• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.506-515
• /
• 2023

This study is to find a method to learn engine sound after the start-up of a diesel generator installed in nuclear power plant with an unsupervised deep learning algorithm (CNN autoencoder) and a new method to predict the failure of a diesel generator using it. In order to learn the sound of a diesel generator with a deep learning algorithm, sound data recorded before and after the start-up of two diesel generators was used. The sound data of 20 min and 2 h were cut into 7 s, and the split sound was converted into a spectrogram image. 1200 and 7200 spectrogram images were created from sound data of 20 min and 2 h, respectively. Using two different deep learning algorithms (CNN autoencoder and binary classification), it was investigated whether the diesel generator post-start sounds were learned as normal. It was possible to accurately determine the post-start sounds as normal and the pre-start sounds as abnormal. It was also confirmed that the deep learning algorithm could detect the virtual abnormal sounds created by mixing the unusual sounds with the post-start sounds. This study showed that the unsupervised anomaly detection algorithm has a good accuracy increased about 3% with comparing to the binary classification algorithm.

• Korean Chemical Engineering Research
• /
• v.50 no.6
• /
• pp.980-987
• /
• 2012

Nonpoint source pollution causes leaks and overtopping, depending on the state of the sewer network as well as aggravates the pollution load of the aqueous water system as it is introduced into the sewer by wash-off. According, the need for efficient sewer monitoring system which can manage the sewage flowrate, water quality, inflow/infiltration and overflow has increased for sewer maintenance and the prevention of environmental pollution. However, the sewer monitoring is not easy since the sewer network is built in underground with the complex nature of its structure and connections. Sewer decontamination mechanism as well as pipe network monitoring and fault diagnosis of water network system on system analysis proposed in this study. First, the pollution removal pattern and behavior of contaminants in the sewer pipe network is analyzed by using sewer process simulation program, stormwater & wastewater management model for expert (XP-SWMM). Second, the sewer network fault diagnosis was performed using the multivariate statistical monitoring to monitor water quality in the sewer and detect the sewer leakage and burst. Sewer decontamination mechanism analysis with static and dynamic state system results showed that loads of total nitrogen (TN) and total phosphorous (TP) during rainfall are greatly increased than non-rainfall, which will aggravate the pollution load of the water system. Accordingly, the sewer outflow in pipe network is analyzed due to the increased flow and inflow of pollutant concentration caused by rainfall. The proposed sewer network monitoring and fault diagnosis technique can be used effectively for the nonpoint source pollution management of the urban watershed as well as continuous monitoring system.

• The Journal of Society for e-Business Studies
• /
• v.23 no.2
• /
• pp.111-121
• /
• 2018

Many problems in rotating machinery such as aircraft engines, wind turbines and motors are caused by bearing defects. The abnormalities of the bearing can be detected by analyzing signal data such as vibration or noise, proper pre-processing through a few signal processing techniques is required to analyze their frequencies. In this paper, we introduce the condition monitoring method for diagnosing the failure of the rotating machines by analyzing the vibration signal of the bearing. From the collected signal data, the normal states are trained, and then normal or abnormal state data are classified based on the trained normal state. For preprocessing, a Hamming window is applied to eliminate leakage generated in this process, and the cepstrum analysis is performed to obtain the original signal of the signal data, called the formant. From the vibration data of the IMS bearing dataset, we have extracted 6 statistic indicators using the cepstral coefficients and showed that the application of the Mahalanobis distance classifier can monitor the bearing status and detect the failure in advance.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.10
• /
• pp.63-68
• /
• 2017

Although automated metro subway systems have the advantage of operating a train without a train driver, it is difficult to detect an immediate fault condition and take countermeasures when an unusual situation occurs. Therefore, it is important to construct a maintenance information system (MIS) that detects the vehicle failure/status information in real time and maintains it efficiently in the depot of the railway's vehicles. This paper proposes a conceptual design method that realizes the interface between the train control system (TCS), the operation control center train control monitoring system (OCC-TCMS) console, and the MIS using wireless communication network in real-time. To transmit a large amount of information on 800,000 occurrences per day during operation, data was collected in a 56 byte data table using a data processing algorithm. This state information was classified into 4 hexadecimal codes and transmitted to the MIS by mapping the status and the fault information on the vehicle during the main line operation. Furthermore, the transmission and reception data were examined in real time between the TCS and MIS, and the implementation of the failure information screen was then displayed.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.25 no.4
• /
• pp.287-292
• /
• 2012

In this study, Multi-Channel Magnetic Flux Leakage(MFL) sensor - based inspection system was applied to monitor the condition of cables. This inspection system measures magnetic flux to detect the local faults(LF) of steel cable. To verify the feasibility of the proposed damage detection technique, an 8-channel MFL sensor head prototype was designed and fabricated. A steel cable bunch specimen with several types of damage was fabricated and scanned by the MFL sensor head to measure the magnetic flux density of the specimen. To interpret the condition of the steel cable, magnetic flux signals were used to determine the locations of the flaws and the level of damage. Measured signals from the damaged specimen were compared with thresholds set for objective decision making. In addition, the magnetic flux density values measured from every channel were summed to focus on the detection of axial location. And, sum of flux density were displayed with threshold. Finally, the results were compared with information on actual inflicted damages to confirm the accuracy and effectiveness of the proposed cable monitoring method.

• Journal of the Korean earth science society
• /
• v.40 no.3
• /
• pp.272-282
• /
• 2019

Although the knowledge of current seismicity is a critical information for making and implementing effective earthquake-related policy, the detailed seismicity information of the metropolitan areas with high-population density has been largely underestimated due to the high-level of cultural noise and small earthquake magnitude. This study presents 12 earthquakes including 2 earthquakes previously known and 10 additional earthquakes occurred from 2010 to 2017 in Busan, but they were unreported by the Korea Meteorological Administration. Matched filter technique is used to detect micro-earthquakes. Although the epicenters of micro-earthquakes though present a distinguished linearity, a correlation with faults in the area is unknown. A repeated micro-seismicity suggests that there are subsurface structures responsible for observed events. If large earthquakes occur along the fault in Busan, they may cause catastrophic natural disasters. Given the fact that the recent earthquakes did not accompany any surface signatures, it is highly recommended that the current micro-seismicity be investigated, and updated seismicity information be incorporated into establishing active fault maps in Korea.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.7
• /
• pp.387-392
• /
• 2019

A technique was researched to detect bolt looseness using a transfer impedance technique (the dual piezoelectric material technique) for monitoring the structural health of a bolt joint. In order to use the single piezoelectric material technique, an expensive impedance analyzer should be used. However, in the transfer impedance technique, low-cost fault detection can be performed using a general function generator and a digital multimeter. A steel plate frame test specimen composed of bolt joints was fabricated, and the tightening torques of the bolts were loosened step by step. By using the transfer impedance method, the damage index was obtained. It was found that the presence of faults could be reasonably estimated using the damage index, which increased with the degree of bolt looseness. An experiment was performed on the same specimen using the single piezoelectric material technique, and the results showed a similar tendency. It could be possible to estimate the damage of a bolt joint at low cost by eliminating the expensive impedance analyzer. This method could be used effectively for structural health monitoring after carrying out a study to estimate the fault location and severity.

• Journal of the Korea Convergence Society
• /
• v.12 no.8
• /
• pp.31-37
• /
• 2021

As the size of buildings increases due to urbanization due to the development of industry, the need to purify the air and maintain a comfortable indoor environment is also increasing. With the development of monitoring technology for refrigeration systems, it has become possible to manage the amount of electricity consumed in buildings. In particular, refrigeration systems account for about 40% of power consumption in commercial buildings. Therefore, in order to develop the refrigeration system failure diagnosis algorithm in this study, the purpose of this study was to understand the structure of the refrigeration system, collect and analyze data generated during the operation of the refrigeration system, and quickly detect and classify failure situations with various types and severity . In particular, in order to improve the classification accuracy of failure types that are difficult to classify, a three-step diagnosis and classification algorithm was developed and proposed. A model based on SVM and LGBM was presented as a classification model suitable for each stage after a number of experiments and hyper-parameter optimization process. In this study, the characteristics affecting failure were preserved as much as possible, and all failure types, including refrigerant-related failures, which had been difficult in previous studies, were derived with excellent results.

• The Journal of Bigdata
• /
• v.7 no.1
• /
• pp.125-137
• /
• 2022

The transportation industry is one of the important industries due to the geographical requirements surrounded by the sea on three sides of Korea and the problem of resource poverty, which relies on imports for most of its resource consumption. Among them, the proportion of the shipping industry is large enough to account for most of the transportation industry, and maintenance in the shipping industry is also important in improving the operational efficiency and reducing costs of ships. However, currently, inspections are conducted every certain period of time for maintenance of ships, resulting in time and cost, and the cause is not properly identified. Therefore, in this study, the proposed methodology, LSTM-AutoEncoder, is used to detect abnormalities that may cause ship failure by considering the time of actual ship operation data. In addition, clustering is performed through clustering, and the potential causes of ship main engine failure are identified by grouping outlier by factor. This enables faster monitoring of various information on the ship and identifies the degree of abnormality. In addition, the current ship's fault monitoring system will be equipped with a concrete alarm point setting and a fault diagnosis system, and it will be able to help find the maintenance time.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.6 no.4
• /
• pp.327-343
• /
• 2004

During tunnel construction, ground failures often occur due to existence of weak zones, such as faults, joints, and cavities, ahead of tunnel face. It is hard to detect effectively weak zones, which can lead underground structure to fail after excavation and before supporting, by using conventional characterization methods. In this study, an enhanced analytical method of predicting weak zones ahead of tunnel face is developed to overcome some problems in the conventional geophysical exploration methods. The analytical method is based on Coulomb's and Gauss' laws with considering the characteristics of electric fields subjected to rock mass. Using the developed method, closed form solutions are obtained to detect a spherical shaped zone and an oriented fault ahead of tunnel face respectively. The analytical results suggest that the presence of weak zones and their sizes, location, and states can be accurately predicted by combining a proper inversion process with resistance measured from several electrodes on the tunnel face. It appears that the skin depth or resistivity in rock mass is affected by the diameter of tunnel face, natural electric potential and noises induced by experimental measurement and spatial distribution of uncertain properties. The developed analytical solution is verified through experimental tests. About 1800 concrete blocks of 5cm by 5cm by 5cm in size are prepared and used to model a joint rock mass around tunnel face. Weak zones are simulated ahead of tunnel face with a material which has relatively higher conductivity than concrete blocks. Experimental results on the model test show a good agreement with analytical results.