• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.25 no.6
• /
• pp.310-316
• /
• 2013

This study was designed to investigate a method for short-term, real-time energy demand prediction, to cope with changing loads for the effective operation and management of buildings. Through a case study, a novel methodology for real-time energy demand prediction with the use of weather forecasting data was suggested. To perform the input and output operations of weather data, and to calculate solar radiation and EnergyPlus, the BCVTB (Building Control Virtual Test Bed) was designed. Through the BCVTB, energy demand prediction for the next 24 hours was carried out, based on 4 real-time weather data and 2 solar radiation calculations. The weather parameters used in a model equation to calculate solar radiation were sourced from the weather data of the KMA (Korea Meteorological Administration). Depending on the local weather forecast data, the results showed their corresponding predicted values. Thus, this methodology was successfully applicable to anywhere that local weather forecast data is available.

• Korean Journal of Agricultural Science
• /
• v.47 no.4
• /
• pp.769-782
• /
• 2020

Interest in fruit and vegetables has increased due to changes in consumer consumption patterns, socioeconomic status, and family structure. This study determined the factors influencing the demand for fruit and vegetables (strawberries, paprika, tomatoes and cherry tomatoes) using a panel of Rural Development Administration household-level purchases from 2010 to 2018 and compared the ability to the prediction performance. An artificial neural network model was constructed, linking household characteristics with final food expenditure. Comparing the analysis results of the artificial neural network with the results of the panel model showed that the artificial neural network accurately predicted the pattern of the consumer panel data rather than the fixed effect model. In addition, the prediction for strawberries was found to be heavily affected by the number of families, retail places and income, while the prediction for paprika was largely affected by income, age and retail conditions. In the case of the prediction for tomatoes, they were greatly affected by age, income and place of purchase, and the prediction for cherry tomatoes was found to be affected by age, number of families and retail conditions. Therefore, a more accurate analysis of the consumer consumption pattern was possible through the artificial neural network model, which could be used as basic data for decision making.

• Korean Journal of Artificial Intelligence
• /
• v.9 no.1
• /
• pp.15-20
• /
• 2021

This paper was conducted to prevent and respond to crimes by predicting crimes based on artificial intelligence. While the quality of life is improving with the recent development of science and technology, various problems such as poverty, unemployment, and crime occur. Among them, in the case of crime problems, the importance of crime prediction increases as they become more intelligent, advanced, and diversified. For all crimes, it is more critical to predict and prevent crimes in advance than to deal with them well after they occur. Therefore, in this paper, we predicted crime types and crime tools using the Multiclass Logistic Regression algorithm and Multiclass Neural Network algorithm of machine learning. Multiclass Logistic Regression algorithm showed higher accuracy, precision, and recall for analysis and prediction than Multiclass Neural Network algorithm. Through these analysis results, it is expected to contribute to a more pleasant and safe life by implementing a crime prediction system that predicts and prevents various crimes. Through further research, this researcher plans to create a model that predicts the probability of a criminal committing a crime again according to the type of offense and deploy it to a web service.

• New & Renewable Energy
• /
• v.18 no.1
• /
• pp.17-28
• /
• 2022

This paper presents an effective Markov transition matrix (EMTM), which will be used to calculate the wind speed at the target site in a wind farm to accurately predict wind energy production. The existing MTS prediction method using a Markov transition matrix (MTM) exhibits a limitation where significant prediction variations are observed owing to random selection errors and its bin width. The proposed method selects the effective states of the MTM and refines its bin width to reduce the error of random selection during a gap filling procedure in MTS. The EMTM reduces the level of variation in the repeated prediction of wind speed by using the coefficient of variations and range of variations. In a case study, MTS exhibited better performance than other MCP models when EMTM was applied to estimate a one-day wind speed, by using mean relative and root mean square errors.

• Korean Journal of Materials Research
• /
• v.33 no.7
• /
• pp.273-278
• /
• 2023

In this study, based on the saturation magnetic flux density experimental values (B_s) of 622 Fe-based bulk metallic glasses (BMGs), regression models were applied to predict B_s using artificial neural networks (ANN), and prediction performance was evaluated. Model performance evaluation was investigated by using the F1 score together with the coefficient of determination (R² score), which is mainly used in regression models. The coefficient of determination can be used as a performance indicator, since it shows the predicted results of the saturation magnetic flux density of full material datasets in a balanced way. However, the BMG alloy contains iron and requires a high saturation magnetic flux density to have excellent applicability as a soft magnetic material, and in this study F1 score was used as a performance indicator to better predict B_s above the threshold value of B_s (1.4 T). After obtaining two ANN models optimized for the R² and F1 score conditions, respectively, their prediction performance was compared for the test data. As a case study to evaluate the prediction performance, new Fe-based BMG datasets that were not included in the training and test datasets were predicted using the two ANN models. The results showed that the model with an excellent F1 score achieved a more accurate prediction for a material with a high saturation magnetic flux density.

• Journal of Environmental Science International
• /
• v.25 no.5
• /
• pp.683-694
• /
• 2016

Sensitivity analysis of the WRF model according to the impact of nudging (e.g., nudging techniques and application domains) was conducted during high nocturnal ozone episode to improve the prediction of the regional ozone concentration in the southeastern coastal area of the Korean peninsula. The analysis was performed by six simulation experiments: (1) without nudging (e.g., CNTL case), (2) with observation nudging (ONE case) to all domains (domain 1~4), (3) with grid nudging (GNE case) to all domains, (4)~(6) with grid nudging to domain 1, domain 1~2 and domain 1~3, respectively (GNE-1, GNE-2, GNE-3 case). The results for nudging techniques showed that the GNE case was in very good agreement with those observed during all analysis periods (e.g., daytime, nighttime, and total), as compared to the ONE case. In particular, the large effect of grid nudging on the near-surface meteorological factors (temperature, relative humidity, and wind fields) was predicted at the coastline and nearby sea during daytime. The results for application domains showed that the effects of nudging were distinguished between the meteorological factors and between the time periods. When applied grid nudging until subdomain, the improvement effects of temperature and relative humidity had differential tendencies. Temperature was increased for all time, but relative humidity was increased in daytime and was decreased in nighttime. Thus, GNE case showed better result than other cases.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.5
• /
• pp.147-156
• /
• 2013

Each field of modern society, industrialization and the development of science and technology are rapidly changing. However, as a side effect of rapid social change has caused various problems. Crime of the side effects of rapid social change is a big problem. In this paper, a model for predicting crime and Markov chains applied to the crime, predictive modeling is proposed. Markov chain modeling of the existing one with the overall status of the case determined the probability of predicting the future, but this paper predict the events to increase the probability of occurrence probability of the prediction and the recent state of the entire state was divided by the probability of the prediction. And the whole state and the probability of the prediction and the recent state by applying the average of the prediction probability and the probability of the prediction model were implemented. Data was applied to the incidence of crime. As a result, the entire state applies only when the probability of the prediction than the entire state and the last state is calculated by dividing the probability value. And that means when applied to predict the probability, close to the crime was concluded that prediction.

• Journal of the Korea Society of Computer and Information
• /
• v.8 no.3
• /
• pp.113-123
• /
• 2003

Many researches on the application of neural networks for making financial index prediction have proven their advantages over statistical and other methods. In this paper, a neural network model is proposed for the Buying, Holding or Selling timing prediction in stocks by the price index of stocks by inputting the closing price and volume of dealing in stocks and the technical indexes(MACD, Psychological Line). This model has an equalized multi-layer arithmetic function as well as the time series prediction function of backpropagation neural network algorithm. In the case that the numbers of learning data are unbalanced among the three categories (Buying, Holding or Selling), the neural network with conventional method has the problem that it tries to improve only the prediction accuracy of the most dominant category. Therefore, this paper, after describing the structure, working and learning algorithm of the neural network, shows the equalized multi-layer arithmetic method controlling the numbers of learning data by using information about the importance of each category for improving prediction accuracy of other category. Experimental results show that the financial index prediction using the equalized multi-layer arithmetic neural network has much higher correctness rate than the other conventional models.

• Korean Journal of Remote Sensing
• /
• v.18 no.1
• /
• pp.1-12
• /
• 2002

Landslide prediction modeling has been regarded as one of the important environmental applications in GIS. While, landslide stability in a certain area as collateral process for prediction modeling can be characterized by DEM-based hydrological features such as flow-direction, flow-accumulation, flow-length, wetness index, and so forth. In this study, Slope-Area plot methodology followed by stability index mapping with these hydrological variables is firstly performed for stability analysis with actual landslide occurrences at Boeun area, Korea, and then Landslide prediction modeling based on likelihood ratio model for landslide potential mapping is carried out; in addition, KOMPSAT EOC imagery is used to detect the locations and scalped scale of Landslide occurrences. These two tasks are independently processed for preparation of unbiased criteria, and then results of those are qualitatively compared. As results of this case study, land stability analysis based on DEM-based hydrological variables directly reflects terrain characteristics; however, the results in the form of land stability map by landslide prediction model are not fully matched with those of hydrologic landslide analysis due to the heuristic scheme based on location of existed landslide occurrences within prediction approach, especially zones of not-investigated occurrences. Therefore, it is expected that the resets on the space-robustness of landslide prediction models in conjunction with DEM-based landslide stability analysis can be effectively utilized to search out unrevealed or hidden landslide occurrences.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1176-1179
• /
• 2020

Vessel path prediction methods generally predict the latitude and longitude of a future location directly. However, in the case of direct prediction, errors could be large since the possible output range is too broad. In addition, error accumulation could occur since recurrent neural networks-based methods employ previous predicted data to forecast future data. In this paper, we propose a vessel path prediction method that does not directly predict the longitude and latitude. Instead, the proposed method predicts the acceleration of the vessel. Then the acceleration is employed to generate the velocity and direction, and the values decide the longitude and latitude of the future location. In the experiment, we show that the proposed method makes smaller errors than the direct prediction method, while both methods employ the same model.