State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China.
State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China.
Sensors (Basel). 2021 Jan 2;21(1):256. doi: 10.3390/s21010256.
Pollutant gases, such as CO, NO, O, and SO affect human health, and low-cost sensors are an important complement to regulatory-grade instruments in pollutant monitoring. Previous studies focused on one or several species, while comprehensive assessments of multiple sensors remain limited. We conducted a 12-month field evaluation of four Alphasense sensors in Beijing and used single linear regression (SLR), multiple linear regression (MLR), random forest regressor (RFR), and neural network (long short-term memory (LSTM)) methods to calibrate and validate the measurements with nearby reference measurements from national monitoring stations. For performances, CO > O > NO > SO for the coefficient of determination (R) and root mean square error (RMSE). The MLR did not increase the R after considering the temperature and relative humidity influences compared with the SLR (with R remaining at approximately 0.6 for O and 0.4 for NO). However, the RFR and LSTM models significantly increased the O, NO, and SO performances, with the R increasing from 0.3-0.5 to >0.7 for O and NO, and the RMSE decreasing from 20.4 to 13.2 ppb for NO. For the SLR, there were relatively larger biases, while the LSTMs maintained a close mean relative bias of approximately zero (e.g., <5% for O and NO), indicating that these sensors combined with the LSTMs are suitable for hot spot detection. We highlight that the performance of LSTM is better than that of random forest and linear methods. This study assessed four electrochemical air quality sensors and different calibration models, and the methodology and results can benefit assessments of other low-cost sensors.
污染物气体，如 CO、NO、O 和 SO，会影响人类健康，而低成本传感器是污染物监测中监管级仪器的重要补充。以前的研究集中在一种或几种物种上，而对多种传感器的综合评估仍然有限。我们在北京对四个 Alphasense 传感器进行了为期 12 个月的现场评估，并使用单一线性回归 (SLR)、多元线性回归 (MLR)、随机森林回归器 (RFR) 和神经网络 (长短期记忆 (LSTM)) 方法，使用附近国家监测站的参考测量值对测量值进行校准和验证。对于性能，CO>O>NO>SO 的决定系数 (R) 和均方根误差 (RMSE) 较高。与 SLR 相比，在考虑温度和相对湿度影响后，MLR 并没有增加 R（O 的 R 保持在 0.6 左右，NO 的 R 保持在 0.4 左右）。然而，RFR 和 LSTM 模型显著提高了 O、NO 和 SO 的性能，O 和 NO 的 R 从 0.3-0.5 增加到>0.7，NO 的 RMSE 从 20.4 降低到 13.2 ppb。对于 SLR，存在相对较大的偏差，而 LSTM 保持接近零的平均相对偏差（例如，O 和 NO 的<5%），这表明这些传感器与 LSTM 结合适用于热点检测。我们强调，LSTM 的性能优于随机森林和线性方法。本研究评估了四个电化学空气质量传感器和不同的校准模型，该方法和结果可以为其他低成本传感器的评估提供参考。
