积极学习优化二进制编码的元表面,由宽带元原子组成
Parvathy Chittur Subramanianprasad1, Yihan Ma1, Achintha Avin Ihalage1
1School of Electronics Engineering and Computer Science, Queen Mary University of London, Mile End Rd, Bethnal Green, London E1 4NS, UK.
Sensors (Basel, Switzerland)
|July 8, 2023
概括
积极学习显著加快了 metasurface 阵列优化,大大减少了与遗传算法相比的计算时间. 这种机器学习方法在最小化雷达截面方面产生了类似的结果,特别是在大型阵列中.
科学领域:
- 电磁学和元材料的使用
- 计算科学与工程 计算科学与工程
- 机器学习应用 机器学习应用
背景情况:
- 雷达横截面最小化的超表面阵列设计是一个关键的研究领域.
- 传统的优化算法,如遗传算法 (GA) 和粒子群优化 (PSO) 是计算密集的,限制了它们对大数组的使用.
- 高的计算复杂性阻碍了复杂的超表面设计的高效优化.
研究的目的:
- 介绍和评估主动学习作为一个机器学习优化技术,用于 metasurface 阵列.
- 证明与传统方法相比,主动学习可以显著减少计算时间.
- 通过使用主动学习实现雷达截面最小化的可比或优异的优化结果.
主要方法:
- 积极学习的应用,一个机器学习优化策略.
- 积极学习性能与遗传算法 (GA) 的比较.
- 在积极学习框架内使用精确训练的代孕模型.
主要成果:
- 积极学习将10x10元表面阵列的优化时间从13,260分钟 (GA) 减少到65分钟.
- 对于60x60元表面阵列,积极学习比GA快24倍,获得类似的结果.
- 机器学习优化大大降低了计算成本,特别是在更大的超表面设计中.
结论:
- 活跃学习提供了一个计算效率高的替代品,用于转移表面数组优化遗传算法.
- 拟议的方法显著加速了设计过程,使大规模的优化成为可能.
- 利用代用模型进一步提高了这个领域积极学习的速度和效率.
相关概念视频
01:27Biasing of Metal-Semiconductor Junctions
284
Biasing metal-semiconductor junctions involves applying a voltage across the junction. Specifically, the metal is connected to a voltage source, while the semiconductor is grounded. This technique is essential for controlling the direction and magnitude of current flow in electronic devices, including diodes, transistors, and photovoltaic cells.
In Schottky junctions, where the semiconductor is n-type, applying a positive voltage to the metal relative to the semiconductor reduces its Fermi...
In Schottky junctions, where the semiconductor is n-type, applying a positive voltage to the metal relative to the semiconductor reduces its Fermi...
284
01:29Plane Electromagnetic Waves II
3.1K
Consider a plane wavefront traveling in position x-direction with a constant speed. This wavefront can be utilized to obtain the relationship between electric and magnetic fields with the help of Faraday's law.
3.1K
01:30Plane Electromagnetic Waves I
3.7K
The existence of combined electric and magnetic fields that propagate through space as electromagnetic (EM) waves is the most significant prediction of Maxwell's equations. As Maxwell's equations hold in free space, the predicted electromagnetic waves do not require a medium for their propagation. An EM wave comprises an electric field, defined as the force per charge on a stationary charge, and a magnetic field, which is the force per charge on a moving charge.
The EM field is assumed...
The EM field is assumed...
3.7K
01:28Standing Waves in a Cavity
962
A household microwave and lasers are examples of standing electromagnetic waves in a cavity. When two conducting metal plates are placed parallel at the nodal planes, it creates a cavity where standing waves are formed. The cavity between the two planes is analogous to a stretched string held at the points x = 0 and x = L. Here, the distance 'L' between the two planes must be an integer multiple of half of the wavelength. The wavelengths that satisfy this condition are given by:
962