Blade stall identification method and device for wind turbine generator set

Abstract

The present invention provides a method and device for identifying blade stall in a wind turbine. The method comprises: monitoring relevant operating parameters of the wind turbine; analyzing the changing trends of the relevant operating parameters of the wind turbine over a period of time; identifying abnormal features related to blade stall from the changing trends; and determining that blade stall has occurred in the wind turbine in response to identifying the abnormal features related to blade stall. The device comprises: an operation monitoring unit configured to monitor relevant operating parameters of the wind turbine; an operation analysis unit configured to analyze the changing trends of the relevant operating parameters of the wind turbine over a period of time; an abnormality identification unit configured to identify abnormal features related to blade stall from the changing trends; and a stall determination unit configured to determine that blade stall has occurred in the wind turbine in response to identifying the abnormal features related to blade stall.

Description

Technical Field

[0001] The present application relates to the field of wind power generation technology, and in particular to a blade stall identification method and device for a wind turbine generator set. Background Art

[0002] In recent years, certain wind turbine models have experienced stalling under certain operating conditions due to environmental factors such as geographical conditions, daytime and seasonal variations. This can cause the actual power output of the wind turbine to fall far below the design value before reaching full power, leading to increased power generation losses. Furthermore, operating in a stalled state can alter the load and aerodynamic characteristics of the wind turbine blades, increasing overall vibration, leading to increased extreme loads and fatigue loads on turbine components, and even causing blade breakage, seriously impacting the safe and stable operation of the wind turbine.

[0003] In related technologies, blade stall conditions are usually identified and judged by estimating the blade angle of attack. This is because when the blade angle of attack is within a certain range (for example, 0 degrees to 12 degrees), the blade lift increases with the increase of the blade angle of attack. However, when the blade angle of attack is greater than a certain angle (the angle is related to the blade airfoil), the blade lift decreases sharply and the blade resistance increases, resulting in a significant reduction in the blade's absorption of wind energy. In this case, the blade can be judged to be stalled and appropriate protective measures can be taken for the wind turbine.

[0004] However, the estimation of blade angle of attack is easily affected by the accuracy of wind speed measurement. Therefore, when the wind speed measurement is inaccurate, the accuracy and reliability of blade stall identification will be seriously affected. Summary of the Invention

[0005] The object of the present invention is to provide a method and device for identifying blade stall of a wind turbine generator set.

[0006] According to one aspect of the present invention, a blade stall identification method for a wind turbine is provided, the blade stall identification method comprising: monitoring relevant operating parameters of the wind turbine; analyzing a changing trend of the relevant operating parameters of the wind turbine over a period of time; identifying abnormal features related to blade stall from the changing trends; and determining that blade stall has occurred in the wind turbine in response to identifying the abnormal features related to blade stall.

[0007] Preferably, the relevant operating parameters of the wind turbine generator set include at least one of the following operating parameters: the wind speed of the wind turbine generator set; the rotational speed of the wind turbine generator set; the acceleration of the wind turbine generator set; and the output power of the wind turbine generator set.

[0008] Preferably, the abnormal characteristics related to blade stall include at least one of the following abnormal characteristics: the speed signal of the wind turbine generator set oscillates at an abnormal frequency; abnormal acceleration shocks appear in the acceleration signal of the wind turbine generator set; the speed of the wind turbine generator set does not reach the expected speed matching the wind speed; and the output power of the wind turbine generator set does not reach the expected power matching the wind speed.

[0009] Preferably, the relevant operating parameters of the wind turbine generator set include multiple relevant operating parameters, wherein determining that blade stall occurs in the wind turbine generator set includes: evaluating the blade stall condition of the wind turbine generator set based on multiple abnormal characteristics related to blade stall identified from the change trends of the multiple relevant operating parameters over a period of time; and determining that blade stall occurs in the wind turbine generator set in response to the blade stall condition of the wind turbine generator set reaching a stall risk level.

[0010] Preferably, the evaluating the stall condition of the blades of the wind turbine generator set includes: determining a stall risk coefficient level for each abnormal feature; assigning a corresponding weight to each abnormal feature according to the impact of each abnormal feature on the stall risk; and determining the stall condition of the blades of the wind turbine generator set using the weight assigned to each abnormal feature and the stall risk coefficient level for each abnormal feature.

[0011] Preferably, the abnormal feature related to blade stall is determined based on historical data of relevant operating parameters of the wind turbine generator set, wherein the historical data are relevant operating parameters of the wind turbine generator set when blade stall previously occurred.

[0012] According to another aspect of the present invention, a blade stall identification device for a wind turbine is provided, the blade stall identification device comprising: an operation monitoring unit configured to monitor relevant operating parameters of the wind turbine; an operation analysis unit configured to analyze a change trend of the relevant operating parameters of the wind turbine over a period of time; an abnormality identification unit configured to identify abnormal features related to blade stall from the change trend; and a stall determination unit configured to determine that blade stall has occurred in the wind turbine in response to identifying the abnormal features related to blade stall.

[0013] Preferably, the relevant operating parameters of the wind turbine generator set include at least one of the following operating parameters: the wind speed of the wind turbine generator set; the rotational speed of the wind turbine generator set; the acceleration of the wind turbine generator set; and the output power of the wind turbine generator set.

[0014] Preferably, the abnormal characteristics related to blade stall include at least one of the following abnormal characteristics: the speed signal of the wind turbine generator set oscillates at an abnormal frequency; abnormal acceleration shocks appear in the acceleration signal of the wind turbine generator set; the speed of the wind turbine generator set does not reach the expected speed matching the wind speed; and the output power of the wind turbine generator set does not reach the expected power matching the wind speed.

[0015] Preferably, the relevant operating parameters of the wind turbine generator set include multiple relevant operating parameters, wherein the stall determination unit includes: a multivariate evaluation unit, configured to: evaluate the blade stall condition of the wind turbine generator set based on multiple abnormal characteristics related to blade stall identified from the change trends of the multiple relevant operating parameters over a period of time; and a stall evaluation unit, configured to: determine that the blade stall of the wind turbine generator set occurs in response to the blade stall condition of the wind turbine generator set reaching a stall risk level.

[0016] Preferably, the multivariate evaluation unit includes: a risk analysis unit configured to determine a stall risk coefficient level for each abnormal feature; a weight allocation unit configured to allocate a corresponding weight to each abnormal feature according to the impact of each abnormal feature on the stall risk; and a condition evaluation unit configured to determine the blade stall condition of the wind turbine using the weight allocated to each abnormal feature and the stall risk coefficient level for each abnormal feature.

[0017] Preferably, the abnormal feature related to blade stall is determined based on historical data of relevant operating parameters of the wind turbine generator set, wherein the historical data are relevant operating parameters of the wind turbine generator set when blade stall previously occurred.

[0018] According to another aspect of the present invention, a computer-readable storage medium storing a computer program is provided. When the computer program is executed by a processor, the blade stall identification method for a wind turbine generator set as described above is implemented.

[0019] According to another aspect of the present invention, a computer device is provided, comprising: a processor; and a memory storing a computer program. When the computer program is executed by the processor, the blade stall identification method for a wind turbine generator set as described above is implemented.

[0020] According to the exemplary embodiments of the present invention, the control method and device for a wind turbine generator set can not only effectively identify and judge the blade stall condition without adding new investment (such as additional hardware equipment), but also comprehensively evaluate the blade stall condition from multiple dimensions to avoid the problem of missed judgment or difficulty in judgment caused by using a single dimension to identify and judge blade stall, thereby further improving the accuracy of blade stall identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above objects and features of the present invention will become more apparent from the following description in conjunction with the accompanying drawings, in which:

[0022] Figure 1 A flow chart of a method for identifying blade stall of a wind turbine generator system according to an exemplary embodiment of the present invention is shown;

[0023] Figure 2 A schematic graph showing a normal rotation speed signal for a wind turbine generator system when the blades are not stalled according to an exemplary embodiment of the present invention is shown;

[0024] Figure 3 A schematic graph showing an abnormal rotation speed signal of a wind turbine generator system in a blade stall situation according to an exemplary embodiment of the present invention;

[0025] Figure 4 A schematic graph showing a normal acceleration signal for a wind turbine generator system when the blades are not stalled according to an exemplary embodiment of the present invention;

[0026] Figure 5 A schematic graph showing an abnormal acceleration signal for a wind turbine in a blade stall situation according to an exemplary embodiment of the present invention;

[0027] Figure 6 A schematic graph showing wind speed and rotation speed matching for a wind turbine generator system when blades are not stalled according to an exemplary embodiment of the present invention is shown;

[0028] Figure 7 A schematic graph showing a mismatch between wind speed and rotation speed for a wind turbine in a blade stall situation according to an exemplary embodiment of the present invention;

[0029] Figure 8 A schematic graph showing wind speed and power matching for a wind turbine generator system when blades are not stalled according to an exemplary embodiment of the present invention is shown;

[0030] Figure 9A schematic graph showing wind speed and power mismatch for a wind turbine generator system in a blade stall situation according to an exemplary embodiment of the present invention is shown;

[0031] Figure 10 shows a schematic process of a blade stall identification method for a wind turbine according to an exemplary embodiment of the present invention;

[0032] Figure 11 A structural block diagram of a blade stall identification device for a wind turbine generator set according to an exemplary embodiment of the present invention is shown;

[0033] Figure 12 A system architecture diagram for blade stall identification of a wind turbine generator system according to an exemplary embodiment of the present invention is shown;

[0034] Figure 13 Another structural block diagram of a blade stall identification device for a wind turbine generator system according to an exemplary embodiment of the present invention is shown; and

[0035] Figure 14 A schematic hardware configuration diagram of blade stall identification for a wind turbine generator system according to an exemplary embodiment of the present invention is shown. DETAILED DESCRIPTION

[0036] In the case of blade stall, in addition to changes in the blade attack angle of the wind turbine, the relevant operating parameters of the wind turbine will also show obvious abnormal changes. These abnormal changes can be manifested as, for example, but not limited to, the speed signal of the wind turbine oscillating at an abnormal frequency, abnormal acceleration shocks appearing in the acceleration signal of the wind turbine, the speed of the wind turbine failing to reach the expected speed matching the wind speed, and the output power of the wind turbine failing to reach the expected power matching the wind speed, etc.

[0037] The present invention utilizes the aforementioned abnormal changes in relevant operating parameters of a wind turbine generator set during blade stall to identify and determine whether blade stall has occurred. This eliminates the need for blade stall identification to be limited to estimating blade angle of attack. Furthermore, a comprehensive assessment of blade stall conditions from multiple dimensions is employed to avoid missed or difficult identifications that can occur when blade stall is identified and determined using a single dimension, further improving the accuracy of blade stall identification.

[0038] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0039] Figure 1 A flow chart 100 of a blade stall identification method for a wind turbine according to an exemplary embodiment of the present invention is shown.

[0040] Reference Figure 1 , Figure 1 The method shown may include the following steps:

[0041] In step 110 , relevant operating parameters of the wind turbine generator system may be monitored.

[0042] In an example, the relevant operating parameters of the wind turbine generator set may include, but are not limited to, one or a combination of the wind speed of the wind turbine generator set, the rotation speed of the wind turbine generator set, the acceleration of the wind turbine generator set, and the output power of the wind turbine generator set.

[0043] In step 120 , the changing trends of the relevant operating parameters of the wind turbine generator set over a period of time may be analyzed.

[0044] Blade stall can cause instability of the wind turbine generator set, and this instability can be manifested as the speed signal oscillating at an abnormal frequency. Therefore, in one example, it is possible to analyze whether the change trend of the speed signal of the wind turbine generator set over a period of time oscillates at an abnormal frequency. As a feasible implementation method, the speed spectrum information of the wind turbine generator set (i.e., the frequency value and amplitude at the point with the maximum frequency amplitude) can be calculated by fast Fourier transform, and the stall risk level can be divided for different modal frequencies. For example, when the frequency value in the speed spectrum information is greater than F and the amplitude in the speed spectrum information is less than a1, the stall risk coefficient level for oscillation at an abnormal frequency can be determined as 1; when a1≤the amplitude in the speed spectrum information is less than a2, the stall risk coefficient level for oscillation at an abnormal frequency can be determined as 2; when the amplitude in the speed spectrum information is ≥a3, the stall risk coefficient level for oscillation at an abnormal frequency can be determined as 3. In this embodiment, the selection of various thresholds including F, a1, a2 and a3 as mentioned above can be derived from a large amount of historical data analysis of the wind turbine generator set. In addition, they can also be set according to the model of the wind turbine generator set. For example, for a 2MW wind turbine generator set, F can be set to 1.35Hz.

[0045] Figure 2 FIG200 shows a schematic graph of a normal rotation speed signal of a wind turbine generator system according to an exemplary embodiment of the present invention when the blades are not stalled. Figure 3 A schematic graph 300 is shown of an abnormal rotation speed signal for a wind turbine in a blade stall situation according to an exemplary embodiment of the present invention.

[0046] Reference Figure 2 , Figure 2 The wind turbine speed signal shown in the figure is relatively stable over a period of time and does not oscillate at an abnormal frequency. This situation can be considered as no blade stall.

[0047] Reference Figure 3 , Figure 3 The wind turbine's speed signal is shown to oscillate at an abnormal frequency for a period of time. This indicates that the wind turbine is in an unstable state due to blade stall and can therefore be considered to have experienced blade stall.

[0048] In addition, the instability of a wind turbine generator set due to blade stall can also manifest as increased vibration of the set. This vibration can cause the acceleration signal of the wind turbine generator set to have abnormal acceleration shocks due to the increased load on the components of the set. Therefore, in another example, the change trend of the acceleration signal of the wind turbine generator set over a period of time can be analyzed to see whether abnormal acceleration shocks occur. As a feasible implementation method, the real-time value of the acceleration can be subtracted from the average value of the acceleration in the previous t seconds along the x-direction and / or y-direction and the absolute value can be taken. Then, the number of points s1 exceeding b1 and the number of points s2 exceeding b2 within t seconds can be counted in real time. When the number of points s1>B, the stall risk coefficient level for the occurrence of abnormal acceleration shocks can be determined as 1; when the number of points s2>B, the stall risk coefficient level for the occurrence of abnormal acceleration shocks can be determined as 2. In this embodiment, the selection of various thresholds including b1, b2 and B as mentioned above can be derived from a large amount of historical data analysis of the wind turbine generator set. In addition, they can also be set according to the model of the wind turbine generator set. For example, for a 2MW wind turbine generator set, t can be set to 30s, b1 can be set to 0.3g, and the set value b1 can be set to 0.45g.

[0049] Figure 4 FIG4 shows a schematic graph 400 of a normal acceleration signal for a wind turbine generator system according to an exemplary embodiment of the present invention when the blades are not stalled. Figure 5 A schematic graph 500 is shown of an abnormal acceleration signal for a wind turbine in a blade stall situation according to an exemplary embodiment of the present invention.

[0050] Reference Figure 4 , Figure 4 The acceleration signal of the wind turbine generator set shown in the figure changes regularly over a period of time, without any sudden and intensive signal changes. This can be considered as the absence of blade stall.

[0051] Reference Figure 5 , Figure 5 The right side of the wind turbine's acceleration signal shows abrupt and frequent abnormal acceleration shocks. This indicates that the wind turbine is in an unstable state due to blade stall and can be considered to have experienced blade stall.

[0052] In addition, instability of a wind turbine generator set due to blade stall can also manifest as the rotation speed of the wind turbine generator set failing to reach the desired rotation speed that matches the wind speed (i.e., a mismatch between the wind speed and the rotation speed). For example, the wind speed of the wind turbine generator set reaches a higher wind speed level (e.g., 20 m / s), but the rotation speed of the wind turbine generator set is lower than the rated speed. Therefore, in another example, the degree of matching between the rotation speed of the wind turbine generator set and the wind speed of the wind turbine generator set over a period of time can be analyzed. As a feasible implementation, when the wind speed of the wind turbine generator set is greater than w1 and the rotation speed of the wind turbine generator set is less than g1, the stall risk coefficient level for the mismatch between the wind speed and the rotation speed can be determined as 1; when the wind speed of the wind turbine generator set is greater than w2 and the rotation speed of the wind turbine generator set is less than g2, the stall risk coefficient level for the mismatch between the wind speed and the rotation speed can be determined as 2; when the wind speed of the wind turbine generator set is greater than w3 and the rotation speed of the wind turbine generator set is less than g3, the stall risk coefficient level for the mismatch between the wind speed and the rotation speed can be determined as 3. In this embodiment, the selection of various thresholds including w1, w2, w3, g1, g2 and g3 as mentioned above can be derived from a large amount of historical data analysis of the wind turbine generator set. In addition, they can also be set according to the model of the wind turbine generator set. For example, for a 2MW wind turbine generator set, w1 can be set to 8m / s, g1 can be set to 10rpm, w2 can be set to 10m / s, g2 can be set to 9rpm, w3 can be set to 15m / s, and g3 can be set to 8rpm.

[0053] Figure 6 600 is a schematic graph showing wind speed and rotation speed matching for a wind turbine generator system in a non-stalled blade state according to an exemplary embodiment of the present invention. Figure 7 A schematic graph 700 illustrating a mismatch between wind speed and rotational speed for a wind turbine in a blade stall situation according to an exemplary embodiment of the present invention is shown.

[0054] Reference Figure 6 , when the wind speed of the wind turbine generator reaches above the rated wind speed (such as Figure 6 7m / s or more as shown in the figure), the speed of the wind turbine generator set is kept above the rated speed (as shown in the figure). Figure 6 This situation indicates that the rotation speed of the wind turbine generator set reaches the desired rotation speed that matches the wind speed. Therefore, it can be regarded as the wind speed and rotation speed matching in the wind turbine generator set.

[0055] Reference Figure 7 , when the wind speed of the wind turbine generator reaches above the rated wind speed (such as Figure 7 10m / s or more as shown in the figure), the speed of the wind turbine generator set is always lower than the rated speed (as shown in the figure). Figure 7This situation indicates that the rotation speed of the wind turbine generator set has not reached the expected rotation speed that matches the wind speed. Therefore, it can be regarded as a mismatch between the wind speed and the rotation speed in the wind turbine generator set.

[0056] In addition, the instability of a wind turbine generator set due to blade stall can also be manifested as the output power of the wind turbine generator set failing to reach the expected power that matches the wind speed (i.e., the wind speed and power do not match). For example, the wind speed of the wind turbine generator set reaches a higher wind speed level (e.g., 20 m / s), but the output power of the wind turbine generator set is far lower than the full power of the wind turbine generator set. Therefore, in another example, the matching degree between the output power of the wind turbine generator set and the wind speed of the wind turbine generator set over a period of time can be analyzed. As a feasible implementation method, the matching relationship between the wind speed and power of the wind turbine generator set can be calculated in real time. For example, the theoretical power value p0(i) can be searched based on the current wind speed value, where i is the wind speed bin in which the current wind speed is located; and then the ratio p between the current actual power and the theoretical power p0(i) is calculated. When the ratio p>0.97, the stall risk coefficient level for wind speed and power mismatch can be determined as no risk; when 0.9<ratio p<0.97, the stall risk coefficient level for wind speed and power mismatch can be determined as 1; when 0.85<ratio p<0.9, the stall risk coefficient level for wind speed and power mismatch can be determined as 2; when the ratio p<0.85, the stall risk coefficient level for wind speed and power mismatch can be determined as 3. In this embodiment, the selection of various thresholds including 0.97 and 0.85 as mentioned above can be derived from a large amount of historical data analysis of the wind turbine generator set, and can also be set according to the model of the wind turbine generator set.

[0057] It should be understood that although the above describes an implementation method of using the ratio p between the actual power and the theoretical power p0(i) to determine the matching degree between wind speed and power, the present invention is not limited to this. For example, the difference between the actual power and the theoretical power p0(i) can also be used to determine the matching degree between wind speed and power, or other methods that can be used to indicate the matching degree between wind speed and power.

[0058] Figure 8 800 is a schematic graph showing a signal of wind speed and power matching for a wind turbine generator system when the blades are not stalled according to an exemplary embodiment of the present invention. Figure 9 A schematic graph 900 illustrates a wind speed to power mismatch signal for a wind turbine in a blade stall situation according to an exemplary embodiment of the present invention.

[0059] Reference Figure 8 , when the wind speed of the wind turbine reaches the full wind speed (such as Figure 810m / s), the output power of the wind turbine generator set remains above the rated power (such as Figure 8 This situation indicates that the output power of the wind turbine generator set reaches the expected power that matches the wind speed. Therefore, it can be regarded as the wind speed and power matching of the wind turbine generator set.

[0060] Reference Figure 9 , when the wind speed of the wind turbine generator reaches above the full wind speed (such as Figure 9 When the output power of the wind turbine generator set is much lower than the rated power (as shown in Figure 1), the output power of the wind turbine generator set is much lower than the rated power (as shown in Figure 1). Figure 9 This situation indicates that the output power of the wind turbine generator set does not reach the expected power matching the wind speed. Therefore, it can be regarded as the wind speed and power mismatch of the wind turbine generator set.

[0061] At step 130 , abnormal features related to blade stall may be identified from the variation trend.

[0062] According to the analysis in step 120, in the case of blade stall, it can be identified from the changing trends of relevant operating parameters of the wind turbine over a period of time, such as, but not limited to, one or a combination of the following abnormal characteristics related to blade stall: the speed signal of the wind turbine oscillates at an abnormal frequency, abnormal acceleration shocks appear in the acceleration signal of the wind turbine, the speed of the wind turbine does not reach the expected speed matching the wind speed, and the output power of the wind turbine does not reach the expected power matching the wind speed.

[0063] It should be understood that while the above descriptions of multiple abnormal characteristics associated with blade stall are merely exemplary, the present invention is not limited thereto. In other words, in addition to using one or a combination of the aforementioned abnormal characteristics associated with blade stall to reflect wind turbine instability information when blade stall occurs, other abnormal characteristics associated with blade stall may also be used to characterize wind turbine instability information.

[0064] At step 140 , it may be determined that blade stall has occurred in the wind turbine in response to identifying an abnormal characteristic associated with blade stall.

[0065] Because the instability information of a wind turbine blade stall can be manifested by a variety of different abnormal characteristics, determining whether a wind turbine blade stall has occurred from a single dimension may result in missed detection or difficulty in determining whether the blade stall has occurred. To further improve the accuracy of blade stall determination, a comprehensive assessment of whether a wind turbine blade stall has occurred from multiple dimensions may be considered. In some examples, when the relevant operating parameters of the wind turbine include multiple relevant operating parameters, the blade stall condition of the wind turbine can be assessed based on multiple abnormal characteristics associated with blade stall identified from the changing trends of the multiple relevant operating parameters over a period of time. In response to the blade stall condition of the wind turbine blade stall reaching a stall risk level, the occurrence of blade stall in the wind turbine blade stall is determined. As a feasible embodiment, a stall risk coefficient level can be determined for each abnormal characteristic, and a corresponding weight can be assigned to each abnormal characteristic based on its impact on the stall risk. The blade stall condition of the wind turbine blade stall can be determined using the weight assigned to each abnormal characteristic and the stall risk coefficient level for each abnormal characteristic. For example, the blade stall condition of the wind turbine generator set may be determined by, but not limited to, the following formula (1):

[0066] R=r1×L1+r2×L2+r3×L3+r4×L4 (1)

[0067] In formula (1), R is the risk value of blade stall in the wind turbine generator set, r1 is the weight of the impact of abnormal frequency oscillation on the stall risk, L1 is the stall risk coefficient level for abnormal frequency oscillation, r2 is the weight of the impact of abnormal acceleration shock on the stall risk, L2 is the stall risk coefficient level for abnormal acceleration shock, r3 is the weight of the impact of wind speed and rotation speed mismatch on the stall risk, L3 is the stall risk coefficient level for wind speed and rotation speed mismatch, r4 is the weight of the impact of wind speed and power mismatch on the stall risk, L4 is the stall risk coefficient level for wind speed and power mismatch. In this embodiment, the various weights including r1, r2, r3, and r4 can be determined based on an analysis of historical data of previous blade stalls in the wind turbine generator set. Optionally, r1 can be set to 0.35, r2 can be set to 0.15, r3 can be set to 0.25, and r4 can be set to 0.25. Various stall risk factor levels including L1, L2, L3 and L4 may be determined using the method described above.

[0068] In this example, when R exceeds the set risk threshold (the risk threshold can be set according to different models), it can be determined that the wind turbine blades have stalled, and then the wind turbine controller is controlled by the control program to implement unit protection measures, such as, but not limited to, shutdown, execution of early warning prompts, triggering pitch protection actions, and pushing operation and maintenance suggestions.

[0069] It should be understood that although the above describes an embodiment of comprehensively evaluating whether a wind turbine generator set has blade stall from multiple dimensions by adopting a weight combination approach, the present invention is not limited thereto.

[0070] Below, we will refer to Figure 10 The above-mentioned processing process for blade stall identification of a wind turbine generator set will be described in further detail.

[0071] Figure 10 A schematic process 1000 of a blade stall identification method for a wind turbine according to an exemplary embodiment of the present invention is shown.

[0072] Reference Figure 10 , process 1000 can be started.

[0073] In steps 1001 to 1004 , the process 1000 may respectively obtain one or a combination of the wind speed, rotation speed, acceleration, and output power of the wind turbine generator set.

[0074] In steps 1005 to 1008, process 1000 may use the aforementioned method to separately identify one or a combination of the following abnormal characteristics associated with blade stall: the speed signal of the wind turbine generator set oscillates at an abnormal frequency, abnormal acceleration shocks appear in the acceleration signal of the wind turbine generator set, the speed of the wind turbine generator set does not reach the expected speed matching the wind speed, and the output power of the wind turbine generator set does not reach the expected power matching the wind speed.

[0075] In step 1009, processing 1000 may adopt the method as described above to respectively determine the stall risk coefficient level for oscillation with abnormal frequency, the stall risk coefficient level for mismatch between wind speed and rotation speed, the stall risk coefficient level for mismatch between wind speed and power, and the stall risk coefficient level for acceleration shock, and on this basis adopt the weight combination method as described above to determine the blade stall condition (or total risk value) of the wind turbine generator set.

[0076] At step 1010 , if the blade stall condition reaches a stall risk level, process 1000 may output a stall identification code (which may typically be set to a Boolean value of 1), otherwise, process 1000 ends.

[0077] In step 1011, processing 1000 can control the wind turbine controller (such as, but not limited to, the main PLC system or pitch control system in the wind turbine) to perform appropriate unit protection measures according to the stall identification code output in step 1010, such as, but not limited to, shutting down, executing early warning prompts, triggering pitch protection actions, and pushing operation and maintenance suggestions.

[0078] After step 1011 , process 1000 may end.

[0079] Figure 11 A structural block diagram 1100 of a blade stall identification device for a wind turbine generator system according to an exemplary embodiment of the present invention is shown.

[0080] Reference Figure 11 , Figure 11 The blade stall identification device shown may include an operation monitoring unit 1110, an operation analysis unit 1120, an abnormality identification unit 1130 and a stall determination unit 1140, wherein the operation monitoring unit 1110 may be configured to monitor relevant operating parameters of the wind turbine generator set; the operation analysis unit 1120 may be configured to analyze the change trend of the relevant operating parameters of the wind turbine generator set over a period of time; the abnormality identification unit 1130 may be configured to identify abnormal features related to blade stall from the change trend; and the stall determination unit 1140 may be configured to determine that blade stall has occurred in the wind turbine generator set in response to identifying the abnormal features related to blade stall.

[0081] exist Figure 11 In the illustrated blade stall identification device, relevant operating parameters of the wind turbine generator set may include, but are not limited to, one or a combination of the wind turbine generator set's wind speed, wind turbine generator set's rotational speed, wind turbine generator set's acceleration, and wind turbine generator set's output power. As previously mentioned, abnormal characteristics associated with blade stall may include, but are not limited to, one or a combination of the wind turbine generator set's rotational speed signal oscillating at an abnormal frequency, abnormal acceleration shocks appearing in the wind turbine generator set's acceleration signal, the wind turbine generator set's rotational speed failing to reach a desired rotational speed that matches the wind speed, and the wind turbine generator set's output power failing to reach a desired power that matches the wind speed.

[0082] As previously mentioned, to further improve the accuracy of blade stall determination, a comprehensive assessment of whether a wind turbine blade stall has occurred may be considered from multiple dimensions. In one example, when the relevant operating parameters of the wind turbine include multiple relevant operating parameters, the stall determination unit 1140 may include a multivariate evaluation unit and a stall assessment unit (neither shown). The multivariate evaluation unit may be configured to assess the blade stall condition of the wind turbine blade based on multiple abnormal features associated with blade stall identified from the changing trends of the multiple relevant operating parameters over a period of time. The stall assessment unit may be configured to determine that the wind turbine blade stall has occurred in response to the blade stall condition of the wind turbine blade reaching a stall risk level.

[0083] As a feasible implementation, the multivariate evaluation unit may further include a risk analysis unit, a weight allocation unit and a condition assessment unit (all not shown), the risk analysis unit may be configured to determine the stall risk coefficient level for each abnormal feature; the weight allocation unit may be configured to assign a corresponding weight to each abnormal feature according to the impact of each abnormal feature on the stall risk; the risk assessment unit may be configured to use the weight assigned to each abnormal feature and the stall risk coefficient level for each abnormal feature to determine the blade stall condition of the wind turbine.

[0084] It should be understood that although the above describes an embodiment of comprehensively evaluating whether a wind turbine generator set has blade stall from multiple dimensions by adopting a weight combination approach, the present invention is not limited thereto.

[0085] Figure 12 A system architecture diagram 1200 for blade stall identification in a wind turbine according to an exemplary embodiment of the present invention is shown.

[0086] Reference Figure 12 , Figure 12 The system architecture shown may include a blade stall identification device 1210 for a wind turbine generator set according to an exemplary embodiment of the present invention, a wind turbine generator set 1220, and a wind turbine generator set controller 1230 (such as, but not limited to, a main control PLC system or a pitch control system in a wind turbine generator set). The blade stall identification method for a wind turbine generator set according to an exemplary embodiment of the present invention may be run as an algorithm on Figure 12 In the computing unit of the blade stall identification device 1210 shown. Figure 12 The blade stall identification device 1210 shown may include the operation monitoring unit 1110 , the operation analysis unit 1120 , the abnormality identification unit 1130 , and the stall determination unit 1140 as described above.

[0087] exist Figure 12 In the illustrated system architecture, wind turbine generator set 620 can transmit relevant operating parameters of the wind turbine generator set as signal A to blade stall identification device 1210. Blade stall identification device 1210 can determine whether blade stall has occurred in the wind turbine generator set based on these relevant operating parameters and transmit the stall identification result as signal B to wind turbine generator set controller 1230. Wind turbine generator set controller 1230 can output signal C to wind turbine generator set 1220 based on received signal B to control wind turbine generator set 1220 to implement unit protection measures, such as, but not limited to, shutdown, execution of early warning prompts, triggering pitch protection actions, and distributing operation and maintenance recommendations. Since the specific implementation process of blade stall identification has been described in detail above, it will not be repeated here.

[0088] It should be understood that although Figure 12 The system architecture for blade stall identification of a wind turbine generator system according to an exemplary embodiment of the present invention is shown, but the present invention is not limited thereto. For example, Figure 12 In addition to being integrated into a separate controller, the blade stall identification device 1210 shown can also be integrated into a wind turbine controller 1230 (such as, but not limited to, a main control PLC system or a pitch control system in a wind turbine) or a background controller for scheduling wind turbines in a wind farm or other control devices that can be connected to the wind turbine controller 1230 or the wind turbine 1220.

[0089] Figure 13 Another structural block diagram 1300 of a blade stall identification device for a wind turbine generator system according to an exemplary embodiment of the present invention is shown.

[0090] Reference Figure 13 , Figure 13 The blade stall identification device shown may include a data unit 1310, a processing unit 1320, a judgment unit 1330 and an early warning unit 1340, wherein the data unit 1310 may be configured to collect, clean, sort and extract data on relevant operating parameters of the wind turbine generator set under various operating conditions; the processing unit 1320 may be configured to perform operating mechanism analysis, feature extraction and feature processing on the data output by the data unit 1320 to obtain available feature quantities including abnormal features related to blade stall; the judgment unit 1330 may be configured to perform boundary conditions, judgment rules, condition optimization and threshold optimization processing on the available feature quantities output by the processing unit 1320 and perform abnormality judgment related to blade stall on the wind turbine generator set based on the available feature quantities obtained thereby; the early warning unit 1340 may be configured to control the wind turbine generator set controller to provide early warning indications, unit protection actions and predictive maintenance when the judgment unit 1330 outputs blade stall, so as to minimize blade stall.

[0091] exist Figure 13In the illustrated blade stall identification device, abnormal characteristics associated with blade stall can be determined based on historical data of relevant operating parameters of the wind turbine generator set (such as, but not limited to, wind speed, rotational speed, acceleration, and output power of the wind turbine generator set). This historical data can be relevant operating parameters of the wind turbine generator set when blade stall has previously occurred. As a feasible implementation, a machine learning and support vector machine (SVM) classification algorithm can be employed to train a model using normal rotational speed signals of the wind turbine generator set when the blades are not stalled as positive samples, and abnormal rotational speed signals of the wind turbine generator set when the blades are stalled as negative samples. When the accuracy of blade stall identification using these rotational speed-specific training samples reaches a high level, these rotational speed-specific training samples can be used to identify whether a wind turbine generator set has experienced blade stall.

[0092] Figure 14 A schematic hardware configuration diagram 1400 for blade stall identification of a wind turbine according to an exemplary embodiment of the present invention is shown.

[0093] Reference Figure 14 , Figure 14 The blade stall identification system shown may include a wind turbine generator set 1410, various data acquisition sensors 1420, and a wind turbine generator set main controller 1430, wherein each data acquisition sensor 1420 and the wind turbine generator set main controller 1430 may be disposed on the wind turbine generator set 1410. Each data acquisition sensor 1420 may be configured to transmit a plurality of relevant operating parameters of the wind turbine generator set collected in real time to the main controller 1430. The wind turbine main controller 1430 can be configured to perform the following operations: receive multiple relevant operating parameters (i.e., multiple variables); identify multiple abnormal characteristics related to blade stall from the multiple relevant operating parameters, calculate the stall risk coefficient level for each abnormal characteristic, and perform weighted combination based on this to obtain the blade stall condition of the wind turbine; identify the operating condition where the blade stall condition reaches the stall risk level as blade stall; output a stall identification code to give a processing instruction; control the wind turbine 1410 to implement unit protection measures according to the processing instruction, such as, but not limited to, shutdown, execution of early warning prompts, triggering pitch protection actions, and pushing operation and maintenance suggestions.

[0094] The blade stall identification method and device for a wind turbine generator set according to an exemplary embodiment of the present invention can not only effectively identify and judge the blade stall condition without adding new investment (such as, additional hardware equipment), but also comprehensively evaluate the blade stall condition from multiple dimensions to avoid the problem of missed judgment or difficulty in judgment caused by using a single dimension to identify and judge blade stall, thereby further improving the accuracy of blade stall identification.

[0095] According to an exemplary embodiment of the present invention, a computer-readable storage medium storing a computer program is also provided. The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform the blade stall identification method for a wind turbine according to the present invention. The computer-readable recording medium is any data storage device that can store data read by a computer system. Examples of computer-readable recording media include read-only memory, random access memory, read-only optical discs, magnetic tapes, floppy disks, optical data storage devices, and carrier waves (such as data storage via the Internet via a wired or wireless transmission path).

[0096] According to an exemplary embodiment of the present invention, a computer device is further provided. The computer device includes a processor and a memory. The memory is configured to store a computer program. The computer program is executed by the processor so that the processor performs the computer program for the blade stall identification method for a wind turbine according to the present invention.

[0097] While the present application has been shown and described with reference to preferred embodiments, it will be understood by those skilled in the art that various modifications and variations can be made to these embodiments without departing from the spirit and scope of the present application as defined by the appended claims.

Claims

1. A method for identifying blade stall in a wind turbine generator set, characterized in that: The blade stall identification method comprises: Monitoring relevant operating parameters of the wind turbine generator set, wherein the relevant operating parameters of the wind turbine generator set include multiple relevant operating parameters; Analyzing the changing trends of relevant operating parameters of the wind turbine generator set over a period of time; identifying abnormal features related to blade stall from the change trend; In response to identifying the abnormal feature associated with blade stall, determining that blade stall occurs in the wind turbine generator set, Wherein, determining that blade stall occurs in the wind turbine generator set includes: identifying a plurality of abnormal features related to blade stall from the change trends of the plurality of relevant operating parameters over a period of time; Determine the stall risk factor level for each abnormal characteristic; Assign a corresponding weight to each abnormal feature according to its impact on the stall risk; Determining a blade stall condition of the wind turbine generator set using a weight assigned to each abnormal feature and a stall risk coefficient level assigned to each abnormal feature; In response to a blade stall condition of the wind turbine generator set reaching a stall risk level, it is determined that blade stall of the wind turbine generator set occurs.

2. The blade stall identification method according to claim 1, characterized in that: The relevant operating parameters of the wind turbine generator set include at least one of the following operating parameters: The wind speed of the wind turbine generator set; The rotational speed of the wind turbine generator set; the acceleration of the wind turbine generator set; and The output power of the wind turbine generator set.

3. The blade stall identification method according to claim 2, characterized in that: The abnormal characteristics related to blade stall include at least one of the following abnormal characteristics: The speed signal of the wind turbine generator set oscillates at an abnormal frequency; An abnormal acceleration shock appears in the acceleration signal of the wind turbine generator set; The rotation speed of the wind turbine generator set does not reach the expected rotation speed matching the wind speed; as well as The output power of the wind turbine generator set does not reach the expected power matching the wind speed.

4. The blade stall identification method according to any one of claims 1 to 3, characterized in that: The abnormal characteristics related to blade stall are determined based on historical data of relevant operating parameters of the wind turbine generator set, wherein the historical data are relevant operating parameters when blade stall previously occurred in the wind turbine generator set.

5. A blade stall identification device for a wind turbine generator set, characterized in that: The blade stall identification device comprises: an operation monitoring unit configured to monitor relevant operation parameters of the wind turbine generator set, wherein the relevant operation parameters of the wind turbine generator set include multiple relevant operation parameters; an operation analysis unit configured to analyze a change trend of relevant operation parameters of the wind turbine generator set over a period of time; an abnormality identification unit configured to identify abnormal features related to blade stall from the change trend; The stall determination unit is configured to: determine that the wind turbine generator set has experienced blade stall in response to identifying the abnormal feature related to blade stall, Wherein, the stall determination unit is further configured to: identifying a plurality of abnormal features related to blade stall from the change trends of the plurality of relevant operating parameters over a period of time; Determine the stall risk factor level for each abnormal characteristic; Assign a corresponding weight to each abnormal feature according to its impact on the stall risk; Determining a blade stall condition of the wind turbine generator set using a weight assigned to each abnormal feature and a stall risk coefficient level assigned to each abnormal feature; In response to a blade stall condition of the wind turbine generator set reaching a stall risk level, it is determined that blade stall of the wind turbine generator set occurs.

6. The blade stall identification device according to claim 5, characterized in that: The relevant operating parameters of the wind turbine generator set include at least one of the following operating parameters: The wind speed of the wind turbine generator set; The rotational speed of the wind turbine generator set; the acceleration of the wind turbine generator set; and The output power of the wind turbine generator set.

7. The blade stall identification device according to claim 6, characterized in that: The abnormal characteristics related to blade stall include at least one of the following abnormal characteristics: The speed signal of the wind turbine generator set oscillates at an abnormal frequency; An abnormal acceleration shock appears in the acceleration signal of the wind turbine generator set; The rotation speed of the wind turbine generator set does not reach the expected rotation speed matching the wind speed; as well as The output power of the wind turbine generator set does not reach the expected power matching the wind speed.

8. The blade stall identification device according to any one of claims 5 to 7, characterized in that: The abnormal characteristics related to blade stall are determined based on historical data of relevant operating parameters of the wind turbine generator set, wherein the historical data are relevant operating parameters when blade stall previously occurred in the wind turbine generator set.

9. A computer-readable storage medium storing a computer program, wherein: When the computer program is executed by a processor, the blade stall identification method for a wind turbine generator set according to any one of claims 1 to 4 is implemented.

10. A computing device comprising: processor; A memory storing a computer program, which, when executed by a processor, implements the blade stall identification method for a wind turbine generator set according to any one of claims 1 to 4.

Images