Due to COVID-19, changes in consumption trends are taking place in the distribution sector, such as an increase in non-face-to-face consumption and a rapid growth in the online shopping market. However, it is difficult for small and medium-sized export sellers to obtain forecast information on the export market by country, compared to large distributors who can easily build a global sales network. This study is about the prediction of export amount and export volume by country and item for market information analysis of small and medium export sellers. A prediction model was developed using Lasso, XGBoost, and MLP models based on supervised learning and deep learning, and export trends for clothing, cosmetics, and household electronic devices were predicted for Korea's major export countries, the United States, China, and Vietnam. As a result of the prediction, the performance of MAE and RMSE for the Lasso model was excellent, and based on the development results, a market analysis system for small and medium sellers was developed.

The large process plant is currently implementing predictive maintenance technology to transition from the traditional Time-Based Maintenance (TBM) approach to the Condition-Based Maintenance (CBM) approach in order to improve equipment maintenance and productivity. The traditional techniques for predictive maintenance involved managing upper/lower thresholds (Set-Point) of equipment signals or identifying anomalies through control charts. Recently, with the development of techniques for big analysis, machine learning-based AAKR (Auto-Associative Kernel Regression) and deep learning-based VAE (Variation Auto-Encoder) techniques are being actively applied for predictive maintenance. However, this predictive maintenance techniques is only effective during steady-state operation of plant equipment, and it is difficult to apply them during start-up and shutdown periods when rises or falls. In addition, unlike processes such as nuclear and thermal power plants, which operate for hundreds of days after a single start-up, because the pumped power plant involves repeated start-ups and shutdowns 4-5 times a day, it is needed the prediction and alarm algorithm suitable for its characteristics. In this study, we aim to propose an approach to apply the optimal predictive alarm algorithm that is suitable for the characteristics of Pumped Storage Power Plant(PSPP) facilities to the system by analyzing the predictive maintenance techniques used in existing nuclear and coal power plants.

Recently in Korea, YouTube stock channels increased rapidly due to the high social interest in the stock market during the COVID-19 period. Accordingly, the role of new media channels such as YouTube is attracting attention in the process of generating and disseminating market information. Nevertheless, prior studies on the market forecasting power of YouTube stock channels remain insignificant. In this study, the market forecasting power of the information from the YouTube stock channel was examined and compared with traditional news media. To measure information from each YouTube stock channel and news media, positive and negative opinions were extracted. As a result of the analysis, opinion in channels operated by media outlets were found to be leading indicators of KOSPI market returns among YouTube stock channels. The prediction accuracy by using logistic regression model show 74%. On the other hand, Sampro TV, a popular YouTube stock channel, and the traditional news media simply reported the market situation of the day or instead showed a tendency to lag behind the market. This study is differentiated from previous studies in that it verified the market predictive power of the information provided by the YouTube stock channel, which has recently shown a growing trend in Korea. In the future, the results of advanced analysis can be confirmed by expanding the research results for individual stocks.

Numerous studies have attempted to predict the energy output of solar-powered vehicles based on different parameters such as road conditions, driver characteristics, and weather. However, since these studies were conducted on stationary vehicles, they are limited in their accuracy when applied to driving vehicles. This study aimed to improve the accuracy of electric power prediction for a solar-powered bus by applying a technique that improves energy efficiency without affecting driving performance. A comparative analysis of power generation and solar irradiance data was conducted for the bus driven on different roads to forecast its power generation, and a high-accuracy power generation prediction equation was derived. A comparison with actual test results revealed that a power generation forecast accuracy of at least 90% was achieved, validating the equation used for forecasting. With this power generation prediction process, it is possible to forecast the amount of energy generated in advance when a solar bus is operated in a specific area.

The cutting process, which is a key processing technology in various industrial fields is achieving continuous growth, and the demand for high-quality cutting surfaces is continuously demanded. Plasma cutting continues to be studied for its excellent workability and productivity, but problems with cutting surface quality such as dross formation occur, so research to secure excellent cutting surface quality through appropriate control of process variables is essential. In this study, we propose a method for predicting surface roughness using real-time current and cutting speed data obtained while performing plasma cutting on A106 B steel pipe. Surface roughness was predicted based on the RBF algorithm applicable to prediction and control models. It was shown that the surface roughness of the plasma cutting surface can be predicted with the arc current waveform and process speed data. This study can be used as a basic study to control the surface roughness of the cut surface in real time.

Aluminum alloy-based additive manufacturing (AM) has emerged as a popular manufacturing process for the fabrication of complex parts in the automotive and aerospace industries. The addition of an inoculant to aluminum alloy powder has been demonstrated to effectively reduce cracking by promoting the formation of equiaxed grains. However, the optimization of the AM process parameters remains challenging owing to their variability. In this study, the response surface methodology (RSM) was used to predict the crack density of AM-processed Al alloy samples. RSM was performed by setting the process parameters and equiaxed grain ratio, which influence crack propagation, as independent variables and designating crack density as a response variable. The RSM-based quadratic polynomial models for crack-density prediction were found to be highly accurate. The relationship among the process parameters, crack density, and equiaxed grain fraction was also investigated using RSM. The findings of this study highlight the efficacy of RSM as a reliable approach for optimizing the properties of AM-processed parts with limited experimental data. These results can contribute to the development of robust AM processing strategies for the fabrication of highquality Al alloy components for various applications.

건설기계와 특장차의 상부와 하부 구조를 연결하는 로터리 조인트는 축과 하우징이 회전하면서 유압을 전달하는데 오일의 유로에 누유를 방지하기 위해 여러 개의 시일이 조립된다. 시일재료는 강성이 커서 조립에 어려움이 있기 때문에 자른 후 조립하는 방법을 모 색하였다. 절단면의 모양은 L형과 /형으로 하였고 유압이 작용할 때 누유 기준은 절단면에 발생하는 접촉압력으로 하였다. 시일의 구 조와 재료는 이중 탄성중합체로 구성되며 강성이 큰 PE 재질만 절단한 경우에 대하여 비선형 접촉 구조해석을 수행하였다. 연구결과 절단 길이가 짧을수록 누유 방지에 유리하며 PE와 하우징이 접촉하는 윗면보다 NBR과 PE가 접촉하는 아랫면으로 누유될 가능성이 큰 것으로 나타났다.

The entire industry is increasing the use of big data analysis using artificial intelligence technology due to the Fourth Industrial Revolution. The value of big data is increasing, and the same is true of the production technology. However, small and medium -sized manufacturers with small size are difficult to use for work due to lack of data management ability, and it is difficult to enter smart factories. Therefore, to help small and medium -sized manufacturing companies use big data, we will predict the gross production time through machine learning. In previous studies, machine learning was conducted as a time and quantity factor for production, and the excellence of the ExtraTree Algorithm was confirmed by predicting gross product time. In this study, the worker's proficiency factors were added to the time and quantity factors necessary for production, and the prediction rate of LightGBM Algorithm knowing was the highest. The results of the study will help to enhance the company's competitiveness and enhance the competitiveness of the company by identifying the possibility of data utilization of the MES system and supporting systematic production schedule management.

Gate valves are hydraulic components used to shut-off the water flow in water distribution systems. Gate valves may fail owing to various aspects such as leakage through seats, wearing of packing, and corrosion. Because it is considerably challenging to detect valve malfunctioning until the operator identifies a significant fault, failure of the gate valve may lead to a severe accident event associated with water distribution systems. In this study, we proposed a methodology to diagnose the faults of gate valves. To measure the pressure difference across a gate valve, two pressure transducers were installed before and after the gate valve in a pilot-scaled water distribution system. The obtained time-series pressure difference data were analyzed using a machine learning algorithm to diagnose faults. The validation of whether the flow rate of the pipeline can be predicted based on the pressure difference between the upstream and downstream sides of the valve was also performed.

A low- and intermediate-level radioactive waste repository contains different types of radionuclides and organic complexing agents. Their chemical interaction in the repository can result in the formation of radionuclide-ligand complexes, leading to their high transport behaviors in the engineered and natural rock barriers. Furthermore, the release of radionuclides from the repository can pose a significant risk to both human health and the environment. This study explores the impact of different experimental conditions on the transport behaviors of 99Tc, 137Cs, and 238U through three types of barrier samples: concrete, sedimentary rock, and granite. To assess the transport behavior of the samples, the geochemical characteristics were determined using X-ray diffraction (XRD), X-ray fluorescence (XRF), Fouriertransform infrared spectroscopy (FTIR), scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDS), and Brunauer-Emmett-Teller (BET) analysis. The adsorption distribution coefficient (Kd) was used as an indicator of transport behavior, and it was determined in batch systems under different conditions such as solution pH (ranging from 7 to 13), temperature (ranging from 10 to 40°C), and with the presence of organic complexing agents such as ethylenediaminetetraacetic acid (EDTA), nitrilotriacetic acid (NTA), and isosaccharinic acid (ISA). A support vector machine (SVM) was used to develop a prediction model for the Kd values. It was found that, regardless of the experimental parameters, 99Tc may migrate easily due to its anionic property. Conversely, 137Cs showed low transport behaviors under all tested conditions. The transport behaviors of 238U were impacted by the order of EDTA > NTA> ISA, particularly with the concrete sample. The SVM models can also be used to predict the Kd values of the radionuclides in the event of an accidental release from the repository.

The organic complexing agents such as ethylenediaminetetraacetic acid (EDTA), nitrilotriacetic acid (NTA), and isosaccharinic acid (ISA) can enhance the radionuclides’ solubility and have the potential to induce the acceleration of radionuclides’ mobility to a far-field from the radioactive waste repository. Hence, it is essential to evaluate the effect of organic complexing agents on radionuclide solubility through experimental analysis under similar conditions to those at the radioactive waste disposal site. In this study, five radionuclides (cesium, cobalt, strontium, iodine, and uranium) and three organic complexing agents (EDTA, NTA, and ISA) were selected as model substances. To simulate environmental conditions, the groundwater was collected near the repository and applied for solubility experiments. The solubility experiments were carried out under various ranges of pHs (7, 9, 11, and 13), temperatures (10°C, 20°C, and 40°C), and concentrations of organic complexing agents (0, 10-5, 10-4, 10-3, and 10-2 M). Experimental results showed that the presence of organic complexing agents significantly increased the solubility of the radionuclides. Cobalt and strontium had high solubility enhancement factors, even at low concentrations of organic complexing agents. We also developed a support vector machine (SVM) model using some of the experimental data and validated it using the rest of the solubility data. The root mean square error (RMSE) in the training and validation sets was 0.012 and 0.016, respectively. The SVM model allowed us to estimate the solubility value under untested conditions (e.g., pH 12, temperature 30°C, ISA 5×10-4 M). Therefore, our experimental solubility data and the SVM model can be used to predict radionuclide solubility and solubility enhancement by organic complexing agents under various conditions.

In the event of a radioactive release, it is essential to quickly detect and locate the source of the release, as well as track the movement of the plume to assess the potential impact on public health and safety. Fixed monitoring posts are limited in their ability to provide a complete picture of the radiation distribution, and the information they provide may not be available in real-time. This is why other types of monitoring systems, such as mobile monitoring, aerial monitoring, and personal dosimeters, are also used in emergency situations to complement the information provided by fixed monitoring posts. Also, the monitoring system can be improved by using the Kriging technique, which is one of the interpolation methods, to predict the radiation dose in the relevant districts. This can be achieved by utilizing both the GPS information and the radiation dose measured at a particular point. The Kriging method involves estimating the value between different measurement points by considering the distance between them. The model used GPS and radiation data that were measured around the Hanbit NPP. The data were collected using a radiation measuring detector on a bus that traveled around the NPP area at 2-second intervals for one day. From the collected data, 200 data points were randomly selected for analysis, excluding the data measured at the bus garage out of a total of 16,550 data points. The average dose of the daily measurement data was 117.94 nSv/h, and the average dose of the 200 randomly extracted data was 119.17 nSv/h. The GPS and radiation dose data were utilized to predict the radiation dose around the Yeonggwang area where the Hanbit NPP is located. In the event of an abnormal release of radioactive material, it can be difficult to accurately determine the dose unless a monitoring measurement point is present. This can delay the rapid evacuation of residents during an emergency situation. By utilizing the Kriging model to make predictions, it is anticipated that more accurate dose predictions can be generated, particularly during accident scenarios. This can aid in the development of appropriate resident protection measures.

Radioactive contaminants, such as 137Cs, are a significant concern for long-term storage of nuclear waste. Migration and retention of these contaminants in various environmental media can pose a risk to the surrounding environment. The distribution coefficient (Kd) is a critical parameter for assessing the behavior of these contaminants and can introduce significant errors in predicting migration and remediation options. Accurate prediction of Kd values is essential to assess the behavior of radioactive contaminants and to ensure environmental safety. In this study, we present machine learning models based on the Japan Atomic Energy Agency Sorption Database (JAEA-SDB) to predict Kd values for Cs in soils. We used three different machine learning models, namely the random forest (RF), artificial neural network (ANN), and convolutional neural network (CNN), to predict Kd values. The models were trained on 14 input variables from the JAEA-SDB, including factors such as Cs concentration, solid phase properties, and solution conditions which are preprocessed by normalization and log transformation. We evaluated the performance of our models using the coefficient of determination (R2) value. The RF, ANN, and CNN models achieved R2 values of over 0.97, 0.86, and 0.88, respectively. Additionally, we analyzed the variable importance of RF using out-of-bag (OOB) and CNN with an attention module. Our results showed that the initial radionuclide concentration and properties of solid phase were important variables for Kd prediction. Our machine learning models provide accurate predictions of Kd values for different soil conditions. The Kd values predicted by our models can be used to assess the behavior of radioactive contaminants in various environmental media. This can help in predicting the potential migration and retention of contaminants in soils and the selection of appropriate site remediation options. Our study provides a reliable and efficient method for predicting Kd values that can be used in environmental risk assessment and waste management.

Commercial operation of KORI Unit 1 ended in 2017, and the final decommissioning plan is currently under approval from the KINS. In order for the dismantling waste to go to the repository, it is judged that the radioactive waste generated during the commercial operation should be treated and disposed in advance. Among these radioactive wastes, spent filters contain various radionuclides. The radiation dose rate from the radiation coming out of the filters ranges from a low dose rate to high dose rate. Therefore, in order to handle the spent filters, a remote processing system is required to reduce the radiation exposure of workers. This paper evaluates the radioactive inventory of filters that are stored in the filter room at the KORI unit #1. For this purpose, a method for predicting the radioactivity of each nuclide in the filter, based on the radiation dose rate, has been described using the MicroShield code, which is a commercial shielding code. The information on the filters in the field has only the creation date, type, size, and surface dose rate. In order to evaluate the radioactivity inventory using such limited data, it is possible to know the nuclide radioactivity ratio in the filter. We took out some of the filters stored on site and measured from using the ISCOS system, a gamma nuclide analyzer. The radioactivity of each nuclide in the filter was inferred by modeling with the MicroShield code, based on the radiation dose rate and the radioactivity value of each nuclide measured in the field.

The high-level nuclear waste (HLW) repository is a 500-1,000 m deep underground structure to dispose high-level nuclear waste. The waste has a very long half-time and is exposed to a number of stresses, including high temperatures, high humidity, high pressure These stresses cause the structure to deteriorate and create cracks. Therefore, structural health monitoring with monitoring sensors is required for safety. However, sensors could also fail due to the stresses, especially high temperature. Given that the sensors are installed in the bentonite buffer and the backfill tunnel, it is impossible to replace them if they fail. That’s why it is necessary to assess the sensors’ durability under the repository’s environmental conditions before installing them. Accelerated life test (ALT) can be used to assess durability or life of the sensors, and it is important to obtain the same failure mode for reliability tests including ALT. Before conducting the test, the proper stress level must be designed first to get reliable data in a short time. After that, acceleration of life reduction with increasing temperature and temperature-life model should be determined with some statistical methods. In this study, a methodology for designing stress levels and predicting the life of the sensor were described.

The intermediate shaft of sliding type is assembled with coated shaft joint and tube joint. Since the intermediate shaft plays a role of absorbing displacement change due to vibration, the intermediate shaft must have a sliding force value in an appropriate range. In this study, an intermediate shaft assembly system for post-processing of defective intermediate shafts was developed. The intermediate shaft assembly system consists of a wear count prediction model and an automatic wear system. A wear count prediction model was created with the initial assembly sliding force, quality, and set values. As a result of applying the intermediate shaft assembly device, the sliding force of the intermediate shaft was induced within the set value range. And it was prevented from the intermediate shaft defect and eliminated manual work.

The bandgap characteristics of semiconductor materials are an important factor when utilizing semiconductor materials for various applications. In this study, based on data provided by AFLOW (Automatic-FLOW for Materials Discovery), the bandgap of a semiconductor material was predicted using only the material’s compositional features. The compositional features were generated using the python module of ‘Pymatgen’ and ‘Matminer’. Pearson’s correlation coefficients (PCC) between the compositional features were calculated and those with a correlation coefficient value larger than 0.95 were removed in order to avoid overfitting. The bandgap prediction performance was compared using the metrics of R2 score and root-mean-squared error. By predicting the bandgap with randomforest and xgboost as representatives of the ensemble algorithm, it was found that xgboost gave better results after cross-validation and hyper-parameter tuning. To investigate the effect of compositional feature selection on the bandgap prediction of the machine learning model, the prediction performance was studied according to the number of features based on feature importance methods. It was found that there were no significant changes in prediction performance beyond the appropriate feature. Furthermore, artificial neural networks were employed to compare the prediction performance by adjusting the number of features guided by the PCC values, resulting in the best R2 score of 0.811. By comparing and analyzing the bandgap distribution and prediction performance according to the material group containing specific elements (F, N, Yb, Eu, Zn, B, Si, Ge, Fe Al), various information for material design was obtained.