相关实验视频
Updated: Jan 6, 2026

Evolution of Staircase Structures in Diffusive Convection
Published on: September 5, 2018
深度强化学习控制解锁了流对流中的增强热传输
1Max Planck Institute for Solar System Research, Göttingen 37077, Germany.
深度强化学习 (DRL) 通过动态调整墙壁温度来优化流对流传热传输. 这种人工智能方法显著提高了传热效率,超过了传统方法,并为智能热管理提供了新的途径.
科学领域:
- 流体动力学和热传递 流体动力学和热传递
- 人工智能和机器学习
背景情况:
- 流对流对于自然和工业系统的热传输至关重要.
- 在流对流中优化热传递,尤其是在极端条件下,是具有挑战性的.
- 传统的控制方法,如温度调节,提供有限的增强 (20-25%).
研究的目的:
- 开发和评估一个深度强化学习 (DRL) 框架,以优化流雷利-贝纳德对流中的热传递.
- 自主发现用于最大化热传输的先进控制策略.
- 将基于DRL的控制与传统方法进行比较,并探索简化控制模型.
主要方法:
- 实施了深度强化学习 (DRL) 代理来控制墙壁温度波动.
- 训练了DRL代理,以最大限度地提高热传递在动荡的雷利-贝纳德对流.
- 将学到的DRL政策提炼成一个简化的大爆炸控制模型.
主要成果:
- DRL剂达到高达38.5%的传热增强,超过了常规限制.
- 学习的控制策略诱导了一个完全调制的边界层制度.
- 简化的Bang-Bang模型实现了可比增强 (高达40.0%) 并将其推广到更高的雷利数字.
结论:
- 深度强化学习为智能流控制提供了一种强大的方法.
- DRL框架成功地优化了超越传统方法的热传递.
- 这项研究为现实世界热传输优化提供了一个有希望的,可通用的框架.
更多相关视频
10:03Uncoupling Coriolis Force and Rotating Buoyancy Effects on Full-Field Heat Transfer Properties of a Rotating Channel
Published on: October 5, 2018
10:29Experimental Methodology for Estimation of Local Heat Fluxes and Burning Rates in Steady Laminar Boundary Layer Diffusion Flames
Published on: June 1, 2016
相关概念视频
Mechanisms of Heat Transfer II
Reynolds Transport Theorem
Mechanisms of Heat Transfer I
Mechanisms of Heat Transfer
Conduction, accounting for approximately 3% of body heat loss at rest, is the process of exchanging heat between molecules of two materials in direct contact. This can result in both heat loss and gain. For instance, when the body is submerged in water, which conducts heat 20 times more effectively than air, it can either lose or gain significant...
Mechanism of heat transfer
Control Volume and System Representations
The control volume approach considers a stationary region in space through which fluid flows. This region is bounded by a control surface. For instance, in the case of water...