用遗传算法实现信号转换电路的非线性校正  

Nonlinear Correction of Signal Conversion Circuit Using Genetic Algorithm

作　　者：温秀兰[1] 宋爱国[1] 崔建伟[1] 高桂本[2] 

机构地区：[1]东南大学仪器科学系,南京210096 [2]内蒙古工业大学机械学院,呼和浩特010062

出　　处：《数据采集与处理》2003年第3期306-309,共4页Journal of Data Acquisition and Processing

基　　金：国家自然科学基金 (69875 0 0 4) ;江苏省自然科学基金 (BK2 0 0 1 40 2 ) ;东南大学-南瑞继保公司学位论文基金资助项目

摘　　要：采用了遗传算法的最新成果 ,提出了一种新颖的基于实数编码的改进遗传算法用以校正信号转换电路的非线性误差。该算法采用基于代沟最小的代选择模型 ,选用 BLX-α混合交叉算子和非一致变异算子 ,算法简单、鲁棒性强、优化效率高。仿真和实验表明 ,用本文提出的方法对信号转换电路进行非线性校正不仅能够自动搜索到最优多项式 ,而且精度高于最小二乘法。该算法同样适用于传感器等其他电子测量元件的非线性校正。The latest achievements of genetic algorithms are adopted and a novel improved genetic algorithms (RCIGA) is proposed based on real coding to correct nonlinear error of signal conversion circuit. That is, a generation-alternation model based the minimal generation gap (MGP), blend crossover (BLX-α) and non-uniform mutation operators are employed. RCIGA has the advantages of simple algorithm, strong robustness and high optimization efficiency. Simulational results and application show that RCIGA has the capability of searching the optimization polynomial and the higher nonlinear correction precision than least square method. And RCIGA provides an important reference value for the nonlinear correction of transducers and other electronical measurement components.

关 键 词：信号转换电路 非线性校正 遗传算法 鲁棒性 非线性误差 测试系统 

分 类 号：TP335[自动化与计算机技术—计算机系统结构] TP274.1[自动化与计算机技术—计算机科学与技术]