Improved Indirect Power Control (IDPC) of Wind Energy Conversion Systems (WECS)

General Introduction

Fayssal Amrane, Chaiba Azeddine

Abstract

In this chapter, a brief general introduction focuses on the well-known topologies of wind energy conversion systems (WECS), on proposed controls and generators by the scientific researchers. One part will be devoted to the latest research that has addressed the performance problems of wind systems and their results (in simulation). There will be also some arguments that reflect the main proposed ideas in this eBook, the proposed selections and their applications in simulation. We present the selecting criteria in particular the type of: generator, controls and theirs application in simulation studies. Also, we discuss in a detailed section on the different contributions of eBook that define the improvement of the proposed algorithms in each chapter. Furthermore, the organization and structure of eBook will be as follow; chapter one is devoted on the state of the art of wind systems and their controls, in particular using the doubly fed induction machine (DFIM). The simulation part is provided in two chapters (3 and 4). The limitations and problems encountered during the realization of this eBook are well described in the following section. After solving problems, very satisfactory simulation results have been found which reflect the quality of the scientific contribution including more papers of conferences; Journal papers were published during this eBook.

Keywords: Doubly Fed Induction Generator (DFIG), Power Electronics (PEs), Wind Energy Conversion Systems (WECS’s), Wind Turbines (WTs).

1. INTRODUCTION

The growing connection of wind turbines has augmented at a quick pace over the last years. Installed wind power production, which is presently higher than 440 GW, is predictable to surpass 760 GW by 2020, creation this form of renewable energy an important element of the current and future energy supply systems [1-3]. The wind energy raises more important than any other renewable energy sources and is becoming really a significant factor in the recent energy supply system [4].

In the 1980, the PEs (Power Electronics) WTs (wind turbines) was a soft starter used to primarily interconnect the induction generator with the electrical grid, only thysistors were used and they did not require to carry the power continuously [5].

In the 1990s the PE technology was essentially used for the rotor resistance control of wound-rotor induction generator (WRIG), where further advanced diode bridges with a chopper were used to control the rotor resistance for generator [6], particularly at rated power process to reduce loading and mechanical stress. Since 2000, the bidirectional power flow have introduced with more progressive voltage source converters; the PEs started to handle the produced power from the WTs, first, by partial scale of power capacity for doubly fed induction generators (DFIGs), and then by the full scale of power capacity for asynchronous or synchronous generators (A/SGs) [5, 6].

Although the WTs can be considered into various structures in terms of the generator type, with/without the gearbox, or the rating of the power electronic converter, it is common to divide the WTs system into a partial-scale power converter equipped with a DFIG and a full-scale power converter together with either a synchronous generator (SG) or an induction generator (IG) [7, 8]. Presently, the DFIG system configuration the occupies close to 50% of the wind energy market, due to its small size, light weight, and cost-effectiveness of the generator, as well as the relatively small and economic power converter [9, 10].

The variable-speed WECSs can be worked in the maximum power point tracking (MPPT) mode to extract the maximum energy from wind. For this raison, good-calibrated mechanical sensors, such as encoders and anemometers/ resolvers, are essential in order to obtain the information of wind speed and generator rotor speed/position. But, the usage of mechanical sensors raises the cost, hardware difficulty of WECSs [11, 12]. These difficulties can be resolved by adopting position/speed sensorless control schemes [13].

The DFIG’ conventional control approaches are generally based on Field oriented control (FOC) algorithms [14, 15]. In the past few years it suffers from the handicap of the generator parameters changement, which comes to compromise the robustness of the control device. Hence, the regulator should accommodate the effects of uncertainties and maintain the system steady against a big variation of system parameters. The traditional PI-based controllers cannot totally fulfill stability and performance necessities [15]. Their optimal PI’s parameters can be defined by other approaches such as genetic algorithm (GA) or particle swarm optimization (PSO) [16-18]. Power converter and drive system have inherent features, such as non-linearities, inaccessibility of an accurate model or excessive complexity, that call for intelligent control approaches such as neural networks (NN), fuzzy logic (FL) [19, 20]. The dynamic performance of a WTS can be substantially enhanced by the application of smart methods for the PES control that are used in WPG systems. Hence, the aims of efficient wind power integration in the power system can be successfully accomplied.

Fuzzy logic (FL) has been applied for WPG control [21, 22]. The FL based controller is able to be implant, in the control strategy, the qualitative knowledge of an operator or field engineer about the system, but has been assessed for its limits, such as the lack of a formal design methodology, the difficulty in predicting stability and robustness of FL controlled systems [23]. The artificial neural networks (ANNs) based controllers have been used as these controllers can be formed straight by using the input-output information of the indefinite system, without requirement any previous model structure. However, to choice an ideal structure, parameter values and the number of training sets are still crucial concerns. To take benefit of their strengths and to mitigate their disadvantages, numerous hybrid methods have been planned [24]. A hybrid system can be achieved by, for example, combining a fuzzy inference system and adaptive neural networks (i.e., the adaptive neuro-fuzzy inference system (ANFIS)) [25]. ANFIS based controllers have been successfully implemented for numerous power systems and PE applications [26, 27].

On the other side, the system is greatly nonlinear. Thus, linearization operating point cannot be applied to design the controller. Nonlinear control methods can be used to efficiently solve this problem [28, 29]. In attempt to reach high performances in the steady and stransient states, a diverse nonlinear control configuration must be applied. In the recent years, several modified nonlinear state feedback such as Input-output feedback linearization control (I/OLC), Sliding Mode Control (SMC); Backstepping Mode Control (BMC) and Model Predictive control (MPC) have been applied to more develop the control performances [30].

2. THE MAIN CONTRIBUTIONS

In the review of the DFIG-based wind system in last decade, it can be seen that the majority relies on the regulation of: speed, flux, torque, current and powers. More than 75% of the published articles (mainly based on "IEEE and Science Direct" databases between 2005 and 2017) concerning the study and development of the DFIG-based wind system are basically focused on three (03) main controls: vector control (rotor flux and torque), predictive control and direct power control (stator active and reactive power). In this eBook, we are interested in power control (in terms of modeling) whose main objective is to improve the quality of energy transmitted into the network by integrating and developing new algorithms in order to overcome or mitigate drawbacks of conventional controls in transient and steady states during the wind speed variation and under robustness tests.

A detailed simulation study in power control using PI (Proportional-Integral) regulators (in order to control the stator powers Ps and Qs and the rotor currents Ird and Irq according to 04 loops respectively) is developed according to three modes, as