ABSTRACT:
The
power from wind varies depending on the environmental factors. Many methods
have been proposed to locate and track the maximum power point (MPPT) of the
wind, such as the fuzzy logic (FL), artificial neural network (ANN) and
neuro-fuzzy. In this paper, a variable-speed wind-generator maximum power- point-tracking
(MPPT) based on adaptative neuro-fuzzy inference system (ANFIS) is presented. It
is designed as a combination of the Sugeno fuzzy model and neural network. The
ANFIS model is used to predict the optimal speed rotation using the variation
of the wind speed as the input. The wind energy conversion system (WECS)
employing a permanent magnet synchronous generator connected to a DC bus using
a power converter is presented. A wind speed step model was used in the design
phase. The performance
of
the WECS with the proposed ANFIS controller is tested for fast wind speed
variation. Simulation results showed the possibility of achieving maximum power
tracking for the wind and output voltage regulation for the DC bus
simultaneously with the ANFIS controller. The results also proved the good
response and robustness of the control system proposed.
KEYWORDS:
1. Wind
energy
2. Power
generation
3. Variable
speed wind generator
4. MPPT
5.
ANFIS
SOFTWARE:
MATLAB/SIMULINK
BLOCK DIAGRAM:
Fig.
1. Wind generation system configuration.
Fig.
2. Training error.
Fig.
3. Setup function of wind speed.
Fig.
4. Rotor speed response.
Fig..5
. Output power response.
Fig.
6. Efficiency.
Fig.
7. DC bus voltage response.
Fig.
8. Wind speed.
Fig.
9. Rotor speed response.
Fig.
10. Output power response.
Fig.
11. Efficiency.
Fig.
12. DC bus voltage response.
Fig.
13. Comparative output power response with ANFIS and FL.v
CONCLUSION:
In
this paper, the WECS was modeled using d-q rotor reference frame. A variable
speed wind generator maximum power point tracking based on an adaptative
neuro-fuzzy-inference-system (ANFIS) was presented. The feasibility of this
controller is demonstrated and the simulation results for both cases proved the
robustness, fast response, and exact maximum power tracking capabilities of the
ANFIS control strategy. The results show also that ANFIS model has a better
response compared to fuzzy logic model.
REFERENCES:
Ansel,
A., & Robyns, B. (2006). Modelling and simulation of an autonomous variable
speed micro hydropower station. Mathematics and Computers in Simulation, 71(4),
320–332.
Avci,
E., & Avci, D. (2007). The performance comparison of discrete wavelet
neural network and discrete wavelet adaptive network based fuzzy inference
system for digital modulation recognition. Expert Systems with Applications,
33, 3.
Avci,
E., Hanbay, D., & Varol, A. (2007). An expert discrete wavelet adaptive
network based fuzzy inference system for digital modulation recognition. Expert
Systems with Applications, 33(3), 582–589.
Aznarte,
M. J. L., Sánchez, J. M. B., Lugilde, D. N., Fernández, C. D. L., Guardia, C.
D. Dl., & Sánchez, F. A. (2007). Forecasting airborne pollen concentration
time series with neural and neuro-fuzzy models. Expert Systems with
Applications, 32(4), 1218–1225.
Avci,
E., & Akpolat, Z. H. (2006). Speech recognition using a wavelet packet
adaptive network based fuzzy inference system. Expert Systems with
Applications, 31(3), 495–503.
