Premium
Stability for ROW‐methods adapted to circuit simulation
Author(s) -
Hoschek Markus
Publication year - 1998
Publication title -
zamm ‐ journal of applied mathematics and mechanics / zeitschrift für angewandte mathematik und mechanik
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.449
H-Index - 51
eISSN - 1521-4001
pISSN - 0044-2267
DOI - 10.1002/zamm.19980781544
Subject(s) - jacobian matrix and determinant , stability (learning theory) , scheme (mathematics) , computer science , function (biology) , core (optical fiber) , integrated circuit , electronic circuit , numerical integration , control theory (sociology) , computer engineering , algorithm , topology (electrical circuits) , control engineering , mathematics , engineering , electrical engineering , telecommunications , artificial intelligence , operating system , mathematical analysis , control (management) , machine learning , evolutionary biology , biology
Automatic generation and numerical integration of network equations are the core of industrial TCAD packages for analog simulation of highly integrated circuits [1]. Many packages supply the Jacobian with every function evaluation at very low additional costs: it is about only 1.4 times more expensive to evaluate both right‐hand side and Jacobian than right‐hand side alone. In this paper we discuss four different approaches to use the cheap Jacobian information within a ROW‐type integration scheme, mainly with respect to stability.