Enhancing the Accuracy of NDIR Sensors for CO2 Monitoring in Flue Gas with Particulate Matter Compensation

The massive emission of carbon dioxide (CO2) has caused serious climate problems. Nondispersive infrared (NDIR) spectroscopy is a commonly used technique for CO2 detection. Temperature, pressure, relative humidity, and particulate matter affect the NDIR sensor indication. Numerous studies have been conducted to explore the effects of temperature, pressure, and relative humidity on CO2 detection by the NDIR technique and compensate to reduce the error. Minimal research exists on the effect of particulate matter on gas detection by NDIR technique and compensation methods, so it is critical to compensate particulate matter for nondispersive infrared (NDIR) sensor concentration measurements. In this paper, the effects of particle concentration and size on the measurement of CO2 concentration by an NDIR sensor were investigated, calculation and particle compensation coefficients were determined through experiments, and the compensation algorithm was completed. In the range of a particle mass concentration of 0–10 mg/m3 and an average particle size D32 of 0–1.4648 μm, the actual concentration values measured by the NDIR CO2 sensor tended to become smaller as the particle concentration increased and the average particle size D32 increased. The proposed compensation method reduced the root-mean-square error of the NDIR CO2 sensor for PM2.5 and PM1 by 40% and 26%, respectively. The relative error of the sensors was reduced by 50% and 28%, respectively, and the relative error was mostly around 1% after compensation. Experimental results verified the reasonableness of the particle compensation algorithm. Subsequently, a CO2 concentration calculation method coupled with particulate matter, temperature, humidity, and pressure compensation was developed, which can be further applied to the high-precision measurement of gas concentration in industrial emission monitoring.