Controller and method for a wind turbine

Abstract

A method of activating a de-icing system of a wind turbine, comprising: monitoring an output power of the wind turbine; monitoring an output from an ice sensor disposed on the wind turbine indicative of icing conditions in the atmosphere; and activating the de-icing system when the output power is lower than an expected output power and the output from the ice sensor indicates that the icing conditions in the atmosphere have ended.

Description

Technical Field

[0001] Several aspects of the present invention relate to techniques, apparatus and methods for monitoring wind turbine structures, particularly but not limited to wind turbine blades. Background Technology

[0002] Wind turbines typically feature a rotor in the form of a rotatable hub, which carries a set of wind turbine blades. Wind acts on the turbine blades, causing the hub to rotate. This rotational motion is transmitted to a generator via a gear train, or, in the case of a so-called direct-drive type of wind turbine, directly to the generator. In the generator, electrical energy is generated, which can then be supplied to the power grid.

[0003] Wind farms, which consist of multiple wind turbines, are typically located in geographical locations where atmospheric temperatures frequently drop below freezing. Such "cold climates" increase the risk of icing on wind turbine components, particularly the rotor blades. This can not only reduce the efficiency of the wind turbine but also increase the risk of ice being ejected from the rotor blades as they rotate, potentially posing a risk to people near the turbine and potentially causing damage to the turbine itself or other nearby wind turbines. For these reasons, wind turbines are typically equipped with systems to detect the presence of ice on the turbine, as well as systems to remove ice from the rotor blades or other components when ice buildup is detected.

[0004] This invention was designed in this context. Summary of the Invention

[0005] In one aspect, the present invention provides a method for activating a de-icing system of a wind turbine. The method includes: monitoring the output power of the wind turbine; monitoring an output from an ice sensor located on or near the wind turbine indicating atmospheric icing conditions near the wind turbine; and activating the de-icing system when the output power is lower than expected and the output from the ice sensor indicates that atmospheric icing conditions have ended.

[0006] In this way, the timing of the de-icing system's activation is optimized, and the need for repeated de-icing is avoided, as the system is only activated after it is certain that the icing event has ended. De-icing uses electricity, thus reducing the efficiency of electricity generated by wind turbines. Therefore, it is beneficial to de-ic the wind turbines once at the end of the ice accumulation phase, rather than repeatedly de-icing them during the period of atmospheric icing.

[0007] This method may include monitoring the ambient temperature and activating the de-icing system only when the ambient temperature is within a predetermined range. This prevents the de-icing system from being triggered in situations where it is too cold to make de-icing operations effective.

[0008] This method may include activating the de-icing system only when the output of the ice sensor indicates that atmospheric icing has ended and has lasted for a predetermined period of time. This helps ensure that atmospheric icing (during which ice buildup occurs) has completely ended and lasted for the predetermined period of time before the de-icing system is triggered.

[0009] This method may include activating the de-icing system only in the following states: when the output power has been determined to be below the expected output power for a first predetermined time period to help ensure prevention of abnormal declines in power production, or when the output power has been substantially zero for a specified percentage of a second predetermined time period. The latter situation indicates that the wind turbine has stopped due to ice accumulation. Advantageously, the second time period is shorter than the first time period, allowing the wind turbine that has stopped “icing” to be de-iced and put back into operation as quickly and effectively as possible.

[0010] Monitoring the output of an ice sensor may include receiving data from the ice sensor at a series of discrete time intervals. Similarly, monitoring the output power of a wind turbine may include receiving data indicating the output power from the wind turbine at a series of discrete time intervals.

[0011] In some examples, monitoring the output power of a wind turbine may include receiving primary power data indicative of the output power from the wind turbine at a series of discrete time intervals within a first window. In this case, the method may include receiving primary power data from multiple consecutive first windows.

[0012] The method may include using the corresponding primary power data for each first window to calculate the average primary output power for each first window. The method may also include determining the average actual power of the wind turbine by calculating the average of the average primary output power.

[0013] Averaging the output power of a wind turbine over a first time window and multiple first time windows is advantageous because it prevents small fluctuations in these parameters from triggering the activation of the de-icing system before it becomes necessary. In other words, averaging the output power in this way prevents the de-icing system from being erroneously triggered.

[0014] Primary power data can be received by the controller in the first window or at the end of each first window.

[0015] The method may include receiving primary wind speed data from at least one wind speed sensor at a series of discrete time intervals in a first window, wherein the primary wind speed data indicates the wind speed near the wind turbine.

[0016] The method may also include receiving primary wind speed data from multiple consecutive first windows. In this case, the method may include using the corresponding primary wind speed data from each first window to calculate an average primary wind speed value for each first window.

[0017] The method may include determining a second average wind speed value by calculating the average of the average primary wind speed values ​​for each first window.

[0018] This method may include determining the expected power of a wind turbine based on a second average wind speed value.

[0019] Similar to the averaging process used to calculate actual output power, averaging wind speed values ​​in this way prevents wind speed fluctuations from causing unrepresentative expected output power, which could otherwise trigger the de-icing system when not needed.

[0020] The method may include comparing the average actual power with the expected power and identifying difference data points, wherein the difference data points are the differences between the average actual power and the expected power.

[0021] The method may include comparing differential data points to a threshold to determine whether the expected power exceeds the average actual power by a greater amount than the threshold; and activating the de-icing system if the expected power exceeds the actual power by a greater amount than the threshold.

[0022] In one example, the threshold depends on a second average wind speed value. Based on this example, the method could include grouping different values ​​of the second average wind speed into two or more groups, each group having a different associated threshold.

[0023] In one example, the primary wind speed data is received by the controller at the end of the first window or each first window.

[0024] In one example, the duration of the first window, or each first window, is 10 minutes.

[0025] The method may include comparing differential data points with a threshold to determine whether the average actual power is less than the expected power by an amount greater than the threshold; and activating the de-icing system if the average actual power is less than the expected power by an amount greater than the threshold.

[0026] Alternatively or additionally, the method may include comparing the average actual power with a threshold; and activating the de-icing system based on the comparison.

[0027] In some embodiments, the threshold may depend on a second average wind speed value. Different values ​​of the second average wind speed may be grouped together into two or more groups, each group having a different associated threshold. In other embodiments, the threshold may be the same for all second average wind speed values, such that the threshold is independent of the second average wind speed value.

[0028] Primary wind speed data may be received by the controller at the end of the first window or each first window. In some embodiments, the duration of the first window or each first window may be 10 minutes, but of course, in other embodiments, the duration of the first window or each first window may vary.

[0029] In another aspect, the present invention relates to a controller for a de-icing system for a wind turbine. The controller includes: an input configured to receive data from an ice sensor indicating the presence of atmospheric icing; an input configured to receive data indicating the power output of the wind turbine; and a processor. The processor is configured to determine whether the period of atmospheric icing has ended and whether the wind turbine's output power is lower than expected. The processor is configured to issue an output control signal to activate the de-icing system upon determining that the output power is lower than expected and that the atmospheric icing has ended. Therefore, the controller is used to trigger the de-icing system when de-icing will be most efficient and effective.

[0030] The controller may include an input configured to receive data indicating ambient temperature from a temperature sensor. The controller may be configured to issue a control signal only when the ambient temperature is determined to be within a predetermined range.

[0031] The controller can be configured to issue a control signal only when ice sensor data indicates that atmospheric icing has ended and continues for a predetermined period of time.

[0032] The controller can be configured to issue a control signal only when the output power is determined to be lower than the expected output power for a predetermined period of time, or when the output power has been substantially zero for a predetermined percentage of a second period of time. Advantageously, the second period of time is shorter than the first period of time.

[0033] The controller can be configured to perform the methods described in any of the preceding sections.

[0034] In another aspect, the invention exists in a control system comprising an ice detector and a controller as described in any of the preceding paragraphs.

[0035] In another aspect, the invention exists in a wind turbine that includes the controller or control system described in the foregoing paragraph. Attached Figure Description

[0036] The invention will now be described by way of example with reference to the accompanying drawings, in which:

[0037] Figure 1 This is a schematic diagram of a wind turbine with a controller according to an embodiment of the present invention;

[0038] Figure 2 It shows Figure 1 A schematic diagram of the controller is shown, along with multiple sensors and a de-icing system; and

[0039] Figure 3 A method is shown. Figure 1 The steps of the method executed by the controller. Detailed Implementation

[0040] The following detailed description refers to the accompanying drawings, which illustrate specific details and embodiments in which the invention may be practiced by way of illustration. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. Other embodiments and structural changes may be used without departing from the scope of the invention as defined in the appended claims.

[0041] Figure 1 A wind turbine 8 according to an embodiment of the present invention is shown. The wind turbine 8 includes a nacelle 12 supported on a generally vertical tower 14, the tower itself being mounted on a foundation (not shown). The nacelle 12 houses multiple functional components, including a gearbox and a generator (not shown), and supports a main rotor assembly 16. The main rotor assembly 16 includes a hub 17 and a plurality of wind turbine blades 18 received in blade receiving holes (not shown) in the hub 17. In this example, the wind turbine 8 includes three wind turbine blades 18.

[0042] The wind turbine 8 also includes a controller 20, multiple sensors 22, and a de-icing system 24, such as Figure 2 The diagram is shown schematically. As will be explained, controller 20 monitors the output power of wind turbine 8 and the output 26 from ice sensor 28 located on top of nacelle 12 of wind turbine 8, and activates de-icing system 24 based on these values. Specifically, as will be explained, controller 20 is configured to trigger activation of de-icing system 24 when it determines that ice is present on wind turbine 8 but no longer accumulating and when the output power of wind turbine 8 is lower than expected.

[0043] Multiple sensors 22 include a current sensor 30a, a voltage sensor 30b, a first wind speed sensor 32a and a second wind speed sensor 32b, a first temperature sensor 34a and a second temperature sensor 34b, a barometric pressure sensor 36, and an ice sensor or ice detector 28.

[0044] A current sensor 30a is configured to measure the output current value of the wind turbine 8 during operation. A voltage sensor 30b is configured to measure the output voltage value of the wind turbine 8 during operation. The output power generated by the wind turbine 8 can be calculated based on the current and voltage values ​​measured by the current sensor 30a and voltage sensor 30b, as will be understood by those skilled in the art. It should also be understood that any sensor or device capable of measuring or determining the output power of the wind turbine 8 can be used, in addition to or replacing the current sensor 30a and voltage sensor 30b of this embodiment.

[0045] Each wind speed sensor 32a, 32b is configured to measure wind speed values ​​near the wind turbine 8. The wind speed sensors 32a, 32b can be in the form of force-torque sensors. Typically, the wind speed sensors are mounted on the top of the nacelle 12 of the wind turbine 8 and are in the form of an anemometer. Anemometers come in various types, such as cup anemometers, blade anemometers, hot-wire anemometers, laser Doppler anemometers, and ultrasonic anemometers. Ultrasonic sensors may be preferred for wind turbines in hard-to-access locations (e.g., at sea) because they do not require recalibration. Specifically, the ultrasonic sensor measures wind speed based on the time of flight of the acoustic pulses between a pair of transducers.

[0046] Each temperature sensor 34a, 34b is configured to measure the ambient temperature value near the wind turbine 8. Temperature sensors 34a, 34b may be mounted, for example, on the nacelle 12 of the wind turbine 8. In different examples, temperature sensors 34a, 34b may be located elsewhere in a wind farm including the wind turbine 8, and the ambient temperature of all wind turbines 8 in the wind farm may be based on data measured by temperature sensors 34a, 34b. The barometric pressure sensor 36 is mounted on the hub 17 in this embodiment, but may be located elsewhere in other embodiments.

[0047] In this embodiment, the ice sensor 28 is located at the top of the nacelle 12, but in other embodiments it may be located at another suitable location. In one embodiment, the ice sensor 28 may be located near the wind turbine 8 or near a wind farm including the wind turbine 8, and the activation of the de-icing system 24 may be based on data measured by the ice sensor 28.

[0048] Ice sensor 28 is capable of operating to determine whether ice has accumulated on wind turbine 8. In this example, ice sensor 28 is based on the principle of ultrasound. When ice is present on the sensor wire of ice sensor 28, the ultrasonic signal attenuates, allowing ice to be detected near the wire. When ice sensor 28 detects ice, the temperature of the wire (which also acts as a heating element) rises to remove the ice. Ice sensor 28 repeatedly measures to determine whether ice is present and removes the detected ice by heating. During an atmospheric (or meteorological) cycle, i.e., an ice accumulation cycle in which ice accumulates on wind turbine 8, ice sensor 28 repeatedly measures the ice and sends a signal to controller 20 to indicate the presence of ice on turbine 8. Because ice sensor 28 repeatedly removes the accumulated ice by heating, ice sensor 28 is configured to identify the time when ice accumulation stops, i.e., the time when atmospheric icing no longer exists. At this point, ice sensor 28 signals to controller 20 to indicate that the atmospheric icing cycle has ended. It should be understood that ice sensor 28 can take many different forms and is not limited to this particular ice sensor 28 or this particular type of ice sensor 28.

[0049] Alternatively or additionally, the ice sensor 28 may be a blade-based ice detection system that measures the vibration mode or frequency response of the wind turbine blade 18. In this case, since the vibration mode or frequency response of the blade 18 depends on the mass of ice accumulated on the blade 18, monitoring these parameters over time can provide an indication that previously existing atmospheric icing conditions no longer exist (and therefore indicate that the atmospheric icing event has ended).

[0050] For example, if it is determined that the measured vibration pattern has stabilized and remains stable over a given time period (e.g., 60 minutes), this could indicate that the mass of ice on blade 18 has stopped increasing, and therefore the atmospheric icing event has ended. The measured vibration pattern could also indicate a decrease in the mass of ice on the blade, thus indicating that the atmospheric icing event has ended.

[0051] Another alternative or additional sensor that can be used in this system is an ice sensor 28 that detects the thickness of ice on blade 18. In this case, the system can determine that the atmospheric icing event has ended when the measured thickness has not increased or has decreased over a given time period (e.g., 60 minutes).

[0052] Now turn to controller 20, which includes multiple inputs 40, a processor 42, and outputs 44. Inputs 40 are configured to receive sensor output data 46 from sensor 22. Although in Figure 2While not explicitly shown, in this embodiment, data output from sensor 22 is indirectly sent to controller 20, which forms part of a Supervisory Control and Data Acquisition (SCADA) system (not shown) that collects and stores data from sensor 22 of wind turbine 8 and sends time-averaged data from desired sensor 22 to controller 20 every ten minutes. Processor 42 is configured to determine whether de-icing system 24 should be activated based on received sensor output data 46, as will be explained. Output 44 is configured to send a control signal 48 in the form of an activation signal to de-icing system 24 to activate it. Controller 20 additionally includes a memory device 50, such as a non-transitory computer-readable medium. The memory device 50 of controller 20 stores a reference power curve of wind turbine 8 indicating the expected power output of wind turbine 8 at different wind speeds. The memory device 50 also stores instructions that, when executed by processor 42, cause processor 42 to perform the methods described below.

[0053] The de-icing system 24 of this embodiment is disposed in the hub 17 of the wind turbine 8 and is capable of operating to capture, heat, and propel air through an outlet in each of the wind turbine blades 18, so that hot air circulates within the blades 18 and heats them internally. It should be understood that the de-icing system 24 can take many different forms and is not limited to this arrangement or to being this particular type of de-icing system 24. As an example, the de-icing system 24 may instead include heating elements integrated into the blades 18.

[0054] Figure 3 The steps of a method performed by controller 20 to determine whether the de-icing system 24 of wind turbine 8 should be activated are shown. It should be noted that, in practice, this method can be performed by controller 20 at any suitable frequency (preferably every ten minutes or every minute) upon receiving data from sensor 22.

[0055] The method begins at step 100, at which point the input 40 of the controller 20 receives sensor outputs 54 from temperature sensors 34a and 34b and wind speed sensors 32a and 32b, respectively. In this embodiment, the controller 20 receives this data through the SCADA system of the wind turbine 8 instead of receiving it directly.

[0056] At step 102, controller 20 determines whether the data received at step 100 is complete and valid. If controller 20 receives all the expected data from sensor 22, it determines that the data is complete. For example, if no data is received from one or more of temperature sensors 34a, 34b and / or wind speed sensors 32a, 32b, then controller 20 determines that the sensor data is incomplete.

[0057] There are many different ways to determine if sensor data is invalid, as will be understood by a person skilled in the art. In this example, the wind turbine 8 includes two temperature sensors 34a and 34b and two wind speed sensors 32a and 32b to enable verification of data received from these sensors. The specific setup or layout of the sensors 32a, 2b, 34a, and 34b, and the specific methods for verifying the measured data, can vary, as will be understood by a person skilled in the art. For example, the first temperature sensor 34a and the second temperature sensor 34b could actually be a single sensor with two channels 34a and 34b. In this case, checks can be performed to ensure that the signals received from each of the channels 34a and 34b are identical, or substantially identical within a given error window, in order to verify the measured data. A similar process can be performed for the data from the first wind speed sensor 32a and the second wind speed sensor 32b, which, in fact, could correspondingly be a single sensor with two channels 32a and 32b.

[0058] If the sensor data is determined by the controller 20 to be incomplete and / or invalid, the method terminates without triggering the activation of the de-icing system 24.

[0059] If the sensor data is determined to be both complete and valid, the method proceeds to step 104. In step 104, the processor 42 determines whether the temperature at the wind turbine 8 is above or below a predetermined temperature, or within a predetermined temperature range. In this example, the processor 42 determines whether the ambient temperature near the wind turbine 8, as measured by temperature sensors 34a and 34b, is greater than or less than a predetermined threshold ambient temperature. The threshold ambient temperature can be a value above which it is unlikely or impossible for ice to form and exist at the wind turbine 8. In this example, the threshold ambient temperature is approximately 0 degrees Celsius; however, any suitable threshold ambient temperature can be used.

[0060] If, at step 104, processor 42 determines that the ambient temperature is above a threshold temperature, then processor 42 determines that there is no ice present at wind turbine 8, and the method is terminated without activating de-icing system 24. If processor 42 determines that the ambient temperature is below a threshold temperature, then processor 42 determines that there may be ice present at wind turbine 8, and the method continues to step 106.

[0061] At step 106, controller 20 receives air pressure data indicating the vicinity of wind turbine 8 from barometric pressure sensor 36, and processor 42 uses data from wind speed sensors 32a, 32b and barometric pressure sensor 36 to calculate the average adjusted wind speed value WS over ninety minutes. avg,90 As described below.

[0062] In this embodiment, each of the wind speed sensors 32a, 32b and the barometric pressure sensor 36 measures every second, and these measurements are transmitted to the controller 20 at a series of discrete time intervals, separated by a series of first time windows tx, each of which has a duration of 10 minutes. At the end of each 10-minute time window tx, the controller receives the primary wind speed data in the form of measurements obtained by the wind speed sensors 32a, 32b and the barometric pressure sensor 26 during that time window tx. Therefore, in this embodiment, the controller receives 600 measurements from each of the wind speed sensors 32a, 32b and the barometric pressure sensor 26 at the end of each time window tx. It should be noted that in other embodiments, the duration of the time window tx may vary, and the number of measurements obtained by each sensor 32a, 32b, 36 during each time window tx may also vary.

[0063] At the end of the first ten-minute time window t1, controller 20 receives measurements obtained during the first time window t1 from wind speed sensors 32a, 32b and barometric pressure sensor 36. Controller 20 calculates the ten-minute average wind speed value WS from the wind speed measurements obtained during t1. avg,10 This ten-minute average wind speed value is also known as the average primary wind speed value. The controller 20 calculates the ten-minute average air pressure value AP from the air pressure measurements obtained during period t1. avg,10 Then, controller 20 calculates the wind speed based on the ten-minute average wind speed value WS during period t1. avg,10 AP value for ten minutes during t1 avg,10 Calculate the ten-minute average adjusted wind speed WS during period t1. avg-adj,10 .

[0064] At the end of the second ten-minute time window t2, which begins immediately after the first ten-minute time window t1 has elapsed, the controller 20 receives measurements obtained during the second time window t2 from wind speed sensors 32a, 32b and barometric pressure sensor 36, and repeats the above process to determine the ten-minute average adjusted wind speed value WS during t2. avg-adj,10 The above process is repeated seven times to determine the ten-minute average adjusted wind speed WS over nine consecutive ten-minute time windows t1 to t9. avg-adj,10 The average adjusted wind speed value every ten minutes is stored in the memory device 50.

[0065] Then, controller 20 adjusts the wind speed value WS by the ten-minute average during the period from t1 to t9. avg-adj,10 The average adjusted wind speed value WS over ninety minutes was calculated by averaging. avg-adj,90 This 90-minute average adjusted wind speed value is also known as the second average wind speed value.

[0066] Then, the method proceeds to step 108, where the 90-minute average adjusted wind speed value WS is calculated. avg-adj,90 This is used to determine the expected power to be generated by the wind turbine 8. That is, referring to the power curve of the wind turbine 8 stored in the memory device 50 of the controller 20, the controller 20 determines, under normal operating conditions (i.e., without ice), for a given ninety-minute average adjusted wind speed value WS avg-adj,90 The power generated by the wind turbine 8 is expected.

[0067] At step 110, the processor 42 generates the ninety-minute average actual power P using the primary power data received from the current sensor 30a and the voltage sensor 30b. avg,90 The primary power data indicates the actual power generated by the wind turbine 8, as described below.

[0068] Current sensor 30a and voltage sensor 30b measure every second during a ten-minute time window t1 to t9 (also known as the first time window), and these measurements (also known as primary power data), like wind speed and air pressure measurements, are transmitted to controller 20 at the same series of discrete time intervals. That is, current and voltage data collected during t1 are transmitted to the controller at the end of t1, current and voltage data collected during t2 are transmitted to the controller at the end of t2, and so on.

[0069] At the end of the first ten-minute time window t1, the controller 20 therefore receives six hundred current measurements (one per second within the ten-minute time window t1) and six hundred voltage measurements (one per second within the ten-minute time window t1). Based on these current and voltage values, the controller 20 calculates the actual output power per second within the ten-minute time window t1, and then calculates the average of these output power values ​​to give the ten-minute average output power value P. avg,10 This ten-minute average output power value is also known as the average primary output power. This process is repeated for time windows t2 to t9, thus determining a ten-minute average actual output power value P for each of the time windows t1 to t9. avg,10 These P avg,10 The value is stored in memory device 50.

[0070] Then, processor 42 calculates the nine P generated during the period from t1 to t9. avg,10 The average value is used to determine the average output power value P over ninety minutes. avg,90 This 90-minute average output power value is also known as the average actual power of a wind turbine.

[0071] Processor 42 determines the expected output power and the actual output power P at step 112. avg,90 The difference between them is recorded, and this information is stored on the memory device 50 as a power difference data point.

[0072] As already noted, although the duration of each time window t1 to t9 is 10 minutes in this embodiment, this may vary in other embodiments. Furthermore, the determination of the expected output power and the actual output power P... avg,90 The total number of time windows tx can vary. As a non-limiting example, in one embodiment, the method utilizes four time windows t1 to t4 to determine the expected and actual output values, each time window having a duration of 10 minutes. In another non-limiting example, three time windows t1 to t3 are used, each time window having a duration of 20 minutes. Many other variations are also possible.

[0073] At step 114, the processor 42 compares the power difference data points with thresholds that are also stored in the memory device 50.

[0074] In this embodiment, the threshold depends on the 90-minute average adjusted wind speed value WS. avg-adj,90 The memory device 50 of the controller 20 stores reference threshold data for the wind turbine 8, including data for different wind speeds. avg-adj,90A list of thresholds is provided. In this embodiment, different wind speed values ​​are grouped, and each group has a different relevant threshold. That is, multiple wind speed values ​​are grouped together into multiple groups, and each group has a different relevant threshold. Groups containing higher wind speeds have higher relevant thresholds compared to groups containing lower wind speed values. It should be noted that in some embodiments of the present invention, a single threshold can be used for all wind speed values ​​WS. avg-adj,90 This makes the threshold independent of wind speed.

[0075] Processor 42 determines whether the measured difference is less than or greater than a threshold.

[0076] If the power difference data point is determined to indicate that the expected output power exceeds the actual output power by an amount greater than a threshold, the method proceeds to step 116, which initiates the first alternative procedure of the method. In other words, if the power difference data point indicates that, based on a comparison with the threshold, the actual output power has been, on average, lower than the expected output power over a first predetermined time period (in this case, ninety minutes), then the method proceeds to step 116. If, based on a comparison with the threshold, the power difference data point is determined to indicate that the expected output power does not exceed the actual output power by an amount greater than the threshold—in other words, the actual output power is not, on average, lower than the expected output power—the method proceeds to step 118, which initiates the second alternative procedure of the method, which will be described later.

[0077] Starting from the first step of the method, at step 116, controller 20 receives data from ice sensor 28 indicating whether ice has accumulated on wind turbine 8, and determines whether the data indicates that ice accumulation on wind turbine 8 has ended. In this embodiment, ice sensor 28 sends data to controller 20 at ten-minute intervals. In other embodiments, data from ice sensor 28 may be sent to controller 20 at different discrete time intervals or continuously. When it is determined at step 116 that ice accumulation on wind turbine 8 has ended, controller 20 then measures the duration for which no ice accumulation has occurred on wind turbine 8. If, after a predetermined time period (in this case, sixty minutes) has elapsed, and it is determined that ice has begun to accumulate on wind turbine 8 again, then the method terminates. However, if processor 42 determines that no further ice accumulation has been detected within the predetermined time period of sixty minutes, then the method proceeds to step 120.

[0078] At step 120, controller 20 receives data from temperature sensors 34a and 34b indicating the current ambient temperature at wind turbine 8. Processor 42 determines whether the measured temperature is within a specified temperature range, or above or below a specified temperature. In this example, processor 42 determines whether the ambient temperature near wind turbine 8 is greater than or less than a specified threshold ambient temperature of approximately 0 degrees Celsius. In other examples, other suitable threshold ambient temperatures or ranges may be used. Because step 120 is a repetition of step 104, this step may be omitted if necessary.

[0079] If the processor 42 determines at step 120 that the ambient temperature is higher than the threshold temperature, then the method terminates, as shown in the figure. If the processor 42 determines at step 120 that the ambient temperature is lower than the threshold temperature, then the controller 20 sends a control signal to the de-icing system 24 at step 122 to activate the de-icing system 24.

[0080] Now proceeding to the second step of the method, at step 118, the controller 20 evaluates the connection status of the wind turbine. To this end, the controller 20 receives data from the current sensor 30a and the voltage sensor 30b at one-second intervals over a second predetermined time period, and uses these values ​​to calculate the corresponding output power for each one-second interval. This second predetermined time period is equal to 30 minutes in this embodiment. If the output power is determined to be substantially zero and persists for at least a predetermined percentage of the second predetermined time period, the method proceeds to steps 124, 126, and 128, which are identical to steps 116, 120, and 122 of the first step, respectively. In this example, the predetermined percentage is 20% of the second predetermined time period, thus equivalent to the controller 20 receiving at least 360 data points from the current sensor 30a and the voltage sensor 30b indicating that the output power from the wind turbine 8 is substantially zero.

[0081] If the output power is determined to be essentially zero and remains less than 20% of a predetermined percentage for a second predetermined time period, then the method terminates.

[0082] In an alternative embodiment of the method, at step 106, the controller receives data from the pressure sensor 36 indicating the air pressure near the wind turbine 8 within a single time or a single time window, and the processor 42 uses the data from the wind speed sensors 32a, 32b and the pressure sensor 36 to calculate an adjusted wind speed value. At step 108 of this alternative method, the adjusted wind speed value is used to determine the expected power generated by the wind turbine 8 by referring to a power curve stored in the memory device 50 of the controller 20.

[0083] In step 110 of this embodiment, the controller 20 receives data from the current sensor 30a and the voltage sensor 30b indicating the output power within a single time period or a single time window, and the processor 42 determines the actual power generated by the wind turbine 8 based on this data. Then, in step 112, the processor 42 determines the difference between the expected output power value and the actual output power value, and this information is stored in the memory device 50 as power difference data points. In this embodiment, the controller 20 receives the output power data at discrete time intervals of ten minutes, and steps 110 and 112 are repeated until nine power difference data points have been collected and stored in the memory device 50.

[0084] At step 114, processor 42 compares the power difference data, comprising nine power difference data points, with a threshold value that is dependent on the adjusted wind speed and also stored in memory device 50. For each data difference point, processor 42 determines whether the measured difference is less than or greater than the threshold.

[0085] If all power difference data points are determined to indicate that the expected output power exceeds the actual output power by an amount greater than a threshold, the method proceeds to step 116, which initiates the first alternative procedure of the method. If not all power difference data points are determined to indicate that the expected output power exceeds the actual output power by an amount greater than a threshold, the method proceeds to step 118, which initiates the second alternative procedure of the method.

[0086] As has already been noted, the described method can be executed at any suitable frequency.

[0087] In some embodiments, such as those in which the method is executed every ten minutes, the expected output power and the actual output power can be determined using a rolling time window. In this case, taking the described embodiment as an example, the first ninety-minute window may include time windows t1 to t9, the second ninety-minute time window may include time windows t2 to t10, and so on.

[0088] In other examples, the expected output power and the actual output power can be determined using back-to-back time windows. For example, during the first run of the method, the expected output power and the actual output power for a first 90-minute window can be determined. During the second run of the method, the expected output power and the actual output power for a second 90-minute window can be determined, where the second 90-minute window begins directly after the first 90-minute window has terminated. In this case, the first 90-minute window comprises time windows t1 to t9, and the second 90-minute window comprises time windows t10 to t19.

[0089] Using the average of the measured wind speed and the actual output power value in this method is advantageous because it prevents small fluctuations in these parameters from unnecessarily triggering the activation of the de-icing system.

[0090] The described method is advantageous because it optimizes the timing of activating the de-icing system 24. That is, the de-icing system 24 is activated only when it is determined that ice has stopped accumulating on the wind turbine 8, and especially when the ice has stopped accumulating for a certain period of time. This provides a more efficient way to solve the de-icing problem because energy is not wasted on de-icing during the period when the ice is still growing. Conversely, the present invention provides a method in which de-icing is only performed on the wind turbine 8 (especially from the wind turbine blades 18) after it has been determined that ice has stopped accumulating. Therefore, the de-icing process can be achieved more efficiently and effectively.

[0091] List of reference numerals in the attached diagram:

[0092] Wind turbine 8

[0093] Cabin 12

[0094] Tower 14

[0095] Main rotor assembly 16

[0096] 17-inch wheels

[0097] 18 wind turbine blades

[0098] Controller 20

[0099] Sensor 22

[0100] De-icing system 24

[0101] Output from ice sensor 26

[0102] Ice sensor 28

[0103] First output power sensor 30a

[0104] Second output power sensor 30b

[0105] First wind speed sensor 32a

[0106] Second wind speed sensor 32b

[0107] First temperature sensor 34a

[0108] Second temperature sensor 34b

[0109] Barometric pressure sensor 36

[0110] Controller input terminal 40

[0111] Processor 42

[0112] Output 44

[0113] Sensor output data 46

[0114] Control signal 47

[0115] Memory device 50

[0116] Temperature sensor output 54

[0117] Method steps 100, 102, 104, 106, 108, 110, 112, 114, 116, 118, 120, 122, 124, 126, 128, 130, 132

Claims

1. A method for activating a de-icing system of a wind turbine, the method comprising: Monitor the output power of the wind turbine; The output monitors the icing conditions of the atmosphere near the wind turbine and indicates the status of the icing. as well as The de-icing system is activated when the output power is lower than the expected output power and the output from the ice sensor indicates that the atmospheric icing condition has ended.

2. The method according to claim 1, wherein, The method includes monitoring the ambient temperature and activating the de-icing system only when the ambient temperature is within a predetermined range.

3. The method according to claim 1 or 2, wherein, The method includes activating the de-icing system only when the output of the ice sensor indicates that the atmospheric icing condition has ended and continues for a predetermined period of time.

4. The method according to claim 1 or 2, wherein, The method includes activating the de-icing system only under the following conditions: When the output power has been determined to be lower than the expected output power and continues for a first predetermined period of time, or When the output power has been substantially zero for a specified percentage of a second predetermined time period.

5. The method according to claim 4, wherein, The second predetermined time period is shorter than the first predetermined time period.

6. The method according to claim 1 or 2, wherein, Monitoring the output from the ice sensor includes receiving data from the ice sensor at a series of discrete time intervals.

7. The method according to claim 1 or 2, wherein, Monitoring the output power of the wind turbine includes receiving data indicating the output power from the wind turbine.

8. The method according to claim 7, wherein, Monitoring the output power of the wind turbine includes receiving primary power data indicating the output power from the wind turbine at a series of discrete time intervals within a first window.

9. The method according to claim 8, wherein, The method includes receiving primary power data from multiple consecutive first windows.

10. The method according to claim 9, wherein, The method includes calculating the average primary output power for each first window using the corresponding primary power data for each first window.

11. The method according to claim 10, wherein, The method includes determining the average actual power of the wind turbine by calculating the average value of the average primary output power.

12. The method according to claim 8, wherein, The primary power data is received by the controller at the end of the first window or each first window.

13. A controller for a de-icing system of a wind turbine, the controller comprising: An input terminal configured to receive data originating from an ice sensor located on or near the wind turbine and indicating atmospheric icing conditions near the wind turbine; An input terminal configured to receive data indicating the output power of the wind turbine; and The processor, which is configured to determine: Has the atmospheric icing ended? Whether the output power of the wind turbine is lower than the expected output power; The processor is configured to issue an output control signal to activate the de-icing system when it determines that the output power is lower than the expected output power and the atmospheric icing condition has ended.

14. A control system comprising an ice detector and a controller according to claim 13.

15. A wind turbine comprising the controller according to claim 13 or the control system according to claim 14.

Images