ELECTRICAL FAULT CONTROL FOR WIND TURBINES BASIC INFORMATION AND TUTORIALS


WIND TURBINE ELECTRICAL FAULT CONTROLS
How To Control Electrical Faults in Wind Turbines

One of the roles of a monitoring and control system in a turbine is that if some fault is detected, the proper action is taken. A fault can be any malfunctioning of a component, including damage and breakage; high or low operating temperature or pressure; or a value outside of the allowed tolerances for a mechanical or electrical parameter such as speed, voltage, current, and the like.

Depending on how serious the fault is, a turbine must be shut down if the fault cannot be corrected. If a turbine has to be shut down, a process of shutting down the turbine must be followed. Some faults may lead to a temporary action that must be checked a number of times to see if it persists.

Th e number of faults that can occur is large and it is not practical to list them, since faults can be diff erent from one machine to another. Some of the more important mechanical faults that can happen are mentioned here:

a. Excessive vibration in various parts of the turbine; this can be in any of the main bearings, in the nacelle, in the tower, and in the gearbox.

b. High temperatures of various fluids in the gearbox, bearings, oil cooling system, or any other hydraulic system.

c. Low level or low pressure of various liquids in various parts of the system.

Also, some faults are not detectable by a turbine controller, unless redundant systems are used. Redundant means that one parameter or variable is measured by two devices, so that if one fails the other shows a (diff erent) value. For instance, suppose that the anemometer malfunctions or freezes in a freezing rain. 

The reading from the anemometer then shows a zero speed for wind, whereas there is wind. In such a case, the controller interprets this as lack of wind and, thus, the turbine is put into halt and with the brakes on. Th is is a typical scenario that calls for a technician to act.

One of the major faults that can happen in a wind turbine and can lead to disastrous results is when a turbine speed increases beyond the allowed limit. Th is is called overspeed. When a turbine rotates, there are dynamic forces that act on blades and all the other related mechanical components.

The magnitude of these forces depends on the speed of rotation. If the speed goes up, the magnitudes of these forces increase as well, and can go beyond the values for which the components are designed. If this happens, these components can break. And if one component in a mechanical system breaks and the cause is still present, this can lead to other component failure and serious damage.

In particular, if the speed of a wind turbine increases beyond its designed value, the blades can fl y away and the whole turbine can fall apart in a short time. If during the normal operation of a wind turbine, the load is suddenly taken off the turbine, this can lead to overspeed. 

This can happen if the generator is disconnected from the grid, for instance, in the case of lightning tripping the overhead breakers and disconnecting the circuit. Even if the turbine controller starts shutting down the turbine in such a condition, the momentum in the rotor, as a result of its existing speed, can speed up the turbine to overspeed. 

In order to prevent this from happening to the turbine, modern turbines are equipped with a set of resistors that can connect to the generator and become a temporary load, until the situation is corrected, or the turbine has slowed down. 

These resistors, called crowbars, are not normally connected, but in the case when such a fault occurs they kick in and connect to generator stator. As a result of taking a large load from the generator, these resistors become hot and their heat must be taken from them as fast as possible. For this reason, they are placed outside of the nacelle where they can be cooled by fresh air.

Control Systems
As we have learned so far, many variables must be controlled in a wind turbine, such as blade pitch angle and nacelle yaw angle. Also, on the electrical side, there are voltage, current, frequency, and other variables that must be controlled to a desired value. Here, we elaborate on how an entity, in general, is controlled. The discussion here applies to any variable that needs to be controlled. But for sake of clarity, we may use the terms for blade pitch control, when necessary.

Suppose that the value of pitch angle for blades at a certain time during the operation of a turbine must be 35°. This value is from a reference point for measurement of blade angle that one may physically count 35°. Moreover, this value is not fixed, since a minute later, it could be altered to, say, 38°. But, for any given instant the figure defines the desired value.

A control action is normally taken in a control loop. In order for better control, feedback control is used and the value to be controlled is continuously checked and compared to the desired value, and corrections are made. In the control loop the desired value is the set point.

An action is taken on the difference between the actual value (that is, the measured value) and the set point. This difference is called error.

The action to be taken based upon the error is performed by the actuator. For a pitch control system, the actuator is an electric motor or a hydraulic piston that turns the blade. The command to the actuator for how much must it rotate the blade can be found by a control law. 

There are various control laws; but the most common and traditional ones are proportional control and PID control (PID stands for proportional, integral, and derivative). Each element in a PID control has a different role.

Proportional control introduces an action proportional to the value of the error. Its role is for stabilization of a system. The integral action is intended to take care of a drift in the controlled value. 

This drift is called offset or steady-state error. If not taken care of, by integral control, the value reached can be slightly different from the desired value. Derivative control has the effect of looking forward and predicting what can be expected to happen, and adding the necessary action.

No comments:

Post a Comment