• The Korean Journal of Applied Statistics
• /
• v.36 no.1
• /
• pp.49-62
• /
• 2023

Solar energy, which is rapidly increasing in proportion, is being continuously developed and invested. As the installation of new and renewable energy policy green new deal and home solar panels increases, the supply of solar energy in Korea is gradually expanding, and research on accurate demand prediction of power generation is actively underway. In addition, the importance of solar radiation prediction was identified in that solar radiation prediction is acting as a factor that most influences power generation demand prediction. In addition, this study can confirm the biggest difference in that it attempted to predict solar radiation using medium-term forecast weather data not used in previous studies. In this paper, we combined the multi-linear regression model, KNN, random fores, and SVR model and the clustering technique, K-means, to predict solar radiation by hour, by calculating the probability density function for each cluster. Before using medium-term forecast data, mean absolute error (MAE) and root mean squared error (RMSE) were used as indicators to compare model prediction results. The data were converted into daily data according to the medium-term forecast data format from March 1, 2017 to February 28, 2022. As a result of comparing the predictive performance of the model, the method showed the best performance by predicting daily solar radiation with random forest, classifying dates with similar climate factors, and calculating the probability density function of solar radiation by cluster. In addition, when the prediction results were checked after fitting the model to the medium-term forecast data using this methodology, it was confirmed that the prediction error increased by date. This seems to be due to a prediction error in the mid-term forecast weather data. In future studies, among the weather factors that can be used in the mid-term forecast data, studies that add exogenous variables such as precipitation or apply time series clustering techniques should be conducted.

• Solar Energy
• /
• v.20 no.1
• /
• pp.45-54
• /
• 2000

Five direct/diffuse decomposition models were validated using the eight years data set of direct normal beam insolation measured in Seoul. The comparison has been performed In terms of the widely used statistical indicators such as MBE, RMSE, CV(RMSE), t-Statistic and Degree of Agreement. Result indicates that most of the correlations exhibit a tendency to underestimate the direct normal beam insolation except Bouguer's model. Most of big discrepancies between the measured and the predicted values was mainly shown in near the sunrising and the sunset period. Even though the investigated five models showed fairly large disagreement for the measured values by 34%$\sim$48% of CV(RMSE), Udagawa's correlation which includes the effect of solar altitude variation appears to performs always better in every statistical error tests.

• Journal of Advanced Navigation Technology
• /
• v.22 no.3
• /
• pp.233-239
• /
• 2018

Since solar power generation is intermittent depending on weather conditions, it is necessary to predict the accurate generation amount of solar power to improve the efficiency and economical efficiency of solar power generation. This study proposes a short - term deep learning prediction model of solar power generation using meteorological data from Mokpo meteorological agency and generation data of Yeongam solar power plant. The meteorological agency forecasts weather factors such as temperature, precipitation, wind direction, wind speed, humidity, and cloudiness for three days. However, sunshine and solar radiation, the most important meteorological factors for forecasting solar power generation, are not predicted. The proposed model predicts solar radiation and solar radiation using forecast meteorological factors. The power generation was also forecasted by adding the forecasted solar and solar factors to the meteorological factors. The forecasted power generation of the proposed model is that the average RMSE and MAE of DNN are 0.177 and 0.095, and RNN is 0.116 and 0.067. Also, LSTM is the best result of 0.100 and 0.054. It is expected that this study will lead to better prediction results by combining various input.

• 한국신재생에너지학회:학술대회논문집
• /
• 2011.11a
• /
• pp.131-131
• /
• 2011

태양광 및 태양열 설비의 효율적인 관리를 위해서는 관련 일사정보가 사전정보로 제공되어 시스템 운용을 위한 입력인자로 활용되어야 한다. 특히 전력그리드에 연계되어 설비가 활용된다고 하면, 그 에너지 공급이 불규칙적인 신재생에너지원의 특성으로 인해 에너지 공급량의 예측이 선행되어 기존의 전력공급체계가 이를 지원할 수 있는 모델과 시스템이 구비되어야 한다. 기존의 다양한 연구들이 한정된 국소지점에 대해 다양한 예측기법을 적용하여 평가를 실시하였지만, 장기간의 결과 축적이 이루어지지 못해 그 신뢰성 확보에 어려움을 겪고 있다. 본 연구에서는 현재 한국에너지기술연구원에서 관리되는 일사정보를 활용하여 청명한 날의 표준 일사 데이터베이스를 생성하고, 기상청에서 RSS(Rich Site Summary) 형태로 지원하는 운량정보를 이용하여 3시간 이상의 미래정보를 계속적으로 산출할 수 있는 시스템을 제작하고자 하였다.

• Plant Journal
• /
• v.17 no.4
• /
• pp.41-52
• /
• 2021

In this study, the PV output and module temperature values were predicted using the Meteorological Agency data and compared with actual data, weather, solar radiation, ambient temperature, and wind speed. The forecast accuracy by weather was the lowest in the data on a clear day, which had the most data of the day when it was snowing or the sun was hit at dawn. The predicted accuracy of the module temperature and the amount of power generation according to the amount of insolation decreased as the amount of insolation increased, and the predicted accuracy according to the ambient temperature decreased as the module temperature increased as the ambient temperature increased and the amount of power generated lowered the ambient temperature. As for wind speed, the predicted accuracy decreased as the wind speed increased for both module temperature and power generation, but it was difficult to define the correlation because wind speed was insignificant than the influence of other weather conditions.

• Journal of Advanced Navigation Technology
• /
• v.25 no.6
• /
• pp.569-575
• /
• 2021

Deep learning shows differences in prediction performance depending on data quality and model. This study uses various input data and multiple deep learning models to build an optimal deep learning model for predicting solar radiation, which has the most influence on power generation prediction. did. As the input data, the weather data of the Korea Meteorological Administration and the clairvoyant meteorological image were used by segmenting the image of the Korea Meteorological Agency. , comparative evaluation, and predicting solar radiation by constructing multiple deep learning models connecting the models with the best error rate in each model. As an experimental result, the RMSE of model A, which is a multiple deep learning model, was 0.0637, the RMSE of model B was 0.07062, and the RMSE of model C was 0.06052, so the error rate of model A and model C was better than that of a single model. In this study, the model that connected two or more models through experiments showed improved prediction rates and stable learning results.

• Journal of Advanced Navigation Technology
• /
• v.21 no.6
• /
• pp.643-650
• /
• 2017

Photovoltaic generation which has unlimited energy sources are very intermittent because they depend on the weather. Therefore, it is necessary to get accurate generation prediction with reducing the uncertainty of photovoltaic generation and improvement of the economics. The Meteorological Agency predicts weather factors for three days, but doesn't predict the sunshine and solar radiation that are most correlated with the prediction of photovoltaic generation. In this study, we predict sunshine and solar radiation using weather, precipitation, wind direction, wind speed, humidity, and cloudiness which is forecasted for three days at Meteorological Agency. The photovoltaic generation forecasting model is proposed by using predicted solar radiation and sunshine. As a result, the proposed model showed better results in the error rate indexes such as MAE, RMSE, and MAPE than the model that predicts photovoltaic generation without radiation and sunshine. In addition, DNN showed a lower error rate index than using SVM, which is a type of machine learning.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.37 no.12
• /
• pp.1105-1112
• /
• 2013

In this study, a method for predicting heating loads using building characteristic coefficients is proposed for heating system control, and a method for predicting hourly temperature and solar insolation, which mainly affect building heating loads, is also proposed. The temperature and solar insolation are predicted by using a fuzzy theory from forecast information at the meteorological agency, and the building characteristic coefficients for the prediction of heating loads are derived from EnergyPlus. The simulated heating loads of the present study show good agreement with those of EnergyPlus. and the variations of the predicted heating loads using the predicted temperature and solar insolation are similar to those using the actual weather data.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.2
• /
• pp.249-258
• /
• 2023

Solar power has the largest proportion of power generation and facility capacity among renewable energy in South Korea. Floating solar power plant is a new way to resolve weakness of land solar power plant. This study analyzes the power generation of the 18.7 MW floating solar power project located in Saemangeum, Gunsan-si. Since the solar power generation has a characteristic that is greatly affected by the climate, various methods have been applied to predict solar power generation. In general, variables necessary for predicting power generation are solar insolation on inclined surfaces, solar generation efficiency, and panel installation area. This study analyzed solar power generation using the monthly solar insolation data from the KMA (Korea Meteorological Administration) over the past 10 years. Monte Carlo simulation (MCS) was applied to predict the solar power generation with the variables including solar panel efficiency and insolation. In the case of Saemangeum solar power project, the most solar power generation was in May, the least was in December, the average solar power generation simulated on MCS is 2.1 GWh per month, the minimum monthly power generation is 0.3 GWh, and the maximum is 5.0 GWh.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.34 no.6
• /
• pp.617-623
• /
• 2001

To predict the seasonal variation of dissolved oxygen (DO) in Hakata bay, Japan, possible 20 time-series of different hourly-solar-radiation intensities were generated based on stochastically changing solar radiation intensity, and a numerical simulation on dissolved oxygen (DO) was carried out for each time series by using the Sediment-Water Ecological Model (SWEM). The model, consisting of two sub-models with hydrodynamic and biological models, simulates the circulation process of nutrient between water column and sediment, such as nutrient regeneration from sediments as well as ecological structures on the growth of phytoplankton and zooplankton, The results of the model calibration followed the seasonal variation of observed water quality well, and generated cumulative-frequency-distribution (CFD) curves of daily solar radiation agreed well with observed ones, The simulation results indicated that the exchange of sea water would have a great influence on the DO concentration, and that the concentration could change more than 1 mg/L in a day. This prediction method seems to be an effective way to examine a solution to minimize fishery damage when DO is depleted.