Loh Byoung Gook, Choi Gi Heung
Department of Applied IT Engineering, Hansung University, Seoul, Republic of Korea.
Department of Mechanical Systems Engineering, Hansung University, Seoul, Republic of Korea.
Saf Health Work. 2019 Dec;10(4):452-460. doi: 10.1016/j.shaw.2019.08.002. Epub 2019 Aug 19.
Monitoring and control of PM are being recognized as key to address health issues attributed to PM. Availability of low-cost PM sensors made it possible to introduce a number of portable PM monitors based on light scattering to the consumer market at an affordable price. Accuracy of light scattering-based PM monitors significantly depends on the method of calibration. Static calibration curve is used as the most popular calibration method for low-cost PM sensors particularly because of ease of application. Drawback in this approach is, however, the lack of accuracy.
This study discussed the calibration of a low-cost PM-monitoring device (PMD) to improve the accuracy and reliability for practical use. The proposed method is based on construction of the PM sensor network using Message Queuing Telemetry Transport (MQTT) protocol and web query of reference measurement data available at government-authorized PM monitoring station (GAMS) in the republic of Korea. Four machine learning (ML) algorithms such as support vector machine, k-nearest neighbors, random forest, and extreme gradient boosting were used as regression models to calibrate the PMD measurements of PM. Performance of each ML algorithm was evaluated using stratified K-fold cross-validation, and a linear regression model was used as a reference.
Based on the performance of ML algorithms used, regression of the output of the PMD to PM concentrations data available from the GAMS through web query was effective. The extreme gradient boosting algorithm showed the best performance with a mean coefficient of determination (R) of 0.78 and standard error of 5.0 μg/m, corresponding to 8% increase in R and 12% decrease in root mean square error in comparison with the linear regression model. Minimum 100 hours of calibration period was found required to calibrate the PMD to its full capacity. Calibration method proposed poses a limitation on the location of the PMD being in the vicinity of the GAMS. As the number of the PMD participating in the sensor network increases, however, calibrated PMDs can be used as reference devices to nearby PMDs that require calibration, forming a calibration chain through MQTT protocol.
Calibration of a low-cost PMD, which is based on construction of PM sensor network using MQTT protocol and web query of reference measurement data available at a GAMS, significantly improves the accuracy and reliability of a PMD, thereby making practical use of the low-cost PMD possible.
对细颗粒物（PM）的监测与控制被视为解决因PM导致的健康问题的关键。低成本PM传感器的出现使得一些基于光散射原理的便携式PM监测仪能够以实惠的价格进入消费市场。基于光散射的PM监测仪的准确性很大程度上取决于校准方法。静态校准曲线是低成本PM传感器最常用的校准方法，尤其是因其易于应用。然而，这种方法的缺点是缺乏准确性。
本研究探讨了对一种低成本PM监测设备（PMD）进行校准，以提高其实际使用中的准确性和可靠性。所提出的方法基于使用消息队列遥测传输（MQTT）协议构建PM传感器网络，并通过网络查询韩国政府授权的PM监测站（GAMS）提供的参考测量数据。使用支持向量机、k近邻、随机森林和极端梯度提升等四种机器学习（ML）算法作为回归模型来校准PMD对PM的测量。使用分层K折交叉验证评估每种ML算法的性能，并将线性回归模型用作参考。
基于所使用的ML算法的性能，通过网络查询将PMD的输出回归到GAMS提供的PM浓度数据是有效的。极端梯度提升算法表现最佳，平均决定系数（R）为0.78，标准误差为5.0μg/m³，与线性回归模型相比，R增加了8%，均方根误差降低了12%。发现将PMD校准到其满量程需要至少100小时的校准期。所提出的校准方法对PMD位于GAMS附近的位置存在限制。然而，随着参与传感器网络的PMD数量增加，已校准的PMD可作为附近需要校准的PMD的参考设备，通过MQTT协议形成校准链。
基于使用MQTT协议构建PM传感器网络并通过网络查询GAMS提供的参考测量数据对低成本PMD进行校准，可显著提高PMD的准确性和可靠性，从而使低成本PMD的实际应用成为可能。
