The importance of renewable energy sources has been growing at a high rate as a result of being environment friendly. In particular, wind power is one of the most successfully utilized of such sources to produce electrical energy. The available wind energy depends on the wind speed, which is a random variable. For the wind-farm operator, this poses difficulty in the system scheduling and energy dispatching, as the schedule of the wind-power availability is not known in advance. This paper proposes to use the two-layered artificial neural networks for predicting the actual wind speed from the previous values of the same variable.
[Show abstract][Hide abstract] ABSTRACT: A number of wind speed forecasting techniques are available in order to predict the uncertainty of the wind, which is key to estimate wind power generation availability for the grid. It is gaining more attention with the recent evolution of smart grid, which throws a challenge of integrating wind power into the grid. Several methods have been proposed to provide wind speed prediction. In the recent years there is a lot of research happening to predict wind speed with several mathematical methods and biologically inspired computing techniques to reduce the prediction error. In this paper a detailed survey of the wind speed forecasting techniques have been provided since past 15 years. An attempt is also made to identify different wind forecasting providers available in the market. This survey will be very useful to the new researchers working in this area of research. This survey will also be helpful to the wind farm owners to understand the current wind prediction model capabilities and will gives them an idea which model will be suitable for predicting the wind speed at their wind farms.
Power System Technology (POWERCON), 2010 International Conference on; 11/2010
[Show abstract][Hide abstract] ABSTRACT: In this paper, to meet the requirement of giving the estimation of wind power in advance, a new method based on the combination of time series and BP-ANN is proposed. With the historical data of wind speed and power for time series data, the prediction model is built up based on time-series and BPANN. The main study in the paper is as follows: the prediction model of wind speed is constructed by time-series, which gives the predicted data of wind speed. Considering the fact that wind power relates to not only wind speed, but also the nonlinearity of wind speed and power, the BP-ANN algorithm whose input and output are nonlinear, can give better mapping between wind speed and power. Therefore, wind power prediction model based on BP-ANN is effective. Consequently, the simulation of the Yantai wind farm shows that the combination prediction model based on time-series and BPANN effectively improves the accuracy of wind power prediction.
[Show abstract][Hide abstract] ABSTRACT: This paper presents modeling and impact analysis of wind farms in Jeju island power system, which is made of wind farms, a current source-type high-voltage, direct current (HVDC) system, and thermal power plants. In this paper, four kinds of major components are modeled: a total of 88 MW wind farms, a 300 MW HVDC system, thermal power plants, and the Jeju power system load. To analyze the impact of the wind power generation to the Jeju power system, simulation is carried out for two case studies by using the PSCAD/EMTDC program. One is for the steady-state operation under different wind speed, and the other is for the transient-state operation when all wind farms are disconnected suddenly from the Jeju power grid due to the wind speed higher than the rated value. These comparative studies have been effective in accessing the impact of wind power generation on the Jeju island power system stability.
IEEE Systems Journal 03/2012; 6(1):134-139. DOI:10.1109/JSYST.2011.2163017 · 1.98 Impact Factor
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.