Wind Turbine - Power generation systems
Fixed speed wind turbines
In these types of turbines induction electrical machines (also known as asynchronous machines), generally used as motors for many industrial applications, are used for the conversion of the mechanical energy extracted from the wind into electrical energy.
In the wind turbines, on the other hand, these electrical machines are used as generators, above all because of their constructional simplicity and toughness, their relative cost-effectiveness and for the simplicity of connection and disconnection from the grid (Figure 6.1).
The stator of an induction machine consists of copper windings for each phase, as the stator of synchronous machines.
On the contrary, the rotor in squirrel-cage motors has no windings, but consists of a series of copper bars set into the grooves of the laminated magnetic core.
Some induction machines can have windings also on the rotor and in this case they are called wound rotor machines. They are expensive and less sturdy than the previous type and are used in variable-speed wind turbines, as better explained in the following paragraphs.
Induction machines require a given quantity of reactive power to function. This power shall be either drawn from the grid or delivered locally through a capacitor bank, which shall beproperly sized so that self-excitement of the synchronous generator can be avoided in case of grid disconnection due to failure.
Besides, these machines need an external source at constant frequency to generate the rotating magnetic field and consequently they are connected to grids with high short-circuit power able to support frequency. When working as a generator, the asynchronous machine is speeded up by the wind rotor up to the synchronous speed and then connected to the grid, or it is at first connected to the grid and started as a motor up to the steady state speed.
If the first starting method is used, the turbine clearly is self-starting and therefore the Pitch control must be present, whereas the second method is used for passive stall-regulated turbines. In this case the control system stores the wind speed and defines the speed range within which the generator is to be started. Once the synchronous speed has been achieved, the wind power extracted makes the rotor run in hypersynchronous operation with negative slip, thus supplying active power to the grid. As a matter of fact, since the slip has values in the order of 2%, the deviation from the rated speed is very limited and that’s why the use of these machines makes the wind turbine run at constant speed. To reduce the starting current, a soft starter is usually interposed between the asynchronous machine and the grid.
Variable speed wind turbines
At least in principle, there are different solutions which allow the rotor to run at variable speed, also keeping constant frequency. These solutions can be both of mechanical as well as electrical nature, even if the most used ones are of electrical type, in particular when using one of the following configurations:
- wound rotor asynchronous generators with external variable resistor
- wound rotor asynchronous generators with a power converter interposed between rotor and grid (doubly-fed configuration)
- asynchronous generators with a power electronic cnverter interposed between stator and grid (full converter configuration)
- synchronous generators (alternators) with a power electronic converter interposed between stator and grid (full converter configuration).
Asynchronous generator with variable resistor
By adding an external variable resistor in series with the rotor windings of a wound rotor asynchronous generator it is possible to get a variation of the electromagnetic torque of the generator and of the speed at which it is delivered (Figure 6.2).
Thus, both the possibility of operating at the optimum TSR point as a function of wind as well as allowing the rotor to accelerate changing the speed due to wind gusts is guaranteed, even though the losses due to joule effect in the external resistor rise. Besides, at high wind speeds, the total resistance of the rotor can be increased to keep constant the current flowing in the rotor (and therefore also in the stator), and with it also the power put into the grid, around the nominal power. The excess of mechanical energy generated by the rotor is therefore dissipated as heat by the additional external resistor.
Through this resistor it is possible to achieve a variation in the speed exceeding the synchronism speed in the range 0-10%. The equivalent electric diagram of an asynchronous generator with variable resistor RX is shown in the Figure 6.3, in which the resistive component RIx / s has been added to the common T circuit of the squirrel-cage asynchronous motor.
In order not to lose the power dissipated as heat in the additional resistor, this power can be put into the grid at the rated frequency by interposing an electronic power converter between the rotor of the asynchronous ring generator and the grid. This device converts the exceeding alternating power at the rotor first into direct power through a controlled rectifier and then reconverts it in alternating current at the rated frequency through an inverter (Figure 6.3).
Thus it is possible to supply the rotor with voltages of the proper width and frequency supplied by the electronic converter with the purpose of compensating for the difference of frequency between the angular velocity of the stator rotating magnetic field and the effective angular velocity of the rotor. The term “doubly-fed” reflects the fact that the stator voltage is applied by the grid, whereas the rotor voltage is applied by the electronic converter. The equivalent electrical diagram of the DFIG is shown in Figure 6.5, where, with the purpose of representing the converter influence, the varying voltage generator (V’r/s), which is a function of slip s, has been added to the common T-circuit of the squirrel-cage asynchronous motor. The active power shall always be going out from the stator and put into the net, independently of the operation state (either hyper- or sub-synchronous), whereas the rotor shall absorb power when operating as motor (sub-synchronism) and deliver it when operating as a generator (hyper-synchronism).
By assuming negligible both the stator as well as the rotor losses, the rotor power Pr managed by the converter, shall be linked to the stator power Ps through the slip s according to the following relation:
By identifying with Pnet the power that the machine delivers to the grid on the whole, determined by the algebraic sum of the stator and rotor powers, Pnet can be expressed as:
- negative s in hyper-synchronous operation
- positive s in sub-synchronous operation.
With this type of configuration, the electric generator supplies the network with 2/3 of the rated power through the stator directly connected and 1/3 through the rotor connected through the converter. Therefore also the converter can be sized for power equal to 1/3 of the rated power of the generator.
Besides, it is possible to control the reactive power production; this allows voltage regulation and magnetization of the machine by the rotor regardless of the grid voltage. By means of the doubly-fed configuration it is possible to obtain 30% speed variation above or below the synchronism speed.
The wound rotor asynchronous generator usually has a synchronism speed up to 2000 rpm and it is connected to the rotor axis through a three-stage gearbox. The connection of the rotor windings to the converter is carried out by means of the slip-rings and relevant brushes.
Asynchronous generator and converter
Squirrel-cage asynchronous generators can be used in wind turbines at variable speed by interposing an electronic converter between the generator and the grid. Such converter de-couples and releases the frequency of the rotating magnetic field from the grid frequency; then the frequency of the rotating magnetic field is modulated to control the rotating speed of the rotor. As shown in Figure 6.6, there is an electronic power system similar to that of the doubly-fed configuration, but positioned on the generator stator. As a consequence, the converter, unlike the previous configuration, must control the total power output. Being an induction generator, it needs however to absorb reactive power to function; this power can be supplied by the converter itself.
Synchronous generator and converter
The most common constructional shape of a synchronous generator (alternator) consists of a rotor, which creates the magnetic field, and of a stator comprisingthe induced windings. The rotor magnetic field (Φ=krIr) is generated by a continuous current (Ir) circulating in the field windings.
Such continuous current is supplied by a dynamo coaxial to the alternator or it is drawn at the stator terminals and then rectified by a diode bridge.
The movement of the rotor magnetic field with respect to the stator windings due to the rotation of the main shaft induces a triad of alternating voltages in the stator windings with an r.m.s. value proportional to the magnetic flow of the rotor and to the rotation speed (n):
Since the frequency of the generated electromotive force is linked to the rotation speed by the relation:
where p is the number of the pole couples of the rotor winding, the r.m.s value of the voltage induced on the stator is proportional to the value of the frequency at which it is generated:
When the generator is connected to a load (island- or grid-connection) and the current is delivered, this generates in the air gaps of the machine a rotating magnetic field synchronized with the induction field, without the relative slip. Besides, if the two magnetic fields are aligned (angle δ=0), ), there is no resistive torque and consequently the active power provided to the grid is null. Otherwise, if there is a displacement due to an external
motive torque, a resistive electrical torque is generated balancing the active power put into the grid (δ>0).
The higher the deviation, the higher the active power put into the grid.
By keeping the angle δ fixed, the active power put into the grid increases linearly with the r.m.s. value of the induced voltage and therefore proportionally to the rotation speed and to the voltage frequency:
On the contrary, by keeping constant the provided active power when the rotation speed and consequently also the frequency and the induced voltage vary, the delivered current changes as shown by the curves in Figure 6.5. As it can be noticed, by assuming as parameter the delivered active power, if the induced voltage exceeds the grid voltage, the alternator provides reactive power, while, if the induced voltage is lower than the grid voltage, the alternator absorbs reactive power. In particular, if the induced power has value equal to the grid voltage (cosϕ=1), the current flowing through the stator takes its minimum value.
Synchronous generators are not intrinsically self-starting. The alternator generally reaches the synchronous speed by means of the prime mover and then it is connected in parallel following a proper procedure. In applications, for which self-starting is needed, the rotor is equipped with dampening copper bars, which start the alternator as an induction machine and during operation they dampen the dynamic oscillations of the machine. In wind applications, turbines with synchronous generators are normally started by the wind itself and a speed control system is used for the synchronization procedure.
Permanent magnet alternators are often used in wind turbines, in which the rotor is without the excitation winding and the induction magnetic field is generated directly by the permanent magnets built-in into the rotor. As a consequence, slip-rings and brushes are not necessary for the supply of the excitation circuit. The operating principle is analogous to that of the alternators with the induction winding, but in the permanent magnet alternators the voltage induced into the stator windings cannot be adjusted through the excitation current; therefore the voltage at the generator terminals is only a function of the rotation speed of the rotor.
Since the frequency generated by the alternator depends on the rotation speed of the rotor and on the number of pole couples, to be able to use the synchronous generator in a variable speed wind turbine keeping constant the frequency on the grid side, it is necessary to interpose a two-stage power converter controlling the whole of the generated electric power (Figure 6.6):
- in the first stage, either a diode or a thyristor-controlled bridge rectifier converts the electrical quantities generated by the alternator from variable frequency alternating quantities into direct quantities;
- in the second stage, through a DC link, supply is given to an inverter which converts the direct electrical quantities of voltage and current into alternating quantities at the grid frequency.
In case of a separated excitation alternator, the regulation of the r.m.s. value of the generated voltage is obtained by acting on the excitation current, while with a permanent magnet alternator the voltage can be adjusted either through a thyristor-controlled bridge rectifier or through a PWM-controlled inverter. The PWM control of the inverter can be carried out through different modalities:
- regulation of the value of the sinusoidal modulating amplitude by comparing the voltage value of the DC-link with the optimum curve P-Vdc;
- MPPT (Maximum Power Point Tracker) by using an anemometer. The power on the dc side is compared with the reference power and from a comparison with the optimum curve, depending on the wind speed, the new voltage on the dc side is determined. The PWM (Pulse Wide Modulation) control signal varies instantaneously as the operating conditions vary;
- MPPT with wind forecast: the energy previously extracted is taken into consideration and, by statistical models, the wind speed in the following moments is forecast. This control system tracks the optimum points as function of the foreseen speeds.
The use of the configuration alternator-power converter allows decoupling the generator from the grid, thus reducing the mechanical shocks on turbines during grid faults. Besides, there is generation also of the desired reactive power and full control on the active power.
There are three full converter concepts for electromechanical conversion: high-speed, medium-speed and low-speed.
High-speed conversion is mechanically similar to the doubly-fed type and normally uses a three-stage gearbox and a turbo alternator (up to 2000 rpm) usually of permanent magnet type or an asynchronous generator (Figure 6.7).
This configuration offers the advantages of using a smallsize and low weight generator and can be used to replace an existing doubly-fed configuration.
With medium-speed conversion instead either a single or a two-stage gearbox is used with a compact permanent magnet alternator (up to 500 rpm) (Figure 6.8).
This concept, with a lower size of the gearbox and lower rotation speeds in comparison with the previous configuration, allows reducing the mechanical stresses and therefore improving reliability. The alternator diameter is larger than in the previous case.
Finally, low-speed conversion eliminates the gearbox and uses an alternator, normally a permanent magnet or a low-speed (up to 30 rpm) separately excited generator, therefore with a number of poles greater than the previous ones (Figure 6.9).
This configuration offers above all the advantages deriving from the absence of the gearbox; they imply a reduction in the mechanical losses, the elimination of total noise contribution and a further increase in the reliability of the wind turbine.