Scholarlink

<p>The Jimbo-Miwa (3 + 1)-dimensional nonlinear equation is a well-known high-dimensional extension of integrable models and is widely used to describe nonlinear wave behavior in plasma physics, fluid dynamics, nonlinear optics, and quantum field theory. In this paper, we present a structured symbolic approach based on the Bilinear Neural Network Method (BNNM) to address the complexities associated with solving this equation. The proposed framework combines two complementary approaches: deriving exact soliton solutions through the Hirota bilinear method and developing a neural network-based scheme to obtain analytical approximations of the Jimbo-Miwa dynamics. Within this framework, we introduce several neural network ansatz structures, including 4-2-1, 4-3-1, and deep 4-2-2-1 architectures, which allow for the systematic construction of broad families of exact analytical solutions. Our results include explicit rational lump solutions, periodic breather solutions, and rich hybrid interactions between lumps and solitons. This combined approach not only recovers known solution forms but also provides greater flexibility for exploring solution spaces that are difficult to handle with conventional methods. Furthermore, the neural network framework offers an effective route to approximate analytical solutions of the Jimbo–Miwa model, improving computational efficiency and enhancing predictive capability in application-oriented settings.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Exact soliton, lump, and breather solutions of the (3 + 1)-dimensional Jimbo-Miwa equation via the bilinear neural network method

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月4日

查看全文
  • Hisham H. Hussein,
  • Hosam Mekawey,
  • Ahmed Elsheikh

摘要

The Jimbo-Miwa (3 + 1)-dimensional nonlinear equation is a well-known high-dimensional extension of integrable models and is widely used to describe nonlinear wave behavior in plasma physics, fluid dynamics, nonlinear optics, and quantum field theory. In this paper, we present a structured symbolic approach based on the Bilinear Neural Network Method (BNNM) to address the complexities associated with solving this equation. The proposed framework combines two complementary approaches: deriving exact soliton solutions through the Hirota bilinear method and developing a neural network-based scheme to obtain analytical approximations of the Jimbo-Miwa dynamics. Within this framework, we introduce several neural network ansatz structures, including 4-2-1, 4-3-1, and deep 4-2-2-1 architectures, which allow for the systematic construction of broad families of exact analytical solutions. Our results include explicit rational lump solutions, periodic breather solutions, and rich hybrid interactions between lumps and solitons. This combined approach not only recovers known solution forms but also provides greater flexibility for exploring solution spaces that are difficult to handle with conventional methods. Furthermore, the neural network framework offers an effective route to approximate analytical solutions of the Jimbo–Miwa model, improving computational efficiency and enhancing predictive capability in application-oriented settings.