PURPOSES : Pavement surface friction depends significantly on pavement surface texture characteristics. The mean texture depth (MTD), which is an index representing pavement surface texture characteristics, is typically used to predict pavement surface friction. However, the MTD may not be sufficient to represent the texture characteristics to predict friction. To enhance the prediction of pavement surface friction, one must select additional variables that can explain complex pavement surface textures. METHODS : In this study, pavement surface texture characteristics that affect pavement surface friction were analyzed based on the friction mechanism. The wavelength, pavement surface texture shape, and pavement texture depth were hypothesized to significantly affect the surface friction of pavement. To verify this, the effects of the three abovementioned pavement surface texture characteristics on pavement surface friction must be investigated. However, because the surface texture of actual pavements is irregular, examining the individual effects of these characteristics is difficult. To achieve this goal, the selected pavement surface texture characteristics were formed quantitatively, and the irregularities of the actual pavement surface texture were improved by artificially forming the pavement surface texture using threedimensionally printed specimens. To reflect the pavement surface texture characteristics in the specimen, the MTD was set as the pavement surface texture depth, and the exposed aggregate number (EAN) was set as a variable. Additionally, the aggregate shape was controlled to reflect the characteristics of the pavement surface texture of the specimen. Subsequently, a shape index was proposed and implemented in a statistical analysis to investigate its effect on pavement friction. The pavement surface friction was measured via the British pendulum test, which enables measurement to be performed in narrow areas, considering the limited size of the three-dimensionally printed specimens. On wet pavement surfaces, the pavement surface friction reduced significantly because of the water film, which intensified the effect of the pavement surface texture. Therefore, the pavement surface friction was measured under wet conditions. Accordingly, a BPN (wet) prediction model was proposed by statistically analyzing the relationship among the MTD, EAN, aggregate shape, and BPN (wet). RESULTS : Pavement surface friction is affected by adhesion and hysteresis, with hysteresis being the predominant factor under wet conditions. Because hysteresis is caused by the deformation of rubber, pavement surface friction can be secured through the formation of a pavement surface texture that causes rubber deformation. Hysteresis occurs through the function of macro-textures among pavement surface textures, and the effects of macro-texture factors such as the EAN, MTD, and aggregate shape on the BPN (wet) are as follows: 1) The MTD ranges set in this study are 0.8, 1.0, and 1.2, and under the experimental conditions, the BPN (wet) increases linearly with the MTD. 2) An optimum EAN is indicated when the BPN (wet) is the maximum, and the BPN decreases after its maximum value is attained. This may be because when the EAN increases excessively, the space for the rubber to penetrate decreases, thereby reducing the hysteresis. 3) The shape of the aggregate is closely related to the EAN; meanwhile, the maximum value of the pavement surface friction and the optimum EAN change depending on the aggregate shape. This is believed to be due to changes in the rubber penetration volume based on the aggregate shape. Based on the results above, a statistical prediction model for the BPN (wet) is proposed using the MTD, EAN, and shape index as variables. CONCLUSIONS : The EAN, MTD, and aggregate shape are crucial factors in predicting skid resistance. Notably, the EAN and aggregate shape, which are not incorporated into existing pavement surface friction prediction models, affect the pavement surface friction. However, the texture of the specimen created via three-dimensional printing differs significantly from the actual pavement surface texture. Therefore, the pavement surface friction prediction model proposed in this study should be supplemented with comparisons with actual pavement surface data in the future.

여름철은 타 계절에 비해 장마와 불안정한 대기 등으로 인하여 빗길 교통사고의 위험성이 크게 증대될 수 있으며, 최근 5년 (2018~2022)간 여름철 빗길 교통사고는 전체 빗길 교통사고의 39%를 차지할 정도로 높은 수준이다. 이러한 빗길 운전은 노면의 배수 불량 및 미끄럼 저항 감소 등으로 인하여 수막현상을 발생시키게 된다. 이에 본 연구에서는 우천 시 도로의 안전성 강화 및 사고 위 험을 최소화하기 위한 요소인 수막두께를 산정하기 위하여 Manning의 평균 유속식을 기반으로 콘크리트 노면의 조도계수 예측 모델을 개발하는 것을 목표로 하였다. 조도계수의 영향인자를 고려하기 위하여 실외 강우 모의 장비를 이용하여 콘크리트를 타설한 뒤 실험 인자로 포장 경사, 배수거리, 강우강도, 노면 조직 특성을 고려하였으며, 이 중 노면 조직 특성은 타이닝 처리를 하지 않은 구간만 고 려한 타 연구의 기존 예측 모델 단점을 보완하기 위하여 16, 25mm 간격의 타이닝 표면 처리한 구간을 추가로 고려하였다. 수막두께 측정은 측정 범위 0.3~5mm의 수막두께 측정 게이지를 제작하여 강우가 모사된 조건에서 배수 거리 1~5m 이내 지점의 노면 조직 상 단과 수면이 접하는 수직 높이를 총 3회 측정하여 평균값을 사용하였다. 실측된 수막두께 데이터베이스를 기반으로 Manning 공식을 이용하여 조도계수를 역산한 결과, 강우강도가 증가함에 따라 조도계수는 감소하였으며, 이는 강우의 증가로 인해 물의 흐름과 콘크리 트 노면 사이의 마찰 저항 감소에 기인한 것으로 판단되었다. 또한 포장 경사가 높고 배수 거리가 짧을수록 배수성이 증가하여 마찰 저항의 지표인 조도계수가 증가하는 것으로 확인되었다. 평균 조직 깊이에 따른 조도계수 영향의 경우, 평균 조직 깊이가 증가할수록 콘크리트 표면에 노출되는 표면적이 증가하여 수막두께가 얕게 생성되고, 얕은 수심으로 인해 물의 흐름 저항이 감소하여 조도계수는 감소하는 것으로 산정되었다. 이후 135개의 데이터를 종합하여 조도계수를 종속변수로 하고 강우강도, 포장경사, 배수거리, 평균 조직 깊이, 수막두께를 독립변수로 하는 회귀분석을 수행하여 조도계수 산정식을 개발하였다.

Existing reinforced concrete buildings with seismically deficient column details affect the overall behavior depending on the failure type of column. This study aims to develop and validate a machine learning-based prediction model for the column failure modes (shear, flexure-shear, and flexure failure modes). For this purpose, artificial neural network (ANN), K-nearest neighbor (KNN), decision tree (DT), and random forest (RF) models were used, considering previously collected experimental data. Using four machine learning methodologies, we developed a classification learning model that can predict the column failure modes in terms of the input variables using concrete compressive strength, steel yield strength, axial load ratio, height-to-dept aspect ratio, longitudinal reinforcement ratio, and transverse reinforcement ratio. The performance of each machine learning model was compared and verified by calculating accuracy, precision, recall, F1-Score, and ROC. Based on the performance measurements of the classification model, the RF model represents the highest average value of the classification model performance measurements among the considered learning methods, and it can conservatively predict the shear failure mode. Thus, the RF model can rapidly predict the column failure modes with simple column details.

This study deals with the application of an artificial neural network (ANN) model to predict power consumption for utilizing seawater source heat pumps of recirculating aquaculture system. An integrated dynamic simulation model was constructed using the TRNSYS program to obtain input and output data for the ANN model to predict the power consumption of the recirculating aquaculture system with a heat pump system. Data obtained from the TRNSYS program were analyzed using linear regression, and converted into optimal data necessary for the ANN model through normalization. To optimize the ANN-based power consumption prediction model, the hyper parameters of ANN were determined using the Bayesian optimization. ANN simulation results showed that ANN models with optimized hyper parameters exhibited acceptably high predictive accuracy conforming to ASHRAE standards.

현대 해양 산업은 기술적 발전을 통해 신속한 발전을 이루고 있다. 이러한 발전을 주도하는 주요 기술 중 하나는 데이터 처리 기술이며, 이 중 자연어 처리 기법은 사람의 언어를 기계가 이해하고 처리할 수 있도록 하는 기술이다. 본 연구는 자연어 처리 기법을 통해 해양안전심판원의 재결서를 분석하여 이미 재결이 이루어진 선박 충돌사고의 원인 제공 비율을 학습한 후, 새로운 재결서를 입력 하면 원인 제공 비율을 예측하는 모델을 개발하고자 하였다. 이 모델은 사고 당시 적용되는 항법과 원인 제공 비율에 영향을 주는 핵심 키워드의 가중치를 이용하여 사고의 원인 제공 비율을 계산하는 방식으로 구성하였다. 이 연구는 이러한 방식을 통해 제작한 모델의 정 확도를 분석하고, 모델의 실무 적용 가능성을 검토함과 동시에 충돌사고 재발 방지 및 해양사고 당사자들의 분쟁 해결에 기여할 것으로 기대한다.

In order to present a predictive drift model, Jeju National University's training ship was tested for about 11 hours and 40 minutes, and 81 samples that selected one of the entire samples at ten-minute intervals were subjected to regression analysis after verifying outliers and influence points. In the outlier and influence point analysis, although there is a part where the wind direction exceeds 1 in the DFBETAS (difference in Betas) value, the CV (cumulative variable) value is 6%, close to 1. Therefore, it was judged that there would be no problem in conducting multiple regression analyses on samples. The standard regression coefficient showed how much current and wind affect the dependent variable. It showed that current speed and direction were the most important variables for drift speed and direction, with values of 47.1% and 58.1%, respectively. The analysis showed that the statistical values indicated the fit of the model at the significance level of 0.05 for multiple regression analysis. The multiple correlation coefficients indicating the degree of influence on the dependent variable were 83.2% and 89.0%, respectively. The determination of coefficients were 69.3% and 79.3%, and the adjusted determination of coefficients were 67.6% and 78.3%, respectively. In this study, a more quantitative prediction model will be presented because it is performed after identifying outliers and influence points of sample data before multiple regression analysis. Therefore, many studies will be active in the future by combining them.

PURPOSES : To enhance the accuracy of predicting the compressive strength of practical concrete mixtures, this study aimed to develop a machine learning model by utilizing the most commonly employed curing age, specifically, the 28-day curing period. The training dataset consisted of concrete mixture sample data at this curing age, along with samples subjected to a total load not exceeding 2,350 kg. The objective was to train a machine learning model to create a more practical predictive model suitable for real-world applications. METHODS : Three machine learning models—random forest, gradient boosting, and AdaBoost—were selected. Subsequently, the prepared dataset was used to train the selected models. Model 1 was trained using concrete sample data from the 28th curing day, followed by a comprehensive analysis of the results. For Model 2, training was conducted using data from the 28th day of curing, focusing specifically on instances where the total load was 2,350 kg or less. The results were systematically analyzed to determine the most suitable machine learning model for predicting the compressive strength of concrete. RESULTS : The machine learning model trained on concrete sample data from the 28th day of curing with a total weight of 2,350 kg or less exhibited higher accuracy than the model trained on weight-unrestricted data from the 28th day of curing. The models were evaluated in terms of accuracy, with the gradient boosting, AdaBoost, and random forest models demonstrating high accuracy, in that order. CONCLUSIONS : Machine learning models trained using concrete mix data based on practical and real-world scenarios demonstrated a higher accuracy than models trained on impractical concrete mix data. This case illustrates the significance of not only the quantity but also the quality of the data during the machine learning training process. Excluding outliers from the data appears to result in better accuracy for machine learning models. This underscores the importance of using high-quality and practical mixed concrete data for reliable and accurate model training.

PURPOSES : In this study, an optimal model for compressive strength prediction was derived by learning and directly comparing several machine learning models based on the same data. METHODS : Approximately 478 pieces of concrete compressive strength data were obtained to compare the performance of the machine learning models. In addition, five machine learning models were trained based on the obtained data. The performance of the learned model was compared using three performance indicators. Finally, the performance of the model trained using additional data was reviewed. RESULTS : As a result of comparing the performance of machine learning models, the XGB(eXtra Gradient Boost) model showed the best performance. In addition, as a result of the verification based on additional data, highly reliable results can be obtained if the XGB model is used to predict the compressive strength of concrete. CONCLUSIONS : If a concrete strength prediction model is derived based on a machine learning model, a highly reliable model can be derived.

PURPOSES : In this study, the applicability of the water content, suction, and suction stress in a resilient modulus prediction model for a subbase was reviewed. METHODS : To compare the applicability of water content, suction, and suction stress models for resilient modulus prediction, the suction stress was determined based on the soil water characteristic curve. The model parameters for each approach were derived from the measured resilient moduli. Finally, the relationships between the degree of saturation and resilient modulus were analyzed using the calculated model parameters. RESULTS : Prediction models of the resilient modulus based on water content and suction demonstrated high correlation with measured values, but overestimated the resilient modulus at saturation levels beyond the laboratory testing range. In contrast, the model accounting for suction stress effectively reduced this overestimation, likely owing to a decrease in suction stress as the suction increased. CONCLUSIONS : Based on the above results, the resilient modulus of subbase materials could be estimated through the change in the degree of saturation and the stress-dependent resilient modulus model using the suction stress proposed in this study.

Occurrence of process environment changes, such as influent load variances and process condition changes, can reduce treatment efficiency, increasing effluent water quality. In order to prevent exceeding effluent standards, it is necessary to manage effluent water quality based on process operation data including influent and process condition before exceeding occur. Accordingly, the development of the effluent water quality prediction system and the application of technology to wastewater treatment processes are getting attention. Therefore, in this study, through the multi-channel measuring instruments in the bio-reactor and smart multi-item water quality sensors (location in bio-reactor influent/effluent) were installed in The Seonam water recycling center #2 treatment plant series 3, it was collected water quality data centering around COD, T-N. Using the collected data, the artificial intelligence-based effluent quality prediction model was developed, and relative errors were compared with effluent TMS measurement data. Through relative error comparison, the applicability of the artificial intelligence-based effluent water quality prediction model in wastewater treatment process was reviewed.

본 연구는 소나무재선충병 방제대상지 선정의 효율성을 높이기 위해 진주시를 대상으로 소나무재선충병 잠재분포를 예측하였다. 예측에 사용된 MaxEnt 모델은 회귀분석을 기반으로 종 발생 확률 평가 및 다양한 잠재분포 예측에 이용되고 있다. 종속변수로는 소나무재선충병 감염목 자료를 사용하였으며, 독립변수로는 지리 ‧ 지형 ‧ 기후적 요인으로 총 15개 인자를 사용하였다. 잠재분포 예측 결과, 모델의 성능은 AUC가 0.801로 우수한 수준의 정확도를 나타냈다. 독립변수 중에는 전년도 감염목과의 거리, 6월 하순 강우량, 5월 강우량, 화목보일러와의 거리 순으로 잠재분포에 영향을 미치는 것으로 나타났다. 이러한 결과는 지속적인 소나무재선충병 감염목 DB 구축과 지리적 요인들에 대한 모니터링의 중요성이 크다는 것을 의미한다.

This study endeavors to enrich investment prospects in cryptocurrency by establishing a rationale for investment decisions. The primary objective involves evaluating the predictability of four prominent cryptocurrencies – Bitcoin, Ethereum, Litecoin, and EOS – and scrutinizing the efficacy of trading strategies developed based on the prediction model. To identify the most effective prediction model for each cryptocurrency annually, we employed three methodologies – AutoRegressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), and Prophet – representing traditional statistics and artificial intelligence. These methods were applied across diverse periods and time intervals. The result suggested that Prophet trained on the previous 28 days' price history at 15-minute intervals generally yielded the highest performance. The results were validated through a random selection of 100 days (20 target dates per year) spanning from January 1st, 2018, to December 31st, 2022. The trading strategies were formulated based on the optimal-performing prediction model, grounded in the simple principle of assigning greater weight to more predictable assets. When the forecasting model indicates an upward trend, it is recommended to acquire the cryptocurrency with the investment amount determined by its performance. Experimental results consistently demonstrated that the proposed trading strategy yields higher returns compared to an equal portfolio employing a buy-and-hold strategy. The cryptocurrency trading model introduced in this paper carries two significant implications. Firstly, it facilitates the evolution of cryptocurrencies from speculative assets to investment instruments. Secondly, it plays a crucial role in advancing deep learning- based investment strategies by providing sound evidence for portfolio allocation. This addresses the black box issue, a notable weakness in deep learning, offering increased transparency to the model.

본 연구는 네트워크 이상 감지 및 예측을 위해 벡터 자기회귀(VAR) 모델의 사용을 비교 분석한다. VAR 모 델에 대한 간략한 개요를 제공하고 네트워크 이상 체크로 사용 가능한 두 가지 버전을 검토하며 두 종류의 VAR 모델을 통한 경험적인 평가를 제시한다. VAR-Filtered moving-common-AR 모델이 단일 노드 이상 감지 성능에서 우수하며, VAR-Adaptive Learning 버전은 몇 개의 노드 간 이상을 효과적으로 식별하는 데 특히 효 과적이며 두 가지 주요VAR 모델의 전반적인 성능 차이에 대한 근본적인 이유도 분석한다. 각 기술의 장단점 을 개요로 제공하고 성능 향상을 위한 제안도 제시하고자 한다.

Numerous factors contribute to the deterioration of reinforced concrete structures. Elevated temperatures significantly alter the composition of the concrete ingredients, consequently diminishing the concrete's strength properties. With the escalation of global CO2 levels, the carbonation of concrete structures has emerged as a critical challenge, substantially affecting concrete durability research. Assessing and predicting concrete degradation due to thermal effects and carbonation are crucial yet intricate tasks. To address this, multiple prediction models for concrete carbonation and compressive strength under thermal impact have been developed. This study employs seven machine learning algorithms—specifically, multiple linear regression, decision trees, random forest, support vector machines, k-nearest neighbors, artificial neural networks, and extreme gradient boosting algorithms—to formulate predictive models for concrete carbonation and thermal impact. Two distinct datasets, derived from reported experimental studies, were utilized for training these predictive models. Performance evaluation relied on metrics like root mean square error, mean square error, mean absolute error, and coefficient of determination. The optimization of hyperparameters was achieved through k-fold cross-validation and grid search techniques. The analytical outcomes demonstrate that neural networks and extreme gradient boosting algorithms outshine the remaining five machine learning approaches, showcasing outstanding predictive performance for concrete carbonation and thermal effect modeling.

In this study, we developed Rapid Enrichment Broth for Vibrio parahaemolyticus (REB-V), a broth capable enriching V. parahaemolyticus from 100 CFU/mL to 106 CFU/mL within 6 hours, which greatly facilitates the rapid detection of V. parahaemolyticus. Using a modified Gompertz model and response surface methodology, we optimized supplement sources to rapidly enrich V. parahaemolyticus. The addition of 0.003 g/10 mL of D-(+)- mannose, 0.002 g/10 mL of L-valine, and 0.002 g/10 mL of magnesium sulfate to 2% (w/v) NaCl BPW was the most effective combination of V. parahaemolyticus enrichment. Optimal V. parahaemolyticus culture conditions using REB-V were at pH 7.84 and 37oC. To confirm REB-V culture efficiency compared to 2% (w/v) NaCl BPW, we assessed the amount of enrichment achieved in 7 hours in each medium and extracted DNA samples from each culture every hour. Real-time PCR was performed using the extracted DNA to verify the applicability of this REB-V culture method to molecular diagnosis. V. parahaemolyticus was enriched to 5.452±0.151 Log CFU/mL in 2% (w/v) NaCl BPW in 7 hours, while in REB-V, it reached 7.831±0.323 Log CFU/mL. This confirmed that REB-V enriched V. parahaemolyticus to more than 106 CFU/mL within 6 hours. The enrichment rate of REB-V was faster than that of 2% (w/v) NaCl BPW, and the amount of enrichment within the same time was greater than that of 2% (w/v) NaCl BPW, indicating that REB-V exhibits excellent enrichment efficiency.

This study aimed to assess and determine the optimal model for predicting the full bloom date of ‘Fuji’ apples across South Korea. We evaluated the performance of four distinct models: the Development Rate Model (DVR)1, DVR2, the Chill Days (CD) model, and a sequentially integrated approach that combined the Dynamic model (DM) and the Growing Degree Hours (GDH) model. The full bloom dates and air temperatures were collected over a three-year period from six orchards located in the major apple production regions of South Korea: Pocheon, Hwaseong, Geochang, Cheongsong, Gunwi, and Chungju. Among these models, the one that combined DM for calculating chilling accumulation and the GDH model for estimating heat accumulation in sequence demonstrated the most accurate predictive performance, in contrast to the CD model that exhibited the lowest predictive precision. Furthermore, the DVR1 model exhibited an underestimation error at orchard located in Hwaseong. It projected a faster progression of the full bloom dates than the actual observations. This area is characterized by minimal diurnal temperature ranges, where the daily minimum temperature is high and the daily maximum temperature is relatively low. Therefore, to achieve a comprehensive prediction of the blooming date of ‘Fuji’ apples across South Korea, it is recommended to integrate a DM model for calculating the necessary chilling accumulation to break dormancy with a GDH model for estimating the requisite heat accumulation for flowering after dormancy release. This results in a combined DM+GDH model recognized as the most effective approach. However, further data collection and evaluation from different regions are needed to further refine its accuracy and applicability.

Determining the size or area of a plant's leaves is an important factor in predicting plant growth and improving the productivity of indoor farms. In this study, we developed a convolutional neural network (CNN)-based model to accurately predict the length and width of lettuce leaves using photographs of the leaves. A callback function was applied to overcome data limitations and overfitting problems, and K-fold cross-validation was used to improve the generalization ability of the model. In addition, ImageDataGenerator function was used to increase the diversity of training data through data augmentation. To compare model performance, we evaluated pre-trained models such as VGG16, Resnet152, and NASNetMobile. As a result, NASNetMobile showed the highest performance, especially in width prediction, with an R_squared value of 0.9436, and RMSE of 0.5659. In length prediction, the R_squared value was 0.9537, and RMSE of 0.8713. The optimized model adopted the NASNetMobile architecture, the RMSprop optimization tool, the MSE loss functions, and the ELU activation functions. The training time of the model averaged 73 minutes per Epoch, and it took the model an average of 0.29 seconds to process a single lettuce leaf photo. In this study, we developed a CNN-based model to predict the leaf length and leaf width of plants in indoor farms, which is expected to enable rapid and accurate assessment of plant growth status by simply taking images. It is also expected to contribute to increasing the productivity and resource efficiency of farms by taking appropriate agricultural measures such as adjusting nutrient solution in real time.

미세구조 특성의 불확실성은 재료 특성에 많은 영향을 준다. 시멘트 기반 재료의 공극 분포 특성은 재료의 역학적 특성에 큰 영향을 미치며, 재료에 랜덤하게 분포되어 있는 많은 공극은 재료의 물성 예측을 어렵게 한다. 공극의 특성 분석과 재료 응답 간의 상관관계 규명에 대한 기존 연구는 통계적 관계 분석에 국한되어 있으며, 그 상관관계가 아직 명확히 규명되어 있지 않다. 본 연구에서는 합성곱 신경망(CNN, convolutional neural network)을 활용한 이미지 기반 데이터 접근법을 통해 시멘트 기반 재료의 역학적 응답을 예측하 고, 공극분포와 재료 응답의 상관관계를 분석하였다. 머신러닝을 위한 데이터는 고해상도 마이크로-CT 이미지와 시멘트 기반 재료의 물성(인장강도)로 구성하였다. 재료의 메시 구조 특성을 분석하였으며, 재료의 응답은 상장균열모델(phase-field fracture model)에 기 반을 둔 2D 직접 인장(direct tension) 유한요소해석 시뮬레이션을 활용하여 평가하였다. 입력 이미지 영역의 기여도를 분석하여 시편 에서 재료 응답 예측에 가장 큰 영향을 미치는 영역을 CNN을 통하여 식별하였다. CNN 과정 중 활성 영역과 공극분포를 비교 분석하 여 공극분포특성과 재료 응답의 상관관계를 분석하여 제시하였다.