• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.9
• /
• pp.1612-1617
• /
• 2008

Since most disturbance observer (DOB) approaches have been limited to minimum-phase systems (or systems having no zero dynamics), we propose a new DOB structure that can be applied to non-minimum phase systems. The new structure features an additional system, which is called as V-filter, whose role is to yield a minimum phase system when connected with the plant in parallel. In order to design the V-filter systematically we first consider a class of linear systems that can be stabilized via PID controller. By inverting the controller's transfer function, we can simply construct the filter. A convenient way of designing V-filter is presented by using an iterative linear matrix inequality (LMI) algorithm. With an illustrative example the simulation result shows that substantial improvement in the performance has been achieved compared with the control system without the DOB.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.9
• /
• pp.1221-1225
• /
• 2012

This paper describes CSC(Current Sourced Converters)-based HVDC operational strategy for voltage stability enhancement in the power system. In case of CSC-based HVDC system, rectifier and inverter consume reactive power up to about 60% of converter rating. Therefore, CSC-based HVDC is basically not useful system for voltage stability even if AC filters and shunt capacitors are attached. But, If the particular power system condition is fulfilled, CSC-based HVDC also can be the rapid reactive power source for voltage stability enhancement using a cooperative control with converter and AC filters/Shunt Capacitors. In this paper, the cooperative control algorithm is presented and simulated to ${\pm}80kV$ 60MW HVDC system in Jeju island.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.1-15
• /
• 2022

The MobileNetV3 is specially designed for mobile devices with limited memory and computing power. To reduce the network parameters and improve the network inference speed, a new lightweight network is proposed based on MobileNetV3. Firstly, to reduce the computation of residual blocks, a partial residual structure is designed by dividing the input feature maps into two parts. The designed partial residual structure is used to replace the residual block in MobileNetV3. Secondly, a dual-path feature extraction structure is designed to further reduce the computation of MobileNetV3. Different convolution kernel sizes are used in the two paths to extract feature maps with different sizes. Besides, a transition layer is also designed for fusing features to reduce the influence of the new structure on accuracy. The CIFAR-100 dataset and Image Net dataset are used to test the performance of the proposed partial residual structure. The ResNet based on the proposed partial residual structure has smaller parameters and FLOPs than the original ResNet. The performance of improved MobileNetV3 is tested on CIFAR-10, CIFAR-100 and ImageNet image classification task dataset. Comparing MobileNetV3, GhostNet and MobileNetV2, the improved MobileNetV3 has smaller parameters and FLOPs. Besides, the improved MobileNetV3 is also tested on CPU and Raspberry Pi. It is faster than other networks

• Proceedings of the KIEE Conference
• /
• 1997.11a
• /
• pp.544-549
• /
• 1997

Injector of high voltage or linear ion accelerator is intended to generate, extract and form beam of certain species with required parameters at the entrance of accelerating structure or, for low energy case, directly in the processing chamber (end station). Injector is the main part defining the ion accelerator performance and reliability. Its power supply and control system (PSCS) features are conditioned by placing the injector equipment at high voltage potential and by complexity of the plasma-beam load. The injector's PSCS should provide: - Transmission of electric power onto high voltage (h/v) terminal; - Obtaining of required output characteristics for injector equipment operation; - Transmission of the operational data and start/stop signals from h/v terminal to control cabinet; - Rremote control of injector; - Withstanding the high voltage breakdowns and X-ray radiation; - Compatibility with other equipment. The paper is concerned with analysis of injectors' PSCS structure and description of the system developed for 50 keV, 20 mA heavy ion injector.

• Proceedings of the KIPE Conference
• /
• 1998.10a
• /
• pp.352-356
• /
• 1998

This paper describes minimum current control method for size reduction and performances improvement at unified inverter-induction motor system. This control method is based on V/F without speed sensor. Through the use of minimum stator current points at required torque during V/F operation it is possible to minimize the size of heat-sink related with the losses of power circuit and to improve overall efficiency compared to conventional V/F control. Using this proposed scheme, it is so much more useful to apply to some fields such a selectric vehicles. air conditioning system and textile mills where the limited space is given and required low cost.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.5 no.4
• /
• pp.276-281
• /
• 2005

This paper presents a low voltage operating IF QPSK receiver block which is consisted of programmable gain amplifier (PGA) and analog to digital converter. This PGA has 6 bit control and 250MHz bandwidth, $0{\sim}20\;dB$ gain range. Using the proposed PGA architecture (low distortion gain control switch block), we can process the continuous fully differential $0.2{\sim}2.5Vpp$ input/output range and 44MHz carrier with 2 MHz bandwidth signal at 1.8V supply voltage. Using the sub-sampling technique (input freq. is $44{\sim}46MHz$, sampling freq. is 25MHz), we can process the IF QPSK signal ($44{\sim}46MHz$) which is the output of the 6 bit PGA. We can get the SNDR 35dB, which is the result of PGA and ADC at full gain mode. We fabricated the PGA and ADC and the digital signal processing block of the IF QPSK with the 0.18um CMOS MIM process 1.8V Supply.

• Journal of Energy Engineering
• /
• v.20 no.3
• /
• pp.216-223
• /
• 2011

This paper presents an effective energy saving algorithm for lighting control systems using M&V method in various buildings and factories. It is important to aggregate a various demand side resource which is controllable at the peak power time to reduce the energy cost. Previous demand side algorithm appropriate for building is based on peak power. In this paper, we develop the new energy saving algorithm using M&V method to reduce the quantity of power consumption. The simulation results show that the proposed method is very effective.

• Proceedings of the KIPE Conference
• /
• 2004.07a
• /
• pp.446-450
• /
• 2004

The amount of electric energy used in a vehicle will be increased continuously. The increment of electric power demand causes interest on new higher power system such as 42V Power Net. Furthermore, the necessity for development of energy storage device is highlighted recently. Bidirectional nm Converter is one of the important parts in 42V power system. Therefore, this paper proposes bidirectional Cascade Buck-Boost DC/DC Converter which can satisfies required specifications in 42V power system The operation principle is described along with simple control method, and experimental results on a 500W prototype are provided.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.55 no.1
• /
• pp.7-12
• /
• 2006

This paper proposes the new voltage event detection method using the weight factor of neural network and describes control system design for the DVR(Dynamic Voltage Restorer) consisted of a rectifier and series inverter applied to 22.9kV distribution system. As this method can express the fault level of each phase, we expect the proposed method can make up for disadvantage of synchronous detection method. Also, in this paper, the control system was designed using double deadbeat controller, As it has an inner current control loop and an outer voltage control loop, we can easily limit the current level during the transient intervals by using the current control loop. Simulation and experiment are performed to prove the analysis of the voltage event detection method and double deadbeat controller.

• Proceedings of the KIPE Conference
• /
• 2014.11a
• /
• pp.137-138
• /
• 2014

우리나라 전력계통은 점점 복잡화, 대형화되어 가고 있으나, 이에 따른 전력설비의 신설은 날로 어려워지고 있다. 이러한 변화에 능동적으로 대처하기 위해 FACTS(Flexible AC Transmission System) 기술이 주목받고 있으며, FACTS는 대용량 전력전자기술을 이용하여 송전선로의 안정도 제어 및 송전용량 증대를 목적으로 설치된다. 현재 우리나라 전력계통에는 SVC, UPFC, STATCOM 등의 기기들이 개발되어 운영되어 있으며, 그 중에 SVC가 대표적으로 변전소에 많이 설치되어 무효전력 보상으로 전압안정도를 향상시키고 있다. 이에 따라서 본 논문은 154kV 송전계통에서 SVC의 여러 제어 방법 중 Susceptance 제어에 관해서 기술하고자 한다.