https://doi.org/10.5370/KIEE.2024.73.9.1499

곽효근(Hyo-Geun Kwak) ; 이효(Hyo Lee) ; 심재웅(Jaewoong Shim)

This paper introduces the method of avoiding control interaction conditions by using virtual impedance control and connecting a R-damper to the capacitor section of an inverter containing an LCL filter connected to the power system. For this purpose, we set the parameters of the inverter and LCL filter. Then, equations were employed to appropriately determine the values of the damping resistor of the filter and virtual impedance. Utilizing these configured values, the overall transfer function was represented using the transfer functions of the inverter and filter. Subsequently, it was examined whether unstable resonance, as compared to the system impedance, could be avoided when using the virtual impedance control and adding resistor to filter. Furthermore, in the frequency domain, We assess how much the voltage harmonics decreased when the virtual impedance and damping resistor was included, compared to when only the LCL filter was present.

https://doi.org/10.5370/KIEE.2024.73.9.1507

윤상철(Sang-Cheol Yun) ; 김병호(Byoungho Kim) ; 김홍래(Hongrae Kim)

In the current situation of increasing power demand, research is needed for better electricity demand forecasting. This study presents a new approach to enhance the accuracy of electricity demand forecasting. The objective of this study is to facilitate more precise Load Forecasting by incorporating the Attention Mechanism into the existing deep learning models, LSTM and GRU. For this purpose, LSTM and GRU models combined with Attention Mechanism were applied to actual power demand data. The experimental results confirmed that the proposed model significantly improved the accuracy of power demand prediction compared to the existing models. These results show that the addition of Attention Mechanism can contribute to improving the performance of deep learning-based power demand prediction models.

https://doi.org/10.5370/KIEE.2024.73.9.1513

유솔의(Solui Yu) ; 허진(Jin Hur)

Wind power generation, as a pivotal resource of renewable energy, is emerging as a leading force in carbon neutral, with its share in the global energy mix experiencing rapid growth. However, the inherent volatility of wind power output poses challenges, leading to potential transmission operating constraint resulting from sudden output fluctuations. This paper presents a methodology for critical transmission operation constraints forecasting (COCF) on a monthly and hourly basis, simulating scenarios involving large-scale wind power generation integrated into the power system. The proposed approach was validated through its application to the Jeju power system, where transmission thermal limits on critical lines were forecasted using actual power load and wind power capacity factor database. The insights gained from this methodology offer valuable perspectives for assessing power grid reliability concerning wind power integration and establishing response systems to uphold overall system stability.

https://doi.org/10.5370/KIEE.2024.73.9.1519

정인주(Injoo Jeong) ; 오승찬(Seungchan Oh) ; 구현근(Hyunkeun Ku) ; 정솔영(Solyoung Jung) ; 이윤선(Yoonseon Lee) ; 이재형(Jaehyeong Lee) ; 이재걸(Jaegul Lee)

High penetration of renewable energy resources affects several stability problems. There are lots of stability concepts, one of them which is strongly related to renewable energy resources is called the voltage stability. As shown in this paper, voltage instability in transient can be occurred by renewable energy generation increasing. It is necessary to analyze this phenomenon and how to solve the problems effectively. This paper shows how this phenomenon can be interpreted and analyzed according to load model such as constant power model and ZIP model, which is using a specific methodology called YV analysis. Also, by evaluating power flow in the power system, it shows the specific cause of the voltage instability in a specific location. Using the above results, it is suggested how voltage instability can be effectively improved and solved through simulation.

https://doi.org/10.5370/KIEE.2024.73.9.1528

이재걸(Jaegul Lee) ; 정솔영(Solyoung Jung) ; 이용승(Yongseung Lee) ; 오승찬(Seungshan Oh) ; 배문성(Moonsung Bae) ; 신정훈(Jeonghoon Shin)

In this paper, previous studies on the method of quantitatively evaluating the grid resilience were investigated. The grid resilience of the test system was evaluated for the large-scale forest fire scenario by applying the some selected grid resilience evaluation methodologies. In addition, the effectiveness of each countermeasure to minimize the power supply disruption due to large-scale forest fires was evaluated as an improvement in the level of grid resilience. The contribution of this paper is a quantitative evaluation of the resilience of the Korean power grid for large-scale forest fire scenarios, and through this, we found that the evaluation indicators developed in previous studies are partially insufficient in evaluating the resilience of the Korean power grid.

https://doi.org/10.5370/KIEE.2024.73.9.1535

신건(Jian Shen) ; 유현상(Hyun-Sang You) ; 박찬욱(Chan-Wook Park) ; 최성문(Sung-Moon Choi) ; 노대석(Dae-Seok Rho)

The estimation method of the lifespan in VRFB for ESS based on general operation data is required as the evaluation on deterioration characteristics is time-consuming and costly. Therefore, this paper implements a deterioration cycle test device of VRFB for ESS, which is composed of VRFB, internal impedance measurement device, monitoring and control sections, and performs accelerated aging test for 2,500 cycles and analyzes the deterioration characteristics of charging-discharging energy efficiency, internal impedance, slope of OCV variation and potentio-metric titration, which are the key index to evaluate the deterioration degree in VRFB. Based on the above-mentioned test results, deterioration prediction models for VRFB by mathematical formalization using linear regression method are proposed. From the simulation result for 3,000 cycles based on the proposed model, it is found that the accuracy of the degradation prediction model is appropriate since the actual measurement values of deterioration index and the simulation results from proposed models are nearly identical, and the proposed deterioration prediction model is calculated as approximately 2,975 cycles compared to the 3,000 cycles measurement.

https://doi.org/10.5370/KIEE.2024.73.9.1545

신명호(Myoung-Ho Shin)

In conventional constant V/f control of a permanent magnet synchronous motor, it is well known that maximum torque at a given frequency remains constant in spite of change of frequency. However, maximum torque at a given frequency does not remain constant as frequency varies since stator resistance is neglected. This paper shows that the maximum torque of conventional constant V/f control decreases as frequency decreases. In the proposed method, a voltage for scalar control of a permanent magnet synchronous motor considering stator resistance is presented and it is shown that maximum torque at a given frequency remains constant in spite of change of frequency. In simulation results, it is shown that speed control under load is not achieved by constant V/f control because of the decrease of torque level as frequency decreases. In addition, it is shown that motor starting is achieved under load and speed is regulated with the change of load torque in the proposed method. Further, it is also seen that speed reversal operation is achieved in the proposed method. In conventional constant V/f control of a permanent magnet synchronous motor, it is well known that maximum torque at a given frequency remains constant in spite of change of frequency. However, maximum torque at a given frequency does not remain constant as frequency varies since stator resistance is neglected. This paper shows that the maximum torque of conventional constant V/f control decreases as frequency decreases. In the proposed method, a voltage for scalar control of a permanent magnet synchronous motor considering stator resistance is presented and it is shown that maximum torque at a given frequency remains constant in spite of change of frequency. In simulation results, it is shown that speed control under load is not achieved by constant V/f control because of the decrease of torque level as frequency decreases. In addition, it is shown that motor starting is achieved under load and speed is regulated with the change of load torque in the proposed method. Further, it is also seen that speed reversal operation is achieved in the proposed method.

https://doi.org/10.5370/KIEE.2024.73.9.1551

김준선(Jun-Seon Kim) ; 서동욱(Dong-Wook Seo)

Due to the characteristic micro-motion of space-targets, a micro-Doppler effect occurs when a radar observes the space-target. The two-dimensional micro-Doppler signature image, which well represents the micro-Doppler effect, is used as a feature in classification using convolutional neural network in many literature. The angle of incidence of electromagnetic waves incident on a space-target is one of the information that can be obtained during radar observation. In this paper, we propose a method to improve the performance of the space-target classifier by using this angle of incidence as a feature in training a convolutional neural network model. The angle of incidence is input to the fully connected layer by concatenating it with the feature maps that are the output of the convolutional layer, and this was applied to ResNet-18 and a simple convolutional neural network model. Although the performance improvement in ResNet-18 was small compared to the simple model, it was clear in all cases, and classification accuracy was especially improved at low SNR. When the weight of the angle of incidence was set large, the F1-score showed a larger increase than when the dwell time was doubled, showing that it can be efficiently applied to space-target classification where decision-making must be completed within a short time.

https://doi.org/10.5370/KIEE.2024.73.9.1561

노태영(Taeyeong Noh) ; 황선남(Sunnam Hwang) ; 김경호(Kyung-Ho Kim)

This design is about research on how to increase portability and practicality by effectively reducing the volume of RF antennas used to detect arcs in the number/distribution panel in the arc generation detection system. By installing RF TRANSFORMER on a BAR-type antenna PCB, the volume of the RF antenna PCB can be effectively reduced and the  value has been reduced. In addition, through arc detection experiments, it was confirmed that the arc data of different RF antenna PCBs were the same, and as a result, the performance of the two was same. This study will be used in future arc generation detection systems and will be further developed to detect arcs more effectively.

https://doi.org/10.5370/KIEE.2024.73.9.1568

김동규(Dong-Gue Kim) ; 조재걸(Jaegeol Cho)

In hospitals, continuous blood pressure is measured by inserting of a catheter into a blood vessel, which is an accurate technique, but leads to damaged blood vessels. To address this drawback, the Volume-Clamping method was introduced as an alternative approach for non-invasive and continuous blood pressure measurement. However, this approach employs complex PID control algorithms for finger cuff, leading to high costs and bulky apparatus. Therefore, we propose a beat-by-beat blood pressure monitoring system in the form of a wrist sphygmomanometer, wherein the pressure-volume relationship of the radial artery is determined through oscillometric measurements. Beat-by-beat blood pressures are estimated based on the artery pressure-volume relationship and artery volume variations obtained from the wrist cuff. Research findings indicate that the errors in systolic and diastolic blood pressure are -3.4 ± 7.6[mmHg], -3.7 ± 5.3[mmHg] respectively, meeting the standards set by the Association for the Advancement of Medical Instrumentation(AAMI). This system has the potential to be reproduced using a simple mechanism involving a solenoid valve, transforming it into a wrist blood pressure sphygmomanometer that offers convenient blood pressure monitoring not only within medical facilities but also in daily life.

https://doi.org/10.5370/KIEE.2024.73.9.1574

홍채희(Chaehui Hong) ; 조현지(Hyunji Cho) ; 유훈(Hoon Yoo)

This paper presents a method for automatically identifying the encoding and language of documents. In the online world, a technique for document encoding and language identification plays an important role in providing users with easy access to the information they need, and in improving the efficiency of data processing and analysis. In this paper, the proposed method first identifies whether a document is encoded in UTF-8 or not by analyzing the bit pattern. For the UTF-8, the language is identified by calculating the percentage of each language in the document through Unicode range analysis. If the document is found to be not UTF-8, it is determined to be a code page document, and the languages in the document are identified by our machine learning technique using N-grams. To evaluate the proposed method, we conducted experiments. The experimental results indicate that the proposed method improves the encoding and language identification performance compared to the existing methods.

https://doi.org/10.5370/KIEE.2024.73.9.1581

심규석(Kyu-Seok Shim) ; 오효석(Hyo-Seok Oh) ; 장진영(Chin-Young Chang) ; 김재문(Jae-Moon Kim)

The maintenance of railway vehicles is currently performed based on Time Based Maintenance (TBM), which is maintained according to a predetermined driving distance and time. However, the TBM system has problems such as breakdowns during vehicle operation due to manufacturing errors and usage environment, or replacement in a state where the remaining life is left. To solve these problems, it is necessary to introduce Condition Based Maintenance (CBM), which can analyze and diagnose the aging or abnormal state of electronic products in real time. Power conversion devices are used to supply propulsion and various loads of railway vehicles, but they have a direct impact on vehicle operation due to frequent failures. The main cause of failure of the power conversion device is a defect in the capacitor, and it is important to estimate the state of the capacitor to prevent failure. In this paper, we have developed a DAQ (Data Acquisition) device that can be installed on existing vehicles for capacitor state estimation and monitoring, and proposed a technique that can estimate the ESR and capacitance of the capacitor through one algorithm in various SIV (Static Inverter) circuits. To verify this, we conducted an experiment by configuring an AC/DC converter, and as a result, we confirmed the validity of the capacitor state estimation system with a maximum error of 2.54%.

https://doi.org/10.5370/KIEE.2024.73.9.1588

강태형(Tae-Hyung Kang) ; 방준호(Jun-Ho Bang) ; 유인호(In-Ho Ryu)

The switchgear is a device which receives the high voltage electricity from the generator it into the useable voltage for the consumers. It is an infrastructure can provide the useable voltage from the households to the big facilities throughout our society. But If a blackout emergency occurs, it can cause serious damage. Altuogh the regular inspection of the switchgear is required to prevent the problems and for the safe maintenance, it is very hard because of some factors like non-standardized equipments, dangerous and narrow working spaces and so on. In this paper, we are about to talk about possibility of adapting the thermal imaging camera for the safer and more effecient monitoring and surveillance. We have developed a deep learning-based thermal diagnostic system which can check the information of the main equipments. It also can detect the status data varying by the temperature and notify an alarm to the manager.

https://doi.org/10.5370/KIEE.2024.73.9.1595

김정주(Jung-Ju Kim) ; 이우열(U-Yeol Lee) ; 조경준(Kyeong-Jun Cho) ; 김동진(Dong-Jin Kim)

As social distancing was lifted after the COVID-19 pandemic, participation in sports for the people increased. Among them, the number of people enjoying tennis increased the most. Tennis is a sport that allows individual lessons or individual practice if not played, and balls must be collected after practice and lessons. The People find it very inconvenient and bothersome to collect the ball after practice. In this paper, a robot that automatically collects tennis balls was developed to solve the inconvenience of collecting tennis balls. It was developed so that the robot could collect the ball through autonomous driving by detecting and recognizing the tennis ball using an image processing algorithm. The developed robot can increase the convenience of people who enjoy tennis.

https://doi.org/10.5370/KIEE.2024.73.9.1602

한정원(Jeong-Won Han) ; 이현재(Hyun-Jae Lee) ; 손진근(Jin-Geun Shon)

This paper, we conducted a study on methods for improving the power production efficiency of photovoltaic power systems using machine learning. The InC method, one of the MPPT methods used to improve the power production efficiency of photovoltaic power systems, determines the power production efficiency by the size of the slope, and to compensate for the disadvantages that occur in this process, we propose an OS MPPT (Optimized Slope MPPT) method that optimizes the constant value that α determines the size of the slope by introducing the optimization method of machine learning. The effects of the OS MPPT method were analyzed through simulation experiments and field experiments. The results of comparing the calculated values by accumulating the produced power showed that the performance of the OS MPPT control method was effective. It is expected to contribute to the promotion of power production when applied to photovoltaic power systems.