基于MPSO-SVM非线性氨氮传感器的数据补偿  

Data compensation based on MPSO-SVM non-linear ammonia nitrogen sensor

作　　者：韩剑[1] 莫德清[2] 汪楠 HAN Jian;MO De-qing;WANG Nan(Institute of Information Technology of Guilin University of Electronic Technology,Guilin 541004,China;School of Life and Environmental Sciences,Guilin University of Electronic Technology,Guilin 541004,China)

机构地区：[1]桂林电子科技大学信息科技学院,广西桂林541004 [2]桂林电子科技大学生命与环境科学学院,广西桂林541004

出　　处：《桂林理工大学学报》2021年第1期225-229,共5页Journal of Guilin University of Technology

基　　金：广西科技重点研发计划项目(桂科AB16380295);桂林市科技局重大专项(20180101-3);广西高校中青年教师科研基础能力提升项目(2020KY57014)。

摘　　要：针对氨氮传感器在信号采集中容易受待测溶液中pH值和温度影响的问题,采用改进的粒子群优化支持向量机方法(MPSO-SVM)对氨氮传感器进行非线性数据补偿校正,构建了氨氮检测系统。系统预处理分别采用递推平均滤波、IIR数字滤波和卡尔曼滤波3种滤波方式,再将滤波后的数值归一化建立基于支持向量机回归模型。研究结果表明,MPSO-SVM算法支持向量机回归模型回归效果好,线性相关度大,数据补偿后氨氮测量值与真实值误差小,系统测试具有良好的稳定性和补偿效果。To solve the problem that the ammonia nitrogen sensor is easily affected by the pH value and temperature of the solution to be measured in the signal acquisition,the improved particle swarm optimization support vector machine(MPSO-SVM)is used to compensate and correct the non-linear data of the ammonia nitrogen sensor.So the ammonia nitrogen detection system is constructed.Three filtering methods,recursive average filtering,IIR digital filtering and Kalman filtering,are used in the system preprocessing.Then the filtered values are normalized to establish the support vector regression model.The results show that MPSO-SVM algorithm support vector regression model has good regression effect:large linear correlation,small error between the measured value and the real value of ammonia nitrogen after data compensation.The system test has good stability and good compensation effect.

关 键 词：氨氮传感器 支持向量机 改进粒子群算法 数字滤波器 非线性数据补偿 回归校正 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]