摘要
根据LSU 4.9宽域氧传感器的特性,采用通用元件、以STM32F103作为控制算法的实现平台,展开了空燃比分析仪的设计与实现。在硬件方面,以LSU4.9宽域氧传感器中氧化锆的负温度系数(NTC)特性为基础,提出了一种通过交流电压比较内阻值的方式控制氧传感器的温度。此方法不仅简化了电路,而且消除了因测量电路的差异性所产生的影响;在软件方面,针对泵电流存在的摄动,提出了采用比例—积分—微分神经网络(PIDNN)算法实现对氧传感器的反馈控制。实验表明:设计的空燃比分析仪可满足快速启动的要求,且能在0%~12.1%的氧气浓度范围内实现准确测量,动态响应时间短,能良好地跟踪环境氧气浓度的变化。
According to characteristics of LSU 4. 9 universal exhaust gas oxygen sensor and set STM32 as implementation platform of the control algorithm,design of air-fuel ratio analyzer is carried out and realized. In hardware aspect,based on negative temperature coefficient( NTC) characteristic of universal exhaust gas oxygen sensor,an internal resistance measurement method based on AC comparison means is proposed which helps to control the temperature of oxygen sensor. Not only eliminate influence of differences in multi-channel measurement circuit. In terms of software,considering perturbation of pump current,proportion integration differentiation neural network( PIDNN) algorithm is proposed to realize feedback control of the oxygen sensor. Experiments show that the designed air-fuel ratio analyzer can meet the requirements of fast start-up,and can achieve accurate measurement in the range of 0 ~ 12. 1 % oxygen concentration,it has short dynamic response time,and can keep track of the change of oxygen concentration.
出处
《传感器与微系统》
CSCD
2017年第12期80-83,共4页
Transducer and Microsystem Technologies
基金
国家自然科学基金面上资助项目(61471210)
国家自然科学基金青年科学基金资助项目(61701267)
浙江省宁波市科技局自然科学基金资助项目(2017A610099)
宁波大学科研启动基金资助项目(421600200)
关键词
空燃比分析仪
宽域氧传感器
控制电路
比例—积分—微分神经网络
air-fuel ratio analyzer
universal exhaust gas oxygen (UEGO) sensor
control circuit
proportion- integration-differentiation neural network (PIDNN)