• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.4
• /
• pp.379-384
• /
• 2014

The most critical rotating parts of a gas turbine engine are turbine blades and disc, given that they must operate under severe conditions such as high turbine inlet temperature, high speeds, and high compression ratios. Owing to theses operating conditions and high rotational speed energy, some failures caused by turbine disks and blades are categorized into catastrophic and critical, respectively. To maximize the margin of structural integrity, we aim to optimize the vulnerable area of disc-blade interface region. Then, to check the robustness of the obtained optimized solution, we evaluated structural reliability under uncertainties such as dimensional tolerance and fatigue life variant. The results highlighted the necessity for and limitations of optimization which is one of deterministic methods, and pointed out the requirement for introducing reliability-based design optimization which is one of stochastic methods. Thermal-structural coupled-filed analysis and contact analysis are performed for them.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.23 no.1
• /
• pp.32-37
• /
• 2011

The present study has been conducted to propose an algorithm regarding real-time load estimation of a gas engine-driven heat pump. In the study, thermal load of an indoor unit is estimated in terms of air-side and refrigerant-side. The air-side estimation is based on a typical heat exchanger model and is found to be in good agreement with experimental data. When it comes to the refrigerant-side load, a pressure difference across a valve must be estimated. For the estimation, it is assumed to be proportional to a bigger pressure difference that is available either by measurement or by estimation. Relative good agreement between the air- and refrigerant-sides suggests that the assumption may be plausible for the load estimation. The summed flow rate of all of indoor units is in good agreement with the throughput of the compressor which are calculated from the manufacturer's software. Accordingly, estimated thermal loads are also in good agreement. The proposed algorithm may be further developed for improved control algorithm and fault diagnosis.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.3
• /
• pp.130-139
• /
• 2024

Artificial intelligence models are being used to detect facility anomalies using physics data such as vibration, current, and temperature for predictive maintenance in the manufacturing industry. Since the types of facility anomalies, such as facility defects and failures, anomaly detection methods using autoencoder-based unsupervised learning models have been mainly applied. Normal or abnormal facility conditions can be effectively classified using the reconstruction error of the autoencoder, but there is a limit to identifying facility anomalies specifically. When facility anomalies such as unbalance, misalignment, and looseness occur, the facility vibration frequency shows a pattern different from the normal state in a specific frequency range. This paper presents an N-segmentation anomaly detection method that performs anomaly detection by dividing the entire vibration frequency range into N regions. Experiments on nine kinds of anomaly data with different frequencies and amplitudes using vibration data from a compressor showed better performance when N-segmentation was applied. The proposed method helps materialize them after detecting facility anomalies.

• Journal of the Korea Organic Resources Recycling Association
• /
• v.27 no.2
• /
• pp.67-73
• /
• 2019

In this study, to optimize the production and utilization of biogas for organic waste resources, the precision monitoring of on-site facilities and the energy balance by facility were analyzed, and the solutions for field problems were investigated, and the design and operation guidelines for pretreatment facilities and generators were presented. Gas pre-treatment is required to solve frequent failures and efficiency degradation in operation of high quality refining facilities, and processing processes such as desulfurization, dehumidification, deoxidization, dust treatment, volatile organic compounds, etc. Since these processes are substances that are also eliminated from the high-quality process, quantitative guidelines are not presented in the gas pretreatment process, but are suggested to operate during the processing process as a qualitative guideline. In particular, dust, siloxane, and volatile organic compounds are the main cause of frequent failure of high-quality processes if they are not removed from the gas pretreatment process. Design of the biogas high-quality process. The operation guidelines provide quality standards [Methane content (including propane) of 95% or more] with 90% or more utilization of the total gas generation, two systems, and a margin of 10% or more. It also proposed installing gas equalization tank, installing thermal automatic control system for controlling equalization of auxiliary fuel, installing dehumidification device at the back of high quality for removing moisture generated in the process of gas compression, installing heat-resisting facilities to prevent freezing of facilities in winter and reducing efficiency, and installing membrane facilities in particular.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.2
• /
• pp.9-17
• /
• 2023

In this paper, we propose a one-class vibration anomaly detection system for bearing defect diagnosis. In order to reduce the economic and time loss caused by bearing failure, an accurate defect diagnosis system is essential, and deep learning-based defect diagnosis systems are widely studied to solve the problem. However, it is difficult to obtain abnormal data in the actual data collection environment for deep learning learning, which causes data bias. Therefore, a one-class classification method using only normal data is used. As a general method, the characteristics of vibration data are extracted by learning the compression and restoration process through AutoEncoder. Anomaly detection is performed by learning a one-class classifier with the extracted features. However, this method cannot efficiently extract the characteristics of the vibration data because it does not consider the frequency characteristics of the vibration data. To solve this problem, we propose an AutoEncoder model that considers the frequency characteristics of vibration data. As for classification performance, accuracy 0.910, precision 1.0, recall 0.820, and f1-score 0.901 were obtained. The network design considering the vibration characteristics confirmed better performance than existing methods.

• Journal of the Korean Institute of Gas
• /
• v.28 no.2
• /
• pp.93-104
• /
• 2024

Facility diagnosis is essential for the smooth operation and life extension of rotating equipment used in industrial sites. Compared to other diagnostic methods, vibration diagnosis can find most of the initial defects, such as unbalance, alignment failure, bearing defects and resonance, compared to other diagnostic methods. Therefore, vibration analysis is the most commonly used facility diagnosis method in industrial sites, and is usefully used as a predictive preservation (PdM) technology to manage the condition of the facility. However, since the vibration diagnosis method is performed based on experience based on the standard, it is carried out by experts. Therefore, it is intended to contribute to the reliability of the facility by establishing a system that anyone can easily judge defects by establishing a vibration diagnosis method performed based on experience as a knowledgeable code system. An algorithm was developed based on the ISO-20816 standard for vibration measurement, and the reliability was verified by comparing the results of vibration measurement at various demonstration sites such as petrochemical plant compressors, hydrogen charging stations, and industrial machinery with the results of analysis using a development system. The developed algorithm can contribute to predictive maintenance (PdM) technology that anyone can diagnose the condition of the rotating machine at industrial sites and identify defects early to replace parts at the exact time of replacement. Furthermore, it is expected that it will contribute to reducing maintenance costs and downtime due to the failure of rotating machines when applied to various industrial sites such as oil refining facilities, transportation, production facilities, and aviation facilities.