Gas Identification Research Articles

PSCFormer: A lightweight hybrid network for gas identification in electronic nose system

Based on their powerful feature extraction capability, a convolutional neural network (CNN) has been gradually applied to gas identification in the electronic nose (e-nose) system. The responses of different intensities in the e-nose system are significantly correlated, and CNN extracts the local gas features by convolution while ignoring their global correlation. Transformer combines different responses and obtains the correlation between global features by self-attention. This paper proposes a lightweight hybrid network called Peak Search-based Convolutional Transformers (PSCFormer). First, combining the data characteristics of gas information, the Local Peak Search and Feature Fusion (LPSF) module is proposed to focus on the key gas features. Second, Transformer Encoder (TE) is proposed to obtain the global correlation between global features, and the parallel Convolution Encoder (CE) is proposed to capture the local dependence. Finally, a reasonable feature complementation mechanism is presented, and the preference of TE is alleviated for the slow-down response while solving the receptive field limitation of CE. This paper has evaluated three different datasets to validate the effectiveness of PSCFormer, all of which show stable and excellent performance with a good tradeoff between efficiency and complexity. The results prove that PSCFormer is an efficient and lightweight gas identification network, which provides a method to promote the engineering application of the e-nose system.

Putting the Common Security and Defense Policy in the Eastern Mediterranean under Scrutiny: Türkiye’s Conflicting Role

The end of the Cold War and the emergence of a new global security and economic environment (Defence budgets’ decline, imminent threats from the Gulf War and NATO’s quest for a new identity), provided a window of opportunity for establishing and designing the EU’s security institution. Thus, what was initially considered farfetched after the 1992 Maastricht Treaty, entered a new phase, which included perception transformation, to be gradually incorporated into daily discourse. Hence, the 2003 “European Security Strategy”, the 2010 “Internal Security Strategy”, and later the 2016 “Global Strategy” reflect Member States’ perception of transformation and their belief that the EU needs a Common Security and Defence Strategy approach. Given these documents that indicate determination for notable integration and institutional restructuring, this analysis delves into the impact of Europeanisation concerning the implementation of institutional reforms and the CSDP. More specifically, this analysis probes the constraints of Europeanisation concerning the EU’s real operational dynamics, especially in light of the Eastern Mediterranean and Aegean Sea security challenges. These challenges are examined in relation to Türkiye’s revisionist public diplomacy, public policies, and the militarization of its foreign policy (including ongoing negotiations centered on migratory flows, assertions, and constructed narratives over Greek and Cypriot islands and seas). The situation has become more pronounced following the identification of gas and oil reserves in the area in 2010. This analysis has a two-fold focus: Firstly, to investigate Europeanisation impact on the operational implementation of CSDP (Common Security and Defence Policy) in time of crisis, and secondly, to evaluate the behavior of Türkiye within the Europeanisation framework, as proposed by Radaelli and Violakis.

Plasma-based identification of gases in a laser-induced cavitation bubble

This study presents a general methodology and an experimental approach to identify the gas components within laser-induced cavitation bubbles. A needle electrode inside the cavitation bubble, which introduces low electric energy into the bubble, produces a homogenous plasma discharge inside the vapor cavity. The primary bubble dynamics remain identical while the rebound bubble becomes about twice as large when a discharge is applied. The effect of non-condensable gases and the electric charge on bubble dynamics is explored theoretically, and the role of the electric charge is found to be significant. Optical emission spectroscopy reveals the evolution of emission lines from gases inside bubbles. H lines and OH lines are persistently observed in all cases, providing a dominant presence of water vapor. The results also confirm that the gases, which are initially present in the water rather than transported from the water, contribute to the optical emission characteristics with different dissolved gases.

Sound Identification Method for Gas and Coal Dust Explosions Based on MLP.

To solve the problems of backward gas and coal dust explosion alarm technology and single monitoring means in coal mines, and to improve the accuracy of gas and coal dust explosion identification in coal mines, a sound identification method for gas and coal dust explosions based on MLP in coal mines is proposed, and the distributions of the mean value of the short-time energy, zero crossing rate, spectral centroid, spectral spread, roll-off, 16-dimensional time-frequency features, MFCC, GFCC, short-time Fourier coefficients of gas explosion sound, coal dust sound, and other underground sounds were analyzed. In order to select the most suitable feature vector to characterize the sound signal, the best feature extraction model of the Relief algorithm was established, and the cross-entropy distribution of the MLP model trained with the different numbers of feature values was analyzed. In order to further optimize the feature value selection, the recognition results of the recognition models trained with the different numbers of sound feature values were compared, and the first 35-dimensional feature values were finally determined as the feature vector to characterize the sound signal. The feature vectors are input into the MLP to establish the sound recognition model of coal mine gas and coal dust explosion. An analysis of the feature extraction, optimal feature extraction, model training, and time consumption for model recognition during the model establishment process shows that the proposed algorithm has high computational efficiency and meets the requirement of the real-time coal mine safety monitoring and alarm system. From the results of recognition experiments, the sound recognition algorithm can distinguish each kind of sound involved in the experiments more accurately. The average recognition rate, recall rate, and accuracy rate of the model can reach 95%, 95%, and 95.8%, respectively, which is obviously better than the comparison algorithm and can meet the requirements of coal mine gas and coal dust explosion sensing and alarming.

Review of Surface Acoustic Wave Sensors for the Detectionand Identification of Toxic Environmental Gases/Vapours

Detection and identification of toxic environmental gases have assumed paramount importance precisely in the defense, industrial and civilian security sector. Numerous methods have been developed for the sensing of toxic gases in the environment ever since surface acoustic wave (SAW) technology came into existence. Such SAW sensors called electronic nose (E-Nose) sensor use the frequency response of a delay line/resonator. SAW device is focused and given importance. The selective coating between input and output interdigital transducers (IDTs) in the SAW device is responsible for corresponding changes in operating frequency of the device for a specific gas/vapour absorbed from the environment. A suitable combination of well-designed SAW delay lines with selective coatings not only help to improve sensor sensitivity and selectivity but also leads to the minimization of false frequency alarms in the E-Nose sensor. This article presents a comprehensive review of design, development, simulation and modelling of a SAW sensor for potential sensing of toxic environmental gases.

An optical nose based on array of metal-doped carbon dots for identification of hazardous amines and assessing meat freshness

There is an obvious pressing demand for the sensitive and rapid, identification of toxic gases as environmental concerns. In this study, a facile approach for the detection and identification of amine solvents is reported. A colorimetric sensor array composed of different ninhydrin-mediated transition metal-doped carbon dots (TMCDs), deposited on a paper substrate, could sniff the odor of 29 different amines and accomplish their accurate identification. This analytical device generates a unique response pattern for each amine. Modeling the response data of the sensor array by linear discriminate analysis yields a discrimination accuracy of 94 % for the amines. Also, this method allows rapid quantification of ethylamine and isopropylamine in the concentration ranges of 0.008–1600 ppm and 0.05–828 ppm with detection limits of 1.4 ppb and 39 ppb, respectively, which are well below the immediately dangerous to life or health (IDLH) limit concentration for these amines. Interestingly, this sensor array provides discrimination between the isomers of some amines. Moreover, binary and ternary mixtures of amine solvents could be well recognized with this sensor. The application of the sensor array was tested for the assessment of the quality and freshness of some kinds of meat.

Investigation of the discharge coefficient in the laminar boundary layer regime of critical flow Venturi nozzles calibrated with different gases including hydrogen

This study represents the first step towards the introduction of critical flow Venturi nozzles into the traceability scheme for gaseous hydrogen and addresses the characterisation of the hydrogen discharge coefficient in the laminar boundary layer regime and the identification of a potential alternative gas for nozzle calibration. The experimental study presented was performed for two nozzles with a throat diameter of 0.175 mm and 0.436 mm. The tests were performed with six different gases, including hydrogen, covering a Reynolds number range from 1.5·103 to 6·104. The experimental results show that the discharge coefficient depends not only on the Reynolds number but also on the isentropic exponent of the gas. Based on the experimental data, theoretical models taking into account the isentropic exponent and the Prandtl number of the gas were evaluated. The study indicates possibilities to use inert gases instead of hydrogen in calibration processes.

A Concentration Prediction and Gas Classification Model Based on LSTM-Attention Multi-task Learning Framework Network

The electronic nose (E-nose), a bionic olfactory system, has been widely used in gas identification and concentration prediction. However, these tasks are usually based on separate systems, leading to high detection costs. To address this issue, a multi-task learning network model based on LSTM-Attention (MTL-LSTMA) as a skeleton has been proposed to simultaneously train both species recognition and concentration prediction tasks. The introduction of an attention mechanism greatly reduces interference in gas feature information, improving the electronic nose pattern recognition algorithm’s efficiency. The MTL-LSTMA model was tested through five sets of comparison experiments, showing the best performance in both concentration prediction and species identification. The proposed model has broad application prospects and may revolutionize gas detection technology.

Harmful gases detection using artificial neural networks of the environment

This work describes a small, low-cost electronic nose device which can detect harmful substances that can harm human health, such as flammable gas like acetone, ethanol, butane as well as methane, among others. An artificial olfactory instrument consists of a set of metal oxide semiconductor sensors as well as a computer-based communications channel for signal gathering, proceeding, and presentation. We used three sensors instead of six, and the results were plotted as a variance, score as well as loading plot with crossvalidation. For gas identification, we use artificial neural network (ANN) and compare them to parallel factor analysis. Electronic nose (e-nose) has provided numerous benefits in a variety of logical study disciplines. Our goal is to create a sensor exhibit framework that can discriminate the most exceedingly contaminated gases while also being extremely responsive, precise, and less power consuming. Thus, for gas detection, we employ an ANN as well as make a comparison of results with parallel factor analysis (PARAFAC).

Characterization of a platform for the gas transport, collection, and identification of fission products in the high-intensity laser environment

Recent progress in the production of laser accelerated high flux proton beams opens new possibilities for the study of fission in unique environments. To this end, we are currently pursuing the development of a platform for the swift gas transport, collection, and identification of fission products. Fission is induced in targets fixed inside a sealed chamber through which a carrier gas flows, transporting fission products to a carbon filter for collection and online spectroscopy. This has been recently demonstrated using target normal sheath accelerated protons at the PHELIX laser facility. There were large discrepancies between measured rates and those expected based on established fission yields and measured beam parameters. These discrepancies prompted a large number of tests at the Idaho Accelerator Center (IAC), where fission in uranium is induced by bremsstrahlung photons generated by 21 MeV electrons from a linear electron accelerator. The results are compared to a series of models that account for the slowing of energetic fission fragments in the carrier gas and the fluid dynamics of the gas flow that transports fission products to the collection filter. Characterization of the apparatus reveals a few mechanisms that together account for a portion of the previously observed discrepancies at the PHELIX laser facility. However, additional research is necessary before high-accuracy experiments can be performed with the apparatus. Work performed under the auspices of the U.S. Department of Energy by LLNL under contract DE-AC52-07NA27344.

Gas Turbine Vibration Detection and Identification based on Dynamic Artificial Neural Networks

Vibration control in rotating machinery is a major challenge in oil and gas facilities that use these machines. In gas turbines, the instability phenomenon is generated by the rotor, and measurements must be made in the axial plane of the turbine. Minor defects can lead to significant vibration amplifications, making it imperative to detect these defects early. The goal of this study is to develop a diagnostic strategy to monitor faults affecting a turbine system using a supervision approach based on artificial neural networks. This strategy allows for early detection of faults, which allows for efficient management of vibration-induced failures, as well as economic gain, by recovering the transported gas used in these machines. By describing the vibration-related parameters and representing the state of the vibratory motion, the proposed approach provides a powerful tool for vibration control in rotating machines.

Classification and identification of mixed gases based on the combination of semiconductor sensor array with SSA-BP neural network

In this paper, pure , , ZnO and Cu/ZnO gas sensitive materials were synthesized by a simple hydrothermal reaction and used to prepare a gas sensor array. The morphological structure and composition of the synthesized materials were characterized using SEM and XRD, respectively. The sensor array was combined with the back propagation neural network algorithm optimized by the sparrow search algorithm (SSA-BPNN) and applied to effectively identify the types of mixed toxic gases in the room, including formaldehyde, ammonia and xylene. The combination of sensor array with optimized neural network algorithms achieved a good classification result for gas mixture and the classification accuracy can reach 93.45% for different classes of mixtures composed of three gases (formaldehyde, ammonia, and xylene). Therefore, the sensor array combined with the SSA-BP algorithm in this study has done a good work in the qualitative identification of ternary gas mixtures and has some application potential.

A Flexible, Low-Cost, Miniature Methane Sensor Based on PEDOT:PSS

The development of methane sensors with high sensitivity is vital for the safe application of methane. In this work, we developed a new flexible, low cost, miniature methane sensor based on PEDOT:PSS coated screen-printed electrochemical electrodes as a resistive sensor. PEDOT:PSS is used as a sensing electrode, and the presence of methane causes a change in the resistance of the PEDOT:PSS layer. The sensor demonstrates a fast detection time of 10 s, excellent sensitivity of 0.5 pA/ppm, and a wide linear detection range of 600 to 0.8x10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> ppm. The sensor shows excellent stability and reliability. The detection error is within 2.8% for a single sample, 7.1% for different samples from the same batch, 16% for different batches and different samples, 3.4% for storage for 4 days, and 14% for temperature increase to 70°C. The sensor is simple to prepare, low cost and highly flexible. We anticipate that this work could open up exciting opportunities for the development and application of methane sensors, such as natural gas exploration and identification of methane leaks in mining and kitchens.

Deep-learning-based gas identification by time-variant illumination of a single micro-LED-embedded gas sensor

Electronic nose (e-nose) technology for selectively identifying a target gas through chemoresistive sensors has gained much attention for various applications, such as smart factory and personal health monitoring. To overcome the cross-reactivity problem of chemoresistive sensors to various gas species, herein, we propose a novel sensing strategy based on a single micro-LED (μLED)-embedded photoactivated (μLP) gas sensor, utilizing the time-variant illumination for identifying the species and concentrations of various target gases. A fast-changing pseudorandom voltage input is applied to the μLED to generate forced transient sensor responses. A deep neural network is employed to analyze the obtained complex transient signals for gas detection and concentration estimation. The proposed sensor system achieves high classification (~96.99%) and quantification (mean absolute percentage error ~ 31.99%) accuracies for various toxic gases (methanol, ethanol, acetone, and nitrogen dioxide) with a single gas sensor consuming 0.53 mW. The proposed method may significantly improve the efficiency of e-nose technology in terms of cost, space, and power consumption.

A Self-Temperature-Modulated Quadrilateral Gas Sensor for Gas Identification

This article proposes a SnO2-based low-power quadrilateral gas sensor, which can realize different responses by self-temperature modulation using a single-layer electrode. Four detection channels are located on four edges and are integrated with a quadrilateral microheater to constitute the completed sensor. Under the premise of ensuring good electrical insulation, the heater and the detection electrodes are designed in parallel on the same layer, eliminating the need for the isolation layer and subsequent dry etching steps to expose the pad. This scheme can greatly simplify the fabrication process and save manufacturing costs. On top of that, temperature modulation is realized by designing the gradational width of the microheater metal line on the four edges to achieve different responses to gases. Six types of food-related gases (trimethylamine, ethanol, hydrogen sulfide, ammonia, ethyl acetate, and formaldehyde) were tested, and the four channels of the sensor show different performances to these stimuli, which can be utilized for gas identification. Moreover, six food species were also tested, showing that the sensor can efficiently differentiate the fruit, vegetables, and meats.

High performance bimetal decorated PtNix-WO3 sensors and the cross-sensitivity investigation

Gas sensors play an essential bridge in connecting humans and health by helping people monitor the gas quality in their living spaces. In this work, PtNi3-WO3 gas sensors were fabricated and demonstrated superior selectivity, sensitivity, and stability to formic acid, which was sufficient for indoor air gas monitoring. The response to 100 ppm formic acid was as high as 591, with a response time of only 3 s, and detection range down to the ppb level. The bimetal PtNi3 decorated sensors showed a 30 times improvement in formic acid response compared to pure WO3 sensors. The cross-sensitivity was investigated, and we observed that the response initially decreased and then increased in a mixed gas with a constant concentration of carboxylic acid and increasing gradually concentration of formaldehyde. The adsorption and desorption mechanisms were studied, which implied that the decomposition and adsorption behavior of the gas could change and produce different intermediate products under different mixing environments. The results of the energy surface testing and density functional theory supported the proposed sensing mechanism. A machine-learning framework was developed for gas identification and concentration prediction, which achieved ultra-high identification accuracy. In addition, Bluetooth-based portable gas sensors were demonstrated, showing promising practical applications.

Crosslinked WO3 nanonet for rapid detection of sulfur mustard gas simulant: Mechanism insights and sensing application

Effective detection of mustard gas is significant for human health and public security on account of its high toxicity and volatility. However, the sluggish response time of current sensors hinders their rapid identification of gases in actual hazardous environments. Herein, crosslinked WO3 nanonet was successfully synthesized on ceramic tubes by a facile in situ solvothermal method. The unique structure with abundant pores could not only facilitate charge transfer but also provide large amounts adsorption sites for gas sensing. The resultant WO3-4:1 sensor exhibited ultra-fast response time (1 s) and high response (58) toward 50 ppm 2-chloroethyl ethyl sulfide (2-CEES). Moreover, a new model of sensing mechanism for the 2-CEES sensing process was proposed through a series of characterization and testing technologies. This work offers guidance for developing other high-performance mustard gas sensors in practical applications.

Machine Learning-Assisted Sensor Based on CsPbBr3@ZnO Nanocrystals for Identifying Methanol in Mixed Environments.

Methanol is a respiratory biomarker for pulmonary diseases, including COVID-19, and is a common chemical that may harm people if they are accidentally exposed to it. It is significant to effectively identify methanol in complex environments, yet few sensors can do so. In this work, the strategy of coating perovskites with metal oxides is proposed to synthesize core-shell CsPbBr3@ZnO nanocrystals. The CsPbBr3@ZnO sensor displays a response/recovery time of 3.27/3.11 s to 10 ppm methanol at room temperature, with a detection limit of 1 ppm. Using machine learning algorithms, the sensor can effectively identify methanol from an unknown gas mixture with 94% accuracy. Meanwhile, density functional theory is used to reveal the formation process of the core-shell structure and the target gas identification mechanism. The strong adsorption between CsPbBr3 and the ligand zinc acetylacetonate lays the foundation for the formation of the core-shell structure. The crystal structure, density of states, and band structure were influenced by different gases, which results in different response/recovery behaviors and makes it possible to identify methanol from mixed environments. Furthermore, due to the formation of type II band alignment, the gas response performance of the sensor is further improved under UV light irradiation.

Rapid Identification Method for CH4/CO/CH4-CO Gas Mixtures Based on Electronic Nose.

The inherent cross-sensitivity of semiconductor gas sensors makes them extremely challenging to accurately detect mixed gases. In order to solve this problem, this paper designed an electronic nose (E-nose) with seven gas sensors and proposed a rapid method for identifying CH4, CO, and their mixtures. Most reported methods for E-nose were based on analyzing the entire response process and employing complex algorithms, such as neural network, which result in long time-consuming processes for gas detection and identification. To overcome these shortcomings, this paper firstly proposes a way to shorten the gas detection time by analyzing only the start stage of the E-nose response instead of the entire response process. Subsequently, two polynomial fitting methods for extracting gas features are designed according to the characteristics of the E-nose response curves. Finally, in order to shorten the time consumption of calculation and reduce the complexity of the identification model, linear discriminant analysis (LDA) is introduced to reduce the dimensionality of the extracted feature datasets, and an XGBoost-based gas identification model is trained using the LDA optimized feature datasets. The experimental results show that the proposed method can shorten the gas detection time, obtain sufficient gas features, and achieve nearly 100% identification accuracy for CH4, CO, and their mixed gases.

Two-parameter Optical Sensing Based on Multilayer Parity-time-symmetric Structure

A two-parameter sensor that can detect the variation of temperature and refractive index is realized in a multilayer dielectric structure obeying parity-time (PT) symmetry. The sensor can operate near exceptional points (EPs), which have been shown to provide dramatic variations of their eigenvalues in response to small parameter changes. The optical sensing behavior is theoretically investigated based on the transfer matrix method. The results show that the sensor can work within the surrounding temperature (tp) ranging from 0 to 30℃, and the refractive index (ng) of incident medium ranging from 1.0 to 1.4. The detectable variation △ng of the sensor can reach 0.02. The sensitivity of ng and tp can reach 372496.53 RIU-1 and 249.18℃-1, respectively. Our structures show great promise in temperature monitoring in cold environment and identification of chemical gases or liquids.