• Journal of Wind Energy
• /
• v.15 no.2
• /
• pp.10-22
• /
• 2024

This research is a comprehensive analysis of wind power prediction sensitivity using a Long Short-Term Memory (LSTM) deep learning neural network model, accounting for the inherent uncertainties in wind speed estimation. Utilizing a year's worth of operational data from an operational wind farm, the study forecasts the power output of both individual wind turbines and the farm collectively. Predictions were made daily at intervals of 10 minutes and 1 hour over a span of three months. The model's forecast accuracy was evaluated by comparing the root mean square error (RMSE), normalized RMSE (NRMSE), and correlation coefficients with actual power output data. Moreover, the research investigated how inaccuracies in wind speed inputs affect the power prediction sensitivity of the model. By simulating wind speed errors within a normal distribution range of 1% to 15%, the study analyzed their influence on the accuracy of power predictions. This investigation provided insights into the required wind speed prediction error rate to achieve an 8% power prediction error threshold, meeting the incentive standards for forecasting systems in renewable energy generation.

• KEPCO Journal on Electric Power and Energy
• /
• v.6 no.4
• /
• pp.433-438
• /
• 2020

Accurate forecasting of wind power is important for grid operation. Wind power has intermittent and nonlinear characteristics, which increases the uncertainty in wind power generation. In order to accurately predict wind power generation with high uncertainty, it is necessary to analyze the factors affecting wind power generation. In this paper, 6 factors out of 11 are selected for more accurate wind power generation forecast. These are wind speed, sine value of wind direction, cosine value of wind direction, local pressure, ground temperature, and history data of wind power generated.

• The Korean Journal of Applied Statistics
• /
• v.22 no.1
• /
• pp.221-235
• /
• 2009

This manuscript illustrates the comparative study between ARIMA and Exponential Smoothing modeling to develop forest fire forecasting system using different weather parameters. In this paper, authors have developed the most suitable and closest forecasting models like ARIMA and Exponential Smoothing techniques using different weather parameters. Authors have considered the extremes of the Wind speed, Radiation, Maximum Temperature and Deviation Temperature of the Summer Season form March to June month for the Ranchi Region in Jharkhand. The data is taken by own resource with the help of Automatic Weather Station. This paper consists a deep study of the effect of extreme values of the different parameters on the weather fluctuations which creates forest fires in the region. In this paper, the numerical illustration has been incorporated to support the present study. Comparative study of different suitable models also incorporated and best fitted model has been tested for these parameters.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.9 no.1
• /
• pp.19-23
• /
• 1973

The method of forecasting advection fog at Busan area in July is developed using the Spreen's scatter-diagraam technique. The used Parameters are (1) air temperature (2) dew-point temperature, (3) sea surface temperature (4) resultantt wind direction (5) resultant wind speed in Busan. The skill score and the pcr cent correct based on 4 yeare of dependent data are 0.79 and 90.3% respectively.

• Atmosphere
• /
• v.14 no.3
• /
• pp.19-28
• /
• 2004

In this study, two years Automatic Weather Station (AWS) data observed near the coast and islands are used to evaluate gust factors only when time averaged wind speed is higher than 5 ms. The gust factors are quite different in spatial and temporal domain according to analysis method. As the averaged time is increased, the gust factors are also increased. But the gust factors are decreased when wind speed is increased. It is because each wind speed is averaged one and a maximum wind is the greatest one for each time interval. The result from t-test is shown that all data are included within the 99% significance level. A sample standard deviation of ten minutes and one minute are 0.137~0.197, 0.067~0.142, respectively. Recently, the gust factor provided at the Korea Meteorological Administration (KMA) Homepage is calculated with one-hour averaged method. All though this method is hard to use directly for forecasting the strong wind over sea and coast, the result will be a great help to express Ocean Storm Flash in the Regional Meteorological Offices and the Meteorological Stations.

• Journal of the Korean Institute of Navigation
• /
• v.20 no.4
• /
• pp.1-6
• /
• 1996

A vessel in distress might be well identified when both the variables affecting the drifting of the vessel and the extent to which the variables affect the drifted vessel are known. And also the disembarking place inside the ship might be easily located if the drifting poised is forecasted. The forecasting method of the drifting poise is resolved by combining the vectors of the current and the wind. It is, however, very hard to forecast the effect of the wind, which should be mainly determined by field survey. This study aims at identifying the drifting characteristics of medium/large ships, considering only the effect of the wind. The experiment revealed the following results. $\circled1$ The drifting poise is determined by the aspect ratio of the ship and the shape of the superstructure of the ship. $\circled2$ Drifting direction is quite stable when wind speed goes over a certain level. $\circled3$ Drifting speed is 3-7% of the wind speed in case of T/S Hannara.

• Journal of Korean Society for Atmospheric Environment
• /
• v.33 no.6
• /
• pp.554-569
• /
• 2017

Sensitivity analysis on $PM_{10}$ forecasting simulations was carried out by using two different initial and boundary conditions of meteorological fields: NCEP/FNL (National Centers for Environmental Prediction/Final Analysis) reanlaysis data and NCEP/GFS (National Centers for Environmental Prediction/Global Forecast System) forecasting data, and the comparisons were made between two different simulations. The two results both yielded lower $PM_{10}$ concentrations than observations, with relatively lower biased results by NCEP/FNL than NCEP/GFS. We explored the detailed individual meteorological variables to associate with $PM_{10}$ prediction performance. With the results of NCEP/FNL outperforming GFS, our conclusion is that no particular significant bias was found in temperature fields between NCEP/FNL and NCEP/GFS data, while the overestimated wind speed by NCEP/GFS data influenced on the lower $PM_{10}$ concentrations simulation than NCEP/FNL, by decreasing the duration time of high-$PM_{10}$ loaded air mass over both coastal and metropolitan areas. These comparative characteristics of FNL against GFS data such as maximum 3~4 m/s weaker wind speed, $PM_{10}$ concentration control with the highest possible factor of 1.3~1.6, and one or two hour difference of peak time for each case in this study, were also reflected into the results of statistical analysis. It is implying that improving the surface wind speed fluctuation is an important controlling factor for the better prediction of $PM_{10}$ over Korean Peninsula.

• International Journal of Computer Science & Network Security
• /
• v.24 no.3
• /
• pp.196-200
• /
• 2024

Weather forecasting is considered to be of utmost important among various important sectors such as flood management and hydro-electricity generation. Although there are various numerical methods for weather forecasting but majority of them are reported to be Mechanistic computationally demanding due to their complexities. Therefore, it is necessary to develop and build models for accurately predicting the weather conditions which are faster as well as efficient in comparison to the prevalent meteorological models. The study has been undertaken to forecast various atmospheric parameters in the city of Bangalore using Naïve Bayes algorithms. The individual parameters analyzed in the study consisted of wind speed (WS), wind direction (WD), relative humidity (RH), solar radiation (SR), black carbon (BC), radiative forcing (RF), air temperature (AT), bar pressure (BP), PM₁₀ and PM_2.5 of the Bangalore city collected from Air Quality Monitoring Station for a period of 5 years from January 2015 to May 2019. The study concluded that Naive Bayes is an easy and efficient classifier that is centered on Bayes theorem, is quite efficient in forecasting the various air pollution parameters of the city of Bangalore.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.19 no.4
• /
• pp.146-156
• /
• 2016

In this study, we developed a large fire forecasting system using critical weather conditions, such as strong winds and effective humidity. We incorporated information on forest type prior to large fires using an incident case study. The case study includes thirty-seven large fires covering more than 100 ha of damaged area over the last 20 years. Dangerous large fire regions were identified as areas of more than 30 ha of Korean red pine and the surrounding two kilometers. Large fires occur when wind speeds average 5.3 m/s with a maximum of 11.6 m/s and standard deviation of 2.5 m/s. Effective humidity for large fires average 30% with a minimum of 13% and standard deviation of 14.5%. In dangerous Korean red pine stand areas, the large fire 'Watch' level is issued when effective humidity is 30-45% for more than two days and average wind speed is 7-10 m/s. The 'Warning' level is issued when effective humidity is less than 30% for more than two days and average wind speed is more than 11 m/s. Therefore, from now on, the large fire forecasting system can be used effectively for forest fire prevention activities based on a selection and concentration strategy in dangerous large fire regions using severe weather conditions.

• Journal of Wind Energy
• /
• v.14 no.3
• /
• pp.25-33
• /
• 2023

In order to secure the safety of increasing offshore activities such as offshore wind farm maintenance and fishing, IMPACT, a mid-term marine weather forecasting system, was established by predicting marine weather up to 7 days in advance. Forecast data from the Korea Hydrographic and Oceanographic Agency (KHOA), which provides the most reliable marine meteorological service in Korea, was used, but wind speed and wave height forecast errors increased as the leading forecast period increased, so improvement of the accuracy of the model results was needed. The Model Output Statistics (MOS) method, a post-correction method using statistical machine learning, was applied to improve the prediction accuracy of wave height, which is an important factor in forecasting the risk of marine activities. Compared with the observed data, the wave height prediction results by the model before correction for 6 to 7 days ahead showed an RMSE of 0.692 m and R of 0.591, and there was a tendency to underestimate high waves. After correction with the MOS technique, RMSE was 0.554 m and R was 0.732, confirming that accuracy was significantly improved.