High-Performance VOC Quantification for IAQ Monitoring Using Advanced Sensor Systems and Deep Learning

Abstract

With air quality being one target in the sustainable development goals set by the United Nations, accurate monitoring also of indoor air quality is more important than ever. Chemiresistive gas sensors are an inexpensive and promising solution for the monitoring of volatile organic compounds, which are of high concern indoors. To fully exploit the potential of these sensors, advanced operating modes, calibration, and data evaluation methods are required. This contribution outlines a systematic approach based on dynamic operation (temperature-cycled operation), randomized calibration (Latin hypercube sampling), and the use of advances in deep neural networks originally developed for natural language processing and computer vision, applying this approach to volatile organic compound measurements for indoor air quality monitoring. This paper discusses the pros and cons of deep neural networks for volatile organic compound monitoring in a laboratory environment by comparing the quantification accuracy of state-of-the-art data evaluation methods with a 10-layer deep convolutional neural network (TCOCNN). The overall performance of both methods was compared for complex gas mixtures with several volatile organic compounds, as well as interfering gases and changing ambient humidity in a comprehensive lab evaluation. Furthermore, both were tested under realistic conditions in the field with additional release tests of volatile organic compounds. The results obtained during field testing were compared with analytical measurements, namely the gold standard gas chromatography mass spectrometry analysis based on Tenax sampling, as well as two mobile systems, a gas chromatograph with photo-ionization detection for volatile organic compound monitoring and a gas chromatograph with a reducing compound photometer for the monitoring of hydrogen. The results showed that the TCOCNN outperforms state-of-the-art data evaluation methods, for example for critical pollutants such as formaldehyde, achieving an uncertainty of around 11 ppb even in complex mixtures, and offers a more robust volatile organic compound quantification in a laboratory environment, as well as in real ambient air for most targets.

Review–Modern Data Analysis in Gas Sensors

“…Feng et al 88 proposed an augmented CNN modeling approach where a feedback network updates the base model with necessary adjustment in case of drifted dataset. Another CNN model with 10 layers was proposed in a prior study 131 for volatile organic compound (VOC) detection which achieved promising results with minimized root mean squared error (RMSE). For specific elimination of the effect of temperature and humidity in sensor signal, a deep back propagation neural network was proposed in a prior study 132 where 14 hidden layers have been used and optimized for analyte classification.…”

5.3 - Überwachung der Luftqualität in Innenräumen mittels komplexer Sensorsysteme und Deep Learning Ansätzen

“…Für dieses Gas kann ein deutlicher Abfall der Konzentration beobachtet werden, welcher deutlich über eine längere Periode anhält. Dieser Effekt kann damit begründet werden, dass der SGP30 keine besonders gute Selektivität für dieses Gas besitzt [7], wie bereits mithilfe der datengetriebenen Modelle in den Laboruntersuchungen gezeigt [13].…”

“…Dieses Vorgehen soll gewährleisten, dass das Model auf jedes Gas angepasst wird und einen minimierten Fehler auf den Testdaten garantiert. Vorhersagen zu generieren, die dem FESR Modell überlegen sind [13]. nahe der theoretisch freigesetzten Menge von 600 ppb vorhersagen.…”

P10 - Propofol Detection with Metal Oxide Semiconductor Gas Sensors

“…By using temperature-cycled operation (TCO) [17] we could show that a single MOS sensor is capable of quantifying single VOCs in the low ppb-range in a complex and varying background of interfering VOCs, hydrogen (H2) and carbon monoxide (CO), e.g. for indoor air quality applications [18]- [20]. Dynamic operation together with signal processing based on machine learning and a complex lab calibration with randomized gas mixtures are the basis to achieve a performance of MOS sensors which is comparable to analytics but with the advantage of being low-cost and offering real time and online monitoring.…”