Multiple Wind Speed Forecast Using Simple Recurrent Unit

Traditional wind speed forecast usually regards wind farm as a point to make forecast, but in a wind farm, wind speed of wind turbines in different geographical locations is not the same. For many wind turbines with wide geographical distribution in a wind farm, this paper gives a forecast method based on convolutional neural network (CNN) to forecast the wind speed at each wind turbine location. In this method, the wind speed and direction characteristics of all wind turbines at different geographical locations are input into the CNN network as variables, and local low-dimensional features of the original data are mapped to high-dimensional features through convolution operation of CNN, thereby realizing the wind speed forecast. The main advantage of this method is that by automatically studying the informative spatial correlation of wind speed, rather than artificial extracting , multi-task forecasts（MTF）are made and the wind speed forecast at different wind turbines locations is more informative and accurate.

ARMA based approaches for forecasting the tuple of wind speed and direction

This paper presents a fuzzy expert system that forecasts the wind speed at a wind energy conversion system (WECS) site and the electrical power that will be generated. The user can define the forecast horizon, which can range from some minutes up to several hours ahead. After training, the system can make reliable wind speed forecasts in less than a second. The system implements wind speed and direction measuring stations that are installed around and in the WECS site. The stations send measurements via wireless modems to a central computer running the fuzzy expert system, which exploits any spatial correlation existing among the measuring stations' wind speed time series. For the training of the fuzzy expert system two genetic algorithm implementations were used and compared.

A novel hybrid algorithm based on Empirical Fourier decomposition and deep learning for wind speed forecasting

Analysis and application of forecasting models in wind power integration: A review of multi-step-ahead wind speed forecasting models

Sensitivity analysis of the system of wind speed distributions

The risk assessment of frequency deviation in power systems caused by the wind power fluctuations is an essential issue on wind power integration, thus a more realistic wind power fluctuation simulation is necessary. To construct wind power fluctuations, not only time scale but also wind direction should be considered for the wind speed model with spatial correlation in a wind farm. In this paper, wind direction is added to the wind speed model on the basis of the second level wind speed with spatial correlation. Based on the Von Mises distribution and the Weibull distribution, the correlation of wind speed and direction is analyzed, and the sequences are obtained according to Monte Carlo Sampling, describing the correlation between wind speed and wind direction clearly. Then the wind power fluctuations with spatial correlation is available. Therefore, the frequency deviation resulted from the uncertainty of wind power fluctuations can be evaluated more accurately.

This work introduces a novel method to generate probabilistic hub‐height wind speed forecasts aimed at power output prediction. We employ state‐of‐the‐art convolutional variational autoencoders (CVAEs) trained with historical wind speed observations, multivariable outputs (wind speed, direction, temperature, pressure, and humidity) from a numerical weather prediction (NWP) model and spatio‐temporal encodings. After training, we exploit the CVAE data generating capabilities to produce probabilistic forecasts from the same deterministic dynamical NWP model. The resulting probabilistic forecast provides an insight into the uncertainty of the original deterministic input that compensate for errors due to numerical discretization, inaccuracies in initial/boundary conditions and parameterizations and, most importantly wind speed fluctuations due to complex terrain features. To show the performance of the proposed model, we validate the approach with forecasted and observed data for fifteen sites in a wind farm in Awaji Island, Japan, in a challenging zone with complex topography and highly fluctuating wind patterns. We show that the proposed method provides improved wind speed forecasts from both deterministic (reduced root mean square error) and probabilistic (reduced continuous ranked probability score) standpoints. We also use ensemble forecast validation methods to assess the statistical properties of the CVAE and conclude that in zones with rapidly changing wind speed dynamics, the CVAE ensemble performs in a superior way when compared to the analog ensemble method. Finally, we also provide an insight on how the results can be used to obtain reliable wind power estimations.

Researching on interaction between wind farm and power system, it usually uses simple model to describe the wind farm. The wind turbines in a wind farm always suppose as a coherent generator group, every wind turbines have identical wind speed. As wind turbine locates in the different site in the wind farm, the wind speed at each wind turbines will not change simultaneously. Wake effect is another factor that should take into consideration. So, it needs to get wind speed equivalent model. The equivalent model of wind speed in the wind farm is based on output of power. Gust wind and ramp wind are discussed. Wind farm will change gust wind into step varying wind and make ramp wind into different type wind. The arrangement of wind turbines will also affect equivalent model of wind speed. The result show that the length of wind farm will smooth the power output. It explains why the power output fluctuation of wind farm is smaller than single wind turbine. The onshore wind farm located in Dongtai Jiangsu province is proposed as an example. The equivalent wind speed model and equivalent impedance connected to grid are calculated. The results illustrate the impact of wind speed model on wind farm output.

As wind energy adoption is growing, ensuring the efficient operation and maintenance of wind turbines becomes essential for maximizing energy production and minimizing costs and downtime. Many AI applications in wind energy, such as in condition monitoring and power forecasting, may benefit from using operational data not only from individual wind turbines but from multiple turbines and multiple wind farms. Collaborative distributed AI which preserves data privacy holds a strong potential for these applications. Federated learning has emerged as a privacy-preserving distributed machine learning approach in this context. We explore federated learning in wind turbine condition monitoring, specifically for fault detection using normal behaviour models. We investigate various federated learning strategies, including collaboration across different wind farms and turbine models, as well as collaboration restricted to the same wind farm and turbine model. Our case study results indicate that federated learning across multiple wind turbines consistently outperforms models trained on a single turbine, especially when training data is scarce. Moreover, the amount of historical data necessary to train an effective model can be significantly reduced by employing a collaborative federated learning strategy. Finally, our findings show that extending the collaboration to multiple wind farms may result in inferior performance compared to restricting learning within a farm, specifically when faced with statistical heterogeneity and imbalanced datasets.

Wind speed and output power forecasting is very important to the utilization of wind energy. In order to improve the forecast precision, a forecasting method based on wavelet transform (WT) and least square support vector machine (LSSVM) is proposed in this paper. The wind speed time series was decomposed into different frequency components. The different LSSVM models to forecast the high frequency and low frequency are built up. These forecasting results of the different frequency bands are combined to obtain the final forecasting results. According to the power characteristics, unit efficiency and the operate condition of the generators, the short-term output power forecasting on the wind farm can be obtained.

Evaluation of hybrid forecasting approaches for wind speed and power generation time series

Simplified non-homogeneous Markov Chain (MC) models for generation of synthetic time series of wind speed (WS) and wind direction (WD) are proposed. The proposed models consider daily, monthly, and seasonal WS and WD characteristics. The accuracy and computational burden of the proposed models are compared with those of a complete model (CM), proposed by the same authors in a previous paper, trough numerical experiments based on field data.

During recent years, wind energy has moved from an emerging technology to a nearly competitive technology. This fact, coupled with an increasing global focus on environmental concern and a political desire of a certain level of diversification in the energy supply, ensures wind energy an important role in the future electricity market. For this challenge to be met in a cost-efficient way, a substantial part of new wind turbine installations is foreseen to be erected in big onshore or offshore wind farms. This fact makes the production, loading and reliability of turbines operating under such conditions of particular interest.

Hybrid wind speed forecasting using ICEEMDAN and transformer model with novel loss function