基于移动最小二乘的传感器数据重构  

Data Reconstruction for Nanosensor by Using Moving Least Square Method

作　　者：宋绍民[1] 王敏[2] 胡新晚[1] 

机构地区：[1]湖南工学院电气与信息工程系,湖南衡阳421002 [2]湖南大学电气与信息工程学院,湖南长沙410082

出　　处：《自动化仪表》2010年第5期16-18,共3页Process Automation Instrumentation

基　　金：湖南省教育厅科研基金资助项目(编号:08C246)

摘　　要：纳米气体传感器的灵敏度易受温度变化的影响。为了有效改善纳米传感器的温度特性,方便确定其最佳工作条件,利用移动最小二乘法对其灵敏度-温度数据进行拟合,并建立相应的近似模型。移动最小二乘法通过紧支性加权的局部近似,使所建模型较常规最小二乘模型更精确、更光滑。实验证实了基于移动最小二乘拟合的近似性能。该方法在传感器数据精确重构领域具有良好的应用前景。Generally,the sensitivity of nano-gas-sensors is easily affected by variation of the temperature.In order to effectively improve the temperature characteristics of the sensor,and be easy to determine its optimal operation conditions,by adopting moving least square method,the sensitivity-temperature data are fitted,and the approximate model is established.Through compact support weighted local approximation,the model established with moving least square（MLS） is more accurate and smooth than the model established with conventional least square method.The experiments prove the approximate performance of the fitting based on MLS.This method possesses excellent prospects in precise reconstruction area of sensor data.

关 键 词：传感器 数据重构 温度特性 移动最小二乘法 近似技术 曲线拟合 

分 类 号：TP212.1[自动化与计算机技术—检测技术与自动化装置]