• Title/Summary/Keyword: Wind prediction

Search Result 942, Processing Time 0.023 seconds

Nonlinear Kalman filter bias correction for wind ramp event forecasts at wind turbine height

  • Xu, Jing-Jing;Xiao, Zi-Niu;Lin, Zhao-Hui
    • 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.

Development of a Deep Learning-based Long-term PredictionGenerative Model of Wind and Sea Conditions for Offshore Wind Farm Maintenance Optimization (해상풍력단지 유지보수 최적화 활용을 위한 풍황 및 해황 장기예측 딥러닝 생성모델 개발)

  • Sang-Hoon Lee;Dae-Ho Kim;Hyuk-Jin Choi;Young-Jin Oh;Seong-Bin Mun
    • Journal of Wind Energy
    • /
    • v.13 no.2
    • /
    • pp.42-52
    • /
    • 2022
  • In this paper, we propose a time-series generation methodology using a generative adversarial network (GAN) for long-term prediction of wind and sea conditions, which are information necessary for operations and maintenance (O&M) planning and optimal plans for offshore wind farms. It is a "Conditional TimeGAN" that is able to control time-series data with monthly conditions while maintaining a time dependency between time-series. For the generated time-series data, the similarity of the statistical distribution by direction was confirmed through wave and wind rose diagram visualization. It was also found that the statistical distribution and feature correlation between the real data and the generated time-series data was similar through PCA, t-SNE, and heat map visualization algorithms. The proposed time-series generation methodology can be applied to monthly or annual marine weather prediction including probabilistic correlations between various features (wind speed, wind direction, wave height, wave direction, wave period and their time-series characteristics). It is expected that it will be able to provide an optimal plan for the maintenance and optimization of offshore wind farms based on more accurate long-term predictions of sea and wind conditions by using the proposed model.

An Estimation of Extreme Wind Speeds Using NCAR Reanalysis Data (NCAR 재해석 자료를 이용한 극한풍속 예측)

  • Kim, Byung-Min;Kim, Hyun-Gi;Kwon, Soon-Yeol;Yoo, Neung-Soo;Paek, In-Su
    • Journal of Industrial Technology
    • /
    • v.35
    • /
    • pp.95-102
    • /
    • 2015
  • Two extreme wind speed prediction models, the EWM(Extreme wind speed model) in IEC61400-1 and the Gumbel method were compared in this study. The two models were used to predict extreme wind speeds of six different sites in Korea and the results were compared with long term wind data. The NCAR reanalysis data were used for inputs to two models. Various periods of input wind data were tried from 1 year to 50 years and the results were compared with the 50 year maximum wind speed of NCAR wind data. It was found that the EWM model underpredicted the extreme wind speed more than 5 % for two sites. Predictions from Gumbel method overpredicted the extreme wind speed or underpredicted it less than 5 % for all cases when the period of the input data is longer than 10 years. The period of the input wind data less than 3 years resulted in large prediction errors for Gumbel method. Predictions from the EWM model were not, however, much affected by the period of the input wind data.

  • PDF

Evaluation of the Performance on WindPRO Prediction in the Northeast Region of Jeju Island (제주 북동부지역을 대상으로 한 WindPRO의 예측성능 평가)

  • Oh, Hyun-Seok;Ko, Kyung-Nam;Huh, Jong-Chul
    • Journal of the Korean Solar Energy Society
    • /
    • v.29 no.2
    • /
    • pp.22-30
    • /
    • 2009
  • In order to clarify predictive accuracy for the wind resource predicted by running WindPRO(Ver. 2.5) which is software for wind farm design developed by EMD from Denmark, an investigation was carried out at the northeast region of Jeju island. The Hangwon, Susan and Hoichun sites of Jeju island were selected for this study. The measurement period of wind at the sites was for one year. As a result, when the sites had different energy roses, though the two Wind Statistics made by STATGEN module were used for the prediction, it was difficult to exactly predict the energy rose at a given site. On the other hand, when the two Wind Statistics were used to predict the average wind speed, the wind power density and the annual energy production, the relative error was under ${\pm}20%$ which improved more than that when using only one Wind Statistics.

Prediction of skewness and kurtosis of pressure coefficients on a low-rise building by deep learning

  • Youqin Huang;Guanheng Ou;Jiyang Fu;Huifan Wu
    • Wind and Structures
    • /
    • v.36 no.6
    • /
    • pp.393-404
    • /
    • 2023
  • Skewness and kurtosis are important higher-order statistics for simulating non-Gaussian wind pressure series on low-rise buildings, but their predictions are less studied in comparison with those of the low order statistics as mean and rms. The distribution gradients of skewness and kurtosis on roofs are evidently higher than those of mean and rms, which increases their prediction difficulty. The conventional artificial neural networks (ANNs) used for predicting mean and rms show unsatisfactory accuracy in predicting skewness and kurtosis owing to the limited capacity of shallow learning of ANNs. In this work, the deep neural networks (DNNs) model with the ability of deep learning is introduced to predict the skewness and kurtosis on a low-rise building. For obtaining the optimal generalization of the DNNs model, the hyper parameters are automatically determined by Bayesian Optimization (BO). Moreover, for providing a benchmark for future studies on predicting higher order statistics, the data sets for training and testing the DNNs model are extracted from the internationally open NIST-UWO database, and the prediction errors of all taps are comprehensively quantified by various error metrices. The results show that the prediction accuracy in this study is apparently better than that in the literature, since the correlation coefficient between the predicted and experimental results is 0.99 and 0.75 in this paper and the literature respectively. In the untrained cornering wind direction, the distributions of skewness and kurtosis are well captured by DNNs on the whole building including the roof corner with strong non-normality, and the correlation coefficients between the predicted and experimental results are 0.99 and 0.95 for skewness and kurtosis respectively.

Medium.Large Horizontal Axis Wind Turbine Noise Analysis Considering Blade Passing Frequency Noise and Retarded Time (블레이드 통과 주파수 소음과 지연시간을 고려한 중.대형 수평축 풍력발전기의 공력소음해석)

  • Kim, Hyun-Jung;Kim, Ho-Geon;Lee, Soo-Gab
    • Proceedings of the Korean Society for Noise and Vibration Engineering Conference
    • /
    • 2007.11a
    • /
    • pp.1490-1493
    • /
    • 2007
  • Aerodynamic noise generated from wind turbines is predicted by it's classified source mechanisms using computational method. BPF noise according to the blade passing motion, is modelled on monopole and dipole sources. They are predicted by Farassat 1A equation. Airfoil self noise and turbulence ingestion noise are modelled upon quadrupole sources and are predicted by semi-empirical formulas composed on the groundwork of Brooks et al. and Lowson. Retarded time is considered, not only in low frequency noise prediction but also in turbulence ingestion noise and airfoil self noise prediction. Wind turbine noise emission of a 3MW wind turbine and a 600 kW wind turbine, standing for large and middle sized wind turbines, is analyzed.

  • PDF

Prediction of downburst-induced wind pressure coefficients on high-rise building surfaces using BP neural network

  • Fang, Zhiyuan;Wang, Zhisong;Li, Zhengliang
    • Wind and Structures
    • /
    • v.30 no.3
    • /
    • pp.289-298
    • /
    • 2020
  • Gusts generated by downburst have caused a great variety of structural damages in many regions around the world. It is of great significance to accurately evaluate the downburst-induced wind load on high-rise building for the wind resistance design. The main objective of this paper is to propose a computational modeling approach which can satisfactorily predict the mean and fluctuating wind pressure coefficients induced by downburst on high-rise building surfaces. In this study, using an impinging jet to simulate downburst-like wind, and simultaneous pressure measurements are obtained on a high-rise building model at different radial locations. The model test data are used as the database for developing back propagation neural network (BPNN) models. Comparisons between the BPNN prediction results and those from impinging jet test demonstrate that the BPNN-based method can satisfactorily and efficiently predict the downburst-induced wind pressure coefficients on single and overall surfaces of high-rise building at various radial locations.

Extreme wind prediction and zoning

  • Holmes, J.D.;Kasperski, M.;Miller, C.A.;Zuranski, J.A.;Choi, E.C.C.
    • Wind and Structures
    • /
    • v.8 no.4
    • /
    • pp.269-281
    • /
    • 2005
  • The paper describes the work of the IAWE Working Group WGF - Extreme Wind Prediction and Zoning, one of the international codification working groups set up in 2000. The topics covered are: the international database of extreme winds, quality assurance and data quality, averaging times, return periods, probability distributions and fitting methods, mixed wind climates, directionality effects, the influence of orography, rare events and simulation methods, long-term climate change, and zoning and mapping. Recommendations are given to promote the future alignment of international codes and standards for wind loading.

Optimization Calculations and Machine Learning Aimed at Reduction of Wind Forces Acting on Tall Buildings and Mitigation of Wind Environment

  • Tanaka, Hideyuki;Matsuoka, Yasutomo;Kawakami, Takuma;Azegami, Yasuhiko;Yamamoto, Masashi;Ohtake, Kazuo;Sone, Takayuki
    • International Journal of High-Rise Buildings
    • /
    • v.8 no.4
    • /
    • pp.291-302
    • /
    • 2019
  • We performed calculations combining optimization technologies and Computational Fluid Dynamics (CFD) aimed at reducing wind forces and mitigating wind environments (local strong winds) around buildings. However, the Reynolds Averaged Navier-stokes Simulation (RANS), which seems somewhat inaccurate, needs to be used to create a realistic CFD optimization tool. Therefore, in this study we explored the possibilities of optimizing calculations using RANS. We were able to demonstrate that building configurations advantageous to wind forces could be predicted even with RANS. We also demonstrated that building layouts was more effective than building configurations in mitigating local strong winds around tall buildings. Additionally, we used the Convolutional Neural Network (CNN) as an airflow prediction method alternative to CFD in order to increase the speed of optimization calculations, and validated its prediction accuracy.

Variation of AEP to wind direction variability (풍향의 변동성에 따른 연간에너지 발전량의 변화)

  • Kim, Hyeon-Gi;Kim, Byeong-Min;Paek, In-Su;Yoo, Neung-Soo;Kim, Hyun-Goo
    • Journal of the Korean Solar Energy Society
    • /
    • v.31 no.5
    • /
    • pp.1-8
    • /
    • 2011
  • In this study, we performed a sensitivity analysis to see how the true north error of a wind direction vane installed to a meteorological mast affects predictions of the annual-average wind speed and the annual energy production. For this study, two meteorological masts were installed with a distance of about 4km on the ridge in complex terrain and the wind speed and direction were measured for one year. Cross predictions of the wind speed and the AEP of a virtual wind turbine for two sites in complex terrain were performed by changing the wind direction from $-45^{\circ}$ to $45^{\circ}$with an interval of $5^{\circ}$. A commercial wind resource prediction program, WindPRO, was used for the study. It was found that the prediction errors in the AEP caused by the wind direction errors occurred up to more than 20% depending on the orography and the main wind direction at that site.