通过深度最小作用方法解决时间演变的初始终端值问题:牛顿动力学和波形方程
Zhipeng Chang1, Jerry Zhijian Yang1, Xiaofei Zhao1
1School of Mathematics and Statistics, and Computational Sciences Hubei Key Laboratory, <a href="https://ror.org/033vjfk17">Wuhan University</a>, Wuhan 430072, China.
我们开发了一种深度最小动作方法 (DLAM) 来解决没有微分方程的进化问题. 这种高效,无监督的方法通过优化基于物理系统动作的神经网络来准确跟踪轨迹.
科学领域:
- 计算物理 计算物理
- 机器学习应用 机器学习应用
- 数学方法 数学方法
背景情况:
- 解决进化问题往往需要复杂的微分方程.
- 现有的方法可能会与非线性,高序列或高维系统扎.
- 最少作用原理为物理系统提供了一个替代的表述.
研究的目的:
- 引入一种新的深度最小动作方法 (DLAM) 来解决轨迹演变问题.
- 为传统的微分方程解法器提供一种高效,无监督的替代方案.
- 证明DLAM对各种物理动态的适用性,包括牛顿式和波形方程.
主要方法:
- 使用最小动作原则制定了这个问题.
- 使用规范化的深度神经网络来满足初始终端值条件.
- 将问题转化为不受约束的优化任务.
- 应用DLAM到牛顿和波动力学,包括复杂的案例.
主要成果:
- DLAM有效地解决了轨迹演变问题.
- 该方法准确处理普通方程和部分微分方程.
- 在非线性,高阶和高维的场景中取得了成功.
- 实现了高效准确的轨迹预测.
结论:
- DLAM提供了一种强大的,没有方程的方法来解决物理进化问题.
- 该方法是多功能,适用于广泛的动态.
- DLAM为计算物理和机器学习集成提供了一个有前途的方向.
更多相关视频
11:03An Analog Macroscopic Technique for Studying Molecular Hydrodynamic Processes in Dense Gases and Liquids
Published on: December 4, 2017
06:45Design and Application of a Fault Detection Method Based on Adaptive Filters and Rotational Speed Estimation for an Electro-Hydrostatic Actuator
Published on: October 28, 2022
相关概念视频
Newtonian Fluid: Problem Solving
A velocity gradient forms within the fluid when a Newtonian fluid is placed between two parallel plates, with...
Linear Approximation in Time Domain
For a simple pendulum with a mass evenly distributed along its length and the center of mass located at half the pendulum's length,...
Kinematic Equations: Problem Solving
Euler Equations of Motion
Kinematic Equations - II
Suppose a car merges into freeway traffic on a 200 m long ramp. If its initial velocity is 10 m/s and it accelerates at 2 m/s2, then the...
Equation of Motion: General Plane motion - Problem Solving
The friction between the roller and the ground is characterized by two coefficients. The static friction coefficient is 0.15, while the kinetic friction coefficient is 0.1. These values are crucial in understanding the interaction between...
