离散化
过程(计算)
计算机科学
基础(线性代数)
物理系统
偏微分方程
简单(哲学)
比例(比率)
工业工程
应用数学
数学优化
算法
数学
工程类
物理
数学分析
量子力学
认识论
操作系统
哲学
几何学
标识
DOI:10.1016/0026-2714(84)90450-5
摘要
The appearance of Very Large Scale Integration caused a pronounced interest in concentrating on process and device modeling. The fundamental properties which represent the basis for all device modeling activities are summarized. The sensible use of physical and technological parameters is discussed and the most important physical phenomena which are required to be taken into account are scrutinized. The assumptions necessary for finding a reasonable trade-off between efficiency and effort for a model synthesis are recollected. Methods to bypass limitations induced by these assumptions are pin-pointed. Formulae that are applicable in a simple and easy way for the physical parameters of major importance are presented. The necessity of a careful parameter-selection, based on physical information, is shown. Some glimpses on the numerical solution of the semiconductor equations are given. The discretisation of the partial differential equations with finite differences is outlined. Linearisation methods and algorithms for the solution of large sparse linear systems are sketched. Results of our two dimensional MOSFET model — MINIMOS — are discussed. Much emphasis is laid on the didactic potential of such a complex high order model.
科研通智能强力驱动
Strongly Powered by AbleSci AI