基于布谷鸟搜索算法的双π电感模型参数优化
发布时间：2018-11-12 12:05
【摘要】：提出了一种新的基于布谷鸟搜索算法的片上电感模型参数优化算法。布谷鸟搜索算法对复杂非线性非凸函数的优化效果显著,这种函数的特点与半导体器件紧凑型模型参数和优化目标之间的非线性特性相近。该算法开发了模型参数交叉操作,将两个解空间的对应参数交换位置,避免敏感度高的参数对敏感度低的参数的过度影响。引入了自适应参数,使算法的搜索步长能自适应调整,既不会因为步长太大跳过最优解,也不会因为步长太小导致收敛速度慢。采用集成电路工艺片上螺旋电感的实测数据对该算法进行验证,得到较好的拟合度。提出的模型优化算法可适用于集成电路器件模型的自动优化。
[Abstract]:A new algorithm for optimizing the parameters of on-chip inductance model based on Cuckoo search algorithm is proposed. The Cuckoo search algorithm has a remarkable effect on the optimization of complex nonlinear nonconvex functions, and the characteristics of this function are similar to the nonlinear characteristics between the parameters of the compact model of semiconductor devices and the optimization objectives. In this algorithm, the cross-operation of model parameters is developed to exchange the corresponding parameters between the two solution spaces so as to avoid the excessive influence of the highly sensitive parameters on the low-sensitivity parameters. By introducing adaptive parameters, the search step size of the algorithm can be adjusted adaptively, neither because the step size is too large to skip the optimal solution, nor because the step size is too small to cause the convergence speed to be slow. The algorithm is verified by the measured data of spiral inductance on chip of integrated circuit, and a good fit is obtained. The proposed model optimization algorithm can be applied to the automatic optimization of integrated circuit device models.
【作者单位】： 杭州电子科技大学射频电路与系统教育部重点实验室;吉林医药学院生物医学工程学院;
【基金】：国家自然科学基金资助项目(61331006)
【分类号】：TN303;TP18
本文编号：2327059
[Abstract]:A new algorithm for optimizing the parameters of on-chip inductance model based on Cuckoo search algorithm is proposed. The Cuckoo search algorithm has a remarkable effect on the optimization of complex nonlinear nonconvex functions, and the characteristics of this function are similar to the nonlinear characteristics between the parameters of the compact model of semiconductor devices and the optimization objectives. In this algorithm, the cross-operation of model parameters is developed to exchange the corresponding parameters between the two solution spaces so as to avoid the excessive influence of the highly sensitive parameters on the low-sensitivity parameters. By introducing adaptive parameters, the search step size of the algorithm can be adjusted adaptively, neither because the step size is too large to skip the optimal solution, nor because the step size is too small to cause the convergence speed to be slow. The algorithm is verified by the measured data of spiral inductance on chip of integrated circuit, and a good fit is obtained. The proposed model optimization algorithm can be applied to the automatic optimization of integrated circuit device models.
【作者单位】： 杭州电子科技大学射频电路与系统教育部重点实验室;吉林医药学院生物医学工程学院;
【基金】：国家自然科学基金资助项目(61331006)
【分类号】：TN303;TP18
【相似文献】
相关期刊论文 前3条
1 朱文兴;程泓;;VLSI电路划分问题的分散搜索算法[J];电子学报;2012年06期
2 李旭,刘军民,刘军;基于菱形搜索算法的运动估计器的FPGA实现[J];电视技术;2005年S1期
3 李德识;曹阳;;基于CDFG的SoC验证方法及其分割与搜索算法[J];计算机工程;2007年02期
相关会议论文 前1条
1 彭明侨;罗先觉;邹晓松;;基于改进概率搜索算法的模拟电路故障诊断[A];第四届中国测试学术会议论文集[C];2006年
相关硕士学位论文 前1条
1 何飞;一种电路设计子图搜索算法的研究与实现[D];西北工业大学;2006年
，本文编号：2327059
本文链接：https://www.wllwen.com/kejilunwen/dianzigongchenglunwen/2327059.html
