Using neural networks to estimate wind turbine power generation

Li, Shuhui; Wunsch, Donald C.; O’Hair, E.; Glesselmann, M.

doi:10.1109/pesw.2001.917195

Cited by 41 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multiple neural networks are constructed with different parameters of BP neural network [6]. The constructor function is: net=newff (P,T,S,TF,BTF,BLF,PF,IPF,OPF,DDF), where P and T are matrix, S is hidden layer nodes, TF is transfer function, BTF is training function and BLF is learning function.…”

Section: Bp Neural Network Parameters Settingmentioning

confidence: 99%

Forecasting Method of Crude Oil Output Based on Adaboost_BP

2017

Proceedings of 2017 the 7th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

In order to improve the prediction accuracy of BP neural network, Adaboost algorithm and BP neural network are combined to propose the Adaboost_BP prediction model. The prediction model firstly preprocesses the training data set, which selects different parameters of BP neural network to construct several BP neural networks. Then, the test data set and its distribution weight are initialized. According to the Adaboost algorithm, the prediction error of each BP weak predictor is calculated and the distribution weight of the test data set is adjusted according to the error, and the weight of the predictor is obtained. Finally, all weak predictors are combined to generate a strong predictor. Based on the original data of oilfield block from 1994 to 2016, the results show that the prediction of Adaboost_BP prediction model is more accurate than the single BP neural network model, and achieves good predictive result, which validates the effectiveness of the proposed model.

show abstract

Section: Bp Neural Network Parameters Settingmentioning

confidence: 99%

Forecasting Method of Crude Oil Output Based on Adaboost_BP

2017

Proceedings of 2017 the 7th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

show abstract

“…Wind farm volatility-output power is mainly caused by wind speed change, so accurate wind speed prediction can effectively reduce the negative impact of the wind farm to power system [73]. Wind speed forecasting is commonly analyzed using artificial neural network [74], [75], time series method [73], [76], [77] and Kalman filtering method [78][79][80][81][82] etc. Considering the accuracy requirements, the combination of several methods for wind speed forecasting has become a trend.…”

Section: Application Of Wind Speed Forecastingmentioning

confidence: 99%

A Research on Significance of Kalman Filter-Approach as Applied in Electrical Power System

Edmund¹

2016

IJRET

View full text Add to dashboard Cite

Recently, AC distribution systems have experienced high harmonic pollution due to the fact that electrical power system parameters are often mixed with noise. In an ideal situation, AC power system is supposed to have a constant frequency at specific voltage but owing to presence of connected nonlinear loads and injection into the grid from non-sinusoidal output active sources etc., have immensely contributed to the total distortion of the both current and voltage waveforms. This has increased the system loses and consequently affected other connected equipment in the system. Therefore there is a need to mitigate these effects if they cannot be eliminated intoto, hence the proposition of Kalman filter. It has been very useful in the aspect of electrical power discipline particularly in harmonic estimation. It has also find it way in the application of power system dynamics, optimal operation and control of motor, relay operation and protection, and also for accurate prediction of short and medium term electrical load forecasting. This paper is to highlight on the significant of Kalman filter methodological approach as adopted in electrical power system.

show abstract

“…However, by the passage of time, more advanced methods have been proposed. To this end, Artificial Neural Network (ANN) in [12], [13], ANN with adaptive Bayesian learning and Gaussian process approximation in [14], combination of ANN with wavelet transform in [15], fuzzy logic methods in [10], [16], Kalman filter in [17], support vector machine in [18] and some hybrid methods in [6] have been proposed for wind power prediction.…”

Section: International Journal Of Computing and Digital Systemsmentioning

confidence: 99%