• Annual Conference of KIPS
• /
• 2019.05a
• /
• pp.458-460
• /
• 2019

최근 클라우드 시스템 환경이 점차 늘어남에 따라 데이터 센터(IDC) 구축이 점차 늘어나가고 있다. 데이터 센터는 최근 부각하고 있는 4 차 산업 영역에서 사물 인터넷(IoT), 자율주행차 등 에서 처리될 대용량 데이터로 인한 이를 처리하는 중요한 역할을 담당하고 있다. 데이터센터 운영에는 대량의 에너지가 필요하다. 수 많은 컴퓨터에서 발생하는 열에너지를 처리하기 위하여 대량의 전력 냉방 에너지를 소비하고 있다. 냉방 공조 운영은 데이터 센터 운영에 중요한 역할을 한다. 이유는 많은 컴퓨터를 가동하는 비용보다 부대 시설로 운영되는 냉방 에너지를 보다 많이 소비하는 현상까지 발생하고 있다. 이에 최근 데이터 센터 냉방 공조 운영을 효율화하는 것에 연구를 맞추고 있다. 본 논문에서는 냉방 공조 운영 효율화 하도록 하기 위해서 다중 기계 학습을 활용한 데이터 센터의 냉방 에너지 절감 시스템을 제안하고자 한다. 기존의 단수 알고리즘을 활용하여 머신 러닝의 모델구현 방식이 아닌 다중의 기계 학습을 통하여 최적화된 모델을 일일 배치로 생성하여 예측을 하는 시스템이다. 본 시스템을 통하여 사전에 최적화된 냉방 운영을 하여 기존 데이터 센터의 운영되는 과다 냉방을 감축 시켜 에너지를 절감해주는 기능을 제공한다. 본 논문 시스템 연구 결과는 폭발적으로 늘어가고 있는 데이터 센터의 에너지 효율화에 기여할 수 있고, 클라우드 사업에서 경쟁력을 줄 수 있는 운영 시스템 방안을 제시한다.

• Journal of the Korean Institute of Gas
• /
• v.7 no.4 s.21
• /
• pp.20-23
• /
• 2003

The analysis of reliabilities of process facilities often uses models based on the Weibull distribution. The parameters in these models are functions of operating conditions, and determined by experiments. Using these values, we calculate the reliability, mean time to failure, and standard deviation. The conventional method assumes that the operating condition is constant, and thus treats the model parameters as constants. In this paper, a reliability function is proposed which is applicable when the scale parameter is a function of time, and an analysis method based on this is also presented. A case study on a cooling fan resulted in a big difference from the conventional method to which the average operating conditions were applied. The proposed method is also applicable to other process facilities, and expected to effectively take into account the effects of changes in the operating conditions on the reliabilities of the facilities.

• Geomechanics and Engineering
• /
• v.13 no.5
• /
• pp.809-823
• /
• 2017

This paper describes lab test results of artificial rock-like material samples having a plane joint. Cyclic shear tests were performed under different normal loads and different shear displacement amplitudes. For this purpose, multi-stage normal loading tests (30 kN, 60 kN, 90 kN, 180 kN, 360 kN and 480 kN) with cyclic excitation at frequency of 1.0 Hz and different shear displacement amplitudes (0.5 mm, 1.0 mm, 2.0 mm, 4.0 mm, 5.0 mm, and 8.0 mm) were conducted using the big shear box device GS-1000. Experimental results show, that shear forces increase with the increase of normal forces and quasi-static friction coefficient is larger than dynamic one. With the increase of normal loads, approaching the peak value of shear forces needs larger shear displacements. During each cycle the normal displacements increase and decrease (rotational behavior in every cycle). Peak angle of inclination increases with the increase of normal load. A phase shift between maximum shear displacement and maximum shear force is observed. The corresponding time shift decreases with increasing normal load and increases with increasing shear displacement amplitudes.

• Korean Journal of Metals and Materials
• /
• v.50 no.2
• /
• pp.146-151
• /
• 2012

The present article shows how the fracture strength of single crystal silicon chips, which are generally used as semiconductor devices, is influenced by loading rate variation during a 3-point bending test. It was found that the fracture strength of the silicon chips slightly increases up to 4% with increasing loading rate for loading rates lower than 20 mm/min. Meanwhile, the fracture strength of the chips hardly increases with increase of loading rate to levels higher than 40 mm/min. However, there was an abrupt transition in the fracture strength within a loading rate range of 20 mm/min to 40 mm/min. This work explains through microscopic examination of the fracture surface of all test chips that such a big transition is related to the deflection of crack propagation direction from the (011) [${\bar{1}}00$] system to the (111) [${\bar{2}}11$] system in a particular loading rate (i.e. from 20 mm/min to 40 mm/min).

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.18 no.9
• /
• pp.52-57
• /
• 2019

The key technology of the fourth industrial revolution is artificial intelligence and machine learning. In this study, FMEA was performed on fuel pumps used as key items in most systems to identify major failure components, and artificial neural networks were built using big data. The main failure mode of the fuel pump identified by the test was coil damage due to overheating. Based on the artificial neural network built, machine learning was conducted to predict the failure and the mean error rate was 4.9% when the number of hidden nodes in the artificial neural network was three and the temperature increased to $140^{\circ}C$ rapidly.

• Journal of the Semiconductor & Display Technology
• /
• v.18 no.4
• /
• pp.40-45
• /
• 2019

The smart factory has developed with the help of several technologies such as automation, artificial intelligence, big data, smart sensors and communication protocols. The Internet of things(IOT) among communication protocols has become the key factor for the seamless integration of various manufacturing equipment. Therefore, it is important that the IOT cooperate with the standards of communication protocols proposed by the SEMI in the semiconductor industry. In this paper, we suggest a novel reference model of the communication protocols for semiconductor equipment by introducing an IOT service layer. With the IOT service layer, we can use the functions and the additional services provided by the IOT standards that give the inter-operability between factory machines and host computers. We implement the standard of the communication protocols for semiconductor equipment with the IOT service layer by using ns3 simulator. It concludes that it is necessary to provide the platform for the IOT service layer to deploy efficiently the proposed reference model of the communication protocols.

• Food Science and Preservation
• /
• v.4 no.1
• /
• pp.17-25
• /
• 1997

The effects of microwave treatment on the perservation of foods, such as a seaweed soup and sea stoned radish shreds, were studied. Microwave treatment of microbial cell suspensions revealed that viable cells decreased dramatically when heated to 6$0^{\circ}C$. However, it was unlikely that microwave treatment to 60 is enough to decrease the viable cell counts efficiently in a seaweed soup and radish shreds. It was thought that microwave heating to at least 7$0^{\circ}C$ as a final temperature was an important factor to reduce microbial cell counts in foods. When foods were heated to 7$0^{\circ}C$ with a repetitive 15 sec "on" followed by 30 sec "off", no big differences were observed in viable counts during storage at 2$0^{\circ}C$ for 3 days, as compared to those treated with a full power. The microwave treatment with three stages was designed to solve problems associated with variations depending on food volumes and difficulties of heat diffusion in a solid food to be irradiated with a microwave oven. The three stage method was found to have a similar efficiency in the reduction of viable cell counts in foods to microwave treatment at a full power and to conventional methods, such as water bath heating or boiling for 3 min with a gas range.in with a gas range.

• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.851-865
• /
• 2021

Enterprise networks in the PyeongChang Winter Olympics were hacked in February 2018. According to a domestic security company's analysis report, attackers destroyed approximately 300 hosts with the aim of interfering with the Olympics. Enterprise have no choice but to rely on digital vaccines since it is overwhelming to analyze all programs executed in the host used by ordinary users. However, traditional vaccines cannot protect the host against variant or new malware because they cannot detect intrusions without signatures for malwares. To overcome this limitation of signature-based detection, there has been much research conducted on the behavior analysis of malwares. However, since most of them rely on a sandbox where only analysis target program is running, we cannot detect malwares intruding the host where many normal programs are running. Therefore, this study proposes a method to detect malware variants in the host through logs rather than the sandbox. The proposed method extracts common behaviors from variants group and finds characteristic behaviors optimized for querying. Through experimentation on 1,584,363 logs, generated by executing 6,430 malware samples, we prove that there exist the common behaviors that variants share and we demonstrate that these behaviors can be used to detect variants.

• Journal of Integrative Natural Science
• /
• v.14 no.1
• /
• pp.27-34
• /
• 2021

전 세계적으로 인공지능(AI)을 구현하려는 움직임이 많아지고 있다. AI구현에서는 많은 양의 데이터, 목적에 맞는 데이터의 분류 등 데이터의 중요성을 뺄 수 없다. 이러한 데이터를 생성하고 가공하는 기술에는 사물인터넷(IOT)과 빅데이터(Big-data) 분석이 있으며 4차 산업을 이끌어 가는 원동력이라 할 수 있다. 또한 이러한 기술은 국가와 개인 차원에서 많이 활용되고 있으며, 특히나 특정분야에 집결되는 데이터를 기준으로 빅데이터 분석에 활용함으로써 새로운 모델을 발견하고, 그 모델로 새로운 값을 추론하고 예측함으로써 미래비전을 제시하려는 시도가 많아지고 있는 추세이다. 데이터 분석을 통한 결론은 데이터가 가지고 있는 정보의 정확성에 따라 많은 변화를 가져올 수 있으며, 그 변화에 따라 잘못된 결과를 발생시킬 수도 있다. 이렇듯 데이터의 분석은 데이터가 가지는 정보 또는 분석 목적에 맞는 데이터 분류가 매우 중요하다는 것을 알 수 있다. 또한 빅데이터 분석결과 통계량의 신뢰성과 정교함을 얻기 위해서는 각 변수의 의미와 변수들 간의 상관관계, 다중공선성 등을 고려하여 분석해야 한다. 즉, 빅데이터 분석에 앞서 분석목적에 맞도록 데이터의 분류가 잘 이루어지도록 해야 한다. 이에 본 고찰에서는 AI기술을 구현하는 머신러닝(machine learning, ML) 기법에 속하는 분류분석(classification analysis, CA) 중 의사결정트리(decision tree, DT)기법, 랜덤포레스트(random forest, RF)기법, 선형분류분석(linear discriminant analysis, LDA), 이차선형분류분석(quadratic discriminant analysis, QDA)을 이용하여 데이터를 분류한 후 데이터의 분류정도를 평가함으로써 데이터의 분류 분석률 향상을 위한 방안을 모색하려 한다.

• Journal of Convergence for Information Technology
• /
• v.9 no.9
• /
• pp.20-26
• /
• 2019

With the development of ICT technology and the promotion of smartphone penetration, purchasing services that purchase various products through online market are being activated. In particular, due to advances in storage and delivery technology, sales of short food materials can be purchased online. Therefore, in this paper, we propose an integrated solution that enables advertisement effect, ordering and delivery through a purchase service even if there is no sales knowledge and sales network in a small shopping mall where only offline sales can be performed. The proposed system is able to efficiently view the product information by category through image search for the product that the user desires, so that the seller of the registered product can efficiently sell without any additional advertisement.