Estimation of wind turbine output power using soft computing models

ABSTRACT

Among renewable energy generation technologies, wind energy has become one of the most outstanding issues, especially in the last decade. Wind speed is the most critical parameter influencing the power obtained from a wind turbine. The unstable structure of the wind causes an impossibility to receive a direct theoretical relation between wind power and speed. Accordingly, obtaining a simulation of the generated turbine power concerning the approaching wind speed has become vitally essential, nowadays. In the current study, generated wind turbine power (P) has been predicted using three forecasting methods. The computer was trained using wind speed (V) and the turbine rotor rotational speed (ṅ) as the inputs of the forecasting methods. The methods used for this purpose were to include adaptive neuro-fuzzy inference system (ANFIS), Elmanneural network (ENN), and feed-forward neural network (FNN) approaches. In the training of the programs, among the cumulative of 43,800 wind speed, turbine rotor rotational speed, and wind power data, 80% and 20% of the total data were used for training and testing stages of the algorithms, respectively. The statistical results of the computations demonstrated that among three methods, ANFIS gave better outcomes when compared to ENN and FNN, in both the training and testing stages of the algorithms. The proposed models of the current study have revealed low and acceptable mean absolute error (MAE) and root mean square error (RMSE) statistical error results of ANFIS tool; corresponding to 52.448 kW and 87.204 kW error, respectively, were obtained at the training stage, whereas 48.675 kW and 78.453 kW error, respectively, were obtained at the testing stage. For the estimation of wind power, while the coefficient of determination (R²) was detected as R² = 0.9948 and 0.9961 in training and testing stages, respectively, with ANFIS model, it was found out as R² = 0.9942 and 0.9957 with ENN model and R² = 0.9943 and 0.9956 with FNN model. Namely, it has been reported that ANFIS tool can be successfully applied in wind output power estimations, as long as wind speed and turbine rotor rotational speed values are provided without the need for numerous experimental measurements, which induces additional time, labor, and measurement expenses.

KEYWORDS:

• Wind turbine power output
• Elman neural network
• feed-forward neural network
• adaptive neuro-fuzzy inference system
• renewable energy

Disclosure statement

No potential conflict of interest was reported by the author(s).

Authorship contribution statement

Sergen Tumse: Investigation, Methodology, Writing, Review, Software and Editing Akin Ilhan: Investigation, Methodology, Writing, Review and Editing Mehmet Bilgili: Investigation, Methodology, Software and Supervision Besir Sahin: Investigation, Methodology, Supervision and Editing

Nomenclature

ANN Artificial neural network

ANFIS Adaptive neuro-fuzzy inference system

ENN Elman neural network

FNN Feed-forward neural network

ARMA Autoregressive moving average

MLP Multilayer perception

LSTM Long short-term memory

LSTMN Long short-term memory networks

EMD Empirical mode decomposition

ARIMA Autoregressive integrated moving average

RNN Recurrent neural network

DNN Deep neural network

VMD Variational mode decomposition

ESN Echo state network

EEMD Ensemble empirical mode decomposition

GPR Gaussian process regression

ELM Extreme learning machine

SVM Support vector machine

NLR Nonlinear logistic regression

WRF Weather research and forecast

SVR Support vector regression

RMSE Root mean square error

MAE Mean absolute error

R² Coefficient of determination

P Wind power (kW)

V Wind speed (m/s)

ṅ Turbine rotational speed (rpm)

x Variable

x_n Normalized value of x variable

x_min Minimum value of x variable

x_max Maximum value of x variable

F Forecasted value

a Bias to layer 2

h Neurons number in hidden layer

v_j Layer weight

k Number of input parameters

w_ij Weight of input layer

b_j Bias to layer 1

x_i Input variable

σ Width of curve

c Center location of curve

A Actual value

n Total observation number