Junchao Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Junchao Chen has a total of 68 co-authors. Here are the top co-authors by number of shared publications:
Yong Chen
Shanghai Key Laboratory of Trustworthy Computing, East China Normal University, Shanghai, 200062, People’s Republic of China
15 shared publicationsBao-Feng Feng
School of Mathematical and Statistical Sciences, The University of Texas Rio Grande Valley, Edinburg, Texas 78541, USA
9 shared publications- Biao Li
School of Mathematics and Statistics, Ningbo University, Ningbo, China
9 shared publications Ken-Ichi Maruno
Department of Applied Mathematics, School of Fundamental Science and Engineering, Waseda University, 3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
6 shared publications- Yong Chen
The Center for Health AI and Synthesis of Evidence (CHASE), University of Pennsylvania Philadelphia Pennsylvania USA
6 shared publications Zhengyi Ma
5 shared publications
Latest Publications
- 17 November 2026
- Published by IOP Publishing in Journal of Physics A: Mathematical and Theoretical
- Vol. 59 (46) , 465201
Tau-functions of integrable discrete massive Thirring model under nonvanishing boundary conditions
The connection between the massive Thirring (MT) model and the KP hierarchy is made clear recently and it is shown that the MT model can be derived from a set of four bilinear equations under nonvanishing boundary conditions. In the present work, we demonstrate that all four bilinear equations originate from a set of four discrete KP equations, thus the integrable discretizations of these four bilinear equations are naturally given. Based on this finding, the semi- and fully discrete MT model equations are re-derived, the dark soliton, breather and first-order rogue wave solutions are constructed.
- 17 November 2026
- Published by IOP Publishing in Journal of Physics A: Mathematical and Theoretical
- Vol. 59 (46) , 465201
Tau-functions of integrable discrete massive Thirring model under nonvanishing boundary conditions
The connection between the massive Thirring (MT) model and the KP hierarchy is made clear recently and it is shown that the MT model can be derived from a set of four bilinear equations under nonvanishing boundary conditions. In the present work, we demonstrate that all four bilinear equations originate from a set of four discrete KP equations, thus the integrable discretizations of these four bilinear equations are naturally given. Based on this finding, the semi- and fully discrete MT model equations are re-derived, the dark soliton, breather and first-order rogue wave solutions are constructed.
- 18 September 2026
- Published by Springer Nature in The European Physical Journal B
- Vol. 99 (9) , 142
Gradient-based spatially weighted physics-informed convolutional recurrent network for solving discrete integrable equations
- 18 September 2026
- Published by Springer Nature in The European Physical Journal B
- Vol. 99 (9) , 142
Gradient-based spatially weighted physics-informed convolutional recurrent network for solving discrete integrable equations
- Top Cited Papers
- 1 June 2026
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 36 (6)
Bayesian hyper-parameter optimization of physics-informed neural networks to a nonlocal nonlinear Schrödinger equation
In previous applications of a physical information neural networks (PINN) algorithm, the adjustment of hyper-parameter combinations for neural networks has predominantly been conducted manually. This manual configuration strategy often suffers from inefficiency, and it cannot guarantee that the network trained with such parameters can achieve optimal performance. To address the hyper-parameter tuning challenge of PINNs, this paper is devoted to introducing a Bayesian hyper-parameter optimization method for PINN (BHPO-PINN) based on a Gaussian process. This approach enhances training performance by identifying optimal combinations of different hyper-parameters and is further applied to solve the nonlocal reverse-time nonlinear Schrödinger equation. Numerical simulations of a data-driven one-soliton solution and four types of two-soliton solutions are performed, and systematic comparative analyses of relative errors and maximum absolute errors are presented to quantify the solution accuracy. To identify the optimal combination of network hyper-parameters, we conduct a thorough investigation by analyzing error variations under different training iterations. The experimental results demonstrate that the BHPO-PINN method achieves significantly enhanced accuracy and stability compared to the traditional PINN algorithm under empirically configured hyper-parameters.
- 1 June 2026
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 36 (6)
Top Cited PapersBayesian hyper-parameter optimization of physics-informed neural networks to a nonlocal nonlinear Schrödinger equation
In previous applications of a physical information neural networks (PINN) algorithm, the adjustment of hyper-parameter combinations for neural networks has predominantly been conducted manually. This manual configuration strategy often suffers from inefficiency, and it cannot guarantee that the network trained with such parameters can achieve optimal performance. To address the hyper-parameter tuning challenge of PINNs, this paper is devoted to introducing a Bayesian hyper-parameter optimization method for PINN (BHPO-PINN) based on a Gaussian process. This approach enhances training performance by identifying optimal combinations of different hyper-parameters and is further applied to solve the nonlocal reverse-time nonlinear Schrödinger equation. Numerical simulations of a data-driven one-soliton solution and four types of two-soliton solutions are performed, and systematic comparative analyses of relative errors and maximum absolute errors are presented to quantify the solution accuracy. To identify the optimal combination of network hyper-parameters, we conduct a thorough investigation by analyzing error variations under different training iterations. The experimental results demonstrate that the BHPO-PINN method achieves significantly enhanced accuracy and stability compared to the traditional PINN algorithm under empirically configured hyper-parameters.
- 29 January 2026
- Published by IOP Publishing in Physica Scripta
- Vol. 101 (7)
Half-horseshoe soliton of the KP-I equation
Since Ohta's systematic summary of the Airy solitons for the KP-I equation in 2025 [Water Waves (2025) 7:349-362], the explicit analytical construction of such solutions has become essentially a closed topic, except for a few remaining intricate integral issues. Building on Ohta's work, this brief study identifies a class of half-horseshoe solitons, whose crests are distributed along parabolic fronts, with amplitudes growing from infinitesimal to infinite. However, the resonant solutions generated by these line-like or lump-like half-horseshoe solitons have amplitudes that remain bounded in space. Although these solutions exhibit intriguing structures, the corresponding transformed cylindrical KP solutions still lack physical relevance, as they are non-periodic with respect to the azimuthal angle.
- 29 January 2026
- Published by IOP Publishing in Physica Scripta
- Vol. 101 (7)
Half-horseshoe soliton of the KP-I equation
Since Ohta's systematic summary of the Airy solitons for the KP-I equation in 2025 [Water Waves (2025) 7:349-362], the explicit analytical construction of such solutions has become essentially a closed topic, except for a few remaining intricate integral issues. Building on Ohta's work, this brief study identifies a class of half-horseshoe solitons, whose crests are distributed along parabolic fronts, with amplitudes growing from infinitesimal to infinite. However, the resonant solutions generated by these line-like or lump-like half-horseshoe solitons have amplitudes that remain bounded in space. Although these solutions exhibit intriguing structures, the corresponding transformed cylindrical KP solutions still lack physical relevance, as they are non-periodic with respect to the azimuthal angle.
- 1 December 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (12)
Physics-informed neural networks for exploring soliton collisions in the non-integrable Schamel equation in plasmas
Physics-informed neural networks (PINNs) have achieved the integration of data with mathematical and physical models and have emerged as a popular method for solving partial differential equations. In this paper, we employ the PINN algorithm to numerically solve the non-integrable Schamel equation that arises in plasmas. By embedding the initial-boundary conditions into this non-integrable equation within the PINN framework, we successfully learn the data-driven single-soliton solution and accurately predict three patterns of two-soliton interaction. These results demonstrate the ability of the PINN algorithm to capture complex dynamical behaviors of non-integrable systems and reveal the unique nonlinear characteristics of soliton collisions in the Schamel equation, in accordance with the results of traditional numerical methods. This study not only extends the applicability of the PINN algorithm to explore solitary wave dynamics in non-integrable models, but it also provides key theoretical support for understanding complex physical processes such as ion acoustic waves and electron capture effects in plasmas.
- 1 December 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (12)
Physics-informed neural networks for exploring soliton collisions in the non-integrable Schamel equation in plasmas
Physics-informed neural networks (PINNs) have achieved the integration of data with mathematical and physical models and have emerged as a popular method for solving partial differential equations. In this paper, we employ the PINN algorithm to numerically solve the non-integrable Schamel equation that arises in plasmas. By embedding the initial-boundary conditions into this non-integrable equation within the PINN framework, we successfully learn the data-driven single-soliton solution and accurately predict three patterns of two-soliton interaction. These results demonstrate the ability of the PINN algorithm to capture complex dynamical behaviors of non-integrable systems and reveal the unique nonlinear characteristics of soliton collisions in the Schamel equation, in accordance with the results of traditional numerical methods. This study not only extends the applicability of the PINN algorithm to explore solitary wave dynamics in non-integrable models, but it also provides key theoretical support for understanding complex physical processes such as ion acoustic waves and electron capture effects in plasmas.
- 5 November 2024
- Published by Springer Nature in Nonlinear Dynamics
- Vol. 113 (8) , 8801-8817
Data-driven soliton solutions and parameter identification of the nonlocal nonlinear Schrödinger equation using the physics-informed neural network algorithm with parameter regularization
- 5 November 2024
- Published by Springer Nature in Nonlinear Dynamics
- Vol. 113 (8) , 8801-8817
Data-driven soliton solutions and parameter identification of the nonlocal nonlinear Schrödinger equation using the physics-informed neural network algorithm with parameter regularization
- 11 July 2023
- Published by Wiley in Studies in Applied Mathematics
- Vol. 151 (3) , 1020-1052
Rogue waves in the massive Thirring model
In this paper, general rogue wave solutions in the massive Thirring (MT) model are derived by using the Kadomtsev–Petviashvili (KP) hierarchy reduction method and these rational solutions are presented explicitly in terms of determinants whose matrix elements are elementary Schur polynomials. In the reduction process, three reduction conditions including one index‐ and two dimension‐ones are proved to be consistent by only one constraint relation on parameters of tau‐functions of the KP‐Toda hierarchy. It is found that the rogue wave solutions in the MT model depend on two background parameters, which influence their orientation and duration. Differing from many other coupled integrable systems, the MT model only admits the rogue waves of bright‐type, and the higher order rogue waves represent the superposition of fundamental ones in which the nonreducible parameters determine the arrangement patterns of fundamental rogue waves. Particularly, the super rogue wave at each order can be achieved simply by setting all internal parameters to be zero, resulting in the amplitude of the sole huge peak of order N being times the background. Finally, rogue wave patterns are discussed when one of the internal parameters is large. Similar to other integrable equations, the patterns are shown to be associated with the root structures of the Yablonskii–Vorob'ev polynomial hierarchy through a linear transformation.
- 11 July 2023
- Published by Wiley in Studies in Applied Mathematics
- Vol. 151 (3) , 1020-1052
Rogue waves in the massive Thirring model
In this paper, general rogue wave solutions in the massive Thirring (MT) model are derived by using the Kadomtsev–Petviashvili (KP) hierarchy reduction method and these rational solutions are presented explicitly in terms of determinants whose matrix elements are elementary Schur polynomials. In the reduction process, three reduction conditions including one index‐ and two dimension‐ones are proved to be consistent by only one constraint relation on parameters of tau‐functions of the KP‐Toda hierarchy. It is found that the rogue wave solutions in the MT model depend on two background parameters, which influence their orientation and duration. Differing from many other coupled integrable systems, the MT model only admits the rogue waves of bright‐type, and the higher order rogue waves represent the superposition of fundamental ones in which the nonreducible parameters determine the arrangement patterns of fundamental rogue waves. Particularly, the super rogue wave at each order can be achieved simply by setting all internal parameters to be zero, resulting in the amplitude of the sole huge peak of order N being times the background. Finally, rogue wave patterns are discussed when one of the internal parameters is large. Similar to other integrable equations, the patterns are shown to be associated with the root structures of the Yablonskii–Vorob'ev polynomial hierarchy through a linear transformation.
- 21 October 2022
- Published by IOP Publishing in Physica Scripta
- Vol. 97 (11) , 115205
Peculiarities of resonant interactions of lump chains within the KP1 equation
Using the Hirota bilinear method, we derive resonant solutions to the KP1 equation. Solutions describe lump chains differently oriented in ( x , y )-plane. We show that resonant solutions arise as the limiting case of more general non-resonant solutions when phase shifts of lump chains caused by their interaction become infinite. Resonant solutions can describe both stationary patterns (for example, Y-shaped patterns consisting of three different lump chains) and non-stationary interacting parallel lump chains. In the latter case, a lump chain can be emitted/absorbed by another lump chain. As the number of lump chains increases, resonance phenomena become more complex and diversified including the cases of exchange of a lump chain by two other lump chains. The method used in this paper can be extended to apply to other integrable systems in two and three spatial dimensions such as, for example, described by Mel’nikov’s equations.
- 21 October 2022
- Published by IOP Publishing in Physica Scripta
- Vol. 97 (11) , 115205
Peculiarities of resonant interactions of lump chains within the KP1 equation
Using the Hirota bilinear method, we derive resonant solutions to the KP1 equation. Solutions describe lump chains differently oriented in ( x , y )-plane. We show that resonant solutions arise as the limiting case of more general non-resonant solutions when phase shifts of lump chains caused by their interaction become infinite. Resonant solutions can describe both stationary patterns (for example, Y-shaped patterns consisting of three different lump chains) and non-stationary interacting parallel lump chains. In the latter case, a lump chain can be emitted/absorbed by another lump chain. As the number of lump chains increases, resonance phenomena become more complex and diversified including the cases of exchange of a lump chain by two other lump chains. The method used in this paper can be extended to apply to other integrable systems in two and three spatial dimensions such as, for example, described by Mel’nikov’s equations.
- 31 August 2022
- Published by Wiley in Studies in Applied Mathematics
- Vol. 150 (1) , 35-68
Tau‐function formulation for bright, dark soliton and breather solutions to the massive Thirring model
In the present paper, we are concerned with the link between the Kadomtsev–Petviashvili–Toda (KP–Toda) hierarchy and the massive Thirring (MT) model. First, we bilinearize the MT model under both the vanishing and nonvanishing boundary conditions. Starting from a set of bilinear equations of two‐component KP–Toda hierarchy, we derive multibright solution to the MT model. Then, considering a set of bilinear equations of the single‐component KP–Toda hierarchy, multidark soliton and multibreather solutions to the MT model are constructed by imposing constraints on the parameters in two types of tau function, respectively. The dynamics and properties of one‐ and two‐soliton for bright, dark soliton and breather solutions are analyzed in details.
- 31 August 2022
- Published by Wiley in Studies in Applied Mathematics
- Vol. 150 (1) , 35-68
Tau‐function formulation for bright, dark soliton and breather solutions to the massive Thirring model
In the present paper, we are concerned with the link between the Kadomtsev–Petviashvili–Toda (KP–Toda) hierarchy and the massive Thirring (MT) model. First, we bilinearize the MT model under both the vanishing and nonvanishing boundary conditions. Starting from a set of bilinear equations of two‐component KP–Toda hierarchy, we derive multibright solution to the MT model. Then, considering a set of bilinear equations of the single‐component KP–Toda hierarchy, multidark soliton and multibreather solutions to the MT model are constructed by imposing constraints on the parameters in two types of tau function, respectively. The dynamics and properties of one‐ and two‐soliton for bright, dark soliton and breather solutions are analyzed in details.
- 10 May 2022
- Published by Elsevier in Communications in Nonlinear Science and Numerical Simulation
- Vol. 112
Degenerate lump interactions within the Kadomtsev–Petviashvili equation
- 10 May 2022
- Published by Elsevier in Communications in Nonlinear Science and Numerical Simulation
- Vol. 112
Degenerate lump interactions within the Kadomtsev–Petviashvili equation