Machine Learning based calibration time reduction for Gas Sensors in Temperature Cycled Operation

Robin, Yannick; Goodarzi, Payman; Baur, Tobias; Schultealbert, Caroline; Schütze, Andreas; Schneider, Tizian

doi:10.1109/i2mtc50364.2021.9459919

Cited by 6 publications

(12 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The second model-building approach then utilizes the TCOCNN architecture (see Figure 3), a 10-layer deep convolutional neural network (CNN) [15]. A similar network was first introduced in [33] and successfully utilized to predict the formaldehyde concentration for the laboratory calibration measurements. For this contribution, the structure of this network was adapted to predict not only one gas concentration at a time, but the concentrations of all gases offered during calibration, i.e., acetone, ethanol, formaldehyde, toluene, the total concentration of all VOCs (VOC sum ), and also the inorganic gases carbon monoxide and hydrogen.…”

Section: Model Buildingmentioning

confidence: 99%

“…For this contribution, the structure of this network was adapted to predict not only one gas concentration at a time, but the concentrations of all gases offered during calibration, i.e., acetone, ethanol, formaldehyde, toluene, the total concentration of all VOCs (VOC sum ), and also the inorganic gases carbon monoxide and hydrogen. The CNN structure was derived from the original ResNet model from [33] to reduce the overall complexity. To build a gas-specific model, the general architecture as illustrated in Figure 3 was used.…”

Section: Model Buildingmentioning

confidence: 99%

“…Additionally, the dropout rate and the number of neurons in the fully connected layers are also parameters worth considering. The ranges for the different target gases were based on the best parameters found in [33]. For carbon monoxide and ethanol, the NAS had to train 60 different TCOCNNs to reach sufficient results as the model building seemed to be more difficult for these gases.…”

Section: Model Buildingmentioning

confidence: 99%

“…As a carcinogenic gas, formaldehyde is of great importance for indoor air quality. Thus, as the first step, we evaluated the suitability of the model for predicting the formaldehyde concentration in the ppb range in a complex mixture of other gases [33]. Here, a model for formaldehyde was trained on the initial calibration dataset with a gas-mixture-based data split of 70% training, 10% validation, and 20% for testing.…”

Section: Data Evaluationmentioning

confidence: 99%

“…To determine the required complexity of the calibration (note that one hundred unique gas mixtures offered for 20 min each resulted in a total calibration time of 33 h; the extended calibration with higher VOC concentrations, therefore, required almost 7 d in total), the same model was built with fewer core samples and/or fewer unique gas mixtures to reproduce the results achieved in [33] and also to show the influence of more core samples compared to more unique gas mixtures (UGM).…”

Section: Data Evaluationmentioning

confidence: 99%

See 4 more Smart Citations

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

et al. 2021

Self Cite

View full text Add to dashboard Cite

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.

show abstract

Section: Model Buildingmentioning

confidence: 99%

Section: Model Buildingmentioning

confidence: 99%

Section: Model Buildingmentioning

confidence: 99%

Section: Data Evaluationmentioning

confidence: 99%

Section: Data Evaluationmentioning

confidence: 99%

See 3 more Smart Citations

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

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

Comparison of different ML methods concerning prediction quality, domain adaptation and robustness

Goodarzi

Schütze

Schneider

2022

Tm - Technisches Messen

View full text Add to dashboard Cite

Nowadays machine learning methods and data-driven models have been used widely in different fields including computer vision, biomedicine, and condition monitoring. However, these models show performance degradation when meeting real-life situations. Domain or dataset shift or out-of-distribution (OOD) prediction is mentioned as the reason for this problem. Especially in industrial condition monitoring, it is not clear when we should be concerned about domain shift and which methods are more robust against this problem. In this paper prediction results are compared for a conventional machine learning workflow based on feature extraction, selection, and classification/regression (FESC/R) and deep neural networks on two publicly available industrial datasets. We show that it is possible to visualize the possible shift in domain using feature extraction and principal component analysis. Also, experimental competition shows that the cross-domain validated results of FESC/R are comparable to the reported state-of-the-art methods. Finally, we show that the results for simple randomly selected validation sets do not correctly represent the model performance in real-world applications.

show abstract

Field Study of Metal Oxide Semiconductor Gas Sensors in Temperature Cycled Operation for Selective VOC Monitoring in Indoor Air

et al. 2021

View full text Add to dashboard Cite

More and more metal oxide semiconductor (MOS) gas sensors with digital interfaces are entering the market for indoor air quality (IAQ) monitoring. These sensors are intended to measure volatile organic compounds (VOCs) in indoor air, an important air quality factor. However, their standard operating mode often does not make full use of their true capabilities. More sophisticated operation modes, extensive calibration and advanced data evaluation can significantly improve VOC measurements and, furthermore, achieve selective measurements of single gases or at least types of VOCs. This study provides an overview of the potential and limits of MOS gas sensors for IAQ monitoring using temperature cycled operation (TCO), calibration with randomized exposure and data-based models trained with advanced machine learning. After lab calibration, a commercial digital gas sensor with four different gas-sensitive layers was tested in the field over several weeks. In addition to monitoring normal ambient air, release tests were performed with compounds that were included in the lab calibration, but also with additional VOCs. The tests were accompanied by different analytical systems (GC-MS with Tenax sampling, mobile GC-PID and GC-RCP). The results show quantitative agreement between analytical systems and the MOS gas sensor system. The study shows that MOS sensors are highly suitable for determining the overall VOC concentrations with high temporal resolution and, with some restrictions, also for selective measurements of individual components.

show abstract

Machine Learning based calibration time reduction for Gas Sensors in Temperature Cycled Operation

Cited by 6 publications

References 17 publications

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

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

Comparison of different ML methods concerning prediction quality, domain adaptation and robustness

Field Study of Metal Oxide Semiconductor Gas Sensors in Temperature Cycled Operation for Selective VOC Monitoring in Indoor Air

Contact Info

Product

Resources

About