郑州师范学院, 数学与统计学院, 河南郑州 450044
| 摘 要: | 本研究聚焦于物理信息神经网络(PINN)求解不同相界厚度变化的Allen-Cahn(AC)方程。传统数值方法在处理复杂相边界条件时,其计算效率与精确度常随问题复杂度而下降;PINN为解决此类问题提供了新途径。本研究首先详细推导了AC方程,进而构建了基于PINN的求解模型,包括设计特定的神经网络架构、构建由偏微分方程残差项、边界条件残差项和初始条件残差项所组成的损失函数,并进行模型训练。计算结果表明,所建立的PINN模型能有效求解不同相界厚度变化的AC方程,准确揭示序参数在不同相界厚度下的分布规律,且能通过改变迭代次数适应相界厚度变化带来的挑战,实现稳定高效的数值求解。此外,将PINN计算结果与有限差分法进行了对比。本研究为深入理解AC方程物理内涵、优化数值计算方法提供支持,也为PINN在材料科学、凝聚态物理等领域解决类似复杂问题奠定基础。未来可进一步拓展PINN在高维相场模型、多物理场耦合问题中的应用,结合其他技术提高相场模拟精确度和效率,并开展更多实际应用案例研究。 |
| 关 键 词: | 物理信息神经网络; Allen-Cahn方程; 相界厚度; 序参数分布; 有限差分法 |
| DOI: | 10.57237/j.se.2026.01.002 |
School of Mathematics and Statistics, Zhengzhou Normal University, Zhengzhou 450044, China
| Abstract: | This study focuses on solving the Allen-Cahn (AC) equation with consideration of varying phase boundary thickness using Physics-Informed Neural Networks (PINN). Traditional numerical methods often exhibit declining computational efficiency and accuracy as problem complexity increases when handling complex phase boundary conditions; PINNs offer a novel approach to addressing such challenges. The study first conducts a detailed derivation of the AC equation and constructs a PINN-based solution model, including three key steps: designing a specific neural network architecture; building a loss function composed of partial differential equation (PDE) residual terms, boundary condition residual terms, and initial condition residual terms; and performing model training. The computational results demonstrate that the established PINN model can effectively solve the AC equation with varying phase boundary thicknesses, accurately revealing the distribution patterns of the order parameter under different phase boundary thicknesses, and adapt to the challenges brought by changes in phase boundary thickness by adjusting the number of iterations to achieve stable and efficient numerical solutions. The phase boundary thickness parameter has a significant impact on order parameter distribution, computational stability, and convergence. Additionally, a discussion is conducted between PINN and the finite difference method. This research provides support for in-depth understanding of the physical connotation of the AC equation and optimization of numerical calculation methods, and also lays a foundation for the application of PINN in solving similar complex problems in fields such as materials science and condensed matter physics. Future work can further expand the application of PINN in high-dimensional phase-field models and multi-physics coupling problems, combine other technologies to improve the accuracy and efficiency of phase-field simulations, and carry out more studies on practical application cases. |
| Keywords: | Physics-Informed Neural Networks; Allen-Cahn Equation; Phase Boundary Thickness; Order Parameter Distribution; Finite Difference Method |
| 1. | 河南省自然科学基金项目 (252300420935) |
| 2. | 郑州师范学院大学生创新创业训练计划目 (DCY2024011) |
| 3. | 郑州师范学院青年骨干教师资助培养计划 (2026GGJS02) |
| 4. | 郑州师范学院科研启动专项经费 (2021-702442) |
| [1] | 苏阳, 吴文华, 杨志刚, 等. 机器学习加速相场计算的发展及应用 [J]. 鞍钢技术, 2024, (06): 50-60+115. |
| [2] | 甘玉凤, 陈静然, 周志华, 等. 石蜡基复合相变材料及其在储能系统中的应用 [J]. 化工进展, 2025, 44(S1): 277-287. |
| [3] | 徐海航. 基于物理信息神经网络的热力耦合数值计算方法 [D]. 大连理工大学, 2024. |
| [4] | 陈金雄, 张敏, 沈丹梅, 等. 有限差分法的一维热传导方程应用 [J]. 武夷学院学报, 2022, 41(03): 53-57. |
| [5] | 曹通. 基于有限元法的低碳钢中马氏体相变相场模拟 [D]. 中北大学, 2022. |
| [6] | 巫诚. 基于深度神经网络的偏微分方程无监督算子学习 [D]. 苏州大学, 2024. |
| [7] | 毛顺元, 董若冰, 王炜琦, 等. 在天文学中基于机器学习的流体力学问题加速求解 [C] // 中国力学学会流体力学专业委员会.第十三届全国流体力学学术会议摘要集 (下), 2024: 623. |
| [8] | 高欣雨. 基于物理信息神经网络的液滴动力学自适应界面捕捉方法研究 [D]. 大理大学, 2025. |
| [9] | 于浩. 基于物理信息神经网络模型的流体力学问题求解 [D]. 华中科技大学, 2022. |
| [10] | 许梦强. 基于物理信息神经网络的多尺度热传导分析 [D]. 合肥工业大学, 2023. |
| [11] | 余波, 许梦强, 高强. 基于物理信息神经网络的功能梯度材料稳态/瞬态热传导分析 [J]. 计算力学学报, 2023, 40(04): 594-601. |
| [12] | 尤红俊, 赵倩倩, 刁佳宇, 等. 加权基因共表达网络分析联合机器学习方法筛选扩张型心肌病生物标记及免疫浸润分析 [J]. 中国循证心血管医学杂志, 2024, 16(05): 517-522. |
| [13] | Fazly M. Qualitative dynamics of a coupled Allen-Cahn system with phase-field interactions [J]. Physica Scripta, 2025, 100(11): 115239-115239. |
| [14] | 谭美华, 范菁. 基于深度学习和物理信息神经网络的微分方程解 [J]. 长春大学学报, 2024, 34(08): 26-30. |
| [15] | 祝鸿伟. 相场方程满足物理规律的神经网络方法 [D]. 广东工业大学, 2025. |
| [16] | 张良承, 郭敏, 张云飞, 等. 小波分析结合多层感知机神经网络模型的PM2.5浓度短期预测 [J/]. 测绘科学, 1-11 (2025-12-05). |
| [17] | 李佳琦. 数据与物理模型双驱动的模型降阶技术 [D]. 电子科技大学, 2025. |
| [18] | 汤双霞. 基于TensorFlow.js和JSDoop的神经网络训练[J].信息技术与信息化, 2021, (07): 68-69+75 |
| [19] | 苏玉刚, 阳剑, 戴欣, 等. 基于TensorFlow神经网络的MCR-WPT系统负载与互感识别方法 [J]. 电力系统自动化, 2021, 45(18): 162-169. |