Information Technology Reference
In-Depth Information
12 Conclusions
Most important success factor of neural network structure is the accessibility of
valuable learning algorithms. Planned approaches optimally control Wind Energy
Conversion Systems with changing Pitch angle and estimates parameters.
In this work, several new methods estimate and predict pitch angle value for a
wind turbine. Introduced proposed controllers are trained using multi-layer per-
ceptron and radial basis function neural networks.
As plotted from simulation results, controllers change pitch angle of turbine
blades to achieve most suitable output power. Since the FRENGA was tested with
the Multi-Layer Perceptron NN, it is independent from the NN construction phase.
These methods used NN to produce a training example for the rule extraction
method. FRENGA can be easily employed for extracting a set of fuzzy if-then rules
directly from data. In Hybrid controller MLP and FRENGA proposed a signal
control to manage output power and torque. Finally online form both methods were
tested on a turbine and one of them that product better respond is selected.
Experimental results admitted that FRENGA, GFS and TFGS work ef
ciently
for both continuous and enumerates attributes. As plotted, wind turbine could
obtain the most suitable performance coef
cient (Cp) and tip speed ratio (TSR) with
change of turbine blades pitch angle. Consequently, simulation results realize which
output power has been regulated successfully.
As results indicated, the new proposed genetic fuzzy rule extraction system with
tuning membership function (TFGS) outperformed one of the best and earliest
approaches in controlling the production through wind
fluctuation.
In future work we will try adjust and improvement our result with another
intelligent methods.
fl
References
Abdelkarim, M., Achraf, A., & Lotfi, K. (2011). Electric power generation based on variable speed
wind turbine under load disturbance. Energy, 36, 5016
5026.
Agrawal, R., Imielinski, T., & Swami, A. (1993). Mining association rules between sets of items in
large databases (pp. 207
-
216). Washington D.C: SIGMOD.
Andrew, K., & Haiyang, Z. (2010). Optimization of wind turbine energy and power factor with an
evolutionary computation algorithm. Energy, 35, 1324
-
1332.
Azar, A. T. (2010b). Adaptive neuro-fuzzy system. In A. T Azar (Ed.) Fuzzy systems. Vienna:
IN-TECH. ISBN 978-953-7619-92-3.
Azar, A. T. (2012). Overview of Type-2 fuzzy logic Systems. International Journal of Fuzzy
System Applications (IJFSA), 2(4), 1 - 28.
Baesens, B., Setiono, R., Mues, C., & Vanthienen, J. (2003). Using neural network rule extraction
and decision tables for credit-risk evaluation. Management Science, 49(3), 313 - 329.
Bianchi, F. D., Battista, H. D., & Mantz, R. J. (2006). Wind turbine control systems: Principles,
modeling and gain scheduling. New York: Springer.
-
Search WWH ::




Custom Search