Calibration of CO, NO2, and O3 Using Airify: A Low-Cost Sensor Cluster for Air Quality Monitoring

Abstract

During the last decade, extensive research has been carried out on the subject of low-cost sensor platforms for air quality monitoring. A key aspect when deploying such systems is the quality of the measured data. Calibration is especially important to improve the data quality of low-cost air monitoring devices. The measured data quality must comply with regulations issued by national or international authorities in order to be used for regulatory purposes. This work discusses the challenges and methods suitable for calibrating a low-cost sensor platform developed by our group, Airify, that has a unit cost five times less expensive than the state-of-the-art solutions (approximately €1000). The evaluated platform can integrate a wide variety of sensors capable of measuring up to 12 parameters, including the regulatory pollutants defined in the European Directive. In this work, we developed new calibration models (multivariate linear regression and random forest) and evaluated their effectiveness in meeting the data quality objective (DQO) for the following parameters: carbon monoxide (CO), ozone (O3), and nitrogen dioxide (NO2). The experimental results show that the proposed calibration managed an improvement of 12% for the CO and O3 gases and a similar accuracy for the NO2 gas compared to similar state-of-the-art studies. The evaluated parameters had different calibration accuracies due to the non-identical levels of gas concentration at which the sensors were exposed during the model’s training phase. After the calibration algorithms were applied to the evaluated platform, its performance met the DQO criteria despite the overall low price level of the platform.

Smart Multi-Sensor Calibration of Low-Cost Particulate Matter Monitors

“…Interestingly, the KNN method presents results close to the best despite its simplicity, although with a higher standard deviation, which can represent unwanted performance variability with small perturbation to the training and test data. These results are in line with other results reported in the literature relating appealing results of ensemble methods in calibrating air quality sensors [ 32 , 33 , 34 , 36 , 37 ], confirming the nonlinear behavior of these devices.…”

A Novel Bike-Mounted Sensing Device with Cloud Connectivity for Dynamic Air-Quality Monitoring by Urban Cyclists

“…An artificial neural network was used to calculate this calibration function. Neural networks are a fundamental tool in the field of low-cost-sensor calibration [ 25 , 26 , 27 , 28 ]. In a previous work [ 29 ], we tested this type of algorithm against other common techniques and concluded that neural networks performed the best.…”

Design and Evaluation of Low-Power Co3O4 Gas Sensing Element as a Part of Cyber Physical Systems

“…80-1542) with the main diffraction peaks at 31.17 • , 36.72 • , 44.66 • , 59.15 • and 65.00 • corresponding to (220), (311), (400), ( 511) and (440) planes, respectively [25,26]. The lattice parameters of the Co 3 O 4 nanoparticles were a = b = c = 0.81099 nm, while the cell volume was 0.53339 nm 3 . Regarding Scherrer's formula applied in the main peak (311), the crystallite size of the Co 3 O 4 nanoparticle was found to be 43.7 nm.…”

“…Previously, only bulky industrial grade gas sensors could be found on the market, however, nowadays, advances in gas sensing technology and materials have sparked the creation of low-cost, low-power, small form factor sensors. Use cases include systems applied in smart cities for the purpose of gas pollution monitoring for health reasons, or in factories for gas-leak detection for safety reasons [3].…”