scholarly journals Decomposition-Based Multistep Sea Wind Speed Forecasting Using Stacked Gated Recurrent Unit Improved by Residual Connections

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-14
Author(s):  
Jupeng Xie ◽  
Huajun Zhang ◽  
Linfan Liu ◽  
Mengchuan Li ◽  
Yixin Su
Keyword(s):  
Wind Speed ◽  
Wind Speed Data ◽  
Wind Speeds ◽  
Mode Decomposition ◽  
Proposed Model ◽  
High Volatility ◽  
Deep Layers ◽  
Sea Wind ◽  
Gated Recurrent Unit ◽  
Degradation Effect

Sea wind speed forecast is important for meteorological navigation system to keep ships in safe areas. The high volatility and uncertainty of wind make it difficult to accurately forecast multistep wind speed. This paper proposes a new decomposition-based model to forecast hourly sea wind speeds. Because mode mixing affects the accuracy of the empirical mode decomposition- (EMD-) based models, this model uses the variational mode decomposition (VMD) to alleviate this problem. To improve the accuracy of predicting subseries with high nonlinearity, this model uses stacked gate recurrent units (GRU) networks. To alleviate the degradation effect of stacked GRU, this model modifies them by adding residual connections to the deep layers. This model decomposes the nonlinear wind speed data into four subseries with different frequencies adaptively. Each stacked GRU predictor has four layers and the residual connections are added to the last two layers. The predictors have 24 inputs and 3 outputs, and the forecast is an ensemble of five predictors’ outputs. The proposed model can predict wind speed in the next 3 hours according to the past 24 hours’ wind speed data. The experiment results on three different sea areas show that the performance of this model surpasses those of a state-of-the-art model, several benchmarks, and decomposition-based models.

scholarly journals A New Hybrid Approach for Wind Speed Forecasting Applying Support Vector Machine with Ensemble Empirical Mode Decomposition and Cuckoo Search Algorithm

2018 ◽  
Vol 8 (10) ◽  
pp. 1754 ◽  
Author(s):  
Tongxiang Liu ◽  
Shenzhong Liu ◽  
Jiani Heng ◽  
Yuyang Gao
Keyword(s):  
Wind Speed ◽  
Search Algorithm ◽  
Wind Farms ◽  
Cuckoo Search ◽  
Ensemble Empirical Mode Decomposition ◽  
Cuckoo Search Algorithm ◽  
Support Vector ◽  
Wind Speed Forecasting ◽  
Mode Decomposition ◽  
Proposed Model

Wind speed forecasting plays a crucial role in improving the efficiency of wind farms, and increases the competitive advantage of wind power in the global electricity market. Many forecasting models have been proposed, aiming to enhance the forecast performance. However, some traditional models used in our experiment have the drawback of ignoring the importance of data preprocessing and the necessity of parameter optimization, which often results in poor forecasting performance. Therefore, in order to achieve a more satisfying performance in forecasting wind speed data, a new short-term wind speed forecasting method which consists of Ensemble Empirical Mode Decomposition (EEMD) for data preprocessing, and the Support Vector Machine (SVM)—whose key parameters are optimized by the Cuckoo Search Algorithm (CSO)—is developed in this paper. This method avoids the shortcomings of some traditional models and effectively enhances the forecasting ability. To test the prediction ability of the proposed model, 10 min wind speed data from wind farms in Shandong Province, China, are used for conducting experiments. The experimental results indicate that the proposed model cannot only improve the forecasting accuracy, but can also be an effective tool in assisting the management of wind power plants.

scholarly journals Dutch Offshore Wind Atlas Validation against Cabauw Meteomast Wind Measurements

Energies ◽  
2020 ◽  
Vol 13 (24) ◽  
pp. 6558
Author(s):  
Steven Knoop ◽  
Pooja Ramakrishnan ◽  
Ine Wijnant
Keyword(s):  
Wind Speed ◽  
Offshore Wind ◽  
Weather Forecasts ◽  
Wind Speeds ◽  
Weather Model ◽  
Significant Difference ◽  
Wind Measurements ◽  
Sea Wind ◽  
Global Datasets ◽  
Meteorological Institute

The Dutch Offshore Wind Atlas (DOWA) is validated against wind speed and direction measurements from the Cabauw meteorological mast for a 10-year period and at heights between 10 m and 200 m. The validation results are compared to the Royal Netherlands Meteorological Institute (KNMI) North Sea Wind (KNW) atlas. It is found that the average difference (bias) between DOWA wind speeds and those measured at Cabauw varies for the different heights between −0.1 m/s to 0.3 m/s. Significant differences between DOWA and KNW are only found at altitudes of 10 m and 20 m, where KNW performs better. For heights above 20 m, there is no significant difference between DOWA and KNW with respect to the 10-year averaged wind speed bias. The diurnal cycle is better captured by DOWA compared to KNW, and the hourly correlation is slightly improved. In addition, a comparison with the global European Center for Medium-Range Weather Forecasts (ECMWF) ERA-Interim and ERA5 reanalyses (used for KNW and DOWA, respectively) is made, highlighting the added skill provided by downscaling those global datasets with the weather model HARMONIE.

scholarly journals Frequency Distribution Model of Wind Speed Based on the Exponential Polynomial for Wind Farms

2019 ◽  
Vol 11 (3) ◽  
pp. 665 ◽  
Author(s):  
Lingzhi Wang ◽  
Jun Liu ◽  
Fucai Qian
Keyword(s):  
Wind Speed ◽  
Frequency Distribution ◽  
Least Squares Method ◽  
Wind Farms ◽  
Rayleigh Distribution ◽  
Distribution Model ◽  
Exponential Polynomial ◽  
Distribution Models ◽  
Wind Speed Data ◽  
Wind Speeds

This study introduces and analyses existing models of wind speed frequency distribution in wind farms, such as the Weibull distribution model, the Rayleigh distribution model, and the lognormal distribution model. Inspired by the shortcomings of these models, we propose a distribution model based on an exponential polynomial, which can describe the actual wind speed frequency distribution. The fitting error of other common distribution models is too large at zero or low wind speeds. The proposed model can solve this problem. The exponential polynomial distribution model can fit multimodal distribution wind speed data as well as unimodal distribution wind speed data. We used the linear-least-squares method to acquire the parameters for the distribution model. Finally, we carried out contrast simulation experiments to validate the effectiveness and advantages of the proposed distribution model.

scholarly journals A New Wind Speed Forecasting Modeling Strategy Using Two-Stage Decomposition, Feature Selection and DAWNN

Energies ◽  
2019 ◽  
Vol 12 (3) ◽  
pp. 334 ◽  
Author(s):  
Sizhou Sun ◽  
Lisheng Wei ◽  
Jie Xu ◽  
Zhenni Jin
Keyword(s):  
Feature Selection ◽  
Wind Speed ◽  
Wind Farms ◽  
Frequency Component ◽  
Ensemble Empirical Mode Decomposition ◽  
Two Stage ◽  
Wind Speed Forecasting ◽  
Wind Speed Data ◽  
Wind Speed Prediction ◽  
Mode Decomposition

Accurate wind speed prediction plays a crucial role on the routine operational management of wind farms. However, the irregular characteristics of wind speed time series makes it hard to predict accurately. This study develops a novel forecasting strategy for multi-step wind speed forecasting (WSF) and illustrates its effectiveness. During the WSF process, a two-stage signal decomposition method combining ensemble empirical mode decomposition (EEMD) and variational mode decomposition (VMD) is exploited to decompose the empirical wind speed data. The EEMD algorithm is firstly employed to disassemble wind speed data into several intrinsic mode function (IMFs) and one residual (Res). The highest frequency component, IMF1, obtained by EEMD is further disassembled into different modes by the VMD algorithm. Then, feature selection is applied to eliminate the illusive components in the input-matrix predetermined by partial autocorrelation function (PACF) and the parameters in the proposed wavelet neural network (WNN) model are optimized for improving the forecasting performance, which are realized by hybrid backtracking search optimization algorithm (HBSA) integrating binary-valued BSA (BBSA) with real-valued BSA (RBSA), simultaneously. Combinations of Morlet function and Mexican hat function by weighted coefficient are constructed as activation functions for WNN, namely DAWNN, to enhance its regression performance. In the end, the final WSF values are obtained by assembling the prediction results of each decomposed components. Two sets of actual wind speed data are applied to evaluate and analyze the proposed forecasting strategy. Forecasting results, comparisons, and analysis illustrate that the proposed EEMD/VMD-HSBA-DAWNN is an effective model when employed in multi-step WSF.

scholarly journals Performance evaluation of Weibull function for wind data analysis in two selected locations in North-Western, Nigeria

2018 ◽  
Vol 6 (1) ◽  
pp. 18
Author(s):  
Boluwaji Olomiyesan
Keyword(s):  
Wind Speed ◽  
Wind Power ◽  
Predictive Ability ◽  
Maximum Energy ◽  
Wind Data ◽  
Weibull Function ◽  
Wind Speed Data ◽  
Wind Speeds ◽  
The North ◽  
North Western

In this study, the predictive ability of two-parameter Weibull distribution function in analyzing wind speed data was assessed in two selected sites with different mean wind speeds in the North-Western region of Nigeria. Twenty-two years wind speed data spanning from 1984 to 2005 was used in the analysis. The data were obtained from the Nigerian Meteorological Agency (NIMET) in Lagos. The results of the analysis show that Weibull function is suitable for analyzing measured wind speed data and in predicting the wind-power density in both locations and that Weibull function is not discriminative between locations with high and low mean wind speeds in analyzing wind data. The annual mean wind speeds for the two sites (Sokoto and Yelwa) are 7.99 ms-1 and 2.59 ms-1 respectively, while the annual values of the most probable wind speed and the maximum, energy-carrying wind speeds are respectively:3.52 and 4.34 ms-1 for Yelwa and 8.33 and 9.02 ms-1 for Sokoto. The estimated annual wind power densities for Yelwa and Sokoto are respectively 36.91 and 359.96 Wm-2. Therefore, Sokoto has a better prospect for wind power generation.

scholarly journals A Hybrid Approach for Short-Term Forecasting of Wind Speed

2013 ◽  
Vol 2013 ◽  
pp. 1-8 ◽  
Author(s):  
Sivanagaraja Tatinati ◽  
Kalyana C. Veluvolu
Keyword(s):  
Wind Speed ◽  
Hybrid Method ◽  
Hybrid Approach ◽  
Adaptive Methods ◽  
Correlation Factor ◽  
Support Vector ◽  
Intrinsic Mode Functions ◽  
Wind Speed Data ◽  
Mode Decomposition ◽  
The Individual

We propose a hybrid method for forecasting the wind speed. The wind speed data is first decomposed into intrinsic mode functions (IMFs) with empirical mode decomposition. Based on the partial autocorrelation factor of the individual IMFs, adaptive methods are then employed for the prediction of IMFs. Least squares-support vector machines are employed for IMFs with weak correlation factor, and autoregressive model with Kalman filter is employed for IMFs with high correlation factor. Multistep prediction with the proposed hybrid method resulted in improved forecasting. Results with wind speed data show that the proposed method provides better forecasting compared to the existing methods.

scholarly journals A "Global Radiosonde and tracked-balloon Archive on Sixteen Pressure levels" (GRASP) going back to 1905 – Part 2: homogeneity adjustments for pilot balloon and radiosonde wind data

2014 ◽  
Vol 6 (2) ◽  
pp. 297-316 ◽  
Author(s):  
L. Ramella Pralungo ◽  
L. Haimberger
Keyword(s):  
Time Series ◽  
Wind Speed ◽  
Wind Direction ◽  
Sampling Bias ◽  
Observation Time ◽  
Early Years ◽  
Wind Speed Data ◽  
Wind Speeds ◽  
Measurement Biases ◽  
The 1960S

Abstract. This paper describes the comprehensive homogenization of the "Global Radiosonde and tracked balloon Archive on Sixteen Pressure levels" (GRASP) wind records. Many of those records suffer from artificial shifts that need to be detected and adjusted before they are suitable for climate studies. Time series of departures between observations and the National Atmospheric and Oceanic Administration 20th-century (NOAA-20CR) surface pressure only reanalysis have been calculated offline by first interpolating the observations to pressure levels and standard synoptic times, if needed, and then interpolating the gridded NOAA-20CR standard pressure level data horizontally to the observation locations. These difference time series are quite sensitive to breaks in the observation time series and can be used for both automatic detection and adjustment of the breaks. Both wind speed and direction show a comparable number of breaks, roughly one break in three stations. More than a hundred artificial shifts in wind direction could be detected at several US stations in the period 1938/1955. From the 1960s onward the wind direction breaks are less frequent. Wind speed data are not affected as much by measurement biases, but one has to be aware of a large fair-weather sampling bias in early years, when high wind speeds were much less likely to be observed than after 1960, when radar tracking was already common practice. This bias has to be taken into account when calculating trends or monthly means from wind speed data. Trends of both wind speed and direction look spatially more homogeneous after adjustment. With the exception of a widespread wind direction bias found in the early US network, no signs of pervasive measurement biases could be found. The adjustments can likely improve observation usage when applied during data assimilation. Alternatively they can serve as a basis for validating variational wind bias adjustment schemes. Certainly, they are expected to improve estimates of global wind trends. All the homogeneity adjustments are available in the PANGAEA archive with associated doi:10.1594/PANGAEA.823617.

scholarly journals A Basic Wind Speed Map for Oman

2014 ◽  
Vol 11 (2) ◽  
pp. 64
Author(s):  
A.S. Alnuaimi ◽  
M.A. Mohsin ◽  
K.H. Al-Riyami
Keyword(s):  
Wind Speed ◽  
Return Period ◽  
Structural Design ◽  
Short Term ◽  
Wind Speed Data ◽  
Wind Speeds ◽  
Mean Wind Speed

The aim of this research was to develop the first basic wind speed map for Oman. Hourly mean wind speed records from 40 metrological stations were used in the calculation. The period of continuous records ranged from 4–37 years. The maximum monthly hourly mean and the maxima annual hourly mean wind speed data were analysed using the Gumbel and Gringorten methods. Both methods gave close results in determining basic wind speeds, with the Gumbel method giving slightly higher values. Due to a lack of long-term records in some regions of Oman, basic wind speeds were extrapolated for some stations with only short-term records, which were defined as those with only 4– 8 years of continuous records; in these cases, monthly maxima were used to predict the long-term basic wind speeds. Accordingly, a basic wind speed map was developed for a 50-year return period. This map was based on basic wind speeds calculated from actual annual maxima records of 29 stations with at least 9 continuous years of records as well as predicted annual maxima wind speeds for 11 short-term record stations. The basic wind speed values ranged from 16 meters/second (m/s) to 31 m/s. The basic wind speed map developed in this research is recommended for use as a guide for structural design in Oman. 

scholarly journals Deterministic prediction of surface wind speed variations

2014 ◽  
Vol 32 (11) ◽  
pp. 1415-1425 ◽  
Author(s):  
G. V. Drisya ◽  
D. C. Kiplangat ◽  
K. Asokan ◽  
K. Satheesh Kumar
Keyword(s):  
Wind Speed ◽  
Arima Model ◽  
Surface Wind ◽  
Surface Wind Speed ◽  
Temporal Variations ◽  
Short Term ◽  
Deterministic Models ◽  
Wind Speed Data ◽  
Wind Speeds ◽  
Past Data

Abstract. Accurate prediction of wind speed is an important aspect of various tasks related to wind energy management such as wind turbine predictive control and wind power scheduling. The most typical characteristic of wind speed data is its persistent temporal variations. Most of the techniques reported in the literature for prediction of wind speed and power are based on statistical methods or probabilistic distribution of wind speed data. In this paper we demonstrate that deterministic forecasting methods can make accurate short-term predictions of wind speed using past data, at locations where the wind dynamics exhibit chaotic behaviour. The predictions are remarkably accurate up to 1 h with a normalised RMSE (root mean square error) of less than 0.02 and reasonably accurate up to 3 h with an error of less than 0.06. Repeated application of these methods at 234 different geographical locations for predicting wind speeds at 30-day intervals for 3 years reveals that the accuracy of prediction is more or less the same across all locations and time periods. Comparison of the results with f-ARIMA model predictions shows that the deterministic models with suitable parameters are capable of returning improved prediction accuracy and capturing the dynamical variations of the actual time series more faithfully. These methods are simple and computationally efficient and require only records of past data for making short-term wind speed forecasts within practically tolerable margin of errors.

scholarly journals A global radiosonde and tracked balloon archive on 16 pressure levels (GRASP) back to 1905 – Part 2: Homogeneity adjustments for PILOT and radiosonde wind data

2014 ◽  
Vol 7 (1) ◽  
pp. 335-383 ◽  
Author(s):  
L. Ramella Pralungo ◽  
L. Haimberger
Keyword(s):  
Time Series ◽  
Wind Speed ◽  
Wind Direction ◽  
Sampling Bias ◽  
Observation Time ◽  
Early Years ◽  
Wind Speed Data ◽  
Wind Speeds ◽  
Measurement Biases ◽  
The 1960S

Abstract. This paper describes the comprehensive homogenization of the GRASP wind records. Many of those records suffer from artificial shifts that need to be detected and adjusted before they are suitable for climate studies. Time series of departures between observations and the National Atmospheric and Oceanic Administration 20th century (NOAA-20CR) surface pressure only reanalysis have been calculated offline by first interpolating the observations to pressure levels and standard synoptic times, if needed, and then interpolating the gridded NOAA-20CR standard pressure level data horizontally to the observation locations. These difference time series are quite sensitive to breaks in the observation time series and can be used for both automatic detection and adjustment of the breaks. Both wind speed and direction show a comparable number of breaks, roughly one break in three stations. More than hundred artificial shifts in wind direction could be detected at several US stations in the period 1938/1955. From the 1960s onward the wind direction breaks are less frequent. Wind speed data are not so much affected by measurement biases but one has to be aware of a large fair weather sampling bias in early years when high wind speeds were much less likely to be observed than after 1960 when RADAR tracking was already common practice. It has to be taken into account when calculating trends or monthly means from wind speed data. Trends of both wind speed and direction look spatially more homogeneous after adjustment. With the exception of a widespread wind direction bias found in the early US network no signs of pervasive measurement biases could be found. The adjustments can likely improve observation usage when applied during data assimilation. Alternatively they can serve as basis for validating variational wind bias adjustment schemes. Certainly they are expected to improve estimates of global wind trends. All the homogeneity adjustments are available in the PANGAEA archive with the associated DOI doi:10.1594/PANGAEA.823617.

Sign in / Sign up

Export Citation Format

Share Document