Automated spiral inductor design by a calibrated PI network with manifold mapping technique

Fan，Xiaolin1,2; Li，Song1,2; Laforge，Paul D.1; Cheng，Qingsha S.2

10.1109/IMS30576.2020.9223860

2020-08-01

0149-645X

978-1-7281-6816-6

IEEE MTT-S International Microwave Symposium Digest

2020-August

76-79

4-6 Aug. 2020

Los Angeles, CA, USA

An automated EM-based method for efficient spiral inductor design is proposed and implemented using Matlab-driven Sonnet em simulator. In the proposed method, a surrogate model is established as a replacement to the computational expensive EM model utilizing a one-step calibrated circuit model and an iterative response correction process. In the design procedure, only one parameter extraction is performed prior to the iterative response correction in which no further parameter extraction is required. A hexdecagon spiral inductor is designed and optimized automatically following the proposed procedure, the algorithm convergent within 4 iterations to satisfy design specifications. A comparison between the fine model and updated surrogate model is also given as a validation to the reliability of the proposed method.

Inductors Manifold mapping Optimization Parameter extraction Space sampling Spiral inductor Surrogate

其他

英语

EI

20204409429489

Electric inductors ; Extraction ; Iterative methods ; Mapping ; Design

Surveying:405.3 ; Data Processing and Image Processing:723.2 ; Computer Applications:723.5 ; Chemical Operations:802.3 ; Numerical Methods:921.6 ; Statistical Methods:922

2-s2.0-85094190536

Scopus

1.University of Regina,Faculty of Engineering Applied Science,Canada
2.Southern University of Science and Technology,Department of Electrical and Electronic Engineering,China

Fan，Xiaolin,Li，Song,Laforge，Paul D.,et al. Automated spiral inductor design by a calibrated PI network with manifold mapping technique[C],2020:76-79.