Calibration based on Polynomial Function for Low-Cost Sensors: A Case Study of Air Purifier Filter

This paper presents a calibration based on a polynomial function to enhance the accuracy of low-cost sensors (LCS) measuring
$P{{M}_{2.5}}$
concentration levels, with a specific focus on the PMS3003 sensor within the LCS category. To process
$P{{M}_{2.5}}$
concentration data recorded by the PMS3003 LCS, an ESP32-based NodeMCU microcontroller is employed. Our approach utilizes polynomial functions for calibrating the PMS3003 LCS, aiming to enhance its accuracy and reliability. The results highlight significant enhancements in accuracy across diverse
$P{{M}_{2.5}}$
concentration ranges. For low (
$P{{M}_{2.5}}\ \le$
200
$\mu g/{{m}^3}$
), medium (200 <
$P{{M}_{2.5}} \le$
400
$\mu g/{{\mathrm{m}}^3}$
), and high (400<
$P{{M}_{2.5}} \le$
600
$\mu g/{{\mathrm{m}}^3}$
)
$P{{M}_{2.5}}$
concentrations, the proposed method demonstrates a maximum error percentage lower than that of the traditional method: approximately 2%, 5.47%, and 6.92%, respectively. Furthermore, we validate the effectiveness of our calibrated PMS3003 LCS by comparing its measurements with those of a reference device (RD)—the air purifier filter. Overall, the results demonstrate substantial accuracy improvements across varying
$P{{M}_{2.5}}$
concentration ranges.