-
搭建红外CO2传感器测试平台对所设计传感器系统进行测试,经过对多种情况下的多组测量数据的分析得出红外CO2传感器的精度与误差,并优化拟合系数,从而提高传感器系统的测量精度[13-15]。
搭建测试平台,将传感器放置到密闭气室内,标定实验在标准大气压和25 ℃环境下进行。为了达到测试的准确性,第一,保证密闭气室的密闭性良好;第二,确保测试环境稳定,浓度标定实验的环境稳定性是整组实验数据拟合的根本所在;第三,当气体浓度达到设定值时,等待1 min后再测量通道的峰峰值,并记录它们对应的浓度值。
-
首先,在测试平台的环境下检测传感器性能的稳定性好坏。在密闭气室内充入浓度为10000 ppm的标准CO2气体,传感器接入上位机,持续工作12 h,每隔30 min保存一次数据,观测传感器采集峰峰值的波动情况,如图9所示。
-
在实验开始之前,首先进行零点标定,利用真空泵将传感器内的气体抽出,密闭气室的注入气体通道接入纯氮气进行零点标定。然后将标气瓶中的浓度98%的标准气体经过气体分析仪器配置成标准的不同浓度的CO2气体,并依序分次注入密闭气室中。之后利用CO2传感器采集的峰峰值与注入的CO2浓度值之间的对应关系图,如图10所示,通过数据拟合得到25 ℃浓度拟合方程,如公式(2)所示。
方程为:
-
实验温湿度箱中的环境温度设置为可调温度分别为0、5、10、15、20、25、30 ℃,将标气瓶中的纯度0%、1%、2%、4%、5%的标准CO2气体注入密闭气室中。由于温湿度箱内空气流通较大,导致箱内不同位置的温度并不完全相同,所以温度值以温湿度传感器的输出值为准。采集温度传感器输出的温度值与CO2传感器的峰峰值,从而得到不同温度下的浓度值与峰峰值之间的关系,如图11所示。
Figure 11. Relationship between the concentration value and the peak-to-peak value at different temperatures
由上图可知,温度主要影响CO2的吸收率,进而影响峰峰值大小,当温度上升时,峰峰值降低,未补偿的情况下测量浓度值偏大。以25 ℃为标准,找寻同一浓度下的峰峰值差值与温度差值的关系,如图12所示。
可以注意到,各个浓度下的峰峰值差值和温度差值近似为一条曲线,符合温度对CO2吸收系数的影响关系与探测器对红外辐射的吸收率的影响关系[6, 9, 13]。故将图10所示曲线拟合为一条曲线,其方程(3)为:
将方程(3)代入方程(2)中,可以得到带有温度补偿功能的浓度计算方程。最终,将带有温度补偿功能的方程编写入STM32的程序中,再次测量,测量结果如表1所示。
Sample gas concentration/ppm Detection of concentration/ppm Absolute error/ppm Relative error 1800 1773 −27 1.5% 12000 11873 −127 1.05% 24000 24269 269 1.12% 30000 30725 725 2.4% 35000 36326 1326 3.7% Table 1. Duplicated testing results
Design of non-dispersive infrared CO2 sensor with temperature compensation
doi: 10.3788/IRLA20210746
- Received Date: 2021-10-11
- Rev Recd Date: 2021-11-29
- Publish Date: 2022-04-07
-
Key words:
- non-dispersive infrared(NDIR) /
- optical simulation of air chamber /
- signal processing circuit /
- temperature compensation concentration algorithm
Abstract: In recent years, the greenhouse effect has become more and more obvious. The side effects brought by increasing of CO2 concentration have seriously affected people's production and life, and are even closely related to everyone's healthy. In defect of the large size, portability, high precision, and modularity of CO2 commercial sensors currently, a pyroelectric-based non-dispersive infrared (NDIR) method measuring CO2 concentration system was designed. The system design was mainly divided into four parts: air chamber structure design, signal circuit design, software control design and data processing. The air chamber structure design adopted a single-channel structure design, which was optically simulated. Finally, the size of the air chamber was determined, which effectively improved the measurement accuracy of the system. In terms of signal circuit design, a small signal amplifying circuit based on the differential method was designed to extract and amplify the effective signal from the noise to increase resolution. The software control design employed digital filtering algorithm to filter out interference and clutter, extract the effective value in the data, and improve the signal-to-noise ratio. In view of data processing, a gas test platform was built, and the temperature and humidity and peak-to-peak compensation formulas were used to compensate for the influence of temperature on the peak-to-peak value. Then, the gas concentration value was calculated using the curve fitting method at 25 ℃, and finally output through the serial port. After testing, the measurement range of the system is 5%, and the relative error is within 1500 ppm, which can meet the safety requirements of fire alarm, underground and other occasions monitoring.