Friday , March 29 2024

NNPID-based Stator Voltage Oriented Vector Control
for DFIG based Wind Turbine Systems

Shanzhi LI1, Haoping WANG1, Yang TIAN1,
Nicolai CHRISTOV2, Abdel AITOUCHE2

1 Sino-French International Joint Laboratory of Automatic Control and Signal Processing (LaFCAS), Nanjing University of Science & Technology (NUST),
Nanjing 210094, China
2 LAGIS- CNRS UMR 8219, LaFCAS, University Lille Nord de France,
Lille France, 59600

Abstract: Doubly fed induction generators are widely adopted for the wind turbine systems since it is cheap and reliable. Based on the traditional stator voltage oriented vector control method, the performance of the proposed vector control is largely influenced by the variations of the DFIG parameters. And the classical PID algorithm cannot achieve the maximum power point tracking (MPPT) in time (owing to the transient wind). Hence, in this paper, to eliminate parameters variations on the power output and capture the MPPT rapidly, we propose a stator voltage oriented vector control which is based on a Neural Network PID (NNPID) technology. The weights which are being similar to the PID coefficients are adapted by Hebb rule to decrease the power error online according to the error gradient descent method, while the classical PID coefficients will be a constant. The effectiveness of the proposed method is demonstrated by corresponding simulation results: even in the case of wind mutation change, the proposed NNPID can track the variation of the wind energy, and robust to the DFIG parameters variations.

Keywords: wind turbine system; maximum power point tracking; DFIG; NNPID.

>>Full text
CITE THIS PAPER AS:
Shanzhi LI, Haoping WANG, Yang TIAN, Nicolai CHRISTOV, Abdel AITOUCHE, NNPID-based Stator Voltage Oriented Vector Control for DFIG based Wind Turbine Systems, Studies in Informatics and Control, ISSN 1220-1766, vol. 23 (1), pp. 5-12, 2014. https://doi.org/10.24846/v23i1y201401

 1. Introduction

As a renewable and clean energy, wind energy which growths rapidly worldwide and attracts widespread attention by governments [1]. Recently, high efficiency and low cost power production control methods and technologies are always an eternal working goal for researchers and engineers.

Doubly fed induction generators (DFIG) which are widely adopted for the wind turbine systems occupy most parts of the global market because of the lower-cost and smaller inverter capacity. However, the DFIG’s characteristics of non-linearties, strong coupling, and varying parameters as well as the wind randomness are intractable problems. It also reduces the efficiency of the wind energy. Moreover, the model of wind is complicated, and affected by many factors. Besides, dynamic and static characteristics of nature wind need to be reflected by the wind model. And wind turbine system also requires an appropriate wind model for different performances. The emitted power usually cannot catch up with capture wind energy because of both the transient wind and the lag inertia of the blades. Considering wind uncertainties and wind turbine systems nonlinearities, it is difficult to selecting appropriate classical PID coefficients to obtain good performances.

In addition, these coefficients usually maintain constant once the PID controller is installed. Recently, artificial intelligent controls have been developed rapidly and applied gradually to wind turbine systems, such as neural network control and fuzzy control are discussed in [2-4]. However, to achieve good control performance needs complex computations in the neural network based control algorithms. A rule table for fuzzy control is also difficult to design [6]. For fuzzy controllers which select this rule table by self-adjusting needs to spend a longer time in calculations

In this paper, to improve the efficiency of capturing wind energy and simplify the referred control methods computations, a particular PID controller which is based on a single-neuron network is developed and called Neural Network PID (NNPID). Known that three layers of neural networks (NNs) can be used to approximate an arbitrary input-output mapping [7]. The weights of the proposed NNPID can be adjusted adaptively online with tracking errors. This used adaptive mechanism is useful to overcome both the wind turbine systems external disturbances and parameters uncertainties.

The rest paper is organized as follows: a wind model is presented in 2nd section. And in 3rd section, a wind turbine model which includes an aerodynamic model of the wind turbine rotor, DFIG model and drive trains model is described.

After that, for the considered system, a fundamental stator voltage oriented vector control method and a NNPID controller is developed in 4th section. Then to validate the proposed method, some numerical simulation results are illustrated in 5th section. Finally, some conclusion remarks are presented in 6th section.

REFERENCES

  1. IZADIAN, A., N. GIRRENS, P. KHAYYER, Renewable Energy Policies: A Brief Review of the Latest U.S. and E.U. Policies, Industrial Electronics Magazine, IEEE, 7(3), 2013, pp. 21-34.
  2. JAFARNEJADSANI, H., J. PIEPER, J. EHLERS, Adaptive Control of a Variable-Speed Variable-Pitch Wind Turbine Using Radial-Basis Function Neural Network, Control Systems Technology, IEEE Transactions on, 21, no. 6, 2013, pp. 2264-2272.
  3. QINGSONG, L., Y. YUFEI, W. JUNXIAO, Simulation and Application of Neural Network PID Auto-Tuning Controller in Servo-System, Database Technology and Applications (DBTA), 2nd International Workshop, 2010, pp. 1-4.
  4. PHAN QUOC, D., B. ANH NGUYEN, L. HONGHEE, New Artificial Neural Network based Direct Virtual Torque Control and Direct Power Control for DFIG in Wind Energy Systems, Power Electronics and Drive Systems (PEDS), 2011 IEEE Ninth International Conference on, 2011, pp. 219-227.
  5. ZHIMING, Y., W. XIAORONG, Fuzzy PID Control of Induction Generators, Cognitive Informatics & Cognitive Computing (ICCI*CC), 10th IEEE Intl. Conference, 2011, pp. 130-134.
  6. ABDELHADY, F., S. ABUELENIN, Design and Simulation of a Fuzzy-supervised PID Controller for a Magnetic Levitation System, Studies in Informatics and Control, 17, no. 3, 2008, pp. 315-328.
  7. SIFAOUI, A., A. ABDELKRIM, S. ALOUANE, On New RBF Neural Network Construction Algorithm for Classification, Studies in Informatics and Control, 18, no. 2, 2009, pp. 103-110.
  8. KAREEM, A., Numerical Simulation of Wind Effects: a Probabilistic Perspective, Journal of Wind Engineering and Industrial Aerodynamics, 96, no. 10, 2008, pp. 1472-1497.
  9. Nichita, C., D. Luca, B. Dakyo, Large band Simulation of the Wind Speed for Real Time Wind Turbine Simulators, Energy Conversion, IEEE Trans., 17, no. 4, 2002, pp. 523-529.
  10. MihetPopa, L., F. Blaabjerg, I. Boldea, Wind Turbine Generator Modeling and Simulation Where Rotational Speed is the Controlled Variable, Industry Applications, IEEE Trans., 40, no. 1, 2004, pp. 3-10.
  11. Tapia, A., G. Tapia, J. X. Ostolaza, Modeling and Control of a Wind Turbine driven Doubly fed Induction Generator, Energy Conversion, IEEE Trans., 18, no. 2, 2003, pp. 194-204.
  12. JIABING, H., H. NIAN, H. XU, Dynamic Modeling and Improved Control of DFIG under Distorted Grid Voltage Conditions, Energy Conversion, IEEE Trans., 26, no. 1, 2011, pp. 163-175.
  13. YAZHOU, L., A. MULLANE, G. LIGHTBODY., Modeling of the Wind Turbine with a Doubly Fed Induction Generator for Grid Integration Studies, Energy Conversion, IEEE Transactions on, 21, no. 1, 2006, pp. 257-264.
  14. IDJDARENE, K., D. REKIOUA, T. REKIOUA, Vector Control of Autonomous Induction Generator taking Saturation Effect into Account, Energy Conversion and Management, 49, no. 10, 2008, pp. 2609-2617.
  15. PAN, G., M. ZHANG, Q. LI, Power Control of DFIG Based on the Nonlinear PID, Wei-Te Dianji (Small & Special Electrical Machines), vol. 41, no. 1, 2013, pp. 50-53.
  16. XINGJIA, Y., W. HEXU, D. YING, Research on Rotor Excitation Neural Network PID Control of Variable Speed Constant Frequency Wind Turbine, Intl. Conference on Electrical Machines and Systems,. ICEMS, 2007, pp. 560-565.
  17. GE, S., C. C. HANG, T. H. LEE, Stable Adaptive Neural Network Control: Springer Publishing Company, Incorporated, 2010.
  18. MARABA, V. A., A. E. KUZUCUOGLU, Speed Control of an Asynchronous Motor using PID Neural Network, Studies in Informatics and Control, 20, no. 3, 2011, pp. 199-208.
  19. PINTEA, A., H. P. WANG, N. CHRISTOV, D. POPESCU, P. BORNE, Modeling and Power Regulation of Horizontal Variable Speed Wind Turbines, Studies in Informatics and Control, vol. 20, no. 3, 2011, pp. 305-312.
  20. WANG, H. P., A. PINTEA, N. CHRISTOV, P. BORNE, Modelling and Recursive Power Control of Horizontal Variable Speed, Control Engineering And Applied Informatics, vol.14, no. 4, 2012, pp. 33-41.