Gas Sensor Array Research Articles

Extrusion‐Printed CNT–Graphene Sensor Array with Embedded MXene/PEDOT:PSS Heater for Enhanced NO2 Sensing at Low Temperature

AbstractWearable sensors are currently one of the top emerging areas with enormous growth potential. Low‐cost fabrication techniques using simple and scalable printing technologies are making a significant impact on their development. Recent advances in high‐performance gas/vapor sensors based on carbon nanomaterials have shown potential applications ranging from disease diagnostics to environmental monitoring and defences. Herein, a hybrid sensing material of 1D carbon nanotubes (CNTs) and 2D graphene is developed, and a conductive ink is formulated, which is applied for fabricating a nitrogen dioxide (NO2) gas sensor array within a compact design utilizing extrusion printing. To improve NO2‐sensing performance and optimal operating temperature, a reverse‐side layer is designed, which combines MXene and poly(3,4‐ethylenedioxythiophene)‐doped poly(styrene sulfonate) (PEDOT:PSS), and functions as a Joule heater. The printed CNT–graphene‐based sensor with an embedded MXene/PEDOT:PSS heater is capable of detecting trace amounts of NO2 gas (1 ppm) at 65 °C. The sensor is able to distinguish between various gases/volatile organic compounds and target NO2 gas based on their chemical affinities. The printed CNT–graphene sensor array also demonstrates a high‐level of recoverability, satisfied stability, durability, and reproducibility, which render this sensor a suitable candidate for practical applications.

Hybrid 3DCNN-RBM Network for Gas Mixture Concentration Estimation With Sensor Array

This work proposes a hybrid 3D Convolutional Neural Network and Restricted Boltzmann Machine (Hybrid 3DCNN-RBM) architecture tailored for gas concentration estimation. The immense success of deep learning in computer vision and natural language processing inspired us to design a deep-learning-based gas concentration estimation network. The proposed network is a joint 3DCNN and RBM network. It uses raw time-series gas sensor array data and provides the concentration of each gas in a mixture of gases. This raw time series data is considered similar to video signal comparing sensor array response as an image, which varies with time. We first time utilized 3DCNN for sensor array data processing and developed a joint network with RBM to design an end-to-end gas concentration estimation model. Although several pattern recognition-based methods have been applied to estimate the gas concentration of the gas mixtures, the effectiveness of these techniques enormously depends on the hand-created feature engineering. Nevertheless, the experiments show that the proposed method is an efficient method for estimating gas concentration in the gas mixture for sensor array data.

Development of Portable E-Nose System for Fast Diagnosis of Whitefly Infestation in Tomato Plant in Greenhouse

An electronic nose (E-nose) system equipped with a gas sensor array and real-time control panel was developed for a fast diagnosis of whitefly infestation in tomato plants. Profile changes of volatile organic compounds (VOCs) released from tomato plants under different treatments (i.e., whitefly infestation, mechanical damage, and no treatment) were successfully determined by the developed E-nose system. A rapid sensor response with high sensitivity towards whitefly-infested tomato plants was observed in the E-nose system. Results of principal component analysis (PCA) and hierarchical clustering analysis (HCA) indicated that the E-nose system was able to provide accurate distinguishment between whitefly-infested plants and healthy plants, with the first three principal components (PCs) accounting for 87.4% of the classification. To reveal the mechanism of whitefly infestation in tomato plants, VOC profiles of whitefly-infested plants and mechanically damaged plants were investigated by using the E-nose system and GC-MS. VOCs of 2-nonanol, oxime-, methoxy-phenyl, and n-hexadecanoic acid were only detected in whitefly-infested plants, while compounds of dodecane and 4,6-dimethyl were only found in mechanically damaged plant samples. Those unique VOC profiles of different tomato plant groups could be considered as bio-markers for diagnosing different damages. Moreover, the E-nose system was demonstrated to have the capability to differentiate whitefly-infested plants and mechanically damaged plants. The relationship between sensor performance and VOC profiles confirmed that the developed E-nose system could be used as a fast and smart device to detect whitefly infestation in greenhouse cultivation.

Design and fabrication of effective gradient temperature sensor array based on bilayer SnO2/Pt for gas classification

Classification of different gases is important, and it is possible to use different gas sensors for this purpose. Electronic noses, for example, combine separated gas sensors into an array for detecting different gases. However, the use of separated sensors in an array suffers from being bulky, high-energy consumption and complex fabrication processes. Generally, gas sensing properties, including gas selectivity, of semiconductor gas sensors are strongly dependent on their working temperature. It is therefore feasible to use a single device composed of identical sensors arranged in a temperature gradient for classification of multiple gases. Herein, we introduce a design for simple fabrication of gas sensor array based on bilayer Pt/SnO2 for real-time monitoring and classification of multiple gases. The study includes design simulation of the sensor array to find an effective gradient temperature, fabrication of the sensors and test of their performance. The array, composed of five sensors, was fabricated on a glass substrate without the need of backside etching to reduce heat loss. A SnO2 thin film sensitized with Pt on top deposited by sputtering was used as sensing material. The sensor array was tested against different gases including ethanol, methanol, isopropanol, acetone, ammonia, and hydrogen. Radar plots and principal component analysis were used to visualize the distinction of the tested gases and to enable effective classification.

Early fire detection based on gas sensor arrays: Multivariate calibration and validation

Smoldering fires are characterized by the production of early gas emissions that can include high levels of CO and Volatile Organic Compounds (VOCs) due to pyrolysis or thermal degradation. Nowadays, standalone CO sensors, smoke detectors, or a combination of these, are standard components for fire alarm systems. While gas sensor arrays together with pattern recognition techniques are a valuable alternative for early fire detection, in practice they have certain drawbacks—they can detect early gas emissions, but can show low immunity to nuisances, and sensor time drift can render calibration models obsolete. In this work, we explore the performance of a gas sensor array for detecting smoldering and plastic fires while ensuring the rejection of a set of nuisances. We conducted variety of fire and nuisance experiments in a validated standard fire room (240 m3). Using PLS-DA and SVM, we evaluate the performance of different multivariate calibration models for this dataset. We show that calibration models remain predictive after several months, but perfect performance is not achieved. For example, 4 months after calibration, a PLS-DA model provides 100% specificity and 85% sensitivity since the system has difficulties in detecting plastic fires, whose signatures are close to nuisance scenarios. Nevertheless, our results show that systems based on gas sensor arrays are able to provide faster fire alarm response than conventional smoke-based fire alarms. We also propose the use of small-scale fire experiments to increase the number of calibration conditions at a reduced cost. Our results show that this is an effective way to increase the performance of the model, even when evaluated on a standard fire room. Finally, the acquired datasets are made publicly available to the community (doi: 10.5281/zenodo.5643074).

A Bioinspired Adhesive-Integrated-Agent Strategy for Constructing Robust Gas-Sensing Arrays.

Gas sensors based on organic molecules are attractive for their tailored molecular structures and controllable functions, but weak interfacial adhesion between sensing materials and supporting substrates has severely hampered their practical applications, particularly in harsh environments. Here, inspired by the combined anchoring-recognizing feature of natural olfactory systems, an adhesive-integrated-agent strategy to integrate the adhesive unit (poly(dimethylsiloxane)) with the sensing unit (organoplatinum(II)) into one chemistry entity, creating robust and sensitive nanobelt array gas sensors is demonstrated. Systematic theoretical and experimental studies reveal that incorporating adhesive units significantly enhances the interfacial adhesion of the array sensors and gas-bridged super-exchange electronic couplings of sensing units ensure their efficient gas-sensing performance. The high shear strength of ≈7.05 × 106 N m-2 allows these arrays to resist aggressive ultrasonication, tape peeling, or repeated bending without compromising their sensing performance. This molecular engineering strategy opens a new guideline to develop robust gas sensors.

Detection of COPD and Lung Cancer with electronic nose using ensemble learning methods

Background and aimsThe chemical gas sensor array based electronic-nose (e-nose) devices with machine learning algorithms can detect and differentiate expelled breath samples of patients with various respiratory ailments and controls. It is by the recognition of levels and variations of volatile organic compounds (VOC) in the exhaled air. Here, we aimed to differentiate chronic obstructive pulmonary disease (COPD) and lung cancer from controls. Materials and methodsThis work presents the details of the developed e-nose system, selection of the study subjects, exhaled breath sampling method and detection, and the data analysis algorithms. The developed device is tested in 199 participants including 93 controls, 55 COPD patients, and 51 lung cancer patients. The main advantage of the device is robustness and portability and cost-effectiveness. ResultsIn the training phase and model validation phase, the ensemble learning method XGBoost outperformed the other two models. In the prediction of lung cancer, XGBoost method attained a classification accuracy of 79.31%. In COPD prediction also the same method had given the better results with 76.67% accuracy. ConclusionThe e-nose system developed with TGS gas sensors was portable, low cost, and gave a rapid response. It has been demonstrated that the VOC profiles of patients with pulmonary diseases and healthy controls are different and hence the e-nose system can be used as a potential diagnostic device for patients with lung diseases.

Gas sensor array to classify the chicken meat with E. coli contaminant by using random forest and support vector machine

Microbes such as Escherichia coli (E. coli) can easily contaminate raw chicken meat in clean conditions, causing decay and unpleasant scents. This study aims to characterize gas patterns by comparing fresh chicken meat and E. coli bacteria contaminated chicken meat based on shelf life using a Gas Sensor Array (GSA) system (MQ2, MQ3, MQ7, MQ8, MQ135, and MQ136) on electronic nose. The findings revealed GSA capability to detect a variety of typical gas patterns formed by the samples. This gas detection property is indicated by the appearance of the variance in the sensors output voltage pattern for each sample variation. The data for fresh and contaminated samples were classified by the random forest (RF) classifier with 99.25% and 98.42% precision, respectively. Furthermore, the support vector machine (SVM) classifier correctly identified the fresh and contaminated samples with 98.61% and 86.66% accuracy, respectively. This finding offers insight for GSA capability in classifying chicken meat contaminated with E. coli using an RF and SVM.

Human Biomonitoring of Environmental and Occupational Exposures by GC-MS and Gas Sensor Systems: A Systematic Review.

Environmental chemicals and contaminants coming from multiple external sources enter the human body, determining a potential risk for human health. Human biomonitoring (HBM), measuring the concentrations of biomarkers in human specimens, has become an emerging approach for assessing population-wide exposure to hazardous chemicals and health risk through large-scale studies in many countries. However, systematic mapping of HBM studies, including their characteristics, targeted hazardous pollutants, analytical techniques, and sample population (general population and occupationally exposed workers), has not been done so far. We conducted a systematic review of the literature related to airborne hazardous pollutants in biofluids to answer the following questions: Which main chemicals have been included in the literature, which bodily fluids have been used, and what are the main findings? Following PRISMA protocol, we summarized the publications published up to 4 February 2021 of studies based on two methods: gas-chromatography/mass spectrometry (GC/MS) and electronic noses (e-noses). We screened 2606 records and 117 publications were included in the analysis, the most based on GC/MS analysis. The selected HBM studies include measurements of biomarkers in different bodily fluids, such as blood, urine, breast milk, and human semen as well as exhaled air. The papers cover numerous airborne hazardous pollutants that we grouped in chemical classes; a lot of hazardous and noxious compounds, mainly persistent organic pollutants (POPs) and volatile organic compounds (VOCs), have been detected in biological fluids at alarming levels. The scenario that emerged from this survey demonstrates the importance of HBM in human exposure to hazardous pollutants and the need to use it as valid tool in health surveillance. This systematic review represents a starting point for researchers who focus on the world of pollutant biomonitoring in the human body and gives them important insights into how to improve the methods based on GC/MS. Moreover, it makes a first overview of the use of gas sensor array and e-noses in HBM studies.

Recent Trends in Exhaled Breath Diagnosis Using an Artificial Olfactory System.

Artificial olfactory systems are needed in various fields that require real-time monitoring, such as healthcare. This review introduces cases of detection of specific volatile organic compounds (VOCs) in a patient’s exhaled breath and discusses trends in disease diagnosis technology development using artificial olfactory technology that analyzes exhaled human breath. We briefly introduce algorithms that classify patterns of odors (VOC profiles) and describe artificial olfactory systems based on nanosensors. On the basis of recently published research results, we describe the development trend of artificial olfactory systems based on the pattern-recognition gas sensor array technology and the prospects of application of this technology to disease diagnostic devices. Medical technologies that enable early monitoring of health conditions and early diagnosis of diseases are crucial in modern healthcare. By regularly monitoring health status, diseases can be prevented or treated at an early stage, thus increasing the human survival rate and reducing the overall treatment costs. This review introduces several promising technical fields with the aim of developing technologies that can monitor health conditions and diagnose diseases early by analyzing exhaled human breath in real time.

Non-injective gas sensor arrays: identifying undetectable composition changes

Metal-organic frameworks (MOFs) are nanoporous materials with good prospects as recognition elements for gas sensors owing to their adsorptive sensitivity and selectivity. A gravimetric, MOF-based sensor functions by measuring the mass of gas adsorbed in a MOF. Changes in the gas composition are expected to produce detectable changes in the mass of gas adsorbed in the MOF. In practical settings, multiple components of the gas adsorb into the MOF and contribute to the sensor response. As a result, there are typically many distinct gas compositions that produce the same single-sensor response. The response vector of a gas sensor array places multiple constraints on the gas composition. Still, if the number of degrees of freedom in the gas composition is greater than the number of MOFs in the sensor array, the map from gas compositions to response vectors will be non-injective (many-to-one). Here, we outline a mathematical method to determine undetectable changes in gas composition to which non-injective gas sensor arrays are unresponsive. This is important for understanding their limitations and vulnerabilities. We focus on gravimetric, MOF-based gas sensor arrays. Our method relies on a mixed-gas adsorption model in the MOFs comprising the sensor array, which gives the mass of gas adsorbed in each MOF as a function of the gas composition. The singular value decomposition of the Jacobian matrix of the adsorption model uncovers (i) the unresponsive directions and (ii) the responsive directions, ranked by sensitivity, in gas composition space. We illustrate the identification of unresponsive subspaces and ranked responsive directions for gas sensor arrays based on Co-MOF-74 and HKUST-1 aimed at quantitative sensing of CH4/N2/CO2/C2H6 mixtures relevant to natural gas sensing.

Urine LOX-1 and Volatilome as Promising Tools towards the Early Detection of Renal Cancer.

Simple SummaryRenal cell carcinoma (RCC) is often late diagnosed at an advanced stage, worsening the prognosis of the patients. Thus, an early marker is desirable. This paper presents an innovative combined approach useful to identify, for the first time, the presence of LOX-1 protein within the urine of clear cell RCC patients. The LOX-1 protein is related to metabolic disorder-associated carcinogenesis and is shown to be quantitatively correlated to tumor grade and stage. The analysis of volatile compounds released by urine shows the diagnostic potentialities of volatilome and indicates that at least one volatile compound is correlated with both LOX-1 and cancer. In this work, we propose the potential use of a noninvasive approach that enables an early, routine ccRCC diagnosis and leads to a better management of the patients.Renal cell carcinoma (RCC) represents around 3% of all cancers, within which clear cell RCC (ccRCC) are the most common type (70–75%). The RCC disease regularly progresses asymptomatically and upon presentation is recurrently metastatic, therefore, an early method of detection is necessary. The identification of one or more specific biomarkers measurable in biofluids (i.e., urine) by combined approaches could surely be appropriate for this kind of cancer, especially due to easy obtainability by noninvasive method. OLR1 is a metabolic gene that encodes for the Lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1), implicated in inflammation, atherosclerosis, ROS, and metabolic disorder-associated carcinogenesis. Specifically, LOX-1 is clearly involved in tumor insurgence and progression of different human cancers. This work reports for the first time the presence of LOX-1 protein in ccRCC urine and its peculiar distribution in tumoral tissues. The urine samples headspace has also been analyzed for the presence of the volatile compounds (VOCs) by SPME-GC/MS and gas sensor array. In particular, it was found by GC/MS analysis that 2-Cyclohexen-1-one,3-methyl-6-(1-methylethyl)- correlates with LOX-1 concentration in urine. The combined approach of VOCs analysis and protein quantification could lead to promising results in terms of diagnostic and prognostic potential for ccRCC tumors.

Gas Sensor Array Using a Hybrid Structure Based on Zeolite and Oxide Semiconductors for Multiple Bio-Gas Detection.

Semiconductor-type gas sensors, composed of metal-oxide semiconductors and porous zeolite materials, are attractive devices for bio-gas detection, particularly when used as bio-gas sensors such as electronic nose application. Previous studies have shown such detection can be obtained with a separate gas concentrator and a sensor device using zeolites and oxide semiconductors of WO3 nanoparticles. By applying the gas concentrator, porous molecular structures alter both the gas sensitivity and the selectivity, and even can be used to define the sensor characteristics. Based on such a gas sensor design, we investigated the properties of an array of three sensors made of a layer of WO3 nanoparticles coated with zeolites with different interactions between gas molecule adsorption and desorption. The array was tested with four volatile organic compounds, each measured at different concentrations. The results confirm that the features of individual zeolites combined with the hybrid gas sensor behavior, along with the differences among the sensors, are sufficient for enabling the discrimination of volatile compounds when disregarding their concentration.

Optimization of gas sensors measurements by dynamic headspace analysis supported by simultaneous direct injection mass spectrometry

Dynamic headspace extraction is frequently used in gas sensors measurements. The procedure may introduce artefacts but its influence in sensor signals interpretation is rarely considered. In this paper, taking advantage of the on-line combination of a quartz microbalance gas sensor array with a proton transfer reaction mass spectrometer, we have been able to track the evolution of the concentration of volatile compounds along 75 s of extraction of the headspace of differently treated tomato pastes.Proton transfer reaction mass spectrometer signals show that VOCs are characterized by a large diversity of the evolution of the concentration. VOCs kinetics has been described by an electric equivalent circuit model.On the other hand, sensor signals continuously grow approaching a steady value. The contrasting behaviour between sensors signals and the concentration of most of VOCs is explained considering that water is the dominant component in the tomato paste sample and that water is one of those compounds whose concentration in the sensor cell steadily grows. Analysis of variance show that sensors signals achieve the largest separation between classes when the concentration of VOCs in the sensor cell reached its peak. Thus, although the sensor signals continue to rise their information content decays. This finding suggests that measurement protocols need to be adjusted according to the properties of the sample and that the actual measurement times could be much shorter than those predicted from the behaviour of sensor signal.

Quantitative detection of formaldehyde and ammonia using a yttrium-doped ZnO sensor array combined with a back-propagation neural network model

This paper describes the fabrication of a yttrium-doped ZnO gas sensor array using a simple hydrothermal reaction process. The sensor array was fabricated by doping different concentrations of yttrium ions, obtaining flower-like ZnO nanosheets. Toxic gases such as formaldehyde and ammonia were detected quantitatively. The morphological structure and crystallinity of ZnO nanomaterials were investigated through SEM and XRD analysis. The optimal response temperature, sensitivity, response/recovery time, and stability of the sensor array were investigated. The sensor array and a back-propagation neural network (BP-NN) algorithm were used to predict the composition of gas mixtures containing formaldehyde and ammonia. The prediction error was <0.8 ppm. The gas-sensing performance for ternary mixtures was also evaluated. The BP-NN model could predict the output concentration value. Therefore, the sensor array (pure ZnO and Y/ZnO sensors) combined with the BP-NN model developed in this study showed promising results for the detection and composition prediction of toxic gas mixtures in the environment.

Artificial Intelligence Based Breath Analysis System for the Diagnosis of lung cancer

Breath analysis has become a promising tool for the detection of pulmonary diseases in recent years. This paper describes the fabrication of an artificial intelligence (AI) based e-nose system for discriminating lung cancer patients from healthy controls. Breath volatile organic compounds (VOC) can be easily analyzed using an electronic nose (e-nose) made of various gas sensor arrays. Metal oxide semiconductor (MOS)-based e-noses are getting popular in the VOC analysis of exhaled breath from humans. Here we have developed the e-nose system with a sensor array system of five MOS gas sensors and incorporated the controller and machine learning algorithms to process the data. The sensor array was designed with MOS sensors developed by Figaro USA and the data acquisition was carried out with the help of the Arduino Uno developer board. 40 healthy control samples and 24 lung cancer patient samples were analyzed using the developed e-nose system. The data analysis was done by two supervised classification algorithms random forest and logistic regression. Among this, random forest with 5-fold cross-validation gave better results with 85.38 % classification accuracy and 0.87 of AUC. This system can be further extended to the diagnosis of various other pulmonary diseases.

Emission source tracing based on bionic algorithm mobile sensors with artificial olfactory system

AbstractThe leakage of hazardous chemicals and toxic volatile substances in the atmosphere may cause serious consequences such as explosion and poisoning. To identify the unknown leakage locations and gas compositions, a mobile robot system to trace the leak source in the outdoor was investigated. First, two bionic searching algorithms, Zigzag and Silkworm algorithms, were tested with outdoor experiments for locating the leak source. The results showed that the locating errors of these two algorithms were within 0.5 m in 10 by 20 m search space, but the failing ratio of Zigzag and Silkworm algorithm was still high (about 40–50%). Therefore, an improved tracing algorithm combining the Silkworm and Zigzag algorithm, called as zigzag–Silkworm algorithm, was proposed. Compared with Silkworm and Zigzag algorithms, zigzag–Silkworm algorithm had a higher success ratio of 80% in outdoor source tracing tests, and the searching efficiency was enhanced, the efficiency parameter L: L0 has improved from 2.58 for Silkworm and 2.66 for Zigzag to 2.17 for zigzag–Silkworm. Then, in order to identify the composition of the leaked gases during the source tracing, an artificial olfaction system (AOS) based on the gas sensor array and support vector machine was set on the mobile robot. The test results in the source tracing experiments with ammonia and ethanol emissions indicated that the recognition accuracy of emission gases reached to 99% with AOS equipped on the robot. Therefore, the mobile robot system equipped with the zigzag–Silkworm algorithm and the AOS is feasible to trace the leakage source and identify the emission composition in the outdoor leakage event with good performance in efficiency and accuracy although some underlying problems still need to be addressed in future work.

Recognition of binary mixture of NO2 and NO gases using a chemiresistive sensors array combined with principal component analysis

Novel room-temperature type chemiresistive gas sensor arrays consisted of two, three or four metal oxide semiconductor (MOS) gas sensors that were constructed using iron oxide/multi-walled carbon nanotubes/tungsten oxide (Fe2O3/MWCNTs/WO3), surface Au-, Pd- and Ag- modified films as sensing materials. Principal component analysis (PCA) coupled with various MOS chemiresistive gas sensor arrays were used to analyze the obtained data under testing conditions of binary mixture gases containing various composition of NO and NO2 at room-temperature. The careful pattern recognition of sensing data obtained with a chemiresistive gas sensor array consisting of three MOS sensors made it possible to qualitatively differentiate the variation in the composition of gas species in a binary mixture of NO2 and NO gases with related to their concentrations.

Development of Gas Sensor Array for Methane Reforming Process Monitoring.

The article presents a new method of monitoring and assessing the course of the dry methane reforming process with the use of a gas sensor array. Nine commercially available TGS chemical gas sensors were used to construct the array (seven metal oxide sensors and two electrochemical ones). Principal Component Regression (PCR) was used as a calibration method. The developed PCR models were used to determine the quantitative parameters of the methane reforming process: Inlet Molar Ratio (IMR) in the range 0.6–1.5, Outlet Molar Ratio (OMR) in the range 0.6–1.0, and Methane Conversion Level (MCL) in the range 80–95%. The tests were performed on model gas mixtures. The mean error in determining the IMR is 0.096 for the range of molar ratios 0.6–1.5. However, in the case of the process range (0.9–1.1), this error is 0.065, which is about 6.5% of the measured value. For the OMR, an average error of 0.008 was obtained (which gives about 0.8% of the measured value), while for the MCL, the average error was 0.8%. Obtained results are very promising. They show that the use of an array of non-selective chemical sensors together with an appropriately selected mathematical model can be used in the monitoring of commonly used industrial processes.

Early-Stage Gas Identification Using Convolutional Long Short-Term Neural Network with Sensor Array Time Series Data.

Gas identification/classification through pattern recognition techniques based on gas sensor arrays often requires the equilibrium responses or the full traces of time-series data of the sensor array. Leveraging upon the diverse gas sensing kinetics behaviors measured via the sensor array, a computational intelligence- based meta-model is proposed to automatically conduct the feature extraction and subsequent gas identification using time-series data during the transitional phase before reaching equilibrium. The time-series data contains implicit temporal dependency/correlation that is worth being characterized to enhance the gas identification performance and reliability. In this context, a tailored approach so-called convolutional long short-term memory (CLSTM) neural network is developed to perform the identification task incorporating temporal characteristics within time-series data. This novel approach shows the enhanced accuracy and robustness as compared to the baseline models, i.e., multilayer perceptron (MLP) and support vector machine (SVM) through the comprehensive statistical examination. Specifically, the classification accuracy of CLSTM reaches as high as 96%, regardless of the operating condition specified. More importantly, the excellent gas identification performance of CLSTM at early stages of gas exposure indicates its practical significance in future real-time applications. The promise of the proposed method has been clearly illustrated through both the internal and external validations in the systematic case investigation.