한국센서학회 학술지영문홈페이지
[ Article ]
JOURNAL OF SENSOR SCIENCE AND TECHNOLOGY - Vol. 32, No. 2, pp.75-81
ISSN: 1225-5475 (Print) 2093-7563 (Online)
Print publication date 31 Mar 2023
Received 21 Mar 2023 Revised 29 Mar 2023 Accepted 31 Mar 2023
DOI: https://doi.org/10.46670/JSST.2023.32.2.75

Gas Distribution Mapping and Source Localization: A Mini-Review

Taehwan Kim1 ; Inkyu Park1,
1Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, South Korea

Correspondence to: + inkyu@kaist.ac.kr

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License(https://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

The significance of gas sensors has been emphasized in various industries and applications, owing to the growing significance of environmental, social, and governance (ESG) management in corporate operations. In particular, the monitoring of hazardous gas leakages and detection of fugitive emissions have recently garnered significant attention across several industrial sectors. As industrial workplaces evolve to ensure the safety of their working environments and reduce greenhouse gas emissions, the demand for high-performance gas sensors in industrial sectors dealing with toxic substances is on the rise. However, conventional gas-sensing systems have limitations in monitoring fugitive gas leakages at both critical and subcritical concentrations in complex environments. To overcome these difficulties, recent studies in the field of gas sensors have employed techniques such as mobile robotic olfaction, remote optical sensing, chemical grid sensing, and remote acoustic sensing. This review highlights the significant progress made in various technologies that have enabled accurate and real-time mapping of gas distribution and localization of hazardous gas sources. These recent advancements in gas-sensing technology have shed light on the future role of gas-detection systems in industrial safety.

Keywords:

Hazardous gas leakages, Fugitive emissions, Greenhouse gas, Mobile robotic Olfaction, Remote optical sensing, Chemical grid, Remote acoustic, Gas distribution mapping, Source localization

1. INTRODUCTION

Gas detection instruments are used in a wide range of industrial applications. The growing demand for smart Internet of Things (IoT) device solutions, coupled with increasing awareness of personal health, industrial safety, and environmental issues, has led to the expansion of the applicability of gas sensors to unconventional areas. These applications include, but are not limited to, monitoring indoor air quality, measuring fugitive greenhouse gas pollutant emissions, analyzing food freshness, diagnosing health conditions, and optimizing agricultural environments.

Although conventional gas-sensing systems can effectively operate in controlled environments in which the target gas flows under steady-state conditions or specific samples are collected for analysis in laboratory settings, their suitability for some newly proposed applications may be limited. The chemical complexity of gas sensing can make operation under harsh and arbitrary conditions crucial for the performance of typical gas sensors. Moreover, individual gas sensors have limited spatial coverage, which can present difficulties in accurately monitoring the temporal progression of gas dispersion and mapping the gas leakage distribution over a large area with limited experimental control, particularly in industrial plants [1].

Nonetheless, the need for a ubiquitous gas monitoring technology has become increasingly important. Monitoring hazardous gas leakages and detecting fugitive emissions have recently garnered attention across several industrial sectors. As environmental, social, and governance (ESG) management has become a significant factor in enterprise valuation, there has been an increased emphasis on safety requirements for both personnel and worksites. Moreover, environmental concerns have forced businesses to monitor and evaluate their carbon footprints and greenhouse gas emissions. Thus, the digital transformation of worksites and incorporation of smart-device solutions for gas sensing are essential in many industries.

Many research groups have developed state-of-the-art technologies to overcome these technical limitations and satisfy the imminent demands for gas distribution mapping and source localization. The scope of this investigation ranges from mobile robotic olfaction techniques to remote optical/acoustic sensing methods and chemical grid-sensing systems. This review provides a comprehensive overview and assessment of the aforementioned gas-monitoring techniques, which are recognized as potential solutions to the current limitations in this field of research. Although various challenges still exist in the implementation of these solutions in industrial settings, recent advancements in gas leakage sensing technology have shed light on the future role of gas detection systems in ensuring industrial safety.


2. TYPES OF GAS DETECTION MECHANISMS

Gas leakage detection methods can be classified into three categories based on their transduction mechanisms: chemical, optical, and acoustic methods. Chemical detection methods directly measure the change in the output signal when a leaked target gas reacts with the active functional layers of the sensor. Optical and acoustic detection methods are based on indirect reactive mechanisms that measure changes in the optical and acoustic properties caused by the presence or movement of target molecules. Finally, this review also discusses various state-of-the-art mobile-platform mechanisms that integrate existing detection methods to achieve more functional applications.

2.1 Chemical type

Gaseous vapors typically consist of volatile compounds. Therefore, gas sensors are designed to have active functional layers that readily react with the target gas analytes. There are various types of chemically reactive sensors, including chemiresistive, electrochemical, metal-oxide semiconductor, and pellistor-type gas sensors. Although, the details of each type of chemical gas sensor differ, the fundamental principle of transducing the chemical reaction caused by the target analyte into electrical signal fluctuations is identical. Moreover, chemically reactive sensors share a common limitation in their effective spatial coverage because these sensors can only measure target gas concentrations by direct contact with gas molecules. As shown in Figure 1, metal-oxide-semiconductor-based gas sensors measure the changes in electrical resistance caused by oxidation and reduction reactions when the target gases bind to the functional layer.

Fig. 1.

Gas sensing mechanism for n-type and p-type metal oxide semiconductor-based gas sensors. Reprinted with permission from Ref [2]. Copyrights (2017) Royal Society of Chemistry.

Fig. 2.

CAD illustration and actual photo of gas test room with metal oxide sensors (in green circles), processing nodes (in orange rectangles), HVAC outlet (in a pick rectangle), gas source, window, and radiators. Reprinted with permission from Ref [3]. Copyrights (2020) Elsevier.

Therefore, it is challenging for chemical gas sensors to localize the source of gas leakage. Unless the sensor is installed near the leakage site, it is almost impossible to trace the gas dispersion. To resolve this inherent limitation, researchers have developed chemical sensor grids to define and characterize the distribution of target gases in various situations. Chemical sensors have also been implemented to characterize gas dispersions within specific areas of interest by monitoring the responses of each sensor and mapping the data analyzed from adjacent sensors.

The use of multiple chemical sensors in batches to compensate for individual spatial coverage can be inefficient for extremely large-scale applications such as continental pipelines. However, for indoor applications and industrial manufacturing sites, the use of low-cost chemical gas sensors can be more economical and efficient than other methods that require expensive equipment and direct inspection.

To activate multiple chemical sensors in batches, the low power consumption of individual sensors is a key feature that must be accompanied by other performance characteristics. Recently, Kang et al. reported a novel suspended microheater platform-based metal-oxide semiconductor gas sensor that minimized the power consumption of the microheater in the milliwatt range by activating a metal oxide layer for gas sensing [4]. The proposed suspended microheater platform continuously maintained the temperature at 250 oC while consuming electrical power of 11 mW. From the same research group, Cho et al. proposed a monolithic micro light-emitting diode (LED) activated metal oxide gas sensor that functions with microwatt-level power consumption [5]. Also, Cho et al. designed a novel nanoarchitecture of periodic 3D TiO2 to create a high-performance light-activated gas sensor. They exploited the various features of 3D nanostructured materials to report an energy efficient light-activated gas sensor to overcome bottlenecks in conventional light-activated gas sensors [6, 7]. Although there are drawbacks to the current light-activated metal oxide semiconductor gas sensors, these studies present solutions from the power consumption perspective of gas sensors used to cover large areas.

Fig. 3.

(a) Glancing angle deposition (GLAD) of nanocolumnar metal oxide. (b) Microscopic top-view image of a suspended microheater platform-based gas sensor. (c) CAD illustration of a gas sensor array and signal processing. Reprinted with permission from Ref [4]. Copyrights (2020) Elsevier.

2.2 Optical type

Optical sensors are widely appreciated for their versatility in various applications. Various optical sensing mechanisms can be applied for both direct and indirect gas sensing [8-9]. Direct sensors operate by detecting changes in optical or electrical properties upon illumination by radiation sources that emit infrared or ultraviolet light. The gases of interest can absorb light in these regions, which is then detected by the sensor.

One of the extremely high-performance optical gas sensors used in industry is the photoionization detector (PID). PID can detect volatile organic compounds (VOCs) over a wide range of concentrations. The working mechanism of PID sensors is analogous to that of the ionization gauges used for measuring vacuum pressures. A deep ultraviolet lamp ionizes the gas molecules and induces a current between the electrodes, and the corresponding voltage output has a linear relationship with the gas concentration. Therefore, handheld PIDs are used at industrial sites to measure ambient gas concentrations. However, PIDs are expensive and lack selectivity for specific gases.

Fig. 4.

Wind condition controlled CH4 leak imaging (a, b) Actual images of experimental apparatus (c-h) Resulting images obtained for different combinations of wind speed and CH4 flow rate. Reprinted with permission from Ref [8]. Copyrights (2022) Elsevier.

Fig. 5.

Industrial environment test of CH4 release measurements with Tunable Diode Lidar (TDLidar) images at Total’s TADI platform (a) Visual camera image overlaid with the infrared signal level image of CH4 plume. (b, c) Sensor signal level image overlaid with CH4 path length (ppm.m) image. Reprinted with permission from Ref [8]. Copyrights (2022) Elsevier.

Tunable diode laser absorption spectroscopy (TDLAS) is another key gas-imaging technology that enables remote gas detection. Titchener et al. proposed a novel gas imager that combines TDLAS and differential absorption Lidar (DIAL) technology with time-correlated single-photon counting (TCSPC) to remotely monitor methane concentrations [8]. This technology can detect leakages at distances greater than 90 m with a minimum leakage rate of 0.012 g/s. They also conducted a methane release test at the total Transverse Anomaly Detection Infrastructure (TADI) platform in France.

The nondispersive infrared (NDIR) technique is also used in many gas sensors owing to its simplicity. The deep-ultraviolet lamp was substituted with an infrared source that illuminated the gas chamber. As the gas molecules enter the chamber, the absorption spectra vary depending on the type of gas and its concentration. Based on this principle, remote gas detectors can also be used in open spaces. These detectors are based on differential optical absorption spectroscopy (DOAS), which emits a projection of light (ultraviolet, visible, or infrared light) across an open space and back towards a detector unit. DOAS is primarily utilized for measuring urban air quality and monitoring emissions in various industrial sectors. Another effective technique for remote optical gas monitoring in open spaces is the differential absorption light detection and ranging (DIAL) method. This method uses a high-power laser source to measure the concentration of a target gas in a remote space. Despite the high cost of implementing DIAL gas monitoring systems, various facilities, such as gas storage plants, petrochemical industries, and environmental and meteorological monitoring systems, have chosen this method for accurate and reliable data collection.

Indirect optical methods can also be used for detecting gas leakages in pressurized pipelines. Infrared thermography is a typical indirect detection technique that detects potential leakage by measuring temperature changes rather than the gas. This technique can be employed remotely to measure continuous and real-time temperature distributions with effective coverage over a relatively wide range of areas compared to other non-optical methods. Thermal image cameras can be installed at fixed locations, or inspectors can easily operate them without special training. However, high-resolution infrared cameras are expensive, and their lack of sensitivity results in the failure to detect minor leakage points.

Yang et al. determined the location and rate of gas leakage by measuring the local temperature gradient using infrared thermal imaging [9]. They characterized the leakage rate based on thermodynamic analysis and numerical methods, and were able to ensure a maximum relative deviation of only 5.6% in the measured leakage rate.

Fig. 6.

(a) Illustration of the experimental container apparatus with a leakage hole (b) Local temperature field at the wall of the container with the leakage hole. Reprinted with permission from Ref [9]. Copyrights (2022) Elsevier.

2.3 Acoustic type

Acoustic gas sensing is typically used to detect leaks in pipelines that operate under high pressure conditions. [10]. When pressurized gas escapes from such pipelines, elastic waves are generated over a wide range of frequencies. Acoustic emission sensors, such as accelerometers, dynamic pressure transducers, and microphones, can capture the airborne or pipeline-transmitted acoustic signals generated by the leakage. Recent studies have demonstrated that acoustic emission sensors installed in high-pressure gas pipelines can detect rupture points and localize source of rupture.

Acoustic emission sensors are usually attached to the external parts of the pipelines to ensure continuous monitoring. As acoustic emission sensors exhibit high sensitivity over longer distances, the distance between individual sensors can be extended beyond that required for chemical sensors [11]. However, in industrial settings, acoustic waves are typically susceptible to background noise. Therefore, noise-removal and feature-extraction techniques were employed to extract the differentiated leakage signals. Recent studies have focused on machine learning algorithms, such as artificial neural networks and random forests, and numerical solutions to increase the efficiency of acoustic leak detection methods [12-15]. Furthermore, cross-correlation methods have been investigated to detect multiple rupture points in pipelines. Elandalibe et al. [16] used a cross-correlation technique to detect multiple leakage points by retrieving two vibration signals from sensors installed at the end of a pipeline.

Source localization technology using acoustic signals is also demonstrated using spherical microphone arrays (SMAs). Studies have shown that SMAs can capture the spatiotemporal information of acoustic signals in a three-dimensional domain. This allows SMAs to remotely and accurately collect acoustic signals, as well as localize and map the source of the acoustic signals into visualizable images. However, the effectiveness of SMAs in masking noise and identifying acoustic signals related to gas leakages is hindered in the presence of various vibration sources and acoustic noise. Ping et al. demonstrated multiple-source localization by using SMAs in a three-dimensional domain [17]. They used sparse Bayesian learning (SBL) and principal component analysis (PCA) to reduce data dimensions and denoise background acoustic data. The proposed approach offers accurate source localization even when three distinct sources are distributed sparsely and in proximity. They could distinguish sources that were only 10 cm apart, although the localization error increased under these conditions.

Fig. 7.

Schematic diagram of experimental apparatus for multi leakage detection measurement. Reprinted with permission from Ref [16]. Copyrights (2022) IEEE.

Fig. 8.

Three-dimensional source localization using sparse Bayesian learning with spherical microphone arrays in triple source condition. Reprinted with permission from Ref [17]. Copyrights (2022) The Journal of the Acoustical Society of America.

2.4 Mobile platform

Mobile platforms provide a distinct approach for gas leakage detection technologies [18]. The three most common mechanisms discussed above generally require bulky equipment and installations at fixed locations. Some handheld instruments are more portable; however, human operation limits their use in extremely vast industrial manufacturing sites. Recent studies demonstrated the implementation of sensors in mobile platforms, such as robots, unmanned aerial vehicles (UAVs), and drones. This allows the sensors to hover on a mobile platform, navigate through a specific area of interest, and collect data continuously.

As shown in Figure 9, Burgués et al. demonstrated the aerial mapping of odorous gases in a wastewater treatment plant using a small drone equipped with an array of metal oxide and electrochemical sensors [19]. By recording multivariate gas sensor signals around a wastewater treatment plant, the proposed system generated a two-dimensional gas concentration heatmap. The resulting hotspot visualization facilitated the identification of potential leakages or gas concentrations.

Fig. 9.

(a) DJI Matrice 600 Pro drone equipped with an array of metal oxide and a Drager X-am 8000 analyzer. (b) Heatmap of the H2S concentration recorded by the sensors on the drone. (c) Gas concentration measurement data signals from the two sensors mounted on the drone. Reprinted with permission from Ref [19]. Copyrights (2021) MDPI Remote Sensing.

Palacin et al. proposed the use of an assistant personal robot with an array of metal-oxide semiconductor gas sensors for early gas leak detection [20]. The mobile robot could detect ethanol and acetone leakage at different locations by incorporating a partial least squares discriminant analysis (PLS-DA) classifier. This research team confirmed the reliability of their mobile robotic system through a series of experiments involving different heating, ventilating, and air conditioning (HVAC) conditions and various gas sources.

In addition to mobile platforms equipped with chemical sensors, studies have used unmanned aerial vehicles equipped with optical infrared spectroscopy to detect environmental gases. Rutkauskas et al. proposed a system that measures propane and carbon dioxide content using a Fourier-transform spectrometer [21]. As shown in Figure 11, the drone scanned over a field of 720 square meters, detected the concentration of ambient propane and carbon dioxide, and localized the source of the gas release. On a comparable note, albeit with subtle distinctions, Nooralishahi et al. presented a drone-enabled approach for gas leak detection to achieve proximity conditions required for reliable optical flow analysis [22].

Fig. 10.

Assistant personal robot and metal oxide semiconductor sensor array used as experimental apparatus. Reprinted with permission from Ref [20]. Copyrights (2019) MDPI Sensors.

Fig. 11.

(a) Top view of experimental site with drone flight plan in red. (b, c) Gas concentration heatmap of propane and carbon dioxide, respectively. (d) On site image of UAV and propane and carbon dioxide gas cylinders with release hoses. (e, f) Gas concentration uncertainty map of propane and carbon dioxide, respectively. Reprinted with permission from Ref [19]. Copyrights (2019) Optics Express.

Mobile platforms provide greater versatility for common gas sensing mechanisms. As demonstrated in the aforementioned studies, unmanned aerial vehicles/drones or mobile robots equipped with metal oxide semiconductor sensors and infrared spectroscopy modules can be used to monitor large spaces. With the development of autonomous system technology, integrated gas-monitoring systems have the potential to address industrial safety and environmental problems. Despite the technological obstacles that must be overcome to develop systematic solutions, the recent studies presented in this paper have revealed promising breakthroughs in hazardous gas monitoring and source localization methods.


3. CONCLUSIONS

This review summarizes recent studies on various advanced techniques for monitoring gas emissions and detecting leaks that are used for various purposes. The methods presented herein, can be categorized according to the implemented sensing mechanism or form factor. Recent research has demonstrated target gas detection using chemical, optical, and acoustic methods based on diverse leakage scenarios. These sensing mechanisms can adopt various form factors at fixed local positions or on mobile platforms to function as state-of-the-art gas detection systems. Because future industrial operations need to be more environmentally friendly and safe, further development of high-performance gas sensors with low power consumption is crucial. In addition to sensor development, integrating these sensors into a comprehensive and intelligent gas detection system for real-world applications is another critical challenge that need to be address in the future.

References

  • P.-S. Murvay and I. Silea, “A survey on gas leak detection and localization techniques”, J. Loss Prev. Process Industries, Vol. 25, No. 6, pp. 966-973, 2012. [https://doi.org/10.1016/j.jlp.2012.05.010]
  • B. Yang, N. V. Myung, and T.-T. Tran, “1D metal oxide semiconductor materials for chemiresistive gas sensors: A review”, Adv. Electron. Mater., Vol. 7, No. 9, p. 2100271, 2021. [https://doi.org/10.1002/aelm.202100271]
  • J. Burgués, V. Hernández, A. J. Lilienthal, and S. Marco, “Gas distribution mapping and source localization using a 3D grid of metal oxide semiconductor sensors”, Sens. Actuators B: Chem., Vol. 304, p. 127309, 2020. [https://doi.org/10.1016/j.snb.2019.127309]
  • M. Kang, I. Cho, J. Park, J. Jeong, K. Lee, B. Lee, D. D. O. Henriquez, K. Yoon, and I. Park, “High Accuracy Real-Time Multi-Gas Identification by a Batch-Uniform Gas Sensor Array and Deep Learning Algorithm”, ACS Sens., Vol. 7, No. 2, pp. 430-440, 2022. [https://doi.org/10.1021/acssensors.1c01204]
  • I. Cho, Y. C. Sim, M. Cho, Y.-H. Cho, and I. Park, “Monolithic Micro Light-Emitting Diode/Metal Oxide Nanowire Gas Sensor with Microwatt-Level Power Consumption”, ACS Sens., Vol. 5, No. 2, pp. 563-570, 2020. [https://doi.org/10.1021/acssensors.9b02487]
  • D. Cho, J. M. Suh, S.-H. Nam, S. Y. Park, M. Park, T. H. Lee, K. S. Choi, J. Lee, C. Ahn, H. W. Jang, Y.-S. Shim, and S. Jeon, “Optically Activated 3D Thin-Shell TiO2 for Super-Sensitive Chemoresistive Responses: Toward Visible Light Activation”, Adv. Sci., Vol. 8, No. 3, p. 2001883, 2020. [https://doi.org/10.1002/advs.202001883]
  • J. Lee, D. Cho, H. Chen, Y.-S. Shim, J. Park, and S. Jeon, “Proximity-field nanopatterning for high-performance chemical and mechanical sensor applications based on 3D nanostructures”, Appl. Phys. Rev., Vol. 9, No. 1, p. 011322, 2022. [https://doi.org/10.1063/5.0081197]
  • J. Titchener, D. Millington-Smith, C. Goldsack, G. Harrison, A. Dunning, X. Ai, and M. Reed, “Single photon Lidar gas imagers for practical and widespread continuous methane monitoring”, Appl. Energy, Vol. 306 p. 118086, 2022. [https://doi.org/10.1016/j.apenergy.2021.118086]
  • H. Yang, X.-F. Yao, S. Wang, L. Yuan, Y.-C. Ke, and Y.-H. Liu, “Simultaneous determination of gas leakage location and leakage rate based on local temperature gradient”, Meas., Vol. 133, pp. 233-240, 2019. [https://doi.org/10.1016/j.measurement.2018.10.017]
  • Z. Hu, S. Tariq, and T. Zayed, “A comprehensive review of acoustic based leak localization method in pressurized pipelines”, Mech. Syst. Signal Process., Vol. 161, p. 107994, 2021. [https://doi.org/10.1016/j.ymssp.2021.107994]
  • S. Li, Y. Wen, P. Li, J. Yang, and L. Yang, “Leak detection and location for gas pipelines using acoustic emission sensors”, 2012 IEEE Int. Ultrason. Symp., pp. 957-960, 2012.
  • V. Asghari, M. H. Kazemi, H.-F. Duan, S.-C. Hsu, and A. Keramat, “Machine learning modeling for spectral transient-based leak detection”, Autom. Constr., Vol. 146, p. 104686, 2023 [https://doi.org/10.1016/j.autcon.2022.104686]
  • J. Zhang, Z. Lian, Z. Zhou, M. Xiong, M. Lian, and J. Zheng,”Acoustic method of high-pressure natural gas pipelines leakage detection: Numerical and applications”, Int. J. Press. Vessel. Piping, Vol. 194, p. 104540, 2021. [https://doi.org/10.1016/j.ijpvp.2021.104540]
  • F. Ning, Z. Cheng, D. Meng, and J. Wei, “A framework combining acoustic features extraction method and random forest algorithm for gas pipeline leak detection and classification”, Appl. Acoust., Vol. 182, p. 108255, 2021. [https://doi.org/10.1016/j.apacoust.2021.108255]
  • W. Wang, X. Mao, H. Liang, D. Yang, J. Zhang, and S. Liu, “Experimental research on in-pipe leaks detection of acoustic signature in gas pipelines based on the artificial neural network”, Meas., Vol. 183, p. 109875, 2021. [https://doi.org/10.1016/j.measurement.2021.109875]
  • K. Elandalibe, A. Jbari, and A. Bourouhou, “Application of cross-correlation technique for multi leakage detection”, Third World Conf. Complex Syst. (WCCS), IEEE, pp. 1-4, 2015. [https://doi.org/10.1109/ICoCS.2015.7483243]
  • G. Ping, E. Fernandez-Grande, P. Gerstoft, and Z. Chu, “Three-dimensional source localization using sparse Bayesian learning on a spherical microphone array”, J. Acoust. Soc. Am., Vol. 147, No. 6, pp. 3895-3904, 2020. [https://doi.org/10.1121/10.0001383]
  • J. Burgués and S. Marco. “Environmental chemical sensing using small drones: A review”, Sci. Total Environ., Vol. 748, p. 141172, 2020. [https://doi.org/10.1016/j.scitotenv.2020.141172]
  • J. Burgués, M. D. Esclapez, S. Doñate, L. Pastor, and S. Marco. “Aerial mapping of odorous gases in a wastewater treatment plant using a small drone”, Remote Sens., Vol. 13, No. 9, p. 1757, 2021. [https://doi.org/10.3390/rs13091757]
  • J. Palacín, D. Martinez, E. Clotet, T. Pallejà, J. Burgués, J. Fonollosa, A. Pardo, and S. Marco. “Application of an array of metal-oxide semiconductor gas sensors in an assistant personal robot for early gas leak detection”, Sens., Vol. 19, No. 9, p. 1957, 2019. [https://doi.org/10.3390/s19091957]
  • M. Rutkauskas, M. Asenov, S. Ramamoorthy, and D. T. Reid. “Autonomous multi-species environmental gas sensing using drone-based Fourier-transform infrared spectroscopy”, Opt. Express, Vol. 27, No. 7, pp. 9578-9587, 2019. [https://doi.org/10.1364/OE.27.009578]
  • P. Nooralishahi, F. López, and X. Maldague, “A Drone-Enabled Approach for Gas Leak Detection Using Optical Flow Analysis”, Appl. Sci., Vol. 11, No. 4, p. 1412, 2021. [https://doi.org/10.3390/app11041412]

Fig. 1.

Fig. 1.
Gas sensing mechanism for n-type and p-type metal oxide semiconductor-based gas sensors. Reprinted with permission from Ref [2]. Copyrights (2017) Royal Society of Chemistry.

Fig. 2.

Fig. 2.
CAD illustration and actual photo of gas test room with metal oxide sensors (in green circles), processing nodes (in orange rectangles), HVAC outlet (in a pick rectangle), gas source, window, and radiators. Reprinted with permission from Ref [3]. Copyrights (2020) Elsevier.

Fig. 3.

Fig. 3.
(a) Glancing angle deposition (GLAD) of nanocolumnar metal oxide. (b) Microscopic top-view image of a suspended microheater platform-based gas sensor. (c) CAD illustration of a gas sensor array and signal processing. Reprinted with permission from Ref [4]. Copyrights (2020) Elsevier.

Fig. 4.

Fig. 4.
Wind condition controlled CH4 leak imaging (a, b) Actual images of experimental apparatus (c-h) Resulting images obtained for different combinations of wind speed and CH4 flow rate. Reprinted with permission from Ref [8]. Copyrights (2022) Elsevier.

Fig. 5.

Fig. 5.
Industrial environment test of CH4 release measurements with Tunable Diode Lidar (TDLidar) images at Total’s TADI platform (a) Visual camera image overlaid with the infrared signal level image of CH4 plume. (b, c) Sensor signal level image overlaid with CH4 path length (ppm.m) image. Reprinted with permission from Ref [8]. Copyrights (2022) Elsevier.

Fig. 6.

Fig. 6.
(a) Illustration of the experimental container apparatus with a leakage hole (b) Local temperature field at the wall of the container with the leakage hole. Reprinted with permission from Ref [9]. Copyrights (2022) Elsevier.

Fig. 7.

Fig. 7.
Schematic diagram of experimental apparatus for multi leakage detection measurement. Reprinted with permission from Ref [16]. Copyrights (2022) IEEE.

Fig. 8.

Fig. 8.
Three-dimensional source localization using sparse Bayesian learning with spherical microphone arrays in triple source condition. Reprinted with permission from Ref [17]. Copyrights (2022) The Journal of the Acoustical Society of America.

Fig. 9.

Fig. 9.
(a) DJI Matrice 600 Pro drone equipped with an array of metal oxide and a Drager X-am 8000 analyzer. (b) Heatmap of the H2S concentration recorded by the sensors on the drone. (c) Gas concentration measurement data signals from the two sensors mounted on the drone. Reprinted with permission from Ref [19]. Copyrights (2021) MDPI Remote Sensing.

Fig. 10.

Fig. 10.
Assistant personal robot and metal oxide semiconductor sensor array used as experimental apparatus. Reprinted with permission from Ref [20]. Copyrights (2019) MDPI Sensors.

Fig. 11.

Fig. 11.
(a) Top view of experimental site with drone flight plan in red. (b, c) Gas concentration heatmap of propane and carbon dioxide, respectively. (d) On site image of UAV and propane and carbon dioxide gas cylinders with release hoses. (e, f) Gas concentration uncertainty map of propane and carbon dioxide, respectively. Reprinted with permission from Ref [19]. Copyrights (2019) Optics Express.