A protection monitoring system and monitoring method for offshore wind turbine blades

Abstract

The present invention discloses a protection monitoring system and monitoring method for offshore wind turbine blades. The present invention relates to the field of wind turbine blade monitoring technology, and solves the problem of not performing abnormal monitoring on blades based on the change characteristics generated during their rotation. The present invention generates a speed change waveform within a monitoring period based on an abnormal signal, and combines the circumference characteristics of the wind turbine blades, the cooperation of the metal sheet and the micro switch, and can accurately lock the characteristic waveform segment of each group of blades. By analyzing and comparing the speed characteristics in the characteristic waveform segments of different blades, blades with large speed differences can be effectively identified, and accurate positioning of abnormal blades can be achieved, greatly improving the efficiency and accuracy of troubleshooting. In subsequent analysis, by confirming the consistency of the angle characteristics of the abnormal segment in different rotation processes, it is further determined whether the blade is truly abnormal, effectively reducing misjudgments caused by interference from other factors, and improving the reliability of the monitoring results.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine blade monitoring, and in particular to a protection monitoring system and a monitoring method for offshore wind turbine blades. Background Art

[0002] Offshore wind turbines are large-scale equipment that use ocean wind energy to generate electricity. Compared with land-based wind turbines, offshore wind turbines can obtain more stable and powerful wind energy resources and generate electricity more efficiently because they are installed at sea. However, they also face complex and harsh environmental challenges such as seawater corrosion and strong wind and huge waves.

[0003] The patent application (CN112780503B) discloses a method and system for monitoring damage of protective paint of fan blades based on audio signals, including a training phase and an online operation phase; the audio signal containing the wind sweeping aerodynamic noise of the fan is collected by an audio device over a period of time, and the sound signal is analyzed to achieve real-time monitoring and fault diagnosis of damage to the protective paint of the fan blades; the application can achieve the purpose of non-destructive testing while being able to monitor the health status of the blades in real time; at the same time, the application can give early warnings to blades that are undergoing deterioration and guide the time point for maintenance;

[0004] Combined with the above monitoring methods, the stable operation of offshore wind turbines urgently needs such precise protection monitoring methods. Although the original wind monitoring method can monitor blade damage, the wind characteristics associated with different wind conditions are different. In the actual monitoring and analysis process, there are too many parameters to be processed, and the detection purpose cannot be completed quickly within the fastest time period. When the corresponding blade is abnormal, its rotation characteristics will also change slightly. The blade is not monitored for abnormalities based on the changing characteristics generated during its rotation, and the timeliness of its monitoring needs to be improved. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a protection monitoring system and a monitoring method for offshore wind turbine blades, which solve the problem of not performing abnormal monitoring on the blades based on the changing characteristics generated during their rotation.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A protection monitoring method for offshore wind turbine blades, comprising the following steps:

[0007] Step 1: Monitor the shaft oscillation of the running fan. According to the displacement sensor set in the shaft, confirm the oscillation displacement generated by the fan during rotation. Then, based on the specific value of the oscillation displacement, evaluate the abnormal signal. The specific method is as follows:

[0008] The oscillation displacement detected at the corresponding moment is calibrated as W i, where i represents different moments, and then the shaft speed associated with the current moment is locked and calibrated as V i ;

[0009] The shaft speed is checked against a preset parameter table, wherein the parameter table includes different speed ranges and displacement standards associated with the corresponding speed ranges, wherein the speed ranges and displacement standards are preset values, the speed range to which the shaft speed belongs is locked, and the displacement standard of the speed range is determined;

[0010] If W i > displacement standard, an abnormal signal is generated. If W i ≤ displacement standard, no processing is performed;

[0011] Step 2: Based on the generated abnormal signal and the current speed of the fan, a set of monitoring periods is determined, and a speed change waveform within the monitoring period is generated. The circumference characteristics and selected characteristics are determined according to the length of the fan blades, and the characteristic waveform segments for each set of blades are locked;

[0012] The specific method of generating the speed change waveform within the monitoring period is:

[0013] The current fan speed is calibrated as V i , and based on the set radius R of the fan blade, lock the circumference Z of the fan blade after one rotation, where R is a preset parameter;

[0014] Using Z÷V i =T i Confirm feature duration T i , taking the current time as the calibration time, extend it to (10×T i ) as a set of monitoring cycles;

[0015] Performing real-time monitoring on different rotation speeds associated with different moments in this monitoring period, and generating a rotation speed change waveform for this monitoring period based on the rotation speed parameters monitored in real time;

[0016] The specific method of locking the characteristic waveform segment of each blade group from the speed change waveform is:

[0017] P1. Based on the metal sheet and micro switch set in the fan, confirm the time when the first group of micro signals is generated in the monitoring period and record it as the characteristic time;

[0018] P2. Based on the associated set radius R, a set of characteristic circles is generated, and based on the positions of the metal sheet and the shaft of each blade, the circumferential points of the metal sheet and each blade on the characteristic circles are confirmed;

[0019] P3. Based on the location of the metal sheet circumference point, confirm the next group of blade circumference points and record them as pending points. The blade rotates in a clockwise direction. The circumference length of the interval between the metal sheet circumference point and the pending point is locked and recorded as L1. Based on the confirmed L1, confirm a group of selected circumference segments. The selected circumference of the selected circumference segments is [L1+X1Z, L1+X2Z], where X1 takes one third and X2 takes two thirds.

[0020] P4. In the speed change waveform, lock the waveform point associated with the characteristic moment and record it as the zero point, and determine the circumference from the zero point in sequence: calibrate the speed associated with several waveform points that appear successively after the zero point as {B1, B2, ..., Bn}, and confirm the walking circumference from the value B1 in sequence, and the walking circumference = (B1+B2+...Bj) × unit time, where the unit time is the preset time. When the walking circumference meets the following conditions for the first time: walking circumference ∈ [L1+X1Z, L1+X2Z], lock The specific moment corresponding to Bj is recorded as the initial moment, and then the walking circumference is confirmed continuously. The walking circumference = (B1+B2+…Bj+…+Bq)×unit time. After the walking circumference∈[L1+X1Z, L1+X2Z], when the walking circumference∉[L1+X1Z, L1+X2Z] is satisfied for the first time, the specific moment corresponding to Bq is locked and recorded as the final moment. The part of the waveform segment associated with the initial moment and the final moment is recorded as the characteristic waveform segment of the next group of leaves, where j and q both belong to [1, n].

[0021] After the characteristic waveform end of the last group of blades is confirmed, the interval circumference lengths L2 and L3 associated with the last two groups of blades and the last three groups of blades are confirmed in step P3, and the selected circumferences of the selected circular segments are confirmed to be [L2+X1Z, L2+X2Z] and [L3+X1Z, L3+X2Z] based on L2 and L3, and then the same processing method as step P4 is adopted to lock the characteristic waveform segments associated with the last two groups of blades and the last three groups of blades;

[0022] Step 3: Based on the different characteristic waveform segments associated with different blades, and according to the different speed characteristics in the different characteristic waveform segments, identify whether there are speed characteristics with large differences, and calibrate the abnormal blades to be determined. The specific method is as follows:

[0023] Based on different characteristic waveform segments associated with different blades, the rotation speeds associated with a number of fluctuation points in the corresponding characteristic waveform segments are averaged to confirm the pending average value;

[0024] Then, based on the specific parameter C of the undetermined mean, identify whether V k Satisfy: |V k-C|≥Y1, if it exists, the corresponding fluctuation point is recorded as an abnormal point, if it does not exist, no calibration is performed, where Y1 is the preset speed fluctuation value, k represents the different moments in the characteristic waveform segment, V k Represents the speed associated with the fluctuation point at the corresponding moment, and marks the continuous segment associated with the abnormal point in the corresponding characteristic waveform segment as the abnormal segment;

[0025] If the line length ratio of the abnormal segment exceeds 10%, the blade associated with this characteristic waveform segment is marked as a pending abnormal blade, and its line length ratio = the total line length of the abnormal segment ÷ the total line length of the characteristic waveform segment;

[0026] If there is only one group of abnormal leaves to be determined, execute step 4 for reanalysis;

[0027] If there are two or three groups of abnormal blades to be determined, a maintenance signal will be directly generated and displayed;

[0028] Step 4: Based on the calibrated abnormal blade to be determined and the abnormal segment position determined in the characteristic waveform segment, confirm the angle feature of the corresponding abnormal segment, and then identify whether the angle feature of the abnormal segment is consistent from the subsequent rotation process. Based on the identification result, confirm whether the abnormal blade to be determined is truly abnormal. The specific method is as follows:

[0029] Based on the abnormal segment determined in the characteristic waveform segment, based on the time before and after the abnormal segment, the circumference segment associated with the abnormal segment relative to the circumference point of the metal sheet is locked, and the position of the circumference point of the metal sheet is taken as the 0° point to confirm the angle interval associated with its circumference segment;

[0030] Confirm the second characteristic waveform segment associated with the abnormal blade to be determined in the next rotation process in the speed change waveform, confirm the abnormal segment of the second characteristic waveform segment based on the same processing method as determining the abnormal segment from the characteristic waveform segment in step 3, and then confirm the angle interval associated with the abnormal segment;

[0031] If the angle intervals associated with the corresponding abnormal segments in the first characteristic waveform segment and the second characteristic waveform segment are completely consistent, an impeller rotation abnormality signal is generated. If they are not completely consistent, an abnormal blade signal is generated and the abnormal blade is marked for external personnel to inspect and check.

[0032] Preferably, a protection monitoring system for offshore wind turbine blades comprises:

[0033] The abnormal monitoring end monitors the shaft oscillation of the running fan. According to the displacement sensor set in the shaft, the oscillation displacement generated by the fan during rotation is confirmed, and then the abnormal signal is evaluated based on the specific value of the oscillation displacement.

[0034] The band feature extraction end determines a set of monitoring periods based on the generated abnormal signal and the current speed of the fan, and generates a speed change waveform within the monitoring period, determines the circumference feature and selects the feature according to the length of the fan blades, and locks the characteristic waveform segment for each set of blades;

[0035] The calibration end of the abnormal blade to be determined, based on the different characteristic waveform segments associated with different blades and the different rotation speed characteristics in the different characteristic waveform segments, identifies whether there are rotation speed characteristics with large differences, and calibrates the abnormal blade to be determined;

[0036] The real abnormality determination end confirms the angle characteristics of the corresponding abnormal segment based on the calibrated abnormal blade to be determined and the abnormal segment position determined in the characteristic waveform segment, and then identifies whether the angle characteristics of the abnormal segment are consistent from the subsequent rotation process. Based on the identification result, it is confirmed whether the abnormal blade to be determined is a real abnormality.

[0037] The present invention provides a protection monitoring system and monitoring method for offshore wind turbine blades. Compared with the prior art, the system has the following beneficial effects:

[0038] The present invention generates a speed change waveform within the monitoring period based on the abnormal signal, and combines the circumference characteristics of the fan blades, the cooperation of the metal sheet and the micro switch to accurately lock the characteristic waveform segment of each group of blades. By analyzing and comparing the speed characteristics within the characteristic waveform segments of different blades, blades with large speed differences can be effectively identified, and the abnormal blades can be accurately located, greatly improving the efficiency and accuracy of troubleshooting;

[0039] During the abnormal blade calibration process, a multi-step judgment method such as mean processing, speed fluctuation value comparison and abnormal segment line length ratio is adopted to carefully calibrate the abnormal blades to be determined; at the same time, in the subsequent analysis, by confirming the consistency of the abnormal segment angle characteristics in different rotation processes, it is further judged whether the blade is truly abnormal, which effectively reduces the misjudgment caused by interference from other factors and improves the reliability of the monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the process of the present invention;

[0041] Figure 2 It is a characteristic schematic diagram of the offshore wind turbine blade of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] First embodiment

[0044] See also Figure 1 The present application provides a protection monitoring method for offshore wind turbine blades, comprising the following steps:

[0045] Step 1: Monitor the shaft oscillation of the running fan. According to the displacement sensor set in the shaft (usually a laser displacement sensor), confirm the oscillation displacement generated by the fan during rotation. Then, based on the specific value of the oscillation displacement, evaluate the abnormal signal. The oscillation displacement standards generated for different operating speeds are different. When the speed is 1r / min, the reference oscillation displacement is generally 10μm, and when the speed is 3r / min, the reference oscillation displacement is generally 30μm. Both are preset parameter values. Determine the oscillation displacement during the oscillation process. The specific method of evaluation is as follows:

[0046] The oscillation displacement detected at the corresponding moment is calibrated as W i (The laser displacement sensor determines the corresponding displacement distance based on the round-trip time of the emitted laser), where i represents different moments, and then the shaft speed associated with the current moment is locked and calibrated as V i ;

[0047] The shaft speed is checked against a preset parameter table, wherein the parameter table includes different speed ranges and displacement standards associated with the corresponding speed ranges, wherein the speed ranges and displacement standards are preset values, the speed range to which the shaft speed belongs is locked, and the displacement standard of the speed range is determined;

[0048] If W i ≤ displacement standard, no processing is performed. If W i > displacement standard, an abnormal signal is generated;

[0049] Specifically, its parameter table is generally: the oscillation displacement associated with 0r / min-1r / min is 10μm, the oscillation displacement associated with 1r / min-2r / min is 20μm, and so on. There is a maximum limit for its oscillation displacement, which is 100μm. The oscillation displacement in other subsequent speed ranges must not exceed 100μm. Under normal circumstances, the fan blades will not rotate too fast. A speed limiter is set inside. No matter how strong the external wind force is, the fan blades can be limited to rotate according to the limited speed.

[0050] Step 2: Based on the generated abnormal signal and the current speed of the fan, a set of monitoring periods is determined, and a speed change waveform within the monitoring period is generated. The circumference characteristics and selected characteristics are determined according to the length of the fan blades, and the characteristic waveform segments of each group of blades are locked. Specifically, when there is a slight problem with a certain blade, the corresponding blade will have specific differences from other blades in the associated speed stage. Then, based on the specific characteristics of such speed differences, the abnormal blade can be locked, and when the corresponding fan completes a circle of rotation, it is generally monitored by the metal sheet set at the end of the shaft. When the metal sheet touches the corresponding micro switch, the micro switch will generate an electrical signal change, and this signal change represents that the fan has completed a circle of rotation; by counting the electrical signal, the number of circles of the fan rotation can be counted to evaluate the corresponding fan's completion of a circle of rotation process. Such settings exist in each fan, all for monitoring the specific speed of the fan blades;

[0051] First, the specific method of generating the speed change waveform within the monitoring period is:

[0052] The current fan speed is calibrated as V i , and based on the set radius R of the fan blade, lock the circumference Z of the fan blade after one rotation, where R is a preset parameter, and Z=2πR;

[0053] Using Z÷V i =T i Confirm feature duration T i (That is, the time required to rotate one circle at the current rotation speed), take the current time as the calibration time, and extend it backward (10×T i ) as a set of monitoring cycles, "10" here is the set value, which can be modified by relevant operators, or other values ​​can be selected. Generally, 5-10 is more appropriate;

[0054] The different speeds associated with different moments in this monitoring period are monitored in real time, and based on the speed parameters monitored in real time, a speed change waveform for this monitoring period is generated, wherein the horizontal coordinate axis of the coordinate system of this waveform is the timeline, and the vertical coordinate axis is the speed value;

[0055] Secondly, the specific method of locking the characteristic waveform segment of each group of blades from the speed change waveform is:

[0056] P1. Based on the metal sheet and micro switch set in the fan, confirm the time when the first group of micro signals is generated in the monitoring period and record it as the characteristic time;

[0057] P2. Based on the associated set radius R, a set of characteristic circles is generated, and based on the positions of the metal sheet and the shaft of each blade, the circumferential points of the metal sheet and each blade on the characteristic circles are confirmed;

[0058] P3. Since the fan is in a clockwise rotation mode, based on the location of the metal sheet circumference point, confirm the latter group of blade circumference points and record them as pending points (the latter here is relative to the clockwise direction, and the clockwise rotation direction is the front), lock the circumference length of the interval between the metal sheet circumference point and the pending point and record it as L1, and confirm a group of selected circumference segments based on the confirmed L1. The selected circumference of the selected circumference segment is [L1+X1Z, L1+X2Z], where X1 and X2 are both preset values, and X1 takes one-third, and X2 takes two-thirds. Specifically, the metal sheet is located at the end of the shaft. When it is fitted to the characteristic circle, it is the vertex of the characteristic circle. In order to identify the most characteristic speed change stage of the corresponding blade, that is, starting from the vertex of the characteristic circle in the circumference area of ​​120°-240°, the associated circumference is the specific area of ​​one-third to two-thirds of the circumference. In order to find the characteristic waveform segment of the corresponding blade, it can be searched based on this method;

[0059] P4. In the speed change waveform, lock the waveform point associated with the characteristic moment and record it as zero point, and determine the circumference from zero point in sequence: calibrate the speed associated with several waveform points that appear successively after the zero point as {B1, B2, ..., Bn}, and confirm the walking circumference from the value B1 in sequence, and the walking circumference = (B1+B2+...Bj) × unit time, where the unit time is the preset time, generally taking 1s. When the walking circumference meets the following conditions for the first time: walking circumference ∈ [L1+X1Z, L1+X2Z], lock Determine the specific time corresponding to Bj and record it as the initial time, and then continue to confirm the walking circumference. The walking circumference = (B1+B2+…Bj+…+Bq)×unit time. When the walking circumference satisfies the walking circumference ∉[L1+X1Z, L1+X2Z] for the first time (the first time not belonging here is based on the premise that the first time belongs), lock the specific time corresponding to Bq and record it as the final time. The part of the waveform segment associated with the initial time and the final time period is recorded as the characteristic waveform segment of the next group of leaves, where j and q both belong to [1, n].

[0060] After the characteristic waveform end of the last group of blades is confirmed, the interval circumference lengths L2 and L3 associated with the last two groups of blades and the last three groups of blades are confirmed in step P3, and the selected circumferences of the selected circular segments are confirmed to be [L2+X1Z, L2+X2Z] and [L3+X1Z, L3+X2Z] based on L2 and L3, and then the same processing method as step P4 is adopted to lock the characteristic waveform segments associated with the last two groups of blades and the last three groups of blades;

[0061] Specifically, Figure 2 As shown in the figure, a set of characteristic circles is constructed according to the radius of the corresponding blade. Then, based on the location of the metal sheet and the location of the corresponding blade, the circumferential points on the characteristic circle can be confirmed. The rotation direction is clockwise. According to the difference in circumference between the circumferential points, the selected circumference associated with the corresponding blade after passing through the circumferential point of the metal sheet can be confirmed. The circumferential area associated with the selected circumference is Figure 2 In the filling area in the figure, in the process of such area change, the speed characteristics associated with the corresponding blades are more obvious, so based on the corresponding perimeter characteristics and the location of the metal sheet, the waveform segment characteristics associated with different blades can be selected in turn, so as to select and confirm the characteristic waveform segment;

[0062] Step 3: Based on the different characteristic waveform segments associated with different blades, and according to the different speed characteristics in the different characteristic waveform segments, identify whether there are large differences in speed characteristics, and calibrate the pending abnormal blades (the reason for calibrating the pending abnormal blades here is to assess errors in the subsequent analysis process. Since such abnormal conditions are likely to be caused by aging of lubricating parts or other abnormalities, abnormal assessment cannot be performed directly here, and only relevant assessments of pending abnormalities can be performed). The specific method for calibrating the pending abnormal blades is as follows:

[0063] Based on different characteristic waveform segments associated with different blades, the rotation speeds associated with a number of fluctuation points in the corresponding characteristic waveform segments are averaged to confirm the pending average value;

[0064] Then, based on the specific parameter C of the undetermined mean, identify whether V k Satisfy: |V k -C|≥Y1, if it exists, the corresponding fluctuation point is recorded as an abnormal point, if it does not exist, no calibration is performed, where Y1 is the preset speed fluctuation value, k represents the different moments in the characteristic waveform segment, V k Represents the speed associated with the fluctuation point at the corresponding moment, and marks the continuous segment associated with the abnormal point in the corresponding characteristic waveform segment as the abnormal segment;

[0065] If the line length ratio of the abnormal segment exceeds 10%, the blade associated with this characteristic waveform segment is marked as a pending abnormal blade, and its line length ratio = the total line length of the abnormal segment ÷ the total line length of the characteristic waveform segment;

[0066] If there is only one group of abnormal leaves to be determined, execute step 4 for reanalysis;

[0067] If there are two or three groups of abnormal blades to be determined, a maintenance signal will be directly generated and displayed. Based on the maintenance signal, the external relevant personnel need to check and repair the fan blades and impeller, assess the specific abnormal cause and perform maintenance.

[0068] Step 4: Based on the calibrated abnormal blade to be determined and the abnormal segment position determined in the characteristic waveform segment, confirm the angle feature of the corresponding abnormal segment, and then identify whether the angle features of the abnormal segment are consistent from the subsequent rotation process. Based on the identification result, confirm whether the abnormal blade to be determined is truly abnormal. The confirmation method is:

[0069] Based on the abnormal segment determined in the characteristic waveform segment, based on the time before and after the abnormal segment, the circumference segment associated with the abnormal segment relative to the circumference point of the metal sheet is locked, and the position of the circumference point of the metal sheet is taken as the 0° point to confirm the angle interval associated with its circumference segment;

[0070] Confirm the second characteristic waveform segment associated with the abnormal blade to be determined in the next rotation process in the speed change waveform (that is, the waveform segment associated after rotating one circle again, that is, adding a circumference Z to the original walking circumference. When the walking circumference associated with the corresponding blade is [L3+X1Z, L3+X2Z], then the next walking circumference is [L3+X1Z+Z, L3+X2Z+Z]. The waveform segment is confirmed based on the position of the original determined metal sheet circumference point, and the second characteristic waveform segment belonging to the same blade can be locked). Based on the same processing method as determining the abnormal segment from the characteristic waveform segment in step 3, confirm the abnormal segment of the second characteristic waveform segment;

[0071] Then confirm the angle interval associated with this abnormal segment;

[0072] If the angle intervals associated with the corresponding abnormal segments in the first characteristic waveform segment and the second characteristic waveform segment are exactly the same, an abnormal impeller rotation signal is generated. There may be aging, wear or other conditions that cause problems at the same point in the rotation process. If they are not exactly the same, an abnormal blade signal is generated and the abnormal blade is marked for external personnel to inspect and check.

[0073] Second embodiment

[0074] A protection monitoring system for offshore wind turbine blades, comprising:

[0075] The abnormal monitoring end monitors the shaft oscillation of the running fan. According to the displacement sensor set in the shaft, the oscillation displacement generated by the fan during rotation is confirmed, and then the abnormal signal is evaluated based on the specific value of the oscillation displacement.

[0076] The band feature extraction end determines a set of monitoring periods based on the generated abnormal signal and the current speed of the fan, and generates a speed change waveform within the monitoring period, determines the circumference feature and selects the feature according to the length of the fan blades, and locks the characteristic waveform segment for each set of blades;

[0077] The calibration end of the abnormal blade to be determined, based on the different characteristic waveform segments associated with different blades and the different rotation speed characteristics in the different characteristic waveform segments, identifies whether there are rotation speed characteristics with large differences, and calibrates the abnormal blade to be determined;

[0078] The real abnormality determination end confirms the angle characteristics of the corresponding abnormal segment based on the calibrated abnormal blade to be determined and the abnormal segment position determined in the characteristic waveform segment, and then identifies whether the angle characteristics of the abnormal segment are consistent from the subsequent rotation process. Based on the identification result, it is confirmed whether the abnormal blade to be determined is a real abnormality.

[0079] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0080] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A protection and monitoring method for offshore wind turbine blades, characterized in that: The following steps are involved: Step 1: Monitor the shaft oscillation of the running fan. According to the displacement sensor set in the shaft, confirm the oscillation displacement generated by the fan during rotation. Then, based on the specific value of the oscillation displacement, evaluate the abnormal signal. The specific method is as follows: The oscillation displacement detected at the corresponding moment is calibrated as W i , where i represents different moments, and then the shaft speed associated with the current moment is locked and calibrated as V i ; The shaft speed is checked against a preset parameter table, wherein the parameter table includes different speed ranges and displacement standards associated with the corresponding speed ranges, wherein the speed ranges and displacement standards are preset values, the speed range to which the shaft speed belongs is locked, and the displacement standard of the speed range is determined; If W i > displacement standard, an abnormal signal is generated. If W i ≤ displacement standard, no processing is performed; Step 2: Based on the abnormal signal generated and the current speed of the fan, determine a set of monitoring periods, generate a speed change waveform within the monitoring period, determine the circumference characteristics and select the characteristics according to the length of the fan blades, and lock the characteristic waveform segment of each set of blades. The specific method is as follows: The shaft speed associated with the current moment is calibrated as V i , and based on the set radius R of the fan blade, lock the circumference Z of the fan blade after one rotation, where R is a preset parameter; Using Z÷V i =T i Confirm feature duration T i , taking the current time as the calibration time, extend it to (10×T i ) as a set of monitoring cycles; Performing real-time monitoring on different rotation speeds associated with different moments in this monitoring period, and generating a rotation speed change waveform for this monitoring period based on the rotation speed parameters monitored in real time; Step 3: Based on the different characteristic waveform segments associated with different blades, and according to the different rotation speed characteristics in the different characteristic waveform segments, identify whether there are rotation speed characteristics with large differences, and calibrate the abnormal blades to be determined; Step 4: Based on the calibrated abnormal blade to be determined and the abnormal segment position determined in the characteristic waveform segment, confirm the angular characteristics of the corresponding abnormal segment, and then identify whether the angular characteristics of the abnormal segment are consistent during the subsequent rotation process. Based on the identification result, confirm whether the abnormal blade to be determined is truly abnormal.

2. A protection and monitoring method for offshore wind turbine blades according to claim 1, characterized in that: In step 2, the specific method of locking the characteristic waveform segment of each group of blades from the speed change waveform is: P1. Based on the metal sheet and micro switch set in the fan, confirm the time when the first group of micro signals is generated in the monitoring period and record it as the characteristic time; P2. Based on the associated set radius R, a set of characteristic circles is generated, and based on the positions of the metal sheet and the shaft of each blade, the circumferential points of the metal sheet and each blade on the characteristic circles are confirmed; P3. Based on the location of the metal sheet circumference point, confirm the next group of blade circumference points and record them as pending points. The blade rotates in a clockwise direction. The circumference length of the interval between the metal sheet circumference point and the pending point is locked and recorded as L1. Based on the confirmed L1, confirm a group of selected circumference segments. The selected circumference of the selected circumference segments is [L1+X1Z, L1+X2Z], where X1 and X2 are both preset values, and X1 takes one third and X2 takes two thirds. P4. In the speed change waveform, lock the waveform point associated with the characteristic moment and record it as the zero point, and determine the circumference from the zero point in sequence: calibrate the speed associated with several waveform points that appear successively after the zero point as {B1, B2, ..., Bn}, and confirm the walking circumference from the value B1 in sequence, and the walking circumference = (B1+B2+...Bj) × unit time, where the unit time is the preset time. When the walking circumference meets the following conditions for the first time: walking circumference ∈ [L1+X1Z, L1+X2Z], lock The specific moment corresponding to Bj is recorded as the initial moment, and then the walking circumference is confirmed continuously. The walking circumference = (B1+B2+…Bj+…+Bq)×unit time. After the walking circumference∈[L1+X1Z, L1+X2Z], when the walking circumference∉[L1+X1Z, L1+X2Z] is satisfied for the first time, the specific moment corresponding to Bq is locked and recorded as the final moment. The part of the waveform segment associated with the initial moment and the final moment is recorded as the characteristic waveform segment of the next group of leaves, where j and q both belong to [1, n]. P5. After the characteristic waveform end of the last group of blades is confirmed, the interval circumference lengths L2 and L3 associated with the last two groups of blades and the last three groups of blades are confirmed in step P3, and based on L2 and L3, the selected circumference of the selected circular segment is confirmed to be [L2+X1Z, L2+X2Z] and [L3+X1Z, L3+X2Z], and then the same processing method as step P4 is adopted to lock the characteristic waveform segments associated with the last two groups of blades and the last three groups of blades.

3. The protection monitoring method for offshore wind turbine blades according to claim 1 is characterized in that: In step 3, the specific method of calibrating the abnormal blade to be determined is: Based on different characteristic waveform segments associated with different blades, the rotation speeds associated with a number of fluctuation points in the corresponding characteristic waveform segments are averaged to confirm the pending average value; Then, based on the specific parameter C of the undetermined mean, identify whether V k Satisfy: |V k -C|≥Y1, if it exists, the corresponding fluctuation point is recorded as an abnormal point, if it does not exist, no calibration is performed, where Y1 is the preset speed fluctuation value, k represents the different moments in the characteristic waveform segment, V k Represents the speed associated with the fluctuation point at the corresponding moment, and marks the continuous segment associated with the abnormal point in the corresponding characteristic waveform segment as the abnormal segment; If the line length ratio of the abnormal segment exceeds 10%, the blade associated with this characteristic waveform segment is marked as a pending abnormal blade, and its line length ratio = the total line length of the abnormal segment ÷ the total line length of the characteristic waveform segment; If there is only one group of abnormal leaves to be determined, execute step 4 for reanalysis; If there are two or three groups of abnormal blades to be determined, a maintenance signal will be directly generated and displayed.

4. The protection monitoring method for offshore wind turbine blades according to claim 1 is characterized in that: In step 4, the specific method of confirming whether the abnormal blade to be determined is truly abnormal is: Based on the abnormal segment determined in the characteristic waveform segment, based on the time before and after the abnormal segment, the circumference segment associated with the abnormal segment relative to the circumference point of the metal sheet is locked, and the position of the circumference point of the metal sheet is taken as the 0° point to confirm the angle interval associated with its circumference segment; Confirm the second characteristic waveform segment associated with the abnormal blade to be determined in the next rotation process in the speed change waveform, confirm the abnormal segment of the second characteristic waveform segment based on the same processing method as determining the abnormal segment from the characteristic waveform segment in step 3, and then confirm the angle interval associated with the abnormal segment; If the angle intervals associated with the corresponding abnormal segments in the first characteristic waveform segment and the second characteristic waveform segment are completely consistent, an impeller rotation abnormality signal is generated. If they are not completely consistent, an abnormal blade signal is generated and the abnormal blade is marked for external personnel to inspect and check.

5. A protection monitoring system for offshore wind turbine blades, the monitoring system being used to execute the protection monitoring method for offshore wind turbine blades according to any one of claims 1 to 4, characterized in that: include: The abnormal monitoring end monitors the shaft oscillation of the running fan. According to the displacement sensor set in the shaft, the oscillation displacement generated by the fan during rotation is confirmed, and then the abnormal signal is evaluated based on the specific value of the oscillation displacement. The band feature extraction end determines a set of monitoring periods based on the generated abnormal signal and the current speed of the fan, and generates a speed change waveform within the monitoring period, determines the circumference feature and selects the feature according to the length of the fan blades, and locks the characteristic waveform segment for each set of blades; The calibration end of the abnormal blade to be determined, based on the different characteristic waveform segments associated with different blades and the different rotation speed characteristics in the different characteristic waveform segments, identifies whether there are rotation speed characteristics with large differences, and calibrates the abnormal blade to be determined; The real abnormality determination end confirms the angle characteristics of the corresponding abnormal segment based on the calibrated abnormal blade to be determined and the abnormal segment position determined in the characteristic waveform segment, and then identifies whether the angle characteristics of the abnormal segment are consistent from the subsequent rotation process. Based on the identification result, it is confirmed whether the abnormal blade to be determined is a real abnormality.

Images