有效的数据采样策略和物理信息的神经网络的边界条件约束,用于识别固体力学中的材料特性
W Wu1,2, M Daneker3, M A Jolley1,2
1Department of Anesthesiology and Critical Care Medicine, Children's Hospital of Philadelphia, Philadelphia, PA 19104, U. S. A.
Applied mathematics and mechanics
|July 28, 2023
概括
这项研究使用物理信息神经网络 (PINNs) 准确识别复杂,非线性材料的材料特性,如生物组织. 新型采样和边界条件策略实现机械工程应用的高精度.
科学领域:
- 连续的固体力学 连续的固体力学
- 计算材料科学 计算材料科学
- 机器学习应用程序 机器学习应用程序
背景情况:
- 材料识别对于理解机械性能和功能至关重要.
- 在生物组织中常见的非线性材料行为,提出了重要的识别挑战.
- 基于物理学的神经网络 (PINNs) 为固体力学的反向问题提供了一个有前途的方法.
研究的目的:
- 开发和应用物理信息的神经网络 (PINNs) 来识别连续固体力学中未知的材料特性.
- 通过新的数据采样和边界条件执行策略,提高PINNs的准确性和效率.
- 在各种时间依赖和时间独立的固体力学问题上验证拟议的方法.
主要方法:
- 利用物理信息的神经网络 (PINNs) 进行反向材料识别.
- 开发了有效的策略,用于观察数据的非均采样,以提高准确性.
- 研究了作为软约束和硬约束的迪里克莱特型边界条件的强制执行.
- 将这些方法应用于线性弹性和超弹性材料模型.
主要成果:
- 在材料参数估计中实现了高精度,相对误差低于1%.
- 证明了拟议的非统一数据采样和边界条件策略的有效性.
- 成功地将这些方法应用于各种时间依赖和时间独立的固体力学实例.
- 验证了PINNs在识别非线性材料特性方面的能力.
结论:
- 开发的基于PINN的方法为连续固体力学中的材料识别提供了准确和高效的方法.
- 数据采样和边界条件强制执行的策略显著提高了PINN的性能.
- 这项工作对于优化结构完整性和推进新材料开发具有广泛的相关性.
相关概念视频
01:28Magnetostatic Boundary Conditions
1.0K
An electric field suffers a discontinuity at a surface charge. Similarly, a magnetic field is discontinuous at a surface current. The perpendicular component of a magnetic field is continuous across the interface of two magnetic mediums. In contrast, its parallel component, perpendicular to the current, is discontinuous by the amount equal to the product of the vacuum permeability and the surface current. Like the scalar potential in electrostatics, the vector potential is also continuous...
1.0K
01:25Boundary Conditions for Current Density
905
Current density becomes discontinuous across an interface of materials with different electrical conductivities. The normal component of the current density is continuous across the boundary.
905
01:29Two-Dimensional Force System: Problem Solving
615
Solving problems related to two-dimensional force systems is an essential aspect of mechanics and engineering. By applying the principles of vector analysis and force equilibrium, one can determine the effect of multiple forces acting on an object in a two-dimensional space.
The first step to solving a two-dimensional force system problem is to draw a free-body diagram of the object under consideration. This diagram helps identify all the external forces acting on the object, including their...
The first step to solving a two-dimensional force system problem is to draw a free-body diagram of the object under consideration. This diagram helps identify all the external forces acting on the object, including their...
615
01:20Maxwell-Boltzmann Distribution: Problem Solving
1.6K
Individual molecules in a gas move in random directions, but a gas containing numerous molecules has a predictable distribution of molecular speeds, which is known as the Maxwell-Boltzmann distribution, f(v).
This distribution function f(v) is defined by saying that the expected number N (v1,v2) of particles with speeds between v1 and v2 is given by
This distribution function f(v) is defined by saying that the expected number N (v1,v2) of particles with speeds between v1 and v2 is given by
1.6K
01:16Electrostatic Boundary Conditions
515
Consider an external electric field propagating through a homogeneous medium. When the electric field crosses the surface boundary of the medium, it undergoes a discontinuity. The electric field can be resolved into normal and tangential components. The amount by which the field changes at any boundary is given by the difference between the field components above and below the surface boundary.
The surface integral of an electric field is given by Gauss's law in integral form and is related to...
The surface integral of an electric field is given by Gauss's law in integral form and is related to...
515
01:15Kinematic Equations: Problem Solving
12.5K
When analyzing one-dimensional motion with constant acceleration, the problem-solving strategy involves identifying the known quantities and choosing the appropriate kinematic equations to solve for the unknowns. Either one or two kinematic equations are needed to solve for the unknowns, depending on the known and unknown quantities. Generally, the number of equations required is the same as the number of unknown quantities in the given example. Two-body pursuit problems always require two...
12.5K