• Journal of Korean Society for Quality Management
• /
• v.49 no.4
• /
• pp.581-594
• /
• 2021

Purpose: The purpose of this study is to compare machine learning models for anomaly detection of mechanical facility equipment and suggest an anomaly detection system for mechanical facility equipment in subway stations. It helps to predict failures and plan the maintenance of facility. Ultimately it aims to improve the quality of facility equipment. Methods: The data collected from Daejeon Metropolitan Rapid Transit Corporation was used in this experiment. The experiment was performed using Python, Scikit-learn, tensorflow 2.0 for preprocessing and machine learning. Also it was conducted in two failure states of the equipment. We compared and analyzed five unsupervised machine learning models focused on model Long Short-Term Memory Variational Autoencoder(LSTM-VAE). Results: In both experiments, change in vibration and current data was observed when there is a defect. When the rotating body failure was happened, the magnitude of vibration has increased but current has decreased. In situation of axis alignment failure, both of vibration and current have increased. In addition, model LSTM-VAE showed superior accuracy than the other four base-line models. Conclusion: According to the results, model LSTM-VAE showed outstanding performance with more than 97% of accuracy in the experiments. Thus, the quality of mechanical facility equipment will be improved if the proposed anomaly detection system is established with this model used.

• Proceedings of the KSME Conference
• /
• 2003.04a
• /
• pp.2012-2017
• /
• 2003

This paper reports on the influence of blade surface roughness on turbine efficiency. The performance of a low speed one-stage axial turbine with roughened blade surfaces was evaluated. Sandpaper with equivalent sandgrain roughness ($k_s$) was used to roughen the blades. Efficiency (${\eta}/{\eta}_0$) decreases by 4.5 % with sandgrain size of 400 ${\mu}m$ on the stator suction surface.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 1996.06c
• /
• pp.1031-1040
• /
• 1996

Spawn bottle cutting and spawn cake shredding machine has been developed to save the farm labor and operating cost for mushroom growing farmers. The prototype can cut the bottom and side of the bottle while shredding the spawn cake by one farmer. The cutting plastic shell is done by two couples of high speed disc saws that can cut cross section and lengthwise two side of the bottle, while spawn cake shredding is done by spike teethed rotating drum and wiremesh concave. The optimum speed of cutting disc saw was observed 1.700rpm both the cross cutting saw and lengthwise cutting saw in consideration of the cutting accuracy. And the location for the instalation of cross cutting disc saw was considered as around 4 mm above the table bottom , while the optimum clearance between two edges of lengthwise cutting disc saw was showed 86mm. For the sawdust spawn cake shredder , proper size of spawn sawdust granule was observed in the 15 x 15mm size withmesh concave with the shredding rum speed of 500rpm. The prototype can be reduce 73 per cent of working hours with the working cost reduction of 49 per cent compare with those of conventional operation.

• Proceedings of the KSME Conference
• /
• 2000.11b
• /
• pp.453-460
• /
• 2000

A test rig is developed for performance test of 1 stage axial-type turbine which is designed by meanline analysis, streamline curvature method, and blade design method using configuration parameters. The purpose of this study is to find the best configuration parameters for designing a high efficiency axial-type turbine blade. To measure the efficiency of turbine stage, a dynamo-meter is installed. Two different stators which are manufactured as an integrated type are developed, and a rotor blade and 5 sets disc are developed for setting different stagger angle. The tip and hub diameters of the test turbine are 300 and 206.4mm, respectively. The rotating speed is 1800RPM, and the extracted power is 2.5kW. Flow coefficient is 1.68 and the reaction factor at meanline is 0.373. The number of stator and rotor of test turbine are 31 and 41, respectively. The Mach number of stator exit flow near hub is 0.164.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.14 no.8
• /
• pp.684-694
• /
• 2004

Presented in this dissertation is a new method of identifying the critical speed of rotor-bearing systems without actually reaching at the critical speed itself. Using the method, it is possible to calculate the critical speed by measuring a series of rotor responses at much lower rotating speeds away from and without reaching at the critical speed of the system. In the course of the procedures illustrated, not only the critical speed but also the damping ratio and the eccentricity of the system can be identified at the same time. Test rotor was tested on the Rotor Dynamics Test Facility at the Korea Institute of Machinery ＆ Materials. Korea, and the theory has been confirmed experimentally. The method can be adopted to monitor changes of the dynamic characteristics of critical rotating machinery before and after overhauls, repairs, exchanges of various parts, or to detect trends or direction of subtle changes in the dynamic characteristic parameters over a long periods of time.

• Journal of Ocean Engineering and Technology
• /
• v.30 no.1
• /
• pp.25-31
• /
• 2016

In this study, hydrodynamic coefficients were obtained from a Rotating Arm (RA) test, which is one of the captive model tests used to provide accurate coefficients in the control motion equation of an underwater vehicle. The RA test was carried out at the RA facility of ADD (Agency for Defense Development), and the forces and moments acting on the underwater vehicle were measured using a six-axis waterproof gage. A multiple regression analysis was used in the analysis of the measured data. The experimental results were also verified by comparison with the theoretical values of the previous linear coefficients. In addition, the stability indices in the horizontal plane were calculated using the linear and nonlinear coefficients, and the dynamic stability of the underwater vehicle was estimated to have a good dynamic performance with a depth ratio of 6.0.

• Journal of Ocean Engineering and Technology
• /
• v.35 no.6
• /
• pp.446-456
• /
• 2021

To build an environment facility of a large-scale ocean basin, various detailed reviews are required, but it is difficult to find data that introduces the related research or construction processes on the environment facility. The current generator facility for offshore structure safety evaluation tests should be implemented by rotating the water of the basin. However, when the water in the large basin rotates, relatively large flow irregularities may occur and the uniformity may not be adequate. In this paper, design and review were conducted to satisfy the performance goals of the DOEB through computational numerical analysis on the shape of the waterway and the flow straightening devices to form the current in the large tank. Based on this, the head loss, which decreases the flow rate when the large tank water rotates through the water channel, was estimated and used as the pump capacity (impeller) design data. The impeller of the DOEB current generator was designed through computational numerical analysis (CFD) based on the lift surface theory from the axial-type impeller shape for satisfying the head loss of the waterway and maximum current velocity. In order to confirm the performance of the designed impeller system, the flow rate and flow velocity performance were checked through factory test operation. And, after installing DOEB, the current flow rate and velocity performance were reviewed compare with the original design target values. Finally, by measuring the current velocity of the test area in DOEB formed through the current generator, the spatial current distribution characteristics in the test area were analyzed. Through the analysis of the current distribution characteristics of the DOEB test area, it was confirmed that the realization of the maximum current velocity and the average flow velocity distribution, the main performance goals in the waterway design process, were satisfied.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.21 no.3
• /
• pp.199-205
• /
• 2011

This paper proposes a framework to predict the mechanical fault of hot rolling mill system (HRMS). The optimum process of HRMS is usually identified by the rotating velocity of working roll. Therefore, observing the velocity of working roll is relevant to early know the HRMS condition. In this paper, we propose the framework which consists of two methods namely spectrum matrix which related to case-based fast Fourier transform(FFT) analysis, and three dimensional condition monitoring based on novel visualization. Validation of the proposed method has been conducted using vibration data acquired from HRMS by accelerometer sensors. The acquired data was also tested by developed software referred as hot rolling mill facility analysis module. The result is plausible and promising, and the developed software will be enhanced to be capable in prediction of remaining useful life of HRMS.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.10
• /
• pp.2724-2737
• /
• 1994

The swirl effect on an axisymmetric turbulence has been investigated along its centerline by the hot wire anemometry. Flow facility to generate and conrol the swirl has been built by using the rotating honeycomb and grid. For the case of internal flows, as the strength of the swirl increases the flow tends to be locally isotropic by modifying the radial and the rotational components mainly. In comparison with those of the plain free jet, the decay of the centerline turbulences seems to be delayed substantially even with a slight swirl component.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.12 no.4
• /
• pp.47-54
• /
• 1998

As higher reliability on generator facilities is required to early detect incipient faults, a diagnostic system for predictive maintenance of the facility is highly recommended. In this paper, an expert system is developed suing the CLIPS - expert system tool - utilizing the knowledge on the insulation diagnosis and the vibration diagnosis on the rotating machines. This expert system, with user - friendly interface for operator, is easy to revise and add diagnostic rules and knowledge.