Performance of a Mid-Infrared Sensor for Simultaneous Trace Detection of Atmospheric CO and N2O Based on PSO-KELM

Abstract

In this article, a field deployable sensor was developed using a self-developed 4.58-µm continuous wave quantum cascade laser (CW-QCL) for the simultaneous detection of carbon monoxide (CO) and nitrous oxide (N2O), both of which have strong fundamental absorption bands in this waveband. The sensor is based on tunable diode laser absorption spectroscopy (TDLAS) technology, which combined a multi-pass gas cell (MPGC) with a 41 m optical path length to achieve high-precision detection. Meanwhile, the particle swarm optimization-kernel extreme learning machine (PSO-KELM) algorithm was applied for CO and N2O concentration prediction. In addition, the self-designed board-level QCL driver circuit and harmonic signal demodulation circuit reduce the sensor cost and size. A series of validation experiments were conducted to verify the sensor performance, and experiments showed that the concentration prediction results of the PSO-KELM algorithm are better than those of the commonly used back propagation (BP) neural networks and partial least regression (PLS), with the smallest root mean square error (RMSE) and linear correlation coefficient closest to 1, which improves the detection precision of the sensor. The limit of detection (LoD) was assessed to be 0.25 parts per billion (ppb) for CO and 0.27 ppb for N2O at the averaging time of 24 and 38 s. Field deployment of the sensor was reported for simultaneous detection of CO and N2O in the air.