• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.493-496
• /
• 2000

We study the problem of designing at minimum cost a two-connected network topology such that the shortest cycle to which each edge belongs does not exceed a given maximum number of hops. This problem is considered as part of network planning and arises in the design of backbone networks. We propose a genetic algorithm approach that uses a solution representation, in which the connectivity and ring constraints can be easily encoded. We also propose a crossover operator that ensures a generated solution is feasible. By doing so, the checking of constraints is avoided and no repair mechanism is required. We carry out experimental evaluations to investigate the solution representation issues and GA operators for the network design problem.

• Nuclear Engineering and Technology
• /
• v.35 no.3
• /
• pp.243-250
• /
• 2003

Nuclear power plants (NPPs) will be highly connected network enabled systems in the future. Using the network and web enabled tools, NPPs will be remotely monitored by operators at any time from any place connected to the network via a general web browser. However, there will be two major issues associated with this implementation. The first is the security issue. Only the authorized persons need to be allowed to access the plant since NPP is a safety-critical system. However, the web technology is open to the public. The second is the network disturbance issue. If operators can not access the plant due to network disturbances, the plant will come into the out-of-control situation. Therefore, in this work, we performed a conceptual design of a web-based remote monitoring and control system (RMCS) considering these issues.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.11
• /
• pp.4426-4442
• /
• 2020

Since scene segmentation is becoming a hot topic in the field of autonomous driving and medical image analysis, researchers are actively trying new methods to improve segmentation accuracy. At present, the main issues in image semantic segmentation are intra-class inconsistency and inter-class indistinction. From our analysis, the lack of global information as well as macroscopic discrimination on the object are the two main reasons. In this paper, we propose a Densely connected residual Attention network (DA-Res2Net) which consists of a dense residual network and channel attention guidance module to deal with these problems and improve the accuracy of image segmentation. Specifically, in order to make the extracted features equipped with stronger multi-scale characteristics, a densely connected residual network is proposed as a feature extractor. Furthermore, to improve the representativeness of each channel feature, we design a Channel-Attention-Guide module to make the model focusing on the high-level semantic features and low-level location features simultaneously. Experimental results show that the method achieves significant performance on various datasets. Compared to other state-of-the-art methods, the proposed method reaches the mean IOU accuracy of 83.2% on PASCAL VOC 2012 and 79.7% on Cityscapes dataset, respectively.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.613-618
• /
• 1992

We propose a new control architecture which consists of a PI controller and a neural network(NN) controller connected together in parallel. This architecture is well adapted to a wide range of uncertainties and variations of systems. The NN controller is learned through weights of the emulator which identify the dynamic chracteristics of the systems. A performance evaluation level of two NN's decides automatically which controller of the two controllers will be used mainly. The PI controller operates mainly during learning phase of the NN controller whereas a good performance is obtained from the NN controller only, when the NN controller is learned sufficiently.

• Proceedings of the KSR Conference
• /
• 2011.10a
• /
• pp.2782-2786
• /
• 2011

In this paper a protocol design for bus network communication between onboard signalling system and MMI(Man Machine Interface) will be presented and illustrated. Recently, many onboard signailling systems adopt hot standby for safety reasons. Hot standby is a method of redundancy in which the primary and secondary systems run simultaneously. It is convenient to use bus network(bus topology) in a hot standby system for communication between onboard signalling system and MMI. Because bus network is the simplest way to connect multiple clients such as onboard signalling system, MMI and etc. However, there are many problems when two clients want to transmit at the same time on the same bus. A effective protocol is necessary to solve that problems. We will describes protocol design which is useful when onboard signalling systems and MMIs are connected via RS485(Bus Network).

• Proceedings of the KIEE Conference
• /
• 1993.07a
• /
• pp.231-234
• /
• 1993

This paper deals with the design of an adaptive controller using neural network. We present RBFMLP Neural Network which consists of serial-connected two networks - Radial Basis Function Network and Multi Layer Perceptron, and then design a controller based on proposed networks with the adaptive control system structure, The plant and parameters of the controller are identified by the neural networks. We use the dynamic backpropagation algorithm for the learning of networks. Simulations represent the superiorities of the proposed network and the controller.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.29 no.2 s.233
• /
• pp.303-310
• /
• 2005

The systems with probe and SIL(Solid Immersion Lens) mechanisms have been researched as the technology to perform NFR(Near Field Recording). Most of them use the flying head mechanism to accomplish high recording density and fast data transfer rate. In this paper, ABS shape of flying head was optimized with the object of securing the maximum compliance ability of OFH. We suggest low different optimization processes to predict the static flying characteristics for the OFH. Two different approximation methods, regression analysis and back propagation neural network were used. And we compared the result of directly connected(between CAE and optimizer) method and two approximated optimization results. Design Optimization Tool(DOT) and ${\mu}GA$ were used as the optimizers.

• Journal of Power Electronics
• /
• v.16 no.5
• /
• pp.1821-1832
• /
• 2016

This study presents a strategy using the synchronized output regulation method (SOR) for controlling inverters operating in stand-alone and grid-connected modes. From the view point of networked dynamic systems, SOR involves nodes with outputs that are synchronized but also display a desirable wave shape. Under the SOR strategy, the inverter and grid are treated as two nodes that comprise a simple network. These two nodes work independently under the stand-alone mode. An intermediate mode, here is named the synchronization mode, is emphasized because the transition from the stand-alone mode to the grid-connected mode can be dealt as a standard SOR problem. In the grid-connected mode, the inverter operates in an independent way, in which the voltage reference changes for generalized synchronization where its output current satisfies the required power injection. Such a relatively independent design leads to a seamless transfer between operation modes. The closed-loop system is analyzed in the state space on the basis of the output regulation theory, which improves the robustness of the design. Simulations and experiments are performed to verify the proposed control strategy.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.4
• /
• pp.832-839
• /
• 2009

In this paper, we introduce an design of fuzzy relation-based fuzzy neural networks with multi-output. Fuzzy relation-based fuzzy neural networks comprise the network structure generated by dividing the entire input space. The premise part of the fuzzy rules of the network reflects the relation of the division space for the entire input space and the consequent part of the fuzzy rules expresses three types of polynomial functions such as constant, linear, and modified quadratic. For the multi-output structure the neurons in the output layer were connected with connection weights. The learning of fuzzy neural networks is realized by adjusting connections of the neurons both in the consequent part of the fuzzy rules and in the output layer, and it follows a back-propagation algorithm. In addition, in order to optimize the network, the parameters of the network such as apexes of membership functions, learning rate and momentum coefficient are automatically optimized by using real-coded genetic algorithm. Two examples are included to evaluate the performance of the proposed network.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.4
• /
• pp.377-384
• /
• 2004

Nowadays, control systems consist of smart sensors, smart actuators, and controllers connected via fieldbus. Some devices such as motors in plant environments generate high degrees of EMI or noise. This noise may cause communication errors and make the successful transmission of data longer. Therefore, the noise condition has to be considered at the design of a reliable control system based on a network. This paper presents a scheduling method of task and message to guarantee the given end-to-end constraints under noise environments. A noise model with multi-sources of noise is used, and the analysis method of message's response time is presented when the noise model is applied to CAN (Controller Area Network). Two kinds of noise models are applied to an example system, and the effect to each control loop s end-to-end response time is analyzed. We believe that the proposed method help system designers design the control system guaranteeing its requirements under noise environment.