foundation of Shandong Province of China(No.ZR2019BA008).
文摘This paper proposes a semi‐analytical and local meshless collocation method,the loca-lized method of fundamental solutions(LMFS),to address three‐dimensional(3D)acoustic inverse problems in complex domains.The proposed approach is a recently developed numerical scheme with the potential of being mathematically simple,nu-merically accurate,and requiring less computational time and storage.In LMFS,an overdetermined sparse linear system is constructed by using the known data at the nodes on the accessible boundary and by making the remaining nodes satisfy the governing equation.In the numerical procedure,the pseudoinverse of a matrix is solved via the truncated singular value decomposition,and thus the regularization techniques are not needed in solving the resulting linear system with a well‐conditioned matrix.Numerical experiments,involving complicated geometry and the high noise level,confirm the ef-fectiveness and performance of the LMFS for solving 3D acoustic inverse problems.
基金supported by US National Science Foundation (Grant No. SES-0631613)the Cowles Foundation for Research in Economics
文摘This paper considers the estimation of an unknown function h that can be characterized as a solution to a nonlinear operator equation mapping between two infinite dimensional Hilbert spaces. The nonlinear operator is unknown but can be consistently estimated, and its inverse is discontinuous, rendering the problem ill-posed. We establish the consistency for the class of estimators that are regularized using general lower semicompact penalty functions. We derive the optimal convergence rates of the estimators under the Hilbert scale norms. We apply our results to two important problems in economics and finance: (1) estimating the parameters of the pricing kernel of defaultable bonds; (2) recovering the volatility surface implied by option prices allowing for measurement error in the option prices and numerical error in the computation of the operator.
基金supported in part by the Fundamental Research Funds for the Central Universities(No.3122019152)the National Natural Science Foundation of China(Grant Nos.11701256,11871258)+2 种基金the Youth Backbone Teacher Foundation of Henan's University(No.2019GGJS196)the China Scholarship Council(Grant No.201908410132)was also supported in part by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada(Grant No.RGPIN 2017-03903).
文摘Let G be a finite abelian group and S be a sequence with elements of G.We say that S is a regular sequence over G if|SH|≤|H|-1 holds for every proper subgroup H of G,where SH denotes the subsequence of S consisting of all terms of S contained in H.We say that S is a zero-sum free sequence over G if 0■Σ(S),where Σ(S)■G denotes the set of group elements which can be expressed as a sum of a nonempty subsequence of S.In this paper,we study the inverse problems associated with Σ(S)when S is a regular sequence or a zero-sum free sequence over G=Cp■Cp,where p is a prime.
基金This work was supported by Simons Foundation Collaboration Grant[351025]。
文摘A general stochastic algorithm for solving mixed linear and nonlinear problems was introduced in[11].We show in this paper how it can be used to solve the fault inverse problem,where a planar fault in elastic half-space and a slip on that fault have to be reconstructed from noisy surface displacement measurements.With the parameter giving the plane containing the fault denoted by m and the regularization parameter for the linear part of the inverse problem denoted by C,both modeled as random variables,we derive a formula for the posterior marginal of m.Modeling C as a random variable allows to sweep through a wide range of possible values which was shown to be superior to selecting a fixed value[11].We prove that this posterior marginal of m is convergent as the number of measurement points and the dimension of the space for discretizing slips increase.Simply put,our proof only assumes that the regularized discrete error functional for processing measurements relates to an order 1 quadrature rule and that the union of the finite-dimensional spaces for discretizing slips is dense.Our proof relies on trace class operator theory to show that an adequate sequence of determinants is uniformly bounded.We also explain how our proof can be extended to a whole class of inverse problems,as long as some basic requirements are met.Finally,we show numerical simulations that illustrate the numerical convergence of our algorithm.
基金Project supported by the National Natural Science Fundation of China
文摘This paper describes a new method which has been developed for the solution of direct and inverse problems of 3-D compressible flows in turbomachinery.Two types of streamfunctions are proposed in the paper and the streamfunction-coordinate system is applied in numerical computations.The algorithm is applied to stator blades and the results are compared with experimental data,It is shown that the comparisons are very satis- factory.
文摘Various works from the literature aimed at accelerating Bayesian inference in inverse problems.Stochastic spectral methods have been recently proposed as surrogate approximations of the forward uncertainty propagation model over the support of the prior distribution.These representations are efficient because they allow affordable simulation of a large number of samples from the posterior distribution.Unfortunately,they do not perform well when the forward model exhibits strong nonlinear behavior with respect to its input.In this work,we first relate the fast(exponential)L2-convergence of the forward approximation to the fast(exponential)convergence(in terms of Kullback-Leibler divergence)of the approximate posterior.In particular,we prove that in case the prior distribution is uniform,the posterior is at least twice as fast as the convergence rate of the forward model in those norms.The Bayesian inference strategy is developed in the framework of a stochastic spectral projection method.The predicted convergence rates are then demonstrated for simple nonlinear inverse problems of varying smoothness.We then propose an efficient numerical approach for the Bayesian solution of inverse problems presenting strongly nonlinear or discontinuous system responses.This comes with the improvement of the forward model that is adaptively approximated by an iterative generalized Polynomial Chaos-based representation.The numerical approximations and predicted convergence rates of the former approach are compared to the new iterative numerical method for nonlinear time-dependent test cases of varying dimension and complexity,which are relevant regarding our hydrodynamics motivations and therefore regarding hyperbolic conservation laws and the apparition of discontinuities in finite time.
基金The authors acknowledge the financial support provided by Natural Science Fund for Excellent Young Scholars of Heilongjiang Province(No.YQ2021E023)Natural Science Foundation of China(No.52076053,No.52106041)+1 种基金China Postdoctoral Science Foundation funded project(2021M690823)National Science and Technology Major Project(No.2017-III-0009-0035,No.2019-11-0010-0030).
文摘To study the feasibility of using machine learning technology to solve the forward problem(prediction of aerodynamic parameters)and the inverse problem(prediction of geometric parameters)of turbine blades,this paper built a forward problem model based on backpropagation artificial neural networks(BP-ANNs)and an inverse problem model based on radial basis function artificial neural networks(RBF-ANNs).The S2(a stream surface obtained by extending a radial curve in turbo blades)calculation program was used to generate the dataset for single-stage turbo blades,and the back propagation algorithm was used to train the model.The parameters of five blade sections in a single-stage turbine were selected as inputs of the forward problem model,including stagger angle,inlet geometric angle,outlet geometric angle,wedge angle of leading edge pressure side,wedge angle of leading edge suction side,wedge angle of trailing edge,rear bending angle,and leading edge diameter.The outputs are efficiency,power,mass flow,relative exit Mach number,absolute exit Mach number,relative exit flow angle,absolute exit flow angle and reaction degree,which are eight aerodynamic parameters.The inputs and outputs of the inverse problem model are the opposite of that of the forward problem model.The models can accurately predict the aerodynamic parameters and geometric parameters,and the mean square errors(MSEs)of the forward problem test set and the inverse problem test set are 0.001 and 0.00035,respectively.This study shows that machine learning technology based on neural networks can be flexibly applied to the design of forward and inverse problems of turbine blades,and the models built by this method have practical application value in regression prediction problems.
基金supported by the Major Project of Humanities Social Science Foundation of Ministry of Education(Grant No. 08JJD910247)Key Project of Chinese Ministry of Education (Grant No. 108120)+4 种基金National Natural Science Foundation of China (Grant No. 10871201)Beijing Natural Science Foundation (Grant No. 1102021)the Fundamental Research Funds for the Central Universitiesthe Research Funds of Renmin University of China(Grant No. 10XNL018)the China Statistical Research Project (Grant No. 2011LZ031)
文摘In this article we consider a sequence of hierarchical space model of inverse problems.The underlying function is estimated from indirect observations over a variety of error distributions including those that are heavy-tailed and may not even possess variances or means.The main contribution of this paper is that we establish some oracle inequalities for the inverse problems by using quantile coupling technique that gives a tight bound for the quantile coupling between an arbitrary sample p-quantile and a normal variable,and an automatic selection principle for the nonrandom filters.This leads to the data-driven choice of weights.We also give an algorithm for its implementation.The quantile coupling inequality developed in this paper is of independent interest,because it includes the median coupling inequality in literature as a special case.
文摘A fractal approximation algorithm is developed to obtain approximate solutions to an inverse initial-value problem IVP(inverse IVP) for the differential equation. Numerical computational results are presented to demonstrate the effectiveness of this algorithm for solving inverse IVP for a class of specific differential equations.
