Finite element methods for design of electromagnetic devices

电磁装置设计的有限元方法

• 批准号: 4906-2013
• 负责人: Webb, Jonathan
• 金额: $ 2.55万
• 依托单位: McGill University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=568380
• 关键词: Finite; element; methods; design; electromagnetic

While all electrical devices are electromagnetic in nature, there is an important class of devices which can only be understood properly by knowing in detail how the electromagnetic field behaves. These are the "electromagnetic devices" referred to in the title of this proposal and they include many microwave components used in telecommunication systems. For a long time engineers have used computers to simulate the fields in order to predict the performance of their designs, but many of the problems that engineers want to analyze push the limits of even modern computers. Devices are becoming more complex and accuracy requirements greater. For this reason, there continues to be a need for better simulation methods and this proposal focuses on one them, the finite element method (FEM). Specifically, it aims to advance the state of the art in two areas: (1) Adaption. This is a way of starting with a simple, fast analysis and progressively increasing the complexity. It can achieve the required accuracy more quickly than the non-adaptive approach. Two types of adaption, called "h" and "p", are well established, but a third, more powerful type, which combines these, is still in its infancy. Effective "hp" methods for microwave devices will be developed. (2) Meshing. Another time consuming and sometimes unreliable part of FEM is automatically dividing the volume occupied by the device into a large number of simple shapes (elements). Usually the elements are made to match the device's surfaces quite carefully, but this adds greatly to the complexity. An alternative strategy will be pursued, based on subdivision into cubic elements.

虽然所有的电气设备在本质上都是电磁的，但有一类重要的设备只能通过详细了解电磁场的行为才能正确理解。这些是本提案标题中提到的“电磁装置”，其中包括电信系统中使用的许多微波部件。长期以来，工程师们一直使用计算机来模拟这些场，以预测其设计的性能，但工程师们想要分析的许多问题甚至突破了现代计算机的极限。设备变得越来越复杂，精度要求越来越高。出于这个原因，仍然需要更好的模拟方法，本提案侧重于其中之一，有限元法（FEM）。具体而言，它旨在推进两个领域的技术水平： (1)适应。这是一种从简单、快速的分析开始，逐步增加复杂性的方法。它可以比非自适应方法更快地达到所需的精度。两种类型的适应，称为“h”和“p”，已经很好地建立起来，但是第三种，更强大的类型，将这些结合起来，仍然处于起步阶段。有效的微波器件的“hp”方法将被开发。 (2)相互配合。FEM的另一个耗时且有时不可靠的部分是自动将设备占用的体积划分为大量简单形状（元素）。通常，这些元件会非常仔细地与设备的表面相匹配，但这大大增加了复杂性。将采用一种基于细分为立方元素的替代策略。