Stateoftheart variable speed wind turbine control

In a variable speed wind turbine, the rotor and the generator are decoupled from the grid by the power electronics implying that the rotor may rotate at (almost) any speed. Consequently, variable speed operation offers more control possibilities than constant rotational speed does. Variable speed operation has two main advantages over constant speed operation: i) additional energy capture at partial load and ii) potential reduction of fatigue loads on the structure by absorbing torque fluctuations in the rotor momentum. Other benefits that have been claimed are enhanced utility grid system compatibility, controllable power factor, reduction of acoustic noise at low wind speeds, adaptation to local conditions or compensating for changing conditions, and avoidance of stall over most the operating range [186].

In the context of maximum power extraction in partial load, the effect of variable speed is easily described. The kinetic energy content of the wind passing through the rotor varies with the cube of the wind velocity as illustrated in Fig. 7.2. Only a fraction of this energy can be extracted by the rotor (and subsequently converted into electrical power) as explained in Section 3.3. This fraction is denoted by the power coefficient Cp, which is a function of the tip-speed ratio A (i.e. the ratio of rotor circumferential speed to wind velocity) and the blade pitch angle 0. Usually, the power coefficient has one distinct maximum at the optimal tip-speed ratio and blade pitch angle. Practically achievable maximum values are in the range of 0.40 < Cp < 0.50, while common optimum values of the tip-speed ratio lie between 5 and 10 depending on the aerodynamic properties of the wind turbine configuration under investigation.

It is thus desirable to let the wind turbine rotational speed vary over a wide range, since maximum energy is extracted if the turbine operates constantly at or

1Most megawatt sized wind turbine models that are installed recently and the ones under development are, however, of the pitch controlled variable speed type

Wind Velocity Wind Turbinev Torque
Figure 7.2: Total power in the wind passing through a 25-meter diameter rotor as function of the wind velocity.

near its optimum tip-speed ratio. Observe that variable speed operation is not possible in case that the generator is directly connected to the utility grid. To allow the generator rotational speed to vary to extract the maximum amount of power, a power electronic interface is thus needed. In such an interface the three-phase generator output is rectified into DC and subsequently interfaced with the three-phase utility source by means of a power electronic converter as outlined in Section 3.5. In general, a rectifier at the generator side as well as an inverter at the utility grid site is required to provide both control and power quality requirements (including power factor).

The control problem of pitch-regulated, variable speed wind turbines significantly differs from that of constant speed wind turbines using pitch regulation. The latter presents a SISO control problem, while variable speed using pitch regulation presents a MIMO control problem as the electromechanical torque control loop is added to the existing pitch control loop. This opens the possibility to exploit the interactions between the inputs and outputs in the system to reduce dynamic loads as well as maintaining a desired amount of energy production. The current control strategy is to use the fast electromechanical torque control to respond to transients, and a slower pitch control loop to follow minute-to-minute fluctuations in wind speed. Both loops, however, often carry out their tasks independently by PID-type controllers [30, 283]. Probably, a more cost-effective strategy is couple the two control loops and make them to interfere constructively. Finally, it must be stressed that in order to make fair comparisons between the industry standard closed-loop controller and more advanced ones, it is necessary to i) establish the best possible PID controller, and ii) define criteria by which the performance can be evaluated.

The control objectives of the variable speed turbine depend on the operating regime. Below rated wind speed (or partial load) a common objective is to operate the turbine at optimum tip-speed ratio until a speed, power or torque limit is reached [245]. Above the rated wind velocity (or full load), power limitation is the main goal. In literature, various controllers are proposed for the aforementioned operating regimes. Multivariable controllers are successfully applied in simulation studies by Steinbuch [281] (at full load only, results based on a rather coarse wind turbine model that has not been confronted with measured data), Bongers [21] (achieved significant fatigue load reduction of both rotor shaft torque and blade root bending moments in one full load operating condition), and Stol [283] (reduced cyclic blade root bending moments by periodic control of the pitch angles). Molenaar [188] examined the effect of a simple filter controller on the fatigue loads on a model of a 2-bladed 500 kW variable speed turbine. It has been shown by comparison of the areas enclosed by the rainflow counts of in total 26 load cases (including power production with or without the occurrence of fault, startups and stops at normal wind conditions and parking cases as prescribed by the IEC-1400-1 standard [112]) that filtering the generator speed before computing a new torque set-point significantly reduces the fatigue loads on blade root (flap direction only), drive-train and tower.

In several other publications (see e.g. [39, 51, 95, 96, 135, 238, 276]) additional promising simulation results are shown. Various control design methodologies have been applied to models of different variable speed wind turbines in order to meet the aforementioned control objectives. However, all studies are performed with fairly basic wind turbine models (either the aerodynamics are approximated by a (static) model stored in a look-up table, or the structural flexibility is neglected). Sensor and actuator dynamics as well as disturbance issues are, in general, not addressed. As a consequence, the obtained results should be handled with care.

At the application side, however, still (too) little convincing results are reported in literature. Bongers applied robust control to a small experimental test-rig and demonstrated, in only one operating point, that a robust controller is capable to reduce fatigue loading while simultaneously maintaining a desired amount of energy production. De Boer and Mortier [16] presented experimental results of the application of static optimal output feedback on a 300 kW wind turbine. They showed by comparing time-series that, at full load, the direct current fluctuations can be reduced with almost a factor three with respect to the existing SISO controller. It is concluded from this that the fatigue loads are reduced as well, although no rain-flow countings or load spectra are shown to illustrate the effect on the fatigue life. Kriiger et al. [41, 42, 142] demonstrated the potential of advanced control on an experimental 33 kW wind turbine. They showed, by comparing load spectra of the blade root bending moments, that i) variable speed operation is capable of reducing mechanical loads when compared to constant speed operation, and ii) advanced control can significantly reduce rotor blade fatigue loads (although at the expense of a slight reduction in the annual electricity yield).

The quick review of the history and state-of-the-art of (closed-loop) wind turbine control shows that:

• The industry standard closed-loop controller is (still) of the PID-type. It is unkown how the PID controller parameters are determined and to what extent they are optimized;

• Simulation studies are often performed with fairly basic wind turbine models, while the proposed control strategies are elaborate and, in general, do not address the real issue: cost-effective control of the whole operating envelope;

• There is a lack of literature reporting on practical field experience obtained with control systems. Many of the applications reported are restricted in the scope to only one operating point;

• For variable speed to become the norm and not the exception in the (near) future, the added cost of power electronics required by most variable-speed designs must be clearly offset by the increased electricity yield, reduction in fatigue loads and other systems costs, impact on the wind turbine design, and the added benefit of providing power conditioning for utilities.

As a consequence, significant improvements are still possible to reach the desired economic viability of wind power. The first step is to show on a simulation level that the competitive position of wind power can be significantly improved by advanced control. The models used in these studies need to be validated against data acquired from experimental tests. Next, the achieved benefits need to be demonstrated in practice on a (prototype of a) commercial wind turbine.

The main question is how these controllers can be designed. The controller design must, in principle, encompass all aspects of both performance and cost, ranging from energy production, quality of power, lifetime and safety, through cost-effectiveness, acoustic noise and reliability. Notice that the relative importance of these aspects may vary from site to site. For example, a key objective for the design of cost-effective offshore wind turbines will be that the operation and maintenance requirements are reduced to a minimum, possibly at the expense of a somewhat higher wind turbine capital cost or lower electricity yield. The challenge for the control system designer is thus to specify an economic control design objective that reveals the financial impact of a proper controlled wind turbine. Before we can develop such an objective, we first have to determine what makes up the cost of generating electricity using wind.

Renewable Energy 101

Renewable Energy 101

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable. The usage of renewable energy sources is very important when considering the sustainability of the existing energy usage of the world. While there is currently an abundance of non-renewable energy sources, such as nuclear fuels, these energy sources are depleting. In addition to being a non-renewable supply, the non-renewable energy sources release emissions into the air, which has an adverse effect on the environment.

Get My Free Ebook


Post a comment