物理学的研究告诉神经网络解决分子静电学中的Poisson-Boltzmann方程
Martín A Achondo1, Jehanzeb H Chaudhry2, Christopher D Cooper1,3
1Department of Mechanical Engineering, Universidad Técnica Federico Santa María, Valparaíso 2390123, Chile.
基于物理学的神经网络 (PINN) 可以准确地解决分子静电学的线性波松-博尔兹曼方程 (PBE). 本研究详细介绍了最佳的PINN架构,并为研究人员提供了一个开源工具.
科学领域:
- 计算化学的计算化学
- 机器学习 机器学习
- 科学计算科学计算
背景情况:
- 基于物理学的神经网络 (PINN) 是新兴的机器学习方法,用于解决微分方程.
- 波桑-博尔兹曼方程 (PBE) 对于模拟分子静电学至关重要.
- 之前的PINN应用在解决各种微分方程方面表现出了前景.
研究的目的:
- 调查PINN在解决线性Poisson-Boltzmann方程 (PBE) 的有效性.
- 确定最佳的神经网络架构,用于准确的静电建模.
- 为分子系统应用PINN提供一个可访问的开源工具.
主要方法:
- 为线性PBE开发了一个多域PINN,包含一个接口.
- 评估建筑特征的影响,如缩放层,富里埃特征和可训练的激活.
- 采用损失平衡算法来优化网络性能.
主要成果:
- 确定了一个优越的PINN架构,包括输入/输出缩放,随机的福里埃特征,可训练的激活和损失平衡.
- 达到10^-2到10^-3的精度,与现有的PINN应用程序相比.
- 展示了整合实验数据的潜力,并讨论了非线性PBE的挑战.
结论:
- 在分子静电学中,PINN提供了一种可行且准确的方法来解决线性PBE.
- 特定的神经网络设计选择显著影响准确性.
- 开源实现促进了PINN在计算化学中的更广泛采用.
更多相关视频
09:17Structure-Based Simulation and Sampling of Transcription Factor Protein Movements along DNA from Atomic-Scale Stepping to Coarse-Grained Diffusion
Published on: March 1, 2022
11:03An Analog Macroscopic Technique for Studying Molecular Hydrodynamic Processes in Dense Gases and Liquids
Published on: December 4, 2017
相关概念视频
Poisson's And Laplace's Equation
Maxwell-Boltzmann Distribution: Problem Solving
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
The Quantum-Mechanical Model of an Atom
The Role of Ion Channels in Neuronal Computation
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
Ampere-Maxwell's Law: Problem-Solving
To solve the problem, we can use the equations from the analysis of an RC circuit and Maxwell's version of Ampère's law.
For the first part of...
Calculations of Electric Potential II
Consider a...
