Volume 2 wind energy 2 11 – wind turbine control systems and power electronics Volume 2 wind energy 2 11 – wind turbine control systems and power electronics Volume 2 wind energy 2 11 – wind turbine control systems and power electronics Volume 2 wind energy 2 11 – wind turbine control systems and power electronics Volume 2 wind energy 2 11 – wind turbine control systems and power electronics Volume 2 wind energy 2 11 – wind turbine control systems and power electronics Volume 2 wind energy 2 11 – wind turbine control systems and power electronics
2.11 Wind Turbine Control Systems and Power Electronics A Pouliezos, Technical University of Crete, Hania, Greece © 2012 Elsevier Ltd All rights reserved 2.11.1 Control Objectives 2.11.2 Wind Turbine Modeling 2.11.2.1 Mechanical Part 2.11.2.2 Electrical Part – Generators and Converters 2.11.2.2.1 Permanent magnet synchronous generators 2.11.2.2.2 Asynchronous (induction) generators 2.11.2.2.3 Doubly fed induction generator 2.11.2.2.4 Squirrel cage generator 2.11.2.2.5 Power converter 2.11.2.3 Full Model 2.11.3 Control 2.11.3.1 Overall Control Strategy 2.11.3.2 Pitch Control 2.11.3.2.1 Collective pitch control 2.11.3.2.2 Individual pitch control 2.11.3.3 Generator Control 2.11.3.3.1 Wound rotor doubly fed induction generator control 2.11.3.3.2 Asynchronous squirrel cage generator control 2.11.3.3.3 Permanent magnet synchronous generator control 2.11.3.4 Coupled Pitch–Generator Control 2.11.3.5 Grid Control 2.11.3.6 Yaw Control 2.11.3.7 Grid Issues 2.11.4 Fault Accommodation 2.11.5 Hardware 2.11.5.1 Sensors 2.11.5.2 Actuators References Further Reading Relevant Websites Glossary Direct drive (DD) technology A design that eliminates the need for gearboxes With fewer moving parts, DD technology can reduce maintenance costs and provide higher wind turbine availability Double fed induction generator (DF) Sometimes referred to as DFIG – has been widely used technology in wind turbines for ten years in thousands of commissioned wind turbines It is based on an induction generator with a multiphase wound rotor and a multiphase slip ring assembly with brushes for access to the rotor Recently, designs without the brushes have been introduced Fault ride-through (FRT) A requirement of network operators, such that the wind turbine remains connected during severe disturbances on the electricity system, and Nomenclature A(γ(t)) turbine rotor swept area (time-varying due to yaw error) (m2) cP power coefficient Comprehensive Renewable Energy, Volume 330 333 334 336 337 339 339 340 340 341 343 343 345 345 348 352 353 354 356 358 358 360 362 364 366 366 368 369 370 370 returns to normal operation very quickly after the disturbance ends Grid-connected A wind turbine is grid-connected when its output is channelled directly into a national grid Power curve The relationship between net electric output of a wind turbine and the wind speed measured at hub height Rated wind speed The lowest steady wind speed at which a wind turbine can produce its rated output power Reactive power An imaginary component of the apparent power It is usually expressed in kilo-vars (kVAr) or mega vars (MVAr) Reactive power is the portion of electricity that establishes and sustains the electric and magnetic fields of alternating-current equipment cT torque coefficient Cd drive-train torsional damping constant iSd(t), iSq(t), iRd(t), iRq(t) stator/rotor (d, q) current components (A) doi:10.1016/B978-0-08-087872-0.00212-2 329 330 Wind Turbine Control Systems and Power Electronics IT lumped rotational inertia of the turbine (rotor, generator, etc.) (kg m2) K lumped stiffness coefficient of the turbine Kd drive-train torsional spring constant Lm stator/rotor mutual inductance (H) LS, LR stator/rotor inductances (H) p number of generator pole pairs qr(t) generator rotor azimuth angle (rad) qt(t) turbine rotor azimuth angle (rad) R turbine rotor radius (m) RS, RR stator/rotor resistances (Ω) Te(t) generator electromagnetic torque (electrical) (N m) Tm(t) rotor aerodynamic torque (mechanical) (N m) VSd(t), VSq(t), VRd(t), VRq(t) stator/rotor (d, q) voltage components (V) w(t) hub-height uniform wind speed across the rotor disk (m s−1) βi(t) blade i pitch angle (rad) γ(t) rotor yaw angle (rad) θS(t) stator flux position (Hz) λ tip speed ratio ρ air density (kg m−3) ψRd(t) = LRiRd(t) stator/rotor (d, q) flux components (weber (m2 kg s−2 A−1)) ψRq(t) = LRiRq(t) stator/rotor (d, q) flux components (weber (m2 kg s−2 A−1)) ψSd(t) = LSiSd(t) stator/rotor (d, q) flux components (weber (m2 kg s−2 A−1)) ψSq(t) = LSiSq(t) stator/rotor (d, q) flux components (weber (m2 kg s−2 A−1)) ω(t) rotor angular velocity (mechanical) (rad s−1) ωe(t) = pωm(t) electrical speed (electrical rad s1) S(t)ẳdS tị=dt stator field frequency (rad s−1) Note: Tip speed ratio (TSR) is nondimensional However, since λ = ωR/v, the units appear to be (rad s−1)(m) (m−1 s) = rad But the SI unit of frequency is given as hertz (Hz), implying the unit cycles per second; the SI unit of angular velocity is given as radian per second Although it would be formally correct to write these units as the reciprocal second, the use of different names emphasizes the different nature of the quantities concerned The use of the unit radian per second for angular velocity, and hertz for frequency, also emphasizes that the numerical value of the angular velocity in radian per second is 2π times the numerical value of the corresponding frequency in hertz The following table summarizes these facts Derived quantity Plane angle Angular velocity Name Symbol Expressed in terms of SI base units Radian Radian per second rad rad s−1 m m−1 m m−1 s−1 = s−1 Hence, λ has units (s−1)(m)(m−1 s): non-dimensional 2.11.1 Control Objectives The conversion of wind energy into electrical power is not as straightforward as it might seem at first glance Wind speed is highly unpredictable and volatile Furthermore, wind potential is not evenly distributed across the globe (Figure 1) Wind generators cannot work optimally in every wind speed, thus they are designed for maximum production in a certain wind speed margin Categorization of wind turbines follows IEC’s 61400 classification as shown in Table [1] In stronger than rated winds, the generator is in danger of being damaged, while at weaker winds the generator produces less than expected To increase power production in these ‘nonrated’ wind speeds, control and supervision systems are employed In short, the most important objectives of a wind turbine control and supervision system are • • • • to maximize efficiency at every operating point; to minimize the structural load on the wind turbine; to meet strict power quality standards (power factor, harmonics, flicker, etc.); and to transfer the electrical power to the grid at an imposed level for a wide range of wind speeds To meet the above objectives, the control and supervisory system of large, variable-speed machines should consist of three main subsystems (Figure 2) Smaller wind turbines which frequently have no blade pitch control have no active speed and power control Instead, passive aerodynamic power control is achieved by exploiting blades stall effect while speed remains almost constant as it is fixed to the system frequency But even in this simpler version, a supervisory control system is necessary for operation monitoring and controlling the operating sequence Control requirements depend mainly on the two ends of the wind power conversion process: the turbine rotor side and the grid side On the turbine generator end, we distinguish between fixed- and variable-speed operation Although these modes require different control strategies, it is common in large megawatt turbines to adopt a discrete strategy, depending on the wind regime that the turbine is actually operating (Figure 3) Region describes start-up when wind speeds are below cut-in speed Region is Wind Turbine Control Systems and Power Electronics Annual 50 m Wind Speed July 1983–June 1993 90 60 30 –30 –60 –90 –180 0.0 –120 1.3 2.7 3.5 –60 4.5 5.0 Region average = 6.808 5.5 6.0 6.5 (m/s) 60 7.0 7.5 120 8.0 8.5 180 9.0 >12.0 NASA /SSE 13 Sep 2004 Figure Global wind potential Reproduced from NASA (www.nasa.gov) Categorization of wind turbines (IEC’s 61400 classification) Table Wind turbine class I II III IV vave (average wind speed at hub height (m s−1)) v50 (extreme 50-year gust (m s−1)) 10.0 70 8.5 59.5 7.5 52.5 6.0 42.0 Aerodynamics Electromagnetic subsystem Grid connection subsystem Pitch Yaw Variable-speed control Output power conditioning Figure Overall control system NREL (www.nrel.gov/wind) Power Wind power Aerodynamic power Rated power wcut-in wrated Region Region Figure Wind turbine operational regions wcutout Region Region Wind speed 331 332 Wind Turbine Control Systems and Power Electronics between cut-in and rated wind speeds, just before the turbine generates rated power The main objective of a controller in this region is to capture the maximum amount of energy from the wind This is achieved by keeping blade pitch approximately constant and using generator torque to vary the rotor speed With small pitch changes about the optimal angle, a controller can also reduce dynamic loads in the structure In Region 3, between rated and cutout wind speeds, wind power must be shed by the rotor to limit output power to the rated value This is usually accomplished by keeping generator torque constant and commanding blade pitch angles Structural fatigue loads can also be reduced in Region via individual pitch commands The overall goal of the control system is to meet different performance objectives in each operating region and make the transition between Regions and proceed smoothly to avoid load spikes Finally in Region the controller should stall the machine This strategy, as shown in the generic block diagram of Figure 4, is supervised by the wind turbine’s supervisor system On the grid side, modern megawatt turbines employ full- or partial-load frequency converters to convert variable-frequency, variable-voltage current into constant-frequency, constant-voltage current (Figure 5) This enables decoupled regulation of active and reactive power, wherever the type of generator allows it Hence, it is acceptable to consider generator control separately from grid control Having said that however, it must be pointed out that the two subsystems may be coupled in the case of systems capable of tolerating grid faults It has been difficult to gather information on the design and field performance of industrial controllers employed in real wind turbines Furthermore, although there is an abundance of research papers on wind turbine controllers, ranging from the simpler proportional-integral-derivative (PID) to more exotic fuzzy or neural versions, the performance of most of these is judged from computer simulations of mathematical models of the wind turbine system Even though these models are sometimes quite complicated and therefore quite accurate, it is impossible for them to capture every detail of the real world Another paradox is the well-known fact that despite an abundance of theoretical work on almost every type of controller available, industry, to the best of my knowledge, still uses the solid PID controller, with various modifications, on all machines Some of the material that follows is taken from the US’s Department of Energy National Renewable Energy Laboratory (NREL) research This is tested both on simulated environments and on their Controls Advanced Research Turbine (CART) (Figure 6) A short description of CART follows The CART is actually a Westinghouse WTG-600 two-bladed, teetered, upwind, active-yaw wind turbine It is of variable speed, and each blade can be independently pitched with its own electromechanical servo The pitch system can pitch the blades up to vcut-in