• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.4
• /
• pp.50-58
• /
• 2017

Recently, some methods of converging heterogeneous fire sensor data have been proposed for effective fire detection, but the rule-based methods have low adaptability and accuracy, and the fuzzy inference methods suffer from detection speed and accuracy by lack of consideration for images. In addition, a few image-based deep learning methods were researched, but it was too difficult to rapidly recognize the fire event in absence of cameras or out of scope of a camera in practical situations. In this paper, we propose a novel fire detection system combining a deep learning algorithm based on CNN and fuzzy inference engine based on heterogeneous fire sensor data including temperature, humidity, gas, and smoke density. we show it is possible for the proposed system to rapidly detect fire by utilizing images and to decide fire in a reliable way by utilizing multi-sensor data. Also, we apply distributed computing architecture to fire detection algorithm in order to avoid concentration of computing power on a server and to enhance scalability as a result. Finally, we prove the performance of the system through two experiments by means of NIST's fire dynamics simulator in both cases of an explosively spreading fire and a gradually growing fire.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.4
• /
• pp.185-192
• /
• 2021

Various methods have been proposed to determine the defect detection of welding bead, and recently sensor data and image data inspection have been steadily announced. There are advantages that sensor data inspection is highly accurate, and two-dimensional-based image data inspection is able to determine the position of the welding bead. However, when analyzing only with sensor data, it is difficult to determine whether the welding has been performed at the correct position. On the other hand, the image data inspection does not have high accuracy due to noise and measurement errors. In this paper, we propose a method that can complement the shortcomings of each inspection method and increase its advantages to improve accuracy and speed up inspection by fusing sensor data inspection which are average current, average volt, and mixed gas data, and image data inspection methods and is implemented as software. In addition, it is intended to allow users to conveniently and intuitively analyze and grasp the results by performing analysis using a graphical user interface(GUI) and checking the data and inspection results used for the inspection. Sensor inspection is performed using the characteristics of each sensor data, and image data is inspected by applying a morphology geodesic active contour algorithm. The experimental results showed 98% accuracy, and when performing the inspection on the four image data, and sensor data the inspection time was about 1.9 seconds, indicating the performance of software that can be used as a real-time inspector in the welding process.

• Korean Journal of Materials Research
• /
• v.20 no.5
• /
• pp.235-240
• /
• 2010

We investigated the carbon monoxide (CO) gas-sensing properties of nanostructured Al-doped zinc oxide thin films deposited on self-assembled Au nanodots (ZnO/Au thin films). The Al-doped ZnO thin film was deposited onto the structure by rf sputtering, resulting in a gas-sensing element comprising a ZnO-based active layer with an embedded Pt/Ti electrode covered by the self-assembled Au nanodots. Prior to the growth of the active ZnO layer, the Au nanodots were formed via annealing a thin Au layer with a thickness of 2 nm at a moderate temperature of $500^{\circ}C$. It was found that the ZnO/Au nanostructured thin film gas sensors showed a high maximum sensitivity to CO gas at $250^{\circ}C$ and a low CO detection limit of 5 ppm in dry air. Furthermore, the ZnO/Au thin film CO gas sensors exhibited fast response and recovery behaviors. The observed excellent CO gas-sensing properties of the nanostructured ZnO/Au thin films can be ascribed to the Au nanodots, acting as both a nucleation layer for the formation of the ZnO nanostructure and a catalyst in the CO surface reaction. These results suggest that the ZnO thin films deposited on self-assembled Au nanodots are promising for practical high-performance CO gas sensors.

• Korean Journal of Materials Research
• /
• v.21 no.10
• /
• pp.546-549
• /
• 2011

Nano-sized $SnO_2$ thick films were prepared by a screen-printing method onto $Al_2O_3$ substrates. The sensing characteristics were investigated by measuring the electrical resistance of each sensor in a test box as a function of the detection gas. The nano-sized $SnO_2$ thick film sensors were treated in a $N_2$ atmosphere. The structural properties of the nano $SnO_2$with a rutile structure according to XRD showed a (110) dominant $SnO_2$ peak. The particle size of $SnO_2$:Ni nano powders at Ni 8 wt% was about 45 nm, and the $SnO_2$ particles were found to contain many pores according to the SEM analysis. The sensitivity of the nano $SnO_2$-based sensors was measured for 5 ppm $CH_4$ gas and $CH_3CH_2CH_3$ gas at room temperature by comparing the resistance in air with that in the target gases. The results showed that the best sensitivity of $SnO_2$:Ni and $SnO_2$:Co sensors for $CH_4$ gas and $CH_3CH_2CH_3$ gas at room temperature was observed in $SnO_2$:Ni sensors doped with 8 wt% Ni. The response time of the $SnO_2$:Ni gas sensors was 10 seconds and recovery time was 15 seconds for the $CH_4$ and $CH_3CH_2CH_3$ gases.

• Journal of the Korea Convergence Society
• /
• v.11 no.12
• /
• pp.23-30
• /
• 2020

In this study, the relationship between natural gas (NG) data and gas-related environmental elements was performed using machine learning algorithms to predict the level of gas leakage risk without directly measuring gas leakage data. The study was based on open data provided by the server using the IoT-based remote control Picarro gas sensor specification. The naturel gas leaks into the air, it is a big problem for air pollution, environment and the health. The proposed method is multivariate outlier removing method based Random Forest (RF) classification for predicting risk of NG leak. After, unsupervised k-means clustering, the experimental dataset has done imbalanced data. Therefore, we focusing our proposed models can predict medium and high risk so best. In this case, we compared the receiver operating characteristic (ROC) curve, accuracy, area under the ROC curve (AUC), and mean standard error (MSE) for each classification model. As a result of our experiments, the evaluation measurements include accuracy, area under the ROC curve (AUC), and MSE; 99.71%, 99.57%, and 0.0016 for MOL_RF respectively.

• Journal of KIISE:Information Networking
• /
• v.36 no.6
• /
• pp.514-527
• /
• 2009

Wireless sensor networks (WSNs) can be used in various applications for military or environmental purpose. Recently, there are lots of on-going researches for detecting and tracking the spread of continuous objects or phenomena such as poisonous gas, wildfires, earthquakes, and so on. Some previous work has proposed techniques to detect edge nodes of such a continuous object based on the information of all the 1-hop neighbor nodes. In those techniques, however, a number of nodes are redundantly selected as edge nodes, and thus, the boundary of the continuous object cannot be presented accurately. In this paper, we propose a new edge detection method in which edge nodes of the continuous object are detected based on the information of the neighbor nodes obtained via the Localized Delaunay Triangulation so that a minimum number of nodes are selected as edge nodes. We also define the sensor behavior rule for tracking continuous objects energy-efficiently. Our simulation results show that the proposed edge detection method provides enhanced performance compared with previous 1-hop neighbor node based methods. On the average, the accuracy is improved by 29.95% while the number of edge nodes, the amount of communication messages and energy consumption are reduced by 54.43%, 79.36% and 72.34%, respectively. Moreover, the number of edge nodes decreases by 48.38% on the average in our field test with MICAz motes.

• Journal of Sensor Science and Technology
• /
• v.29 no.3
• /
• pp.156-161
• /
• 2020

In this study, hierarchical mesoporous CuO spheres, ZnO flowers, and heterojunction CuO/ZnO nanostructures were fabricated via a facile hydrothermal method. The as-prepared materials were characterized in detail using various analytical methods such as powder X-ray diffraction, micro Raman spectroscopy, X-ray photoelectron spectroscopy, field-emission scanning electron microscopy, and transmission electron microscopy. The obtained results are consistent with each other. The H₂S-sensing characteristics of the sensors fabricated based on the CuO spheres, ZnO flowers, and CuO/ZnO heterojunction were investigated at different temperatures and gas concentrations. The sensor based on ZnO flowers showed a maximum response of ~141 at 225 ℃. The sensor based on CuO spheres exhibited a maximum response of 218 at 175 ℃, whereas the sensor based on the CuO/ZnO nano-heterostructure composite showed a maximum response of 344 at 150 ℃. The detection limit (DL) of the sensor based on the CuO/ZnO heterojunction was ~120 ppb at 150 ℃. The CuO/ZnO sensor showed the maximum response to H₂S compared with other interfering gases such as ethanol, methanol, and CO, indicating its high selectivity.

• Journal of Sensor Science and Technology
• /
• v.13 no.4
• /
• pp.298-302
• /
• 2004

Thick films of $LaNiO_{3}$ having perovskite structure impregnated with indium and bismuth oxides have been used as sensing material for acetonitrile ($CH_{3}CN$) gas. The sensor response for $CH_{3}CN$ is quite good with an excellent recovery for partial pressure from 3 ppm to 20 ppm between 200 and $250^{\circ}C$. $LaNiO_{3}$ alone has exhibited low response, but after impregnation of $In_{2}O_{3}$ and $Bi_{2}O_{3}$ have given increased sensitivity even with 3 ppm partial pressure of $CH_{3}CN$ at $200^{\circ}C$. It is assumed that $CH_{3}CN$ is undergoing oxidation reaction on surface of the film.

• Journal of the Korean Vacuum Society
• /
• v.8 no.4B
• /
• pp.507-513
• /
• 1999

8mol%-yttria-stabilized zirconia(YSZ) thin films as oxygen ion conductor were deposited by rf-magnetron sputtering, and the oxygen gas sensors with the structure of $SiO_2$ substrate/Ni-NiO mixed reference layer/Pt/YSZ/Pt were fabricated and their oxygen sensing properties were investigated. The steady-state electro-motive force (EMF) values were measured as a function of oxygen partial pressure ($PO_2;form 1.013\times10^3 \textrm{Pa \;to}\; 1.013\times10^5$Pa) and operating temperature ($300^{\circ}C$ to $700^{\circ}C$). The fabricated YSZ oxygen sensor showed the best oxygen sensing properties at 50$0^{\circ}C$. However, oxygen sensing properties were very low at the temperature lower than 30$0^{\circ}C$ due to the lack of oxygen ion mobility and at the temperature higher than $700^{\circ}C$ due 새 intermixing of materials between the layers. Especially, the YSZ sensor operating at $500^{\circ}C$ and oxygen partial pressure above $1.565\times10^4$Pa showed the oxygen sensing properties close to the values predicted by ideal Nernst equation.

• Proceedings of the KIEE Conference
• /
• 2003.07c
• /
• pp.1638-1640
• /
• 2003

In the recent years, UHF PD measuring technique for detecting partial discharges was proved the effective method for Gas Insulated Switchgear (GIS). However, in case of GIS with a metal flange around insulating spacer, UHF PD measurement using typical external UHF PD sensor is difficult. In this paper, a novel hole-type UHF PD sensor based on Archimedean spiral antenna theory has been proposed and realized. All spacers with metal flange have small hole in order to inject epoxy. Using the novel hole-type UHF PD sensor, it makes detection possible to PD signal that are emitted through the epoxy injection hole. Additionally, the measuring characteristic of UHF PD signals from several artificial defects in GIS and the novel ${\Phi}$-f-q pattern analysis technology are discussed.