We propose a reverse-space nonlocal nonlinear self-dual network equation under special symmetry reduction,which may have potential applications in electric circuits.Nonlocal infinitely many conservation laws are const...We propose a reverse-space nonlocal nonlinear self-dual network equation under special symmetry reduction,which may have potential applications in electric circuits.Nonlocal infinitely many conservation laws are constructed based on its Lax pair.Nonlocal discrete generalized(m,N−m)-fold Darboux transformation is extended and applied to solve this system.As an application of the method,we obtain multi-soliton solutions in zero seed background via the nonlocal discrete N-fold Darboux transformation and rational solutions from nonzero-seed background via the nonlocal discrete generalized(1,N−1)-fold Darboux transformation,respectively.By using the asymptotic and graphic analysis,structures of one-,two-,three-and four-soliton solutions are shown and discussed graphically.We find that single component field in this nonlocal system displays unstable soliton structure whereas the combined potential terms exhibit stable soliton structures.It is shown that the soliton structures are quite different between discrete local and nonlocal systems.Results given in this paper may be helpful for understanding the electrical signals propagation.展开更多
The N-fold Darboux transformation(DT) T_n^([N]) of the nonlinear self-dual network equation is given in terms of the determinant representation. The elements in determinants are composed of the eigenvalues λ_j(j = 1,...The N-fold Darboux transformation(DT) T_n^([N]) of the nonlinear self-dual network equation is given in terms of the determinant representation. The elements in determinants are composed of the eigenvalues λ_j(j = 1, 2..., N)and the corresponding eigenfunctions of the associated Lax equation. Using this representation, the N-soliton solutions of the nonlinear self-dual network equation are given from the zero "seed" solution by the N-fold DT. A general form of the N-degenerate soliton is constructed from the determinants of N-soliton by a special limit λ_j →λ_1 and by using the higher-order Taylor expansion. For 2-degenerate and 3-degenerate solitons, approximate orbits are given analytically,which provide excellent fit of exact trajectories. These orbits have a time-dependent "phase shift", namely ln(t^2).展开更多
In this paper,we propose a symmetric difference data enhancement physics-informed neural network(SDE-PINN)to study soliton solutions for discrete nonlinear lattice equations(NLEs).By considering known and unknown symm...In this paper,we propose a symmetric difference data enhancement physics-informed neural network(SDE-PINN)to study soliton solutions for discrete nonlinear lattice equations(NLEs).By considering known and unknown symmetric points,numerical simulations are conducted to one-soliton and two-soliton solutions of a discrete KdV equation,as well as a one-soliton solution of a discrete Toda lattice equation.Compared with the existing discrete deep learning approach,the numerical results reveal that within the specified spatiotemporal domain,the prediction accuracy by SDE-PINN is excellent regardless of the interior or extrapolation prediction,with a significant reduction in training time.The proposed data enhancement technique and symmetric structure development provides a new perspective for the deep learning approach to solve discrete NLEs.The newly proposed SDE-PINN can also be applied to solve continuous nonlinear equations and other discrete NLEs numerically.展开更多
The closed form of solutions of Kac-van Moerbeke lattice and self-dual network equations are considered by proposing transformations based on Riccati equation, using symbolic computation. In contrast to the numerical ...The closed form of solutions of Kac-van Moerbeke lattice and self-dual network equations are considered by proposing transformations based on Riccati equation, using symbolic computation. In contrast to the numerical computation of travelling wave solutions for differential difference equations, our method obtains exact solutions which have physical relevance.展开更多
Intelligent Adaptive Control(AC) has remarkable advantages in the control system design of aero-engine which has strong nonlinearity and uncertainty. Inspired by the Nonlinear Autoregressive Moving Average(NARMA)-L2 a...Intelligent Adaptive Control(AC) has remarkable advantages in the control system design of aero-engine which has strong nonlinearity and uncertainty. Inspired by the Nonlinear Autoregressive Moving Average(NARMA)-L2 adaptive control, a novel Nonlinear State Space Equation(NSSE) based Adaptive neural network Control(NSSE-AC) method is proposed for the turbo-shaft engine control system design. The proposed NSSE model is derived from a special neural network with an extra layer, and the rotor speed of the gas turbine is taken as the main state variable which makes the NSSE model be able to capture the system dynamic better than the NARMA-L2 model. A hybrid Recursive Least-Square and Levenberg-Marquardt(RLS-LM) algorithm is advanced to perform the online learning of the neural network, which further enhances both the accuracy of the NSSE model and the performance of the adaptive controller. The feedback correction is also utilized in the NSSE-AC system to eliminate the steady-state tracking error. Simulation results show that, compared with the NARMA-L2 model, the NSSE model of the turboshaft engine is more accurate. The maximum modeling error is decreased from 5.92% to 0.97%when the LM algorithm is introduced to optimize the neural network parameters. The NSSE-AC method can not only achieve a better main control loop performance than the traditional controller but also limit all the constraint parameters efficiently with quick and accurate switching responses even if component degradation exists. Thus, the effectiveness of the NSSE-AC method is validated.展开更多
In this paper,based on physics-informed neural networks(PINNs),a good deep learning neural network framework that can be used to effectively solve the nonlinear evolution partial differential equations(PDEs)and other ...In this paper,based on physics-informed neural networks(PINNs),a good deep learning neural network framework that can be used to effectively solve the nonlinear evolution partial differential equations(PDEs)and other types of nonlinear physical models,we study the nonlinear Schrodinger equation(NLSE)with the generalized PT-symmetric Scarf-Ⅱpotential,which is an important physical model in many fields of nonlinear physics.Firstly,we choose three different initial values and the same Dinchlet boundaiy conditions to solve the NLSE with the generalized PT-symmetric Scarf-Ⅱpotential via the PINN deep learning method,and the obtained results are compared with ttose denved by the toditional numencal methods.Then,we mvestigate effect of two factors(optimization steps and activation functions)on the performance of the PINN deep learning method in the NLSE with the generalized PT-symmetric Scarf-Ⅱpotential.Ultimately,the data-driven coefficient discovery of the generalized PT-symmetric Scarf-Ⅱpotential or the dispersion and nonlinear items of the NLSE with the generalized PT-symmetric Scarf-Ⅱpotential can be approximately ascertained by using the PINN deep learning method.Our results may be meaningful for further investigation of the nonlinear Schrodmger equation with the generalized PT-symmetric Scarf-Ⅱpotential in the deep learning.展开更多
We propose an effective scheme of the deep learning method for high-order nonlinear soliton equations and explore the influence of activation functions on the calculation results for higherorder nonlinear soliton equa...We propose an effective scheme of the deep learning method for high-order nonlinear soliton equations and explore the influence of activation functions on the calculation results for higherorder nonlinear soliton equations. The physics-informed neural networks approximate the solution of the equation under the conditions of differential operator, initial condition and boundary condition. We apply this method to high-order nonlinear soliton equations, and verify its efficiency by solving the fourth-order Boussinesq equation and the fifth-order Korteweg–de Vries equation. The results show that the deep learning method can be used to solve high-order nonlinear soliton equations and reveal the interaction between solitons.展开更多
The novel multisoliton solutions for the nonlinear lumped self-dual network equations, Toda lattice and KP equation were obtained by using the Hirota direct method.
A heuristic technique is developed for a nonlinear magnetohydrodynamics （MHD） Jeffery-Hamel problem with the help of the feed-forward artificial neural net- work （ANN） optimized with the genetic algorithm （GA） a...A heuristic technique is developed for a nonlinear magnetohydrodynamics （MHD） Jeffery-Hamel problem with the help of the feed-forward artificial neural net- work （ANN） optimized with the genetic algorithm （GA） and the sequential quadratic programming （SQP） method. The twodimensional （2D） MHD Jeffery-Hamel problem is transformed into a higher order boundary value problem （BVP） of ordinary differential equations （ODEs）. The mathematical model of the transformed BVP is formulated with the ANN in an unsupervised manner. The training of the weights of the ANN is carried out with the evolutionary calculation based on the GA hybridized with the SQP method for the rapid local convergence. The proposed scheme is evaluated on the variants of the Jeffery-Hamel flow by varying the Reynold number, the Hartmann number, and the an- gles of the walls. A large number of simulations are performed with an extensive analysis to validate the accuracy, convergence, and effectiveness of the scheme. The comparison of the standard numerical solution and the analytic solution establishes the correctness of the proposed designed methodologies.展开更多
This paper presents a design method of H<sub>2</sub> and H<sub>∞</sub>-feedback control loop for nonlinear smooth gene networks that are in control affine form. Formulaic solution methodology ...This paper presents a design method of H<sub>2</sub> and H<sub>∞</sub>-feedback control loop for nonlinear smooth gene networks that are in control affine form. Formulaic solution methodology for solving the nonlinear partial differential equations, namely the Hamilton-Jacobi-Bellman and Hamilton-Jacobi-Isaacs equations through successive Galerkin’s approximation is implemented and the results are compared. Throughout the implementation, there were several caveats that need to be further resolved for practical applications in general cases. Such issues and the clarification of causes are mathematically established and reviewed.展开更多
The defocusing action ground state of the nonlinear Schrodinger equation can be characterized via three different but equivalent minimization formulations.In this work,we propose some deep neural network(DNN)approache...The defocusing action ground state of the nonlinear Schrodinger equation can be characterized via three different but equivalent minimization formulations.In this work,we propose some deep neural network(DNN)approaches to compute the action ground state through the three formulations.We first consider the unconstrained formulation,where we propose the DNN with a shift layer and demonstrate its necessity towards finding the correct ground state.The other two formulations involve the L^(p+1)-normalization or the Nehari manifold constraint.We enforce them as hard constraints into the networks by further proposing a normalization layer or a projection layer to the DNN.Our DNNs can then be trained in an unconstrained and unsupervised manner.Systematical numerical experiments are conducted to demonstrate the effectiveness and superiority of the approaches.展开更多
Optical solitons,as self-sustaining waveforms in a nonlinear medium where dispersion and nonlinear effects are balanced,have key applications in ultrafast laser systems and optical communications.Physics-informed neur...Optical solitons,as self-sustaining waveforms in a nonlinear medium where dispersion and nonlinear effects are balanced,have key applications in ultrafast laser systems and optical communications.Physics-informed neural networks(PINN)provide a new way to solve the nonlinear Schrodinger equation describing the soliton evolution by fusing data-driven and physical constraints.However,the grid point sampling strategy of traditional PINN suffers from high computational complexity and unstable gradient flow,which makes it difficult to capture the physical details efficiently.In this paper,we propose a residual-based adaptive multi-distribution(RAMD)sampling method to optimize the PINN training process by dynamically constructing a multi-modal loss distribution.With a 50%reduction in the number of grid points,RAMD significantly reduces the relative error of PINN and,in particular,optimizes the solution error of the(2+1)Ginzburg–Landau equation from 4.55%to 1.98%.RAMD breaks through the lack of physical constraints in the purely data-driven model by the innovative combination of multi-modal distribution modeling and autonomous sampling control for the design of all-optical communication devices.RAMD provides a high-precision numerical simulation tool for the design of all-optical communication devices,optimization of nonlinear laser devices,and other studies.展开更多
In this paper,physics-informed liquid networks(PILNs)are proposed based on liquid time-constant networks(LTC)for solving nonlinear partial differential equations(PDEs).In this approach,the network state is controlled ...In this paper,physics-informed liquid networks(PILNs)are proposed based on liquid time-constant networks(LTC)for solving nonlinear partial differential equations(PDEs).In this approach,the network state is controlled via ordinary differential equations(ODEs).The significant advantage is that neurons controlled by ODEs are more expressive compared to simple activation functions.In addition,the PILNs use difference schemes instead of automatic differentiation to construct the residuals of PDEs,which avoid information loss in the neighborhood of sampling points.As this method draws on both the traveling wave method and physics-informed neural networks(PINNs),it has a better physical interpretation.Finally,the KdV equation and the nonlinear Schr¨odinger equation are solved to test the generalization ability of the PILNs.To the best of the authors’knowledge,this is the first deep learning method that uses ODEs to simulate the numerical solutions of PDEs.展开更多
With the advent of physics informed neural networks(PINNs),deep learning has gained interest for solving nonlinear partial differential equations(PDEs)in recent years.In this paper,physics informed memory networks(PIM...With the advent of physics informed neural networks(PINNs),deep learning has gained interest for solving nonlinear partial differential equations(PDEs)in recent years.In this paper,physics informed memory networks(PIMNs)are proposed as a new approach to solving PDEs by using physical laws and dynamic behavior of PDEs.Unlike the fully connected structure of the PINNs,the PIMNs construct the long-term dependence of the dynamics behavior with the help of the long short-term memory network.Meanwhile,the PDEs residuals are approximated using difference schemes in the form of convolution filter,which avoids information loss at the neighborhood of the sampling points.Finally,the performance of the PIMNs is assessed by solving the Kd V equation and the nonlinear Schr?dinger equation,and the effects of difference schemes,boundary conditions,network structure and mesh size on the solutions are discussed.Experiments show that the PIMNs are insensitive to boundary conditions and have excellent solution accuracy even with only the initial conditions.展开更多
This paper presents a solution methodology for H<sub>∞</sub>-feedback control design problem of Heparin controlled blood clotting network under the presence of stochastic noise. The formulaic solution pro...This paper presents a solution methodology for H<sub>∞</sub>-feedback control design problem of Heparin controlled blood clotting network under the presence of stochastic noise. The formulaic solution procedure to solve nonlinear partial differential equation, the Hamilton-Jacobi-Isaacs equation with Successive Galrkin’s Approximation is sketched and validity is proved. According to Lyapunov’s theory, with solutions of the nonlinear PDEs, robust feedback control is designed. To confirm the performance and robustness of the designed controller, numerical and Monte-Carlo simulation results by Simulink software on MATLAB are provided.展开更多
1 Introduction and Main Results Consider the following differential equation x(t)=f(x(t)),x(0)=x0,where x∈Rn denotes the state variable of system(1.1),f:Rn→Rn is a nonlinear vector field and x0 is the initial value ...1 Introduction and Main Results Consider the following differential equation x(t)=f(x(t)),x(0)=x0,where x∈Rn denotes the state variable of system(1.1),f:Rn→Rn is a nonlinear vector field and x0 is the initial value of the system.展开更多
The transient simulation technology of natural gas pipeline networks plays an increasingly prominent role in the scheduling management of natural gas pipeline network system.The increasingly large and complex natural ...The transient simulation technology of natural gas pipeline networks plays an increasingly prominent role in the scheduling management of natural gas pipeline network system.The increasingly large and complex natural gas pipeline network requires more strictly on the calculation efficiency of transient simulation.To this end,this paper proposes a new method for the transient simulation of natural gas pipeline networks based on fracture-dimension-reduction algorithm.Firstly,a pipeline network model is abstracted into a station model,inter-station pipeline network model and connection node model.Secondlly,the pressure at the connection node connecting the station and the inter-station pipeline network is used as the basic variable to solve the general solution of the flow rate at the connection node,reconstruct the simulation model of the inter-station pipeline network,and reduce the equation set dimension of the inter-station pipeline network model.Thirdly,the transient simulation model of the natural gas pipeline network system is constructed based on the reconstructed simulation model of the inter-station pipeline network.Fnally,the calculation accuracy and efficiency of the proposed algorithm are compared and analyzed for the two working conditions of slow change of compressor speed and rapid shutdown of the compressor.And the following research results are obtained.First,the fracture-dimension-reduction algorithm has a high calculation accuracy,and the relative error of compressor outlet pressure and user pressure is less than 0.1%.Second,the calculation efficiency of the new fracture-dimension-reduction algorithm is high,and compared with the nonlinear equations solv ing method,the speed-up ratios under the two conditions are high up to 17.3 and 12.2 respectively.Third,the speed-up ratio of the fracture-dimension-reduction algorithm is linearly related to the equation set dimension of the transient simulation model of the pipeline network system.The larger the equation set dimension,the higher the speed-up ratio,which means the more complex the pipeline network model,the more remarkable the calculation speed-up effect.In conclusion,this new method improves the calculation speed while keeping the calculation accuracy,which is of great theoretical value and reference significance for improving the calculation efficiency of the transient simulation of complex natural gas pipeline network systems.展开更多
基金Project supported by the National Natural Science Foundation of China(Grant Nos.12071042 and 61471406)the Beijing Natural Science Foundation,China(Grant No.1202006)Qin Xin Talents Cultivation Program of Beijing Information Science and Technology University(QXTCP-B201704).
文摘We propose a reverse-space nonlocal nonlinear self-dual network equation under special symmetry reduction,which may have potential applications in electric circuits.Nonlocal infinitely many conservation laws are constructed based on its Lax pair.Nonlocal discrete generalized(m,N−m)-fold Darboux transformation is extended and applied to solve this system.As an application of the method,we obtain multi-soliton solutions in zero seed background via the nonlocal discrete N-fold Darboux transformation and rational solutions from nonzero-seed background via the nonlocal discrete generalized(1,N−1)-fold Darboux transformation,respectively.By using the asymptotic and graphic analysis,structures of one-,two-,three-and four-soliton solutions are shown and discussed graphically.We find that single component field in this nonlocal system displays unstable soliton structure whereas the combined potential terms exhibit stable soliton structures.It is shown that the soliton structures are quite different between discrete local and nonlocal systems.Results given in this paper may be helpful for understanding the electrical signals propagation.
基金Supported by the Natural Science Foundation of Zhejiang Province under Grant No.LY15A010005the Natural Science Foundation of Ningbo under Grant No.2018A610197+1 种基金the NSF of China under Grant No.11671219K.C.Wong Magna Fund in Ningbo University
文摘The N-fold Darboux transformation(DT) T_n^([N]) of the nonlinear self-dual network equation is given in terms of the determinant representation. The elements in determinants are composed of the eigenvalues λ_j(j = 1, 2..., N)and the corresponding eigenfunctions of the associated Lax equation. Using this representation, the N-soliton solutions of the nonlinear self-dual network equation are given from the zero "seed" solution by the N-fold DT. A general form of the N-degenerate soliton is constructed from the determinants of N-soliton by a special limit λ_j →λ_1 and by using the higher-order Taylor expansion. For 2-degenerate and 3-degenerate solitons, approximate orbits are given analytically,which provide excellent fit of exact trajectories. These orbits have a time-dependent "phase shift", namely ln(t^2).
基金supported by the National Natural Science Foundation of China(Grant No.12071042)the Beijing Natural Science Foundation(Grant No.1202004)Promoting the Development of University Classification-Student Innovation and Entrepreneurship Training Programme(Grant No.5112410857)。
文摘In this paper,we propose a symmetric difference data enhancement physics-informed neural network(SDE-PINN)to study soliton solutions for discrete nonlinear lattice equations(NLEs).By considering known and unknown symmetric points,numerical simulations are conducted to one-soliton and two-soliton solutions of a discrete KdV equation,as well as a one-soliton solution of a discrete Toda lattice equation.Compared with the existing discrete deep learning approach,the numerical results reveal that within the specified spatiotemporal domain,the prediction accuracy by SDE-PINN is excellent regardless of the interior or extrapolation prediction,with a significant reduction in training time.The proposed data enhancement technique and symmetric structure development provides a new perspective for the deep learning approach to solve discrete NLEs.The newly proposed SDE-PINN can also be applied to solve continuous nonlinear equations and other discrete NLEs numerically.
基金The project supported by ＂973＂ Project under Grant No.2004CB318000, the Doctor Start-up Foundation of Liaoning Province of China under Grant No. 20041066, and the Science Research Plan of Liaoning Education Bureau under Grant No. 2004F099
文摘The closed form of solutions of Kac-van Moerbeke lattice and self-dual network equations are considered by proposing transformations based on Riccati equation, using symbolic computation. In contrast to the numerical computation of travelling wave solutions for differential difference equations, our method obtains exact solutions which have physical relevance.
基金co-supported by the National Science and Technology Major Project, China (No. J2019-Ⅰ-0010-0010)the Project funded by China Postdoctoral Science Foundation (No. 2021M701692)+3 种基金the Fundamental Research Funds for the Central Universities, China (No. NS2022029)the Postgraduate Research & Practice Innovation Program of NUAA, China (No. xcxjh20220206)the National Natural Science Foundation of China (No. 51976089)Jiangsu Funding Program for Excellent Postdoctoral Talent, China (No. 2022ZB202)。
文摘Intelligent Adaptive Control(AC) has remarkable advantages in the control system design of aero-engine which has strong nonlinearity and uncertainty. Inspired by the Nonlinear Autoregressive Moving Average(NARMA)-L2 adaptive control, a novel Nonlinear State Space Equation(NSSE) based Adaptive neural network Control(NSSE-AC) method is proposed for the turbo-shaft engine control system design. The proposed NSSE model is derived from a special neural network with an extra layer, and the rotor speed of the gas turbine is taken as the main state variable which makes the NSSE model be able to capture the system dynamic better than the NARMA-L2 model. A hybrid Recursive Least-Square and Levenberg-Marquardt(RLS-LM) algorithm is advanced to perform the online learning of the neural network, which further enhances both the accuracy of the NSSE model and the performance of the adaptive controller. The feedback correction is also utilized in the NSSE-AC system to eliminate the steady-state tracking error. Simulation results show that, compared with the NARMA-L2 model, the NSSE model of the turboshaft engine is more accurate. The maximum modeling error is decreased from 5.92% to 0.97%when the LM algorithm is introduced to optimize the neural network parameters. The NSSE-AC method can not only achieve a better main control loop performance than the traditional controller but also limit all the constraint parameters efficiently with quick and accurate switching responses even if component degradation exists. Thus, the effectiveness of the NSSE-AC method is validated.
基金supported by the National Natural Science Foundation of China under Grant Nos.11775121,11435005the K.C.Wong Magna Fund of Ningbo University。
文摘In this paper,based on physics-informed neural networks(PINNs),a good deep learning neural network framework that can be used to effectively solve the nonlinear evolution partial differential equations(PDEs)and other types of nonlinear physical models,we study the nonlinear Schrodinger equation(NLSE)with the generalized PT-symmetric Scarf-Ⅱpotential,which is an important physical model in many fields of nonlinear physics.Firstly,we choose three different initial values and the same Dinchlet boundaiy conditions to solve the NLSE with the generalized PT-symmetric Scarf-Ⅱpotential via the PINN deep learning method,and the obtained results are compared with ttose denved by the toditional numencal methods.Then,we mvestigate effect of two factors(optimization steps and activation functions)on the performance of the PINN deep learning method in the NLSE with the generalized PT-symmetric Scarf-Ⅱpotential.Ultimately,the data-driven coefficient discovery of the generalized PT-symmetric Scarf-Ⅱpotential or the dispersion and nonlinear items of the NLSE with the generalized PT-symmetric Scarf-Ⅱpotential can be approximately ascertained by using the PINN deep learning method.Our results may be meaningful for further investigation of the nonlinear Schrodmger equation with the generalized PT-symmetric Scarf-Ⅱpotential in the deep learning.
基金supported by National Science Foundation of China(52171251)Liao Ning Revitalization Talents Program(XLYC1907014)+2 种基金the Fundamental Research Funds for the Central Universities(DUT21ZD205)Ministry of Industry and Information Technology(2019-357)the Project of State Key Laboratory of Satellite Ocean Environment Dynamics,Second Institute of Oceanography,MNR(QNHX2112)。
文摘We propose an effective scheme of the deep learning method for high-order nonlinear soliton equations and explore the influence of activation functions on the calculation results for higherorder nonlinear soliton equations. The physics-informed neural networks approximate the solution of the equation under the conditions of differential operator, initial condition and boundary condition. We apply this method to high-order nonlinear soliton equations, and verify its efficiency by solving the fourth-order Boussinesq equation and the fifth-order Korteweg–de Vries equation. The results show that the deep learning method can be used to solve high-order nonlinear soliton equations and reveal the interaction between solitons.
文摘The novel multisoliton solutions for the nonlinear lumped self-dual network equations, Toda lattice and KP equation were obtained by using the Hirota direct method.
文摘A heuristic technique is developed for a nonlinear magnetohydrodynamics （MHD） Jeffery-Hamel problem with the help of the feed-forward artificial neural net- work （ANN） optimized with the genetic algorithm （GA） and the sequential quadratic programming （SQP） method. The twodimensional （2D） MHD Jeffery-Hamel problem is transformed into a higher order boundary value problem （BVP） of ordinary differential equations （ODEs）. The mathematical model of the transformed BVP is formulated with the ANN in an unsupervised manner. The training of the weights of the ANN is carried out with the evolutionary calculation based on the GA hybridized with the SQP method for the rapid local convergence. The proposed scheme is evaluated on the variants of the Jeffery-Hamel flow by varying the Reynold number, the Hartmann number, and the an- gles of the walls. A large number of simulations are performed with an extensive analysis to validate the accuracy, convergence, and effectiveness of the scheme. The comparison of the standard numerical solution and the analytic solution establishes the correctness of the proposed designed methodologies.
文摘This paper presents a design method of H<sub>2</sub> and H<sub>∞</sub>-feedback control loop for nonlinear smooth gene networks that are in control affine form. Formulaic solution methodology for solving the nonlinear partial differential equations, namely the Hamilton-Jacobi-Bellman and Hamilton-Jacobi-Isaacs equations through successive Galerkin’s approximation is implemented and the results are compared. Throughout the implementation, there were several caveats that need to be further resolved for practical applications in general cases. Such issues and the clarification of causes are mathematically established and reviewed.
基金supported by the National Natural Science Foundation of China No.12271413the Natural Science Foundation of Hubei Province No.2019CFA007the Fundamental Research Funds for the Central Universities No.2042024kf0016。
文摘The defocusing action ground state of the nonlinear Schrodinger equation can be characterized via three different but equivalent minimization formulations.In this work,we propose some deep neural network(DNN)approaches to compute the action ground state through the three formulations.We first consider the unconstrained formulation,where we propose the DNN with a shift layer and demonstrate its necessity towards finding the correct ground state.The other two formulations involve the L^(p+1)-normalization or the Nehari manifold constraint.We enforce them as hard constraints into the networks by further proposing a normalization layer or a projection layer to the DNN.Our DNNs can then be trained in an unconstrained and unsupervised manner.Systematical numerical experiments are conducted to demonstrate the effectiveness and superiority of the approaches.
基金supported by the National Key R&D Program of China(Grant No.2022YFA1604200)National Natural Science Foundation of China(Grant No.12261131495)+1 种基金Beijing Municipal Science and Technology Commission,Adminitrative Commission of Zhongguancun Science Park(Grant No.Z231100006623006)Institute of Systems Science,Beijing Wuzi University(Grant No.BWUISS21)。
文摘Optical solitons,as self-sustaining waveforms in a nonlinear medium where dispersion and nonlinear effects are balanced,have key applications in ultrafast laser systems and optical communications.Physics-informed neural networks(PINN)provide a new way to solve the nonlinear Schrodinger equation describing the soliton evolution by fusing data-driven and physical constraints.However,the grid point sampling strategy of traditional PINN suffers from high computational complexity and unstable gradient flow,which makes it difficult to capture the physical details efficiently.In this paper,we propose a residual-based adaptive multi-distribution(RAMD)sampling method to optimize the PINN training process by dynamically constructing a multi-modal loss distribution.With a 50%reduction in the number of grid points,RAMD significantly reduces the relative error of PINN and,in particular,optimizes the solution error of the(2+1)Ginzburg–Landau equation from 4.55%to 1.98%.RAMD breaks through the lack of physical constraints in the purely data-driven model by the innovative combination of multi-modal distribution modeling and autonomous sampling control for the design of all-optical communication devices.RAMD provides a high-precision numerical simulation tool for the design of all-optical communication devices,optimization of nonlinear laser devices,and other studies.
基金supported by the National Natural Science Foundation of China under Grant Nos.11975143 and 12105161.
文摘In this paper,physics-informed liquid networks(PILNs)are proposed based on liquid time-constant networks(LTC)for solving nonlinear partial differential equations(PDEs).In this approach,the network state is controlled via ordinary differential equations(ODEs).The significant advantage is that neurons controlled by ODEs are more expressive compared to simple activation functions.In addition,the PILNs use difference schemes instead of automatic differentiation to construct the residuals of PDEs,which avoid information loss in the neighborhood of sampling points.As this method draws on both the traveling wave method and physics-informed neural networks(PINNs),it has a better physical interpretation.Finally,the KdV equation and the nonlinear Schr¨odinger equation are solved to test the generalization ability of the PILNs.To the best of the authors’knowledge,this is the first deep learning method that uses ODEs to simulate the numerical solutions of PDEs.
文摘With the advent of physics informed neural networks(PINNs),deep learning has gained interest for solving nonlinear partial differential equations(PDEs)in recent years.In this paper,physics informed memory networks(PIMNs)are proposed as a new approach to solving PDEs by using physical laws and dynamic behavior of PDEs.Unlike the fully connected structure of the PINNs,the PIMNs construct the long-term dependence of the dynamics behavior with the help of the long short-term memory network.Meanwhile,the PDEs residuals are approximated using difference schemes in the form of convolution filter,which avoids information loss at the neighborhood of the sampling points.Finally,the performance of the PIMNs is assessed by solving the Kd V equation and the nonlinear Schr?dinger equation,and the effects of difference schemes,boundary conditions,network structure and mesh size on the solutions are discussed.Experiments show that the PIMNs are insensitive to boundary conditions and have excellent solution accuracy even with only the initial conditions.
文摘This paper presents a solution methodology for H<sub>∞</sub>-feedback control design problem of Heparin controlled blood clotting network under the presence of stochastic noise. The formulaic solution procedure to solve nonlinear partial differential equation, the Hamilton-Jacobi-Isaacs equation with Successive Galrkin’s Approximation is sketched and validity is proved. According to Lyapunov’s theory, with solutions of the nonlinear PDEs, robust feedback control is designed. To confirm the performance and robustness of the designed controller, numerical and Monte-Carlo simulation results by Simulink software on MATLAB are provided.
基金Supported by National Natural Science Foundation of China(61973241).
文摘1 Introduction and Main Results Consider the following differential equation x(t)=f(x(t)),x(0)=x0,where x∈Rn denotes the state variable of system(1.1),f:Rn→Rn is a nonlinear vector field and x0 is the initial value of the system.
文摘The transient simulation technology of natural gas pipeline networks plays an increasingly prominent role in the scheduling management of natural gas pipeline network system.The increasingly large and complex natural gas pipeline network requires more strictly on the calculation efficiency of transient simulation.To this end,this paper proposes a new method for the transient simulation of natural gas pipeline networks based on fracture-dimension-reduction algorithm.Firstly,a pipeline network model is abstracted into a station model,inter-station pipeline network model and connection node model.Secondlly,the pressure at the connection node connecting the station and the inter-station pipeline network is used as the basic variable to solve the general solution of the flow rate at the connection node,reconstruct the simulation model of the inter-station pipeline network,and reduce the equation set dimension of the inter-station pipeline network model.Thirdly,the transient simulation model of the natural gas pipeline network system is constructed based on the reconstructed simulation model of the inter-station pipeline network.Fnally,the calculation accuracy and efficiency of the proposed algorithm are compared and analyzed for the two working conditions of slow change of compressor speed and rapid shutdown of the compressor.And the following research results are obtained.First,the fracture-dimension-reduction algorithm has a high calculation accuracy,and the relative error of compressor outlet pressure and user pressure is less than 0.1%.Second,the calculation efficiency of the new fracture-dimension-reduction algorithm is high,and compared with the nonlinear equations solv ing method,the speed-up ratios under the two conditions are high up to 17.3 and 12.2 respectively.Third,the speed-up ratio of the fracture-dimension-reduction algorithm is linearly related to the equation set dimension of the transient simulation model of the pipeline network system.The larger the equation set dimension,the higher the speed-up ratio,which means the more complex the pipeline network model,the more remarkable the calculation speed-up effect.In conclusion,this new method improves the calculation speed while keeping the calculation accuracy,which is of great theoretical value and reference significance for improving the calculation efficiency of the transient simulation of complex natural gas pipeline network systems.
