• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.5
• /
• pp.1085-1092
• /
• 2017

In this paper, we propose a wind forecasting method that reflects wind characteristics to improve the accuracy of wind power prediction. The proposed method consists of extracting wind characteristics and predicting power generation. The part that extracts the characteristics of the wind uses correlation analysis of power generation amount, wind direction and wind speed. Based on the correlation between the wind direction and the wind speed, the feature vector is extracted by clustering using the K-means method. In the prediction part, machine learning is performed using the SVR that generalizes the SVM so that an arbitrary real value can be predicted. Machine learning was compared with the proposed method which reflects the characteristics of wind and the conventional method which does not reflect wind characteristics. To verify the accuracy and feasibility of the proposed method, we used the data collected from three different locations of Jeju Island wind farm. Experimental results show that the error of the proposed method is better than that of general wind power generation.

• KIEAE Journal
• /
• v.12 no.1
• /
• pp.127-132
• /
• 2012

The energy consumption in the world is growing rapidly. And the environmental issues of climate become a important task. The interest in renewable energy like wind and solar is increasing now. Especially, by reducing power transmission loss, a small wind power is getting attention at the residential areas and campus of university. In this study, we attempted to estimate and compare the wind energy density using wind data of AWS (Automatic Weather Station) of H University. In this case of a campus, the weibull distribution parameter C is 2.27, and K is 0.88. According to the data, the energy density of the small wind power is 12.7 W/m2. We did CFD(Computational Fluid Dynamics) simulations at H University campus by 7 wind directions(ENE, ESE, SE, NW, WNW, W, WSW). In the results, we suggest 4 small wind powers. The small wind power generating system can produce 4,514kWh annually.

• Journal of Power System Engineering
• /
• v.21 no.6
• /
• pp.31-39
• /
• 2017

In order to examine if met-masts wind data can exchange each other for wind resource assessment, an investigation was carried out in Kimnyeong and Haengwon regions of Jeju Island. The two regions are both simple terrain and 4.31 km away from each other. The one-year wind speed data measured by 70 m-high anemometers of each met-mast of the two regions were analysed in detail. Measure-Correlate-Predict (MCP) method was applied to the two regions using the 10-year Automatic Weather System (AWS) wind data of Gujwa region for creating 10-year Wind Statistics by running WindPRO software. The two 10-year Wind Statistics were applied to the self-met mast point for self prediction of Annual Energy Production (AEP) and Capacity Factor (CF) and the each other's met mast point for mutual prediction of them. As a result, when self-prediction values were reference, relative errors of mutual prediction values were less than 1% for AEP and CF so that met masts wind data under the same condition of this study could exchange each other for estimating accurate wind resource.

• ETRI Journal
• /
• v.37 no.2
• /
• pp.283-294
• /
• 2015

A model that precisely forecasts how much wind power is generated is critical for making decisions on power generation and infrastructure updates. Existing studies have estimated wind power from wind speed using forecasting models such as ANFIS, SMO, k-NN, and ANN. This study applies a projected clustering technique to identify wind power patterns of wind turbines; profiles the resulting characteristics; and defines hourly and daily power patterns using wind power data collected over a year-long period. A wind power pattern prediction stage uses a time interval feature that is essential for producing representative patterns through a projected clustering technique along with the existing temperature and wind direction from the classifier input. During this stage, this feature is applied to the wind speed, which is the most significant input of a forecasting model. As the test results show, nine hourly power patterns and seven daily power patterns are produced with respect to the Korean wind turbines used in this study. As a result of forecasting the hourly and daily power patterns using the temperature, wind direction, and time interval features for the wind speed, the ANFIS and SMO models show an excellent performance.

• 한국태양에너지학회:학술대회논문집
• /
• 2008.11a
• /
• pp.300-305
• /
• 2008

Using WindPRO that was software for windfarm design developed by EMD from Denmark, wind resources for the western Jeju island were analyzed, and the performance of WindPRO prediction was evaluated in detail. The Hansu site and the Yongdang site that were located in coastal region were selected, and wind data for one year at the two sites were analyzed using WindPRO. As a result, the relative error of the Prediction for annual energy Production and capacity factor was about ${\pm}20%$. For evaluating wind energy more accurately, it is necessary to obtain lots of wind data and real electric power production data from real windfarm.

• 한국신재생에너지학회:학술대회논문집
• /
• 2005.06a
• /
• pp.42-46
• /
• 2005

In spite of fast growing of prediction codes, there is still not negligible uncertainty in their results. This uncertainty affects on the turbine structural design and power production prediction. With the growing size of wind turbine, reducing this uncertainty is becoming one of critical issues for high performance and efficient wind turbine design. In this respect, there are international efforts to evaluate and tune prediction codes of wind turbine. As the reference data for this purpose, field test data is not appropriate because of its uncontrollable wind characteristics and its inherent uncertainty. Wind tunnel can provide controllable wind. For this reason, NREL has done the full scale test of the 10m turbine at NASA-Ames. With this reference data, a blind comparison has been done with participation of 18 organizations with 19 modeling tools. The results were not favorable. In Europe, a similar project is going on. Nine organizations from five countries are participating in the MEXICO project to do full scale wind tunnel tests and calculation with prediction codes. In this study. these two projects were reviewed in respect of wind tunnel test and its contribution. As a conclusion, it is suggested that scale model wind tunnel tests can be a complementary tool to calculation codes which were evaluated worse than expected.

• Journal of the Korean Solar Energy Society
• /
• v.30 no.6
• /
• pp.66-72
• /
• 2010

In this study, we analyzed the characteristics of wind speed and wind direction at different locations in Jeju area using past 10 years observed data and used them in our wind power forecasting model. Generally the strongest hourly wind speeds were observed during daytime(13KST~15KST) whilst the strongest monthly wind speeds were measured during January and February. The analysis with regards to the available wind speeds for power generation gave percentages of 83%, 67%, 65% and 59% of wind speeds over 4m/s for the locations Gosan, Sungsan, Jeju site and Seogwipo site, respectively. Consequently the most favorable periods for power generation in Jeju area are in the winter season and generally during daytime. The predicted wind speed from the forecast model was in average lower(0.7m/s) than the observed wind speed and the correlation coefficient was decreasing with longer prediction times(0.84 for 1h, 0.77 for 12h, 0.72 for 24h and 0.67 for 48h). For the 12hour prediction horizon prediction errors were about 22~23%, increased gradually up to 25~29% for 48 hours predictions.

• Journal of Advanced Marine Engineering and Technology
• /
• v.30 no.6
• /
• pp.669-676
• /
• 2006

The optimum design and the performance analysis software called POSEIDON for the HAWT (Horizontal Axis Wind Turbine) is developed by use of BEMT, which is the standard computational technique for prediction of power curves of wind turbines. The Prandtl's tip loss theory is adopted to consider the blade tip loss. The lift and the drag coefficient of S-809 airfoil are predicted via X-FOIL and the post stall characteristics of S-809 also are estimated by the Viterna's equations.$^{[13]}$ All the predicted aerodynamic characteristics are fairly well agreed with the wind tunnel test results. performed by Sommers in Delft university of technology. The rated power of the testing rotor is 20kW(FIL-20) at design conditions. The experimental aerodynamic parameters and the X-FOIL data are used for the power Prediction of the FIL-20 respectively The comparison results shows good agreement in power prediction.

• Wind and Structures
• /
• v.30 no.4
• /
• pp.393-403
• /
• 2020

One of the growing concerns of the wind energy production is wind ramp events. To improve the wind ramp event forecasts, the nonlinear Kalman filter bias correction method was applied to 24-h wind speed forecasts issued from the WRF model at 70-m height in Zhangbei wind farm, Hebei Province, China for a two-year period. The Kalman filter shows the remarkable ability of improving forecast skill for real-time wind speed forecasts by decreasing RMSE by 32% from 3.26 m s_-1 to 2.21 m s_-1, reducing BIAS almost to zero, and improving correlation from 0.58 to 0.82. The bias correction improves the forecast skill especially in wind speed intervals sensitive to wind power prediction. The fact shows that the Kalman filter is especially suitable for wind power prediction. Moreover, the bias correction method performs well under abrupt weather transition. As to the overall performance for improving the forecast skill of ramp events, the Kalman filter shows noticeable improvements based on POD and TSS. The bias correction increases the POD score of up-ramps from 0.27 to 0.39 and from 0.26 to 0.38 for down-ramps. After bias correction, the TSS score is significantly promoted from 0.12 to 0.26 for up-ramps and from 0.13 to 0.25 for down-ramps.

• New & Renewable Energy
• /
• v.10 no.2
• /
• pp.19-28
• /
• 2014

Wind power generation of 5 MW wind turbine was predicted by using wind measurement data from HeMOSU-1 which is at south west coast of Korea. Time histories of turbulent wind was generated from 10-min mean wind speed and then they were used as input to Bladed to estimated electric power. Those estimated powers are used in both polynominal regression and neural network training. They were compared with each other for daily production and yearly production. Effect of mean wind speed and turbulence intensity were quantitatively analyzed and discussed. This technique further can be used to assess lifetime power of wind turbine.