Calibration model of a low-cost air quality sensor using an adaptive neuro-fuzzy inference system

Kemal Maulana Alhasa, Mohd Shahrul Mohd Nadzir, Popoola Olalekan, Mohd Talib Latif, Yusri Yusup, Mohammad Rashed Iqbal Faruque, Fatimah PK Ahamad, Haris Hafizal Abd Hamid, Kadaruddin Aiyub, Sawal Hamid Md Ali, Firoz Khan, Azizan Abu Samah, Imran Yusuff, Murnira Othman, Tengku Mohd Farid Tengku Hassim, Nor Eliani Ezani

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Conventional air quality monitoring systems, such as gas analysers, are commonly used in many developed and developing countries to monitor air quality. However, these techniques have high costs associated with both installation and maintenance. One possible solution to complement these techniques is the application of low-cost air quality sensors (LAQSs), which have the potential to give higher spatial and temporal data of gas pollutants with high precision and accuracy. In this paper, we present DiracSense, a custom-made LAQS that monitors the gas pollutants ozone (O3), nitrogen dioxide (NO2), and carbon monoxide (CO). The aim of this study is to investigate its performance based on laboratory calibration and field experiments. Several model calibrations were developed to improve the accuracy and performance of the LAQS. Laboratory calibrations were carried out to determine the zero offset and sensitivities of each sensor. The results showed that the sensor performed with a highly linear correlation with the reference instrument with a response-time rangefrom 0.5 to 1.7 min. The performance of several calibration models including a calibrated simple equation and supervised learning algorithms (adaptive neuro-fuzzy inference system or ANFIS and the multilayer feed-forward perceptron or MLP) were compared. The field calibration focused on O3 measurements due to the lack of a reference instrument for CO and NO2. Combinations of inputs were evaluated during the development of the supervised learning algorithm. The validation results demonstrated that the ANFIS model with four inputs (WE OX, AE OX, T, and NO2) had the lowest error in terms of statistical performance and the highest correlation coefficients with respect to the reference instrument (0.8 < r < 0.95). These results suggest that the ANFIS model is promising as a calibration tool since it has the capability to improve the accuracy and performance of the low-cost electrochemical sensor.

Original languageEnglish
Article number4380
JournalSensors (Switzerland)
Volume18
Issue number12
DOIs
Publication statusPublished - 1 Dec 2018

Fingerprint

air quality
Fuzzy inference
inference
Air quality
Calibration
Air
Costs and Cost Analysis
sensors
Sensors
Costs
Gases
Supervised learning
Carbon Monoxide
Carbon monoxide
carbon monoxide
Learning algorithms
learning
contaminants
gases
Learning

Keywords

  • Air quality monitoring
  • Low-cost sensor
  • Machine learning
  • Quality control

ASJC Scopus subject areas

  • Analytical Chemistry
  • Atomic and Molecular Physics, and Optics
  • Biochemistry
  • Instrumentation
  • Electrical and Electronic Engineering

Cite this

Calibration model of a low-cost air quality sensor using an adaptive neuro-fuzzy inference system. / Alhasa, Kemal Maulana; Mohd Nadzir, Mohd Shahrul; Olalekan, Popoola; Latif, Mohd Talib; Yusup, Yusri; Faruque, Mohammad Rashed Iqbal; PK Ahamad, Fatimah; Hamid, Haris Hafizal Abd; Aiyub, Kadaruddin; Md Ali, Sawal Hamid; Khan, Firoz; Samah, Azizan Abu; Yusuff, Imran; Othman, Murnira; Hassim, Tengku Mohd Farid Tengku; Ezani, Nor Eliani.

In: Sensors (Switzerland), Vol. 18, No. 12, 4380, 01.12.2018.

Research output: Contribution to journalArticle

Alhasa, Kemal Maulana ; Mohd Nadzir, Mohd Shahrul ; Olalekan, Popoola ; Latif, Mohd Talib ; Yusup, Yusri ; Faruque, Mohammad Rashed Iqbal ; PK Ahamad, Fatimah ; Hamid, Haris Hafizal Abd ; Aiyub, Kadaruddin ; Md Ali, Sawal Hamid ; Khan, Firoz ; Samah, Azizan Abu ; Yusuff, Imran ; Othman, Murnira ; Hassim, Tengku Mohd Farid Tengku ; Ezani, Nor Eliani. / Calibration model of a low-cost air quality sensor using an adaptive neuro-fuzzy inference system. In: Sensors (Switzerland). 2018 ; Vol. 18, No. 12.
@article{2eba70b9651d462aab40b3c1494db54d,
title = "Calibration model of a low-cost air quality sensor using an adaptive neuro-fuzzy inference system",
abstract = "Conventional air quality monitoring systems, such as gas analysers, are commonly used in many developed and developing countries to monitor air quality. However, these techniques have high costs associated with both installation and maintenance. One possible solution to complement these techniques is the application of low-cost air quality sensors (LAQSs), which have the potential to give higher spatial and temporal data of gas pollutants with high precision and accuracy. In this paper, we present DiracSense, a custom-made LAQS that monitors the gas pollutants ozone (O3), nitrogen dioxide (NO2), and carbon monoxide (CO). The aim of this study is to investigate its performance based on laboratory calibration and field experiments. Several model calibrations were developed to improve the accuracy and performance of the LAQS. Laboratory calibrations were carried out to determine the zero offset and sensitivities of each sensor. The results showed that the sensor performed with a highly linear correlation with the reference instrument with a response-time rangefrom 0.5 to 1.7 min. The performance of several calibration models including a calibrated simple equation and supervised learning algorithms (adaptive neuro-fuzzy inference system or ANFIS and the multilayer feed-forward perceptron or MLP) were compared. The field calibration focused on O3 measurements due to the lack of a reference instrument for CO and NO2. Combinations of inputs were evaluated during the development of the supervised learning algorithm. The validation results demonstrated that the ANFIS model with four inputs (WE OX, AE OX, T, and NO2) had the lowest error in terms of statistical performance and the highest correlation coefficients with respect to the reference instrument (0.8 < r < 0.95). These results suggest that the ANFIS model is promising as a calibration tool since it has the capability to improve the accuracy and performance of the low-cost electrochemical sensor.",
keywords = "Air quality monitoring, Low-cost sensor, Machine learning, Quality control",
author = "Alhasa, {Kemal Maulana} and {Mohd Nadzir}, {Mohd Shahrul} and Popoola Olalekan and Latif, {Mohd Talib} and Yusri Yusup and Faruque, {Mohammad Rashed Iqbal} and {PK Ahamad}, Fatimah and Hamid, {Haris Hafizal Abd} and Kadaruddin Aiyub and {Md Ali}, {Sawal Hamid} and Firoz Khan and Samah, {Azizan Abu} and Imran Yusuff and Murnira Othman and Hassim, {Tengku Mohd Farid Tengku} and Ezani, {Nor Eliani}",
year = "2018",
month = "12",
day = "1",
doi = "10.3390/s18124380",
language = "English",
volume = "18",
journal = "Sensors (Switzerland)",
issn = "1424-8220",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "12",

}

TY - JOUR

T1 - Calibration model of a low-cost air quality sensor using an adaptive neuro-fuzzy inference system

AU - Alhasa, Kemal Maulana

AU - Mohd Nadzir, Mohd Shahrul

AU - Olalekan, Popoola

AU - Latif, Mohd Talib

AU - Yusup, Yusri

AU - Faruque, Mohammad Rashed Iqbal

AU - PK Ahamad, Fatimah

AU - Hamid, Haris Hafizal Abd

AU - Aiyub, Kadaruddin

AU - Md Ali, Sawal Hamid

AU - Khan, Firoz

AU - Samah, Azizan Abu

AU - Yusuff, Imran

AU - Othman, Murnira

AU - Hassim, Tengku Mohd Farid Tengku

AU - Ezani, Nor Eliani

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Conventional air quality monitoring systems, such as gas analysers, are commonly used in many developed and developing countries to monitor air quality. However, these techniques have high costs associated with both installation and maintenance. One possible solution to complement these techniques is the application of low-cost air quality sensors (LAQSs), which have the potential to give higher spatial and temporal data of gas pollutants with high precision and accuracy. In this paper, we present DiracSense, a custom-made LAQS that monitors the gas pollutants ozone (O3), nitrogen dioxide (NO2), and carbon monoxide (CO). The aim of this study is to investigate its performance based on laboratory calibration and field experiments. Several model calibrations were developed to improve the accuracy and performance of the LAQS. Laboratory calibrations were carried out to determine the zero offset and sensitivities of each sensor. The results showed that the sensor performed with a highly linear correlation with the reference instrument with a response-time rangefrom 0.5 to 1.7 min. The performance of several calibration models including a calibrated simple equation and supervised learning algorithms (adaptive neuro-fuzzy inference system or ANFIS and the multilayer feed-forward perceptron or MLP) were compared. The field calibration focused on O3 measurements due to the lack of a reference instrument for CO and NO2. Combinations of inputs were evaluated during the development of the supervised learning algorithm. The validation results demonstrated that the ANFIS model with four inputs (WE OX, AE OX, T, and NO2) had the lowest error in terms of statistical performance and the highest correlation coefficients with respect to the reference instrument (0.8 < r < 0.95). These results suggest that the ANFIS model is promising as a calibration tool since it has the capability to improve the accuracy and performance of the low-cost electrochemical sensor.

AB - Conventional air quality monitoring systems, such as gas analysers, are commonly used in many developed and developing countries to monitor air quality. However, these techniques have high costs associated with both installation and maintenance. One possible solution to complement these techniques is the application of low-cost air quality sensors (LAQSs), which have the potential to give higher spatial and temporal data of gas pollutants with high precision and accuracy. In this paper, we present DiracSense, a custom-made LAQS that monitors the gas pollutants ozone (O3), nitrogen dioxide (NO2), and carbon monoxide (CO). The aim of this study is to investigate its performance based on laboratory calibration and field experiments. Several model calibrations were developed to improve the accuracy and performance of the LAQS. Laboratory calibrations were carried out to determine the zero offset and sensitivities of each sensor. The results showed that the sensor performed with a highly linear correlation with the reference instrument with a response-time rangefrom 0.5 to 1.7 min. The performance of several calibration models including a calibrated simple equation and supervised learning algorithms (adaptive neuro-fuzzy inference system or ANFIS and the multilayer feed-forward perceptron or MLP) were compared. The field calibration focused on O3 measurements due to the lack of a reference instrument for CO and NO2. Combinations of inputs were evaluated during the development of the supervised learning algorithm. The validation results demonstrated that the ANFIS model with four inputs (WE OX, AE OX, T, and NO2) had the lowest error in terms of statistical performance and the highest correlation coefficients with respect to the reference instrument (0.8 < r < 0.95). These results suggest that the ANFIS model is promising as a calibration tool since it has the capability to improve the accuracy and performance of the low-cost electrochemical sensor.

KW - Air quality monitoring

KW - Low-cost sensor

KW - Machine learning

KW - Quality control

UR - http://www.scopus.com/inward/record.url?scp=85058515642&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058515642&partnerID=8YFLogxK

U2 - 10.3390/s18124380

DO - 10.3390/s18124380

M3 - Article

VL - 18

JO - Sensors (Switzerland)

JF - Sensors (Switzerland)

SN - 1424-8220

IS - 12

M1 - 4380

ER -