Research on a Mixed Gas Recognition and Concentration Detection Algorithm Based on a Metal Oxide Semiconductor Olfactory System Sensor Array

Xu, Yong; Zhao, Xibin; Chen, Yinsheng; Zhao, Wei

doi:10.3390/s18103264

Cited by 42 publications

(22 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…All of this was done in [ 11 ] using classical methods. In [ 12 ], kernel principal component analysis was applied to raw data to extract non-linear characteristics from it, and then the K-nearest neighbor algorithm was used for recognition tasks.…”

Section: Discussionmentioning

confidence: 99%

“…The kernel principal component analysis was used [ 12 ] to extract the nonlinear mixed gas characteristics of different components, and the K-nearest neighbor algorithm was used to recognize the target gas. In [ 12 ], a gas identification and concentration detection method was presented and a multivariable relevance vector machine was used to detect the concentration of the hybrid gas. The aforementioned method was validated by using CO and methane (CH 4 ).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust Estimation of Carbon Monoxide Measurements

Hernández

Méndez

2020

Sensors

View full text Add to dashboard Cite

This paper presents a robust analysis of carbon monoxide (CO) concentration measurements conducted at the Belisario air-quality monitoring station (Quito, Ecuador). For the analysis, the data collected from 1 January 2008 to 31 December 2019 were considered. Additionally, each of the twelve years analyzed was considered as a random variable, and robust location and scale estimators were used to estimate the central tendency and dispersion of the data. Furthermore, classic, nonparametric, bootstrap, and robust confidence intervals were used to group the variables into categories. Then, differences between categories were quantified using confidence intervals and it was shown that the trend of CO concentration at the Belisario station in the last twelve years is downward. The latter was proven with the precision provided by both nonparametric and robust statistical methods. The results of the research work robustly proved that the CO concentration at Belisario station in the last twelve years is not considered a health risk, according to the criteria established by the Quito Air Quality Index.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Robust Estimation of Carbon Monoxide Measurements

Hernández

Méndez

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…The difference in baseline values for different sensors is slightly larger. In order to eliminate the impact of the baseline on the data and ensure the reliability of the data, the general pattern recognition algorithm first take the standardized operation to the data [12]. The method is as shown in Equation (4), where X Standard value is the value obtained after subtracting the baseline treatment, which we call: “standard value”, X Response value is the true value of the sensor during the response, X Baseline value is the baseline value of the sensor in air or standard gas.…”

Section: Data Processingmentioning

confidence: 99%

“…(3) Fusion method. A variety of statistical or machine learning methods are combined to classify signals, which can improve the classification accuracy to a certain extent [12]. Table 1 summarizes the advantages and disadvantages of the three methods for identifying components in a gas mixture.…”

Section: Introductionmentioning

confidence: 99%

A New Method of Mixed Gas Identification Based on a Convolutional Neural Network for Time Series Classification

Han

Xiao

et al. 2019

Sensors

View full text Add to dashboard Cite

This paper proposes a new method of mixed gas identification based on a convolutional neural network for time series classification. In view of the superiority of convolutional neural networks in the field of computer vision, we applied the concept to the classification of five mixed gas time series data collected by an array of eight MOX gas sensors. Existing convolutional neural networks are mostly used for processing visual data, and are rarely used in gas data classification and have great limitations. Therefore, the idea of mapping time series data into an analogous-image matrix data is proposed. Then, five kinds of convolutional neural networks—VGG-16, VGG-19, ResNet18, ResNet34 and ResNet50—were used to classify and compare five kinds of mixed gases. By adjusting the parameters of the convolutional neural networks, the final gas recognition rate is 96.67%. The experimental results show that the method can classify the gas data quickly and effectively, and effectively combine the gas time series data with classical convolutional neural networks, which provides a new idea for the identification of mixed gases.

show abstract

“…Machine olfaction is a new type of biomimetic detection technology, which can be used to simulate the working mechanism of biological olfaction. It is often used in the analysis and detection of various gases, such as in pollution control [1,2], medical technology [3,4], and oil exploration [5]. Researchers have achieved good results in the study of machine olfaction [6][7][8].…”

Section: Introductionmentioning

confidence: 99%

Research on a Mixed Gas Classification Algorithm Based on Extreme Random Tree

Zhao

Chen

et al. 2019

Applied Sciences

Self Cite

View full text Add to dashboard Cite

Because of the low accuracy of the current machine olfactory algorithms in detecting two mixed gases, this study proposes a hybrid gas detection algorithm based on an extreme random tree to greatly improve the classification accuracy and time efficiency. The method mainly uses the dynamic time warping algorithm (DTW) to perform data pre-processing and then extracts the gas characteristics from gas signals at different concentrations by applying a principal component analysis (PCA). Finally, the model is established by using a new extreme random tree algorithm to achieve the target gas classification. The sample data collected by the experiment was verified by comparison experiments with the proposed algorithm. The analysis results show that the proposed DTW algorithm improves the gas classification accuracy by 26.87%. Compared with the random forest algorithm, extreme gradient boosting (XGBoost) algorithm and gradient boosting decision tree (GBDT) algorithm, the accuracy rate increased by 4.53%, 5.11% and 8.10%, respectively, reaching 99.28%. In terms of the time efficiency of the algorithms, the actual runtime of the extreme random tree algorithm is 66.85%, 90.27%, and 81.61% lower than that of the random forest algorithm, XGBoost algorithm, and GBDT algorithm, respectively, reaching 103.2568 s.

show abstract

Research on a Mixed Gas Recognition and Concentration Detection Algorithm Based on a Metal Oxide Semiconductor Olfactory System Sensor Array

Cited by 42 publications

References 45 publications

Robust Estimation of Carbon Monoxide Measurements

Robust Estimation of Carbon Monoxide Measurements

A New Method of Mixed Gas Identification Based on a Convolutional Neural Network for Time Series Classification

Research on a Mixed Gas Classification Algorithm Based on Extreme Random Tree

Contact Info

Product

Resources

About