Narrow Results

Day-ahead prediction of wind speed is a basic and key problem of large-scale wind power penetration. Many current techniques fail to satisfy practical engineering requirements because of wind speed's strong nonlinear features, influenced by many complex factors, and the general model's inability to automatically learn features. It is well recognized that wind speed varies in different patterns. In this paper, we propose a deep feature learning (DFL) approach to wind speed forecasting because of its advantages at both multi-layer feature extraction and unsupervised learning. A deep belief network (DBN) model for regression with an architecture of 144 input and 144 output nodes was constructed using a restricted Boltzmann machine (RBM). Day-ahead prediction experiments were then carried out. By comparing the experimental results, it was found that the prediction errors with respect to both size and stability of a DBN model with only three hidden layers were less than those of the other three typical approaches including support vector regression (SVR), single hidden layer neural networks (SHL-NN), and neural networks with three hidden layers (THL-NN). In addition, the DBN model can learn and obtain complex features of wind speed through its strong nonlinear mapping ability, which effectively improves its prediction precision. In addition, prediction errors are minimized when the number of DBN model's hidden layers reaches a threshold value. Above this number, it is not possible to improve the prediction accuracy by further increasing the number of hidden layers. Thus, the DBN method has a high practical value for wind speed prediction.

Short-term wind speed prediction is an essential task for wind resource and wind energy planning. However, most of this literature does not take into account the spatio-termporal correlation of wind data from the geographical field. For this reason, we propose an integrated spatio-temporal kriging and functional kriging strategy to exploit such spatio-temporal correlation into the wind speed prediction. First, the deterministic trend component in wind data is estimated to be removed. The residuals are used for spatio-temporal modeling and prediction. Based on the spatio-temporal kriging framework, four spatio-temporal covariance models (product-sum model, separable exponential product model, separable and nonseparable Gneiting models) are considered which describe the spatio-temporal correlation of wind data. In particular, the flexibility of using the nonseparable Gneiting model is highlighted. More specifically, four spatio-temporal random fields are modeled from the 12 wind monitoring stations over Ireland. We also use an involved weighted least squares method for estimating parameters of the four covariance models involved in the spatio-temporal kriging strategy. We apply the fitted covariance models to generate day-ahead wind speed predictions at both observed and nonobserved locations where wind station already exist but also to nearby locations. Leave-one-out cross-validation is applied to check the significance of the difference among the four models, these spatio-temporal ordinary kriging (STOK), functional ordinary kriging (FOK) and autoregressive integrated moving average (ARIMA) methods are compared for day-ahead wind speed predictions. Forecasting results indicate that the predicting accuracy is improved almost 33.5% using FOK compared with three approaches which confirm the effectiveness of the functional kriging method in the paper.

It is essential to enhance the ability of wind speeds forecasting for wind energy and wind resource planning. For this purpose, a hybrid strategy has been proposed based on spatio-temporal covariance model which combined the spatio-temporal ordinary kriging (STOK) technology with autoregressive integrated moving average (ARIMA) regression smoothing method. This is because wind speed time series exhibits a long-term dependency. In the case study, both STOK method and ARIMA method are employed and their performances are compared. The ARIMA model can obtain a necessary and sufficient smoothing condition for them to be smoothed. Meanwhile, further theoretical analysis is provided to discuss why the STOK method is potentially more accurate than the ARIMA method for wind speed time series prediction. Results show that the proposed method outperforms the Non-Sep-Gneiting model by 9% and 7.2% in terms of mean absolute error (MAE) and root-mean-square error (RMSE).

In this paper, an analysis of temporal variation of wind speed and wind direction recorded at 10 min intervals are presented. The measurements were carried out at Hambanthota, a site located in the southern coastal belt of Sri Lanka which has a high potential for wind power generation. The multifractal detrended fluctuation analysis was used to analyze the temporal scaling properties of wind speeds and wind directions. The analysis was carried out for seasonal variation of wind speed and wind direction. It was observed that the scaling behavior of wind speed in Hambanthota is similar to the scaling behavior observed in previous studies which were carried out in other parts of the world. The seasonal wind and wind direction change exhibits different scaling behavior. No difference in scaling behavior was observed with heights. The degree of multifractality is high for wind direction when compared with wind speed for each season.

This paper aims to develop a wind speed prediction model by utilizing deep learning and neural networks. The analysis of weather data using a neural network architecture has been completed. The Long Short-Term Memory (LSTM) architecture is a type of artificial Recurrent Neural Network (RNN) used in deep learning is the first method plots the predicting Wind Speed based on the dataset and predicts the future spread. A dataset from a real-time weather station is used in the implementation model. The dataset consists of information from the weather station implements of the recurrent neural network model that plots the past spread and predicts the future stretch of the weather. The performance of the recurrent neural network model is presented and compared with Adaline neural network, Autoregressive Neural Network (NAR), and Group Method of Data Handling (GMDH). The NAR used three hidden layers. The performance of the model is analyzed by presenting the Wind Speeds of Erbil city. The dataset consists of the Wind Speed of (1992-2020) years, and each year consist of twelve months (from January to December).

This paper proposes a novel method for estimating the evolutionary power spectral density (EPSD) of a nonstationary process based on a single sample. In the proposed method, a sample of a nonstationary process is decomposed into several components with a new binomial fitting decomposition (BFD). The EPSD of each component can be estimated using a newly proposed time-varying standard deviation estimation method and short-time Thomson multiple-window spectrum estimation method. The EPSD of the analyzed nonstationary sample is obtained by combining the EPSDs of all components. Via a comprehensive numerical study, the applicability of the proposed EPSD estimation method (for estimating the EPSD of a nonstationary process) is analyzed and compared with those by the Priestley method and wavelet-based method. The numerical results indicate that the estimated EPSD by the proposed method is more consistent with the corresponding theoretical one than those by the other two methods. Finally, the EPSDs of Storm Ampil, measured atop the Shanghai World Financial Center, are analyzed by the proposed method.

The complexity of the atmosphere endows it with the property of turbulence by virtue of which, wind speed variations in the atmospheric boundary layer (ABL) exhibit highly irregular fluctuations that persist over a wide range of temporal and spatial scales. Despite the large and significant body of work on microscale turbulence, understanding the statistics of atmospheric wind speed variations has proved to be elusive and challenging. Knowledge about the nature of wind speed at ABL has far reaching impact on several fields of research such as meteorology, hydrology, agriculture, pollutant dispersion, and more importantly wind energy generation. In the present study, temporal wind speed records from twenty eight stations distributed through out the state of North Dakota (ND, USA), (~ 70,000 square-miles) and spanning a period of nearly eight years are analyzed. We show that these records exhibit a characteristic broad multifractal spectrum irrespective of the geographical location and topography. The rapid progression of air masses with distinct qualitative characteristics originating from Polar regions, Gulf of Mexico and Northern Pacific account for irregular changes in the local weather system in ND. We hypothesize that one of the primary reasons for the observed multifractal structure could be the irregular recurrence and confluence of these three air masses.

Wind energy has been the focus of attention by governments over the past several decades because it is clean and renewable energy source that offers many advantages compared to others. This paper reports a particle swarm optimization based flowchart used to identify the most appropriate wind speed distribution function. In this procedure, the PSO algorithm modifies and estimates the parameters of the distribution functions examined. The distributions of Nakagami, Normal and Weibull are judged for right-skewed, non-skewed and left-skewed distributions, respectively. The outcomes showed that each function offers impressive performance compared to the others in its defined conditions. Despite the changes of one-component distribution’s parameters made by the PSO algorithm are not significant, the same changes will affect the function’s performance greatly. Because of the sufficient flexibility offered by the nominated functions, the proposed flowchart is applicable for all wind regimes.

Surface velocimetry has been steadily increasing interest and research on its applicability to stationary discharge measurement due to its measurement efficiency and economy. However, it has not yet been widely used due to some uncertainties in the measurement of surface velocity. The first is the uncertainty of the relation between the surface and mean velocity, and the others is for wind effect on surface velocity.

In the case of measurement of water surface velocity, the direction and height of the wave on water surface can be changed by wind, which may cause an error in the surface velocity value. For higher velocity in flood flow, the influence of the wind speed is negligible on surface velocity, but as the velocity decreases, the influence of the wind effect on the surface velocity increases relatively.

In this study, in order to analyze the effect of wind speed and direction on the surface velocity under constant flow conditions when measuring discharge using surface velocity meter, wind direction anemometer(150WX of AIRMAR Co.) was installed with radar surface velocimetry(RQ-30 of SOMMER co.) in an test river.

From the results, speed and direction of wind was found to affect the measured surface velocity in low flow condition. Wind speeds of 2.5 m/s or more caused fluctuations of 20.0~71.4 % of the surface velocity depending on the wind direction, while the effect was not significant as 2.9~8.6 % at less than 1 m/s of wind speed. Therefore, for calculating the discharge by measuring the surface velocity, the effect of wind speed and direction should be considered, especially in low flow conditions where the influence of wind is relatively large.

Based on grid icing observation data of Erlang Mountain region in the winter during 2013-2014, the characteristics of the grid icing in the regional Erlang Mountain and time variations were researched. The regional power grid and icing intensity classified into mild, moderate, severe icing, and ice characteristics of the growth process were also studied. Air temperature, wind direction, wind speed and other meteorological elements were discussed in the influence of the strength of the power grid ice. The results showed that: (1) in Erlang mountain areas, most icing are mild to moderate icing, severe icing phenomenon is relatively rare, except for the special effects of the weather system. (2) When the temperature is low, roughly -5 °C ∼ -8 °C is the most conducive to ice crystals due to the air in the water, constant humidity and the lower the air temperature, the faster the ice formation. Certain temperature and humidity result in longer freezing and thicker ice. Wind speed plays a role in transporting water vapor and water droplets have an important influence on the formation of ice. This study demonstrates when the wind speed is at 2 ∼ 6m/s, ice forms the fastest. (3) Temperature, wind speed and quantitative relationship of ice thickness are linear correlation.

In this chapter, support vector regression and whale optimization algorithm (WOA) are presented for long-term prediction of wind speed. The WOA is adapted for optimizing the support vector regression (SVR) parameters so that the prediction error can be reduced. The rendering of the proposed algorithm is evaluated using three different measurements including forecasting based-measurements, statistical analyses, and stability. The daily average wind speed data from Space Weather Monitoring Center (SWMC) in Egypt was selected in the experiments. The experimental results showed that the suggested WOA algorithm is eligible for finding the best values of SVR parameters, avoiding local optima issue and it is competitive for wind speed forecasting. The result also demonstrates lower classification error rates compared with traditional SVR algorithm. The experimental result also proved that the WOA-SVR algorithm utilizing Linear kernel accomplished lesser classification fault rates than RBF kernel.

This paper investigates the spectral characteristics of low frequency fluctuations of natural wind velocity and atmospheric pressure, and the resonant evolution of long period waves by low frequency fluctuations of winds in a laboratory experiment. Laboratory experiments on the responses of wind waves under periodically fluctuating winds have been conducted to find an alternative generation mechanism for long period waves. The experiments show that the long period waves developed at the peak frequency of the low frequency fluctuations of wind speed and air pressure. Through the spectral analysis of fluctuating components of wind speed, air pressure and water surface elevation, the long period waves are found to be generated and resonantly developed by the fluctuating components of surface shear stress due to periodically fluctuating wind speed.

Seasonal variation of residual currents in the Meghna Estuary, located at the northern part of the Bay of Bengal, has been investigated through the use of a 3D numerical model. Residual current in the Meghna Estuary appears to be strongly influenced by tidal currents and Coriolis Effect under average meteorological and hydrological conditions of four different seasons considered,. Average seasonal variation of wind speed and direction as well as fresh water inflow does not seem to have significant influence on residual current. Only under the influence of average of maximum wind speeds of different seasons, residual currents in the Meghna Estuary show their dependency on wind stress. In general, at the surface layer northward and northwestward flow is created during the pre-monsoon and monsoon periods and southwest and southeastward flow is created during the post-monsoon and winter periods.