Automatic Optimal Design Method for Circuit Sizing Based on CNN Surrogate Model Assisted Differential Evolution Algorithm

被引:0
作者
Tang, Chaoying [1 ]
Chen, Xiaofei [1 ]
Luo, Yanshen [1 ]
Zeng, Yanhan [1 ]
机构
[1] Guangzhou Univ, Sch Elect & Commun Engn, Guangzhou 510006, Peoples R China
关键词
Analog circuit design; automated circuit design; convolutional neural network; differential evolution algorithm; simulation-based optimization; VOLTAGE REFERENCE; ANALOG; OPTIMIZATION;
D O I
10.1109/ACCESS.2024.3462952
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The physical characteristics of analog ICs are intricate, leading to prolonged design and development cycles. Currently, the automatic optimization of analog integrated circuit size heavily relies on simulations, which unfortunately results in a significant waste of simulation resources during the optimization process. To address this issue, this paper introduces a novel optimization framework named "MODE-CNN", which integrates Multi-Objective Differential Evolution (MODE) algorithm and Convolutional Neural Network (CNN) surrogate models. The MODE-CNN framework is designed to reduce unnecessary consumption of simulation resources while maintaining high optimization accuracy. Within the MODE-CNN framework, we first enhance the predictive capabilities of the surrogate models using Latin Hypercube Sampling and Quantile Transformation. Subsequently, we employ CNN surrogate models to evaluate circuit performance during the optimization process, effectively screening out promising designs and substantially reducing the number of simulations required. Moreover, we propose an improved fast non-dominated sorting method that further enhances the global optimization performance of the MODE algorithm. Through testing on three circuit design cases and comparing MODE-CNN with existing mainstream methods and expert manual designs, we find that MODE-CNN not only significantly reduces the number of simulations (by up to 63%), but also surpasses traditional methods in optimization outcomes, with performance improvements of up to 95.09% for TC and 58.75% for LS. The experimental results validate the effectiveness and advancement of the MODE-CNN framework in the optimization of analog integrated circuits.
引用
收藏
页码:136238 / 136247
页数:10
相关论文
共 50 条
[41]   Fast Variation-aware Circuit Sizing Approach for Analog Design with ML-Assisted Evolutionary Algorithm [J].
Song, Ling-Yen ;
Kuo, Tung-Chieh ;
Wang, Ming-Hung ;
Liu, Chien-Nan Jimmy ;
Huang, Juinn-Dar .
27TH ASIA AND SOUTH PACIFIC DESIGN AUTOMATION CONFERENCE, ASP-DAC 2022, 2022, :80-85
[42]   Optimal Filter Design Using Differential Evolution Algorithm and Its Advanced Variants [J].
Durmus, Burhanettin ;
Unlu, Aybuke ;
Ozyon, Serdar ;
Temurtas, Hasan .
CIRCUITS SYSTEMS AND SIGNAL PROCESSING, 2025,
[43]   Optimal Design of PCS Ceramic Microwave Filters using the Differential Evolution Algorithm [J].
Kosmanis, Theodoros I. ;
Rekanos, Ioannis T. ;
Tsitsos, Stelios P. .
APPLIED COMPUTATIONAL ELECTROMAGNETICS SOCIETY JOURNAL, 2016, 31 (04) :361-365
[44]   The Improved Intelligent Optimal Algorithm Based on the Artificial Bee Colony Algorithm and the Differential Evolution Algorithm [J].
Li, Jingyi ;
Cheng, Ju ;
Mo, Lipo .
PROCEEDINGS OF 2019 CHINESE INTELLIGENT SYSTEMS CONFERENCE, VOL I, 2020, 592 :111-121
[45]   Optimization Design Method of Spherical Magnetic Field Generation Coil Based on Differential Evolution Algorithm [J].
Xu, Wei ;
Ge, Jian ;
Yu, Hong ;
Xiao, Min .
JOURNAL OF ADVANCED COMPUTATIONAL INTELLIGENCE AND INTELLIGENT INFORMATICS, 2023, 27 (03) :490-495
[46]   An Approach for Learning the Optimal "Tuned" Masks Based on Differential Evolution Algorithm [J].
Zhang, Xu ;
Ye, Zhiwei ;
Yang, Juan ;
Liu, Wei ;
Jin, Huazhong .
2017 INTERNATIONAL CONFERENCE ON SECURITY, PATTERN ANALYSIS, AND CYBERNETICS (SPAC), 2017, :585-590
[47]   Optimal Analysis of the Weighted Matrices in LQR Based on the Differential Evolution Algorithm [J].
Deng, Xiongfeng ;
Sun, Xiuxia ;
Liu, Ri ;
Wei, Wei .
2017 29TH CHINESE CONTROL AND DECISION CONFERENCE (CCDC), 2017, :832-836
[48]   INDOOR SPACE LAYOUT DESIGN BASED ON DIFFERENTIAL EVOLUTION ALGORITHM [J].
Meng S. .
Scalable Computing, 2024, 25 (04) :3074-3085
[49]   Logistics Network Design Optimization Based on Differential Evolution Algorithm [J].
Ding, SiBo .
PROCEEDINGS OF 2010 INTERNATIONAL CONFERENCE ON LOGISTICS SYSTEMS AND INTELLIGENT MANAGEMENT, VOLS 1-3, 2010, :1064-1068
[50]   Feature Image-Based Automatic Modulation Classification Method Using CNN Algorithm [J].
Lee, Jung Ho ;
Kim, Kwang-Yul ;
Shin, Yoan .
2019 1ST INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE IN INFORMATION AND COMMUNICATION (ICAIIC 2019), 2019, :560-563