Non-dispersive Infrared Flat Cone Cavity CO₂ Gas Sensor With Temperature And Humidity Compensation

The year-on-year increase in the concentration of carbon dioxide?CO₂?not only causes global warming,but also gradually threatens human health.Therefore,it is particularly urgent to develop a real-time and accurate CO₂ concentration measurement system.Based on this,a non-dispersive infrared?NDIR?flat cone cavity CO₂ gas sensor is designed and its performance is systematically studied.Firstly,a novel infrared CO₂ gas sensor is designed by non-spectral infrared detection technology.It chooses incandescent infrared light source,dual-channel thermopile detector and flat cone cavity as the main components.The distribution of light intensity on the detection surface of the novel CO₂ gas sensor was studied by using Zemax software,and the reasonable placement position of the detector is given.The fluid dynamics simulation calculation?CFD?is used to focus on the relationship between the infrared radiation absorption efficiency in the sensor cavity and the cavity length.The calculated results show that the cavity length has an optimal value of 8 cm when the taper angle of the flat cone cavity is 5°,the diameter of the detecting surface is 1cm,and the wall reflectance is 96%.As this time,the sensor's infrared absorption efficiency reaches maximum and the sensor has the highest detection sensitivity.Based on the above sensor optical system,a matched circuit system is designed.The STM32 single chip is applied as the core component to achieve the electrical modulation of the light source,and the ICL7650 is used to amplify the electrical signal output from the detector,and then STM32 controls the ADC to perform signal acquisition.This circuit can effectively improve the signal to noise ratio of the system.Secondly,based on the above simulation results,the flat cone cavity was designed and processed,and the NDIR flat cone cavity infrared CO₂ concentration measurement system was developed.The different coating chambers were calibrated and tested in the range of0²000 ppm.The experimental results show that the gold-plated gas chamber has higher sensitivity.Then the sensor was used to measure four different concentrations of samples,and the errors of the four groups of experiments are all less than 10 ppm in the dry environment of25°C.The repeatability error and long-term stability error are within 5.5%and 3.5%,respectively,and the concentration inversion results are consistent with the measured values of the standard instrument,the slope of the fitted line is close to 1.Finally,the temperature and humidity experiments were carried out on the CO₂concentration measurement system.The results of temperature and humidity test show that the temperature difference between the ambient temperature and the system calibration temperature is larger in the temperature range of 5°C⁴5°C,and the measurement error of CO₂ concentration is larger.In the humidity range of 40%⁸0%,the sensor Both the measurement channel and the reference channel voltage value exhibit an exponential decay with an increase in relative humidity,resulting in a decrease in measurement accuracy.Aiming at the above phenomenon,the neural network algorithm?GA-BP?based on genetic algorithm is used to compensate the temperature and humidity of the sensor.The temperature and humidity drift of the sensor system are well suppressed.The CO₂ concentration measurement error compensated by the algorithm is respectively within±60 ppm and±110ppm.This paper studies the need to meet the real-time monitoring of CO₂ concentration in indoor environment,and has important reference value for structural optimization and measurement accuracy improvement of infrared CO₂ gas sensor.