Application of Machine Learning for the in-Field Correction of a PM2.5 Low-Cost Sensor Network

Abstract

Many low-cost sensors (LCSs) are distributed for air monitoring without any rigorous calibrations. This work applies machine learning with PM2.5 from Taiwan monitoring stations to conduct in-field corrections on a network of 39 PM2.5 LCSs from July 2017 to December 2018. Three candidate models were evaluated: Multiple linear regression (MLR), support vector regression (SVR), and random forest regression (RFR). The model-corrected PM2.5 levels were compared with those of GRIMM-calibrated PM2.5. RFR was superior to MLR and SVR in its correction accuracy and computing efficiency. Compared to SVR, the root mean square errors (RMSEs) of RFR were 35% and 85% lower for the training and validation sets, respectively, and the computational speed was 35 times faster. An RFR with 300 decision trees was chosen as the optimal setting considering both the correction performance and the modeling time. An RFR with a nighttime pattern was established as the optimal correction model, and the RMSEs were 5.9 ± 2.0 μg/m3, reduced from 18.4 ± 6.5 μg/m3 before correction. This is the first work to correct LCSs at locations without monitoring stations, validated using laboratory-calibrated data. Similar models could be established in other countries to greatly enhance the usefulness of their PM2.5 sensor networks.

Building Low-Cost Sensing Infrastructure for Air Quality Monitoring in Urban Areas Based on Fog Computing

“…The analysis presented in [13] has shown that from different evaluated ML models, the random forest algorithm expresses the best performances, in most of the observed cases, while in some cases artificial neural networks (ANN) could improve the performance as well. In [14][15][16][17], various kinds of machine learning algorithms were proposed and evaluated for different air pollution monitoring scenarios. Due to aging, sensors gradually begin to lose their sensitivity or start to show a drift in their measurements.…”

Multi-Sensor Information Ensemble-Based Automatic Parking System for Vehicle Parallel/Nonparallel Initial State

“…The value measured by the sensor is usually affected by external conditions, so the sensor requires a separate correction equation. In mathematical models, multiple regression models [ 40 ] that can consider surrounding environmental factors have been used to establish relevant correction formulas. According to the results, the mathematical model of multiple linear regression is given as follows: where is the measured average error of double sensors; and and are the horizontal distance and driving speed, respectively.…”

“…According to the results, the mathematical model of multiple linear regression is given as follows: The value measured by the sensor is usually affected by external conditions, so the sensor requires a separate correction equation. In mathematical models, multiple regression models [40] that can consider surrounding environmental factors have been used to establish relevant correction formulas. According to the results, the mathematical model of multiple linear regression is given as follows:…”

Analysis of Air Pollution Migration during COVID-19 Lockdown in Krakow, Poland

“…Airly is not sharing its calibration algorithms with the public (Airly, 2021b). However, the use of ML techniques for this purpose is well established and has been published in many papers (Zimmerman et al, 2018;Okafor et al, 2020;Wang et al, 2020), as well as discussed in the official World Health Organisation and World Meteorological Organization report (Peltier et al, 2021). It was also shown that during the COVID-19 pandemic spring the migration of air pollutions from solid fuels heating from surrounding municipalities to Krakow was the main source of PMs in the city as transportation was significantly reduced due to lockdown.…”