The dynamic development of wind power in recent years has generated the demand for production forecasting tools in wind farms. The data obtained from mathematical models is useful both for wind farm owners and distribution and transmission system operators. The predictions of production allow the wind farm operator to control the operation of the turbine in real time or plan future repairs and maintenance work in the long run. In turn, the results of the forecasting model allow the transmission system operator to plan the operation of the power system and to decide whether to reduce the load of conventional power plants or to start the reserve units.
The presented article is a review of the currently applied methods of wind power generation forecasting. Due to the nature of the input data, physical and statistical methods are distinguished. The physical approach is based on the use of data related to atmospheric conditions, terrain, and wind farm characteristics. It is usually based on numerical weather prediction models (NWP). In turn, the statistical approach uses historical data sets to determine the dependence of output variables on input parameters. However, the most favorable, from the point of view of the quality of the results, are models that use hybrid approaches. Determining the best model turns out to be a complicated task, because its usefulness depends on many factors. The applied model may be highly accurate under given conditions, but it may be completely unsuitable for another wind farm.
In this paper a system of a grid side and a generator side converters, both working with a common capacitor, is presented. The 6-phase asymmetric inset-type SMPMSM generator is used. A large pole pair number of this generator enables a gearless wind turbine operation. The fundamental and 3rd harmonic cooperation is used to increase the generator performance. This is accomplished by means of the 3rd harmonic current injection. For that reason the generator side converter must have a neutral connection.
Wind and solar radiation are intermittent with stochastic fluctuations, which can influence the stability of operation of the hybrid system in the grid integrated mode of operation. In this research work, a smoothing control method for mitigating output power variations for a grid integrated wind/PV hybrid system using a battery and electric double layer capacitor (EDLC) is investigated. The power fluctuations of the hybrid system are absorbed by a battery and EDLC during wide variations in power generated from the solar and wind system, subsequently, the power supplied to the grid is smoothened. This makes higher penetration and incorporation of renewable energy resources to the utility system possible. The control strategy of the inverter is realized to inject the power to the utility system with the unity power factor and a constant DC bus voltage. Both photovoltaic (PV) and wind systems are controlled for extracting maximum output power. In order to observe the performance of the hybrid system under practical situations in smoothing the output power fluctuations, one-day practical site wind velocity and irradiation data are considered. The dynamic modeling and effectiveness of this control method
Against the background of increasing installed capacity of wind power in the power generation system, high-precision ultra-short-term wind power prediction is significant for safe and reliable operation of the power generation system. We present a method for ultra-short-term wind power prediction based on a copula function, bivariate empirical mode decomposition (BEMD) algorithm and gated recurrent unit (GRU) neural network. First we use the copula function to analyze the nonlinear correlation between wind power and external factors to extract the key factors influencing wind power generation. Then the joint data composed of the key factors and wind power are decomposed into a series of stationary subsequence data by a BEMD algorithm which can decompose the bivariate data jointly. Finally, the prediction model based on a GRU network uses the decomposed data as the input to predict the power output in the next four hours. The experimental results show that the proposed method can effectively improve the accuracy of ultra-short-term wind power prediction.