MODELLING AND CONTROL OF DOUBLY FED INDUCTION GENERATOR USED FOR WIND POWER APPLICATIONS

Abstract

Energy sources development is the ongoing effort to provide sufficient energy to fulfill the ever-growing demand of energy. However, the availability of fossil fuel and nuclear resources is limited. The extensive use of fossil fuel and nuclear resources is causing serious environmental pollution and safety problems, which are now becoming vital issues in our society. Presently, world is turning more and more towards environmentally clean and safe renewable energy sources (wind, photovoltaic, fuel cells etc.). The wind power generation is emerging as one of the most significant and promising source of renewable energy in the power industry and its capacity is increasing globally. Unlike any other form of energy generation, the production of wind energy does not contribute to the greenhouse effect. Today, India has one of the largest programs in the renewable energy resources and out of these programs, wind power program is extremely successful. In this area, India has now occupied the fourth position after Germany, USA and Spain. Wind turbine generators are broadly classified into three categories; fixed speed asynchronous wind turbines employing self-excited generators or wound rotor induction generators with external resistance connected to rotor terminals, variable speed doubly fed induction generators (DFIG) with back-toback voltage source power electronic converters, and synchronous generators with full size power electronic converter connected to its stator. Fixed-speed wind turbines operate near a constant rotor speed, and are directly connected to the power system. The rotor speed is dictated by the fundamental frequency of the power system. The major benefit incurred from this concept is that it requires simple construction of the whole wind turbine with very low investment and maintenance costs. The major disadvantage is the fixed rotational speed. In variable speed wind energy conversion system (WECS), the generator is connected to the grid through a power electronic converter system, or the generator excitation windings are fed by an external frequency from a power electronic interface. The rotational speed of the generator and thus of the rotor is decoupled from the fundamental frequency of the power system. Hence, the rotor can operate with variable speed adjusted to the actual wind speed situation. Presently, the variable speed WECS equipped with DFIGs are widely used in large wind farms due to its wide operating range from sub-synchronous to super-synchronous speeds with the ability to supply power at constant voltage and frequency. The DFIG concept also provides a possibility to control the overall system power factor. Other advantages include reduction of the mechanical stresses on the wind turbines, reduced noise, reduced requirements for the pitch angle controllers, low rated power electronic converters, and thus an improved overall efficiency. Modern wind power generation systems are almost exclusively using variable speed electronically controlled generators. The growing penetration levels of WECS equipped with DFIGs in electric networks pose significant challenges on a wide range of issues. This thesis addresses some of those issues, namely modeling, control, inertial response and stability studies. These are summarized as the following: • The mathematical modeling of DFIG connected WECS is done for designing the controllers and for a more accurate and realistic assessment of small signal stability performance of the system. Reduced order models are also derived for simplifying the controller design. • The controllers implemented in DFIG schemes are designed using Internal Model Control (IMC) structure, which is based on the linearized equations. The tuning module of the current controller parameters, for desired performance in nonlinear system, is derived using particle swarm optimization (PSO). • The system has been analyzed for frequency changes due to abrupt variations in generation or load. An additional frequency response is provided to reintroduce inertia response by adding a supplementary control loop. The optimal tuning of the controller parameter is carried out using PSO. • The other significant objective of this thesis is to investigate the impact of a WECS equipped with DFIG on the small signal dynamics of a single machine infinite bus system (SMIB) and a multi machine power system. The influence of power system stabilizer (PSS) on the small signal behavior of DFIG connected multi machine power system is also an important aspect of this work. An optimization process tunes the PSS parameters. A general model can not be introduced that would represent the dynamic behavior of all schemes of variable speed WECS with sufficient accuracy. Therefore, each configuration requires an elaboration of the individual model, depending on the type of generator, converters and control systems used, as well as on the modeling requirements, i.e. on the time scale and nature of the phenomena to be reproduced. Thus, one of the objectives of this thesis is to derive an adequate model of the WECS equipped with DFIG for investigating its dynamic performance on the power system. A model order reduction technique is also proposed to reduce the complexity of the model that limits the computational speed in dynamic simulation of power systems as well as difficulties arising in controller design. One of the reasons for the increased use of DFIG in WECS is its distinct advantage of decoupled control of active and reactive powers using back-toback voltage source inverters in the rotor-grid interaction circuit. This would require the proper design of the controllers for smooth and stable operation of the system. The ability to generate electricity with different power factors would reduce the costs of introducing additional capacitors for reactive power regulation. This would be especially advantageous for both producers and distributors in charge of transmission systems. In this thesis, different aspects of designing control systems for DFIG are treated. Owing to the fact that DFIG controls have a significant influence on the system dynamics, vector control is applied for both grid-side and rotor-side converters to increase the degree of controllability. The IMC scheme based controllers are precisely designed and tested for the operation of a WECS equipped with DFIG. PSO, which is a robust stochastic evolutionary computational algorithm based on the movement and intelligence of swarms, is used to tune the parameters of the controllers. The increasing share of variable speed WECS, such as DFIGs, in power generation would result in the reduction of connected conventional generator in power plants. The drawback of using DFIG for variable speed operation is its negligible inertial response to frequency control. This is because; the DFIG control systems decouple the mechanical and electrical systems. However, the standard fixed speed wind turbines with directly connected generators contribute to power system inertia using the inertia of the generator and the inertia of wind turbine rotor. This contribution is due to the linking of rotor speed with the system frequency. Lower system inertia will result in larger and faster frequency deviations after the occurrence of abrupt variations in generation and load. This thesis presents the design of supplementary control loop with different control structures to provide frequency support in DFIG based WECS during the system frequency changes. The parameter of the best controller is tuned using PSO. As long as the wind power penetration level is low, the synchronous generators in power plants will mainly determine the overall dynamic behavior of the power system. The increasing share of wind turbines in electric power generation may begin to influence the overall power system behavior and it would no longer be possible to run a power system by controlling only largescale power plants. It is therefore important to study the behavior of wind turbines in an electrical power system. An attempt has been made to investigate the impact of a WECS equipped with DFIG on the small signal dynamics of a SMIB and a multi machine power system. The study on a multi machine power system has been carried on Western System Coordinating Council (WSCC) 3-machine, 9-bus system with constant loads. The influence of PSS on the small signal behavior of DFIG connected multi machine power system has been investigated. The PSS parameters have been tuned by formulating an optimization objective and solved using the PSO technique. In summary, the present work focuses on modeling and control of DFIG used for wind power generation. Simple and logical decoupled vector control systems for DFIG are designed using reduced order modeling and IMC based control philosophy. The inertial response of the DFIG connected power system is analyzed and controllers are proposed for better frequency support during frequency excursions. The thesis also examines the small signal stability performance of DFIG connected SMIB and multi-machine power systems.