• Nuclear Engineering and Technology
• /
• v.54 no.12
• /
• pp.4751-4758
• /
• 2022

Two-phase flow may almost exist in every branch of the energy industry. For the corresponding engineering design, it is very essential and crucial to monitor flow patterns and their transitions accurately. With the high-speed development and success of deep learning based on convolutional neural network (CNN), the study of flow pattern identification recently almost focused on this methodology. Additionally, the photographing technique has attractive implementation features as well, since it is normally considerably less expensive than other techniques. The development of such a two-phase flow pattern online monitoring system is the objective of this work, which seldom studied before. The ongoing preliminary engineering design (including hardware and software) of the system are introduced. The flow pattern identification method based on CNNs and transfer learning was discussed in detail. Several potential CNN candidates such as ALexNet, VggNet16 and ResNets were introduced and compared with each other based on a flow pattern dataset. According to the results, ResNet50 is the most promising CNN network for the system owing to its high precision, fast classification and strong robustness. This work can be a reference for the online monitoring system design in the energy system.

• Tribology and Lubricants
• /
• v.22 no.1
• /
• pp.47-52
• /
• 2006

It is very important to control pressure and flow rate distribution on each component of engine lubrication network. Sometimes many kinds of orifice are used to control flow rate in the hydraulic lubrication field. In this study orifices were adopted on the lubrication network to control oil flow rate distribution. And unsteady transient flow network analysis was carried out to find out the effects of orifices on the engine oil circuit system.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.234-237
• /
• 2014

This paper provides insights into the sFlow monitoring system of OF@KREONET. OF@KREONET is software defined network (SDN) testbed adapted by KREONET (Korea Research Environment Open NETwork). OF@KREONET uses SDN-based network virtualization to slice the network among multiple concurrent experimenter. Flow Monitoring of OF@KREONET using sFlow. sFlow and OpenFlow can be used to provide an integrated flow monitoring system where OpenFlow controller can be used to define flows to be monitored by sFlow. OF@KREONET flow monitoring system supports monitoring of per slice FlowSpace. An Experimental can monitor his/her own FlowSpace while network administrator can monitor all spaces.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.4
• /
• pp.83-90
• /
• 1998

In this paper, we present neural network for control of Left Ventricular Assist Device(LVAD) system with a pneumatically driven mock circulation system. Beat rate(BR), Systole-Diastole Rate(SDR) and flow rate are collected as the main variables of the LVAD system. System modeling is completed using the neural network with input variables(BR, SBR, their derivatives, actual flow) and output variable(actual flow). It is necessary to apply high perfomance control techniques, since the LVAD system represent nonlinear and time-varing characteristics. Fortunately. the neural network can be applied to control of a nonlinear dynamic system by learning capability In this study, we identify the LVAD system with neural network and control the LVAD system by PID controller and neural network feedforward controller. The ability and effectiveness of controlling the LVAD system using the proposed algorithm will be demonstrated by experiment.

• Korean Institute of Interior Design Journal
• /
• v.16 no.3 s.62
• /
• pp.58-65
• /
• 2007

The residential space of super high-rise buildings, having a form of a huge three-dimensional vertical city, affect the residents psychologically and qualitatively according to the line of flow. Because of these affects, the system of the line of flows is a very important factor. In this study, we recognize the super high-rise apartment's line of flow as a semantic network system based on case studies. And we also understand the mutual relationship by analyzing each space to recognize what effect it does on the residential environment. Furthermore, to bring up a better semantic network system for super high-rise apartment's line of flows is our goal. According to the case studies, the semantic network of the line of flow consists of 3 parts: the functional network, economical network and unit network. The functional network is composed of the 'need' and 'has', while the economical network includes variable walls that can be changed following the user's taste and eccentric positioned living rooms that protect personal privacy. Therefore the economical network started to appear while the personal value changed according to the improvement of the social condition. Finally, the unit network is a network that effects each unit that has ambiguous boundaries due to the appropriate arrangement between transitional spaces. And the unit network is based on the functional network.

• IE interfaces
• /
• v.10 no.3
• /
• pp.179-187
• /
• 1997

The efficiency of material flow systems in terms of optimal network flow and minimum cost flow has always been an important design and operational goal in material handling and distribution system. In this research, an attempt was made to develop a new algorithm and the model to solve a stochastic material flow network with bidirectional and uncertain flows. A stochastic material flow network with bidirectional flows can be considered from a finite set with unknown demand probabilities of each node. This problem can be formulated as a special case of a two-stage linear programming problem which can be converted into an equivalent linear program. To find the optimal solution of proposed stochastic material flow network, some terminologies and algorithms together with theories are developed based on the partitioning and subgradient techniques. A computer program applying the proposed method was developed and was applied to various problems.

• Proceedings of the Korean Nuclear Society Conference
• /
• 1998.05a
• /
• pp.641-646
• /
• 1998

In order to enhance the dynamic and interactive simulation capability of a system thermal hydraulic code for nuclear power plant, applicability of flow network models in SINDA/FLUIN $T^{™}$ has been tested by modeling feedwater system and coupling to DSNP which is one of a system thermal hydraulic simulation code for a pressurized heavy water reactor. The feedwater system is selected since it is one of the most important balance of plant systems with a potential to greatly affect the behavior of nuclear steam supply system. The flow network model of this feedwater system consists of condenser, condensate pumps, low and high pressure heaters, deaerator, feedwater pumps, and control valves. This complicated flow network is modeled and coupled to DSNP and it is tested for several normal and abnormal transient conditions such turbine load maneuvering, turbine trip, and loss of class IV power. The results show reasonable behavior of the coupled code and also gives a good dynamic and interactive simulation capabilities for the several mild transient conditions. It has been found that coupling system thermal hydraulic code with a flow network code is a proper way of upgrading simulation capability of DSNP to mature nuclear plant analyzer (NPA).

• Journal of Korea Multimedia Society
• /
• v.17 no.8
• /
• pp.988-994
• /
• 2014

Software Defined Network (SDN) is a new technology in computer network area which enables user to centralize control plane. The security issue is important in computer network to protect system from attackers. SYN flooding attack is one of Distributed Denial of Service attack methods which are popular to degrade availability of targeted service on Internet. There are many methods to protect system from attackers, i.e. firewall and IDS. Even though firewall is designed to protect network system, but it cannot mitigate DDoS attack well because it is not designed to do so. To improve performance of DDOS mitigation we utilize another mechanism by using SDN technology such as OpenFlow and sFlow. The methodology of sFlow to detect attacker is by capturing and sum cumulative traffic from each agent to send to sFlow collector to analyze. When sFlow collector detect some traffics as attacker, OpenFlow controller will modify the rule in OpenFlow table to mitigate attacks by blocking attack traffic. Hence, by combining sum cumulative traffic use sFlow and blocking traffic use OpenFlow we can detect and mitigate SYN flooding attack quickly and cheaply.

• Convergence Security Journal
• /
• v.8 no.3
• /
• pp.65-71
• /
• 2008

Network Flow has been playing important role in the modern industrial society. No matter what people or company run the network, both of them can't avoid circulating goods among the many branches. But in practical situations, not only the price rising in network increase the transportation costs, but the huge traffic flow volumes increase the transportation costs. Given to such a network environment, how to flow goods in the network is very important. In this paper, the MAX-Flow algorithm will be applied to network flow in order to maximize the network flow volumes. As far as the functions of network are correctly provided, the optimized network system always can make the flow process efficiently.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2008.03a
• /
• pp.22-24
• /
• 2008

This paper presents a numerical model of internal flows in a lubricant supplying and scavenging flow path of an aero-engine lubrication system. The numerical model was built in the General Analysis Software of Aero-engine Lubrication System, GASLS, developed by Northwestern Polytechnical University. The lubricant flow flux, pressure and temperature distribution at steady state were calculated. GASLS is a general purpose computer program employed a 1-D steady state network algorithm for analyzing flowrates, pressures and temperatures in a complex flow network. All kinds of aero-engine lubrication systems can be divided into finite correlative typical elements and nodes from which the calculation network be developed in GASLS. Special emphasis is on how to use combined elements which is a type of typical elements to replace some complex components like bearing bores, accessory gearboxes or heat exchangers. This method can reduce network complexity and improve calculation efficiency. Final computational results show good agreement with experimental data.