Wind turbine generator gearbox fault alarm and early warning method and system

Abstract

The invention discloses a wind turbine generator gearbox fault alarm early warning method comprising the following steps: obtaining historical operation data of a plurality of indexes of a gearbox in a preset time period, and generating corresponding standard monitoring data according to the historical operation data; the method comprises the following steps: acquiring real-time operation data of a plurality of indexes of a gearbox, and processing the plurality of real-time data by using an anti-pulse interference average filtering method to obtain corresponding target operation data; based on the multiple pieces of target operation data, establishing a monitoring data matrix of the gearbox; and performing comparative analysis on the target operation data in the monitoring data matrix and a plurality of preset thresholds of the corresponding indexes, and determining whether to start a fault alarm and early warning function or not. According to the method, the gearbox parameters are comprehensively evaluated and analyzed, different operation monitoring data can be combined, systematic fault early warning analysis of the gearbox is achieved, the method is low in complexity and high in calculation efficiency, and the operation quality of the wind turbine generator can be greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind turbine gearbox fault diagnosis, and in particular relates to a wind turbine gearbox fault alarm and early warning method and system. Background Art

[0002] Most wind turbines are installed in remote areas such as mountains, hills, and Gobi Desert. The operating environment is extremely harsh, with large temperature differences between day and night, large changes in wind conditions, strong winds and sand, strong ultraviolet rays, and complex grid access environment. In addition, wind turbines are unattended for a long time during normal operation. The complexity of the environment has a great impact on the operation of wind turbines.

[0003] The wind turbine gearbox is one of the key components of the transmission chain of the doubly-fed wind turbine generator. Its main function is to act as a speed increaser to convert the low speed of the wind wheel into the high speed required by the generator for power generation, so that the generator can generate electricity. The gearbox is installed in the nacelle at the top of the tower. Once a failure occurs, it is very difficult to repair and replace. Therefore, its reliability and service life are much higher than those of general traditional machinery. Since the wind turbine is unattended for a long time during normal operation, it is difficult to find the hidden dangers of the components in time, which can easily lead to the magnification of small problems and increase the difficulty of maintenance. Therefore, the health status and service life of the gearbox have a great impact on whether the wind turbine can operate stably and generate electricity normally. As a transmission system, the gearbox should have good lubrication conditions and operation status monitoring equipment in addition to conventional mechanical properties during use to ensure that the gearbox is in the best condition for a long time during operation. At present, the main maintenance methods for wind turbine gearboxes are planned inspection and post-maintenance. Planned inspection mainly relies on the operating characteristics of the gearbox and the experience of technicians to conduct regular inspections on the gearbox. This inspection work has the disadvantages of high maintenance cost, power generation loss and inadequate inspection of key components for wind turbines. It also requires high sense of responsibility and technical ability of maintenance personnel, and the maintenance effect varies from person to person. At the same time, it is difficult to find gear and bearing failures inside the gearbox. Post-maintenance cannot play a role in sudden failures of the gearbox, and it is difficult to track and find hidden faults with a long latent period. Once a fault occurs, it is a major fault. Due to insufficient preparation beforehand, maintenance work is often prolonged and losses are heavy.

[0004] At present, the gearbox early warning system in the industry only determines the potential fault sources of the wind turbine gearbox by collecting single-category data of the gearbox. The vibration monitoring system only analyzes the vibration data, the oil detection data only analyzes the oil status, and the bearing temperature and lubricating oil of the gearbox only detect the status of each component of the gearbox. There is a lack of systematic fault early warning analysis for the gearbox, and it will only issue a fault alarm when a fault occurs. It is impossible to conduct a comprehensive analysis of the gearbox operating status, achieve early warning of faults, and predict maintenance. Summary of the invention

[0005] (I) Purpose of the invention

[0006] The purpose of the present invention is to provide a wind turbine gearbox fault alarm and early warning method and system, which can accurately and systematically judge the operating health status of the gearbox, and perform fault alarm or early warning according to the fault occurrence condition, so as to avoid the problem that the hidden dangers of gearbox faults have a long incubation period and are difficult to detect, which eventually causes major faults.

[0007] (II) Technical solution

[0008] In order to solve the above problems, a first aspect of the present invention provides an alarm and early warning method for a wind turbine gearbox failure, comprising:

[0009] Acquire historical operating data of multiple indicators of the gearbox within a preset time period, and generate corresponding standard monitoring data according to the historical operating data;

[0010] Collecting real-time operating data of multiple indicators of the gearbox, and processing the multiple real-time operating data by using an anti-pulse interference average filtering method to obtain corresponding target operating data, specifically including: sorting the collected real-time operating data from small to large, removing the maximum value and the minimum value, and calculating the average value of the remaining real-time operating data to obtain the corresponding target operating data;

[0011] Establishing a monitoring data matrix of the gearbox based on a plurality of the target operating data;

[0012] The target operation data in the monitoring data matrix is ​​compared and analyzed with a plurality of preset thresholds of corresponding indicators to determine whether to start the fault alarm warning function, wherein the plurality of preset thresholds include an alarm preset threshold, a similarity threshold and an evaluation parameter threshold.

[0013] Furthermore, comparing and analyzing the target operation data in the monitoring data matrix with a plurality of preset thresholds of corresponding indicators to determine whether to start the fault alarm warning function includes:

[0014] First, the target operation data is compared with the preset alarm threshold, and whether to activate the alarm function is determined according to the comparison result;

[0015] If the alarm function is not activated, the target operation data is subtracted from the corresponding standard monitoring data to obtain a similarity analysis result, and whether to activate the early warning function is determined based on the similarity analysis result.

[0016] Further, determining whether to start the early warning function according to the similarity analysis result includes:

[0017] If the similarity result exceeds the similarity threshold, a comprehensive evaluation is performed on the operating status of the gearbox and it is determined whether to activate the early warning function;

[0018] If the similarity analysis result does not exceed the similarity threshold, it is determined that the gearbox operating state is normal.

[0019] Further, the comprehensive evaluation of the operating status of the gearbox and determining whether to start the early warning function includes:

[0020] Calculating a comprehensive evaluation parameter of the gearbox based on the similarity analysis result, and activating an early warning function if the comprehensive evaluation parameter exceeds the evaluation parameter threshold;

[0021] If the comprehensive evaluation parameter does not exceed the evaluation parameter threshold, it is determined that the gearbox operating state is normal.

[0022] Furthermore, the multiple indicators include: vibration monitoring data of the gearbox, oil cleanliness data, temperature of bearings of various components of the gearbox and gear oil, and pressure of the lubrication system of the gearbox.

[0023] Furthermore, the step of comparing the target operation data with the preset alarm threshold and determining whether to start the alarm function according to the comparison result includes:

[0024] When the indicators are the vibration monitoring data of the gearbox and the oil cleanliness data, the determination is as follows:

[0025] If the target operating data is less than the first alarm threshold, the alarm function is not activated;

[0026] If the first alarm threshold value is less than the target operating data and less than the second alarm threshold value, the alarm function is activated and the unit does not stop;

[0027] If the target operating data is greater than the second alarm threshold, the alarm function is activated and the unit is shut down; wherein the alarm preset threshold includes a first alarm threshold and a second alarm threshold, and the second alarm threshold is greater than the first alarm threshold.

[0028] Furthermore, the step of comparing the target operation data with the preset alarm threshold and determining whether to activate the alarm function according to the comparison result further includes:

[0029] When the indicators are the temperature of the bearings of the gearbox components and the gear oil and the lubrication system pressure of the gearbox, the determination is as follows:

[0030] If the first alarm threshold ≤ the target operating data ≤ the second alarm threshold, the alarm function is not activated;

[0031] If the target operating data is greater than the second alarm threshold or the target operating data is less than the first alarm threshold, the alarm function is activated and the unit is shut down.

[0032] Another aspect of the present invention provides an alarm and early warning system for a wind turbine gearbox failure, comprising:

[0033] A data acquisition module is used to acquire historical operating data of multiple indicators of the gearbox within a preset time period, and generate corresponding standard monitoring data according to the historical operating data;

[0034] Generate a target data module, which is used to collect real-time operating data of multiple indicators of the gearbox, and process the multiple real-time data using an anti-pulse interference average filtering method to obtain multiple target operating data;

[0035] Establishing a monitoring matrix module, for establishing a monitoring data matrix of the gearbox based on the target operation data;

[0036] The alarm and warning module is used to compare and analyze the target operating data in the monitoring data matrix with multiple preset thresholds of corresponding indicators to determine whether to start the fault alarm and warning function, where the multiple preset thresholds include alarm preset thresholds, similarity thresholds and evaluation parameter thresholds.

[0037] Furthermore, the alarm and warning module is specifically used for:

[0038] First, the target operation data is compared with the preset alarm threshold, and whether to activate the alarm function is determined according to the comparison result;

[0039] If the alarm function is not activated, the target operation data is subtracted from the corresponding standard monitoring data to obtain a similarity analysis result, and whether to activate the early warning function is determined based on the similarity analysis result.

[0040] Furthermore, the alarm module is also used for:

[0041] If the similarity result exceeds the similarity threshold, a comprehensive evaluation is performed on the operating status of the gearbox and it is determined whether to activate the early warning function;

[0042] If the similarity analysis result does not exceed the similarity threshold, it is determined that the gearbox operating state is normal.

[0043] (III) Beneficial effects

[0044] The above technical scheme of the present invention has the following beneficial technical effects: the present invention provides a method and system for alarming and warning the fault of a gearbox of a wind turbine generator set, which obtains the historical operation data of multiple indicators of the gearbox within a preset time period, analyzes and stores the historical data of the gearbox of the wind turbine generator set, and determines the standard monitoring data of the gearbox under different environmental temperatures and wind speeds according to the historical operation data, thereby analyzing the operation data of the gearbox and judging whether there is an abnormality in the operation state of the gearbox; and the present invention simultaneously collects the real-time operation data of multiple indicators of the gearbox, and processes the multiple real-time data by using the anti-pulse interference average filtering method through the vibration data, oil cleanliness data, gearbox temperature data, and gearbox lubrication system pressure data of the wind turbine generator set gearbox obtained in real time, and obtains multiple target operation data, and then processes and analyzes the target operation data to realize the fault warning of the gearbox. Finally, the target operation data in the monitoring data matrix is ​​compared and analyzed with the multiple preset thresholds of the corresponding indicators to determine whether to start the fault alarm or warning function, wherein the multiple preset thresholds include the alarm preset threshold, the similarity threshold and the evaluation parameter threshold. The present invention can realize dynamic comparison of gearbox data of different power units and different manufacturers, avoid setting a single fault threshold, and avoid misjudgment of faults of different power units and different manufacturers' gearboxes, so as to realize flexible control and dynamic comparison of gearbox operation data monitoring. In the process of judging whether there is an abnormality in operation, the present invention also judges the gearbox operation status by performing similarity analysis on the gearbox operation data, and timely warns of the occurrence of unit failure before the unit has a serious failure, so as to realize safe and stable operation of the wind turbine unit when it is unattended for a long time. By comprehensively evaluating and analyzing the gearbox parameters, different operation monitoring data can be combined to realize systematic fault warning analysis of the gearbox. This method has low complexity and high calculation efficiency, and can greatly improve the operation quality of the wind turbine unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of a wind turbine gearbox fault alarm and early warning method of the present invention;

[0046] Figure 2 It is a specific embodiment of the wind turbine gearbox fault alarm and early warning method of the present invention;

[0047] Figure 3 This is a flow chart of a wind turbine gearbox fault alarm and early warning system of the present invention;

[0048] Figure 4 It is a schematic diagram of the gear box structure of the present invention.

[0049] Figure numbers: 1, main shaft; 2, main bearing; 3, first-stage planetary carrier; 4, first-stage planetary carrier bearing. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0051] In the description of this application, it is necessary to understand that the "axial direction" is parallel to the main axis, and the "radial direction" is perpendicular to the main axis. The "upwind direction" is the direction where the wind blows first, and the "downwind direction" is the direction where the wind blows later. In a wind turbine, since the wind rotor always faces the direction from which the wind comes and is in front of the tower, it can also be understood that the direction closer to the wind rotor is the upwind direction, and the direction away from the wind rotor is the downwind direction.

[0052] The gearbox structure monitored by the embodiment of the present invention is an integrated spindle gearbox, that is, the spindle and the main bearing are integrated inside the gearbox, the main bearing 2 is a double-row tapered bearing that provides front support for the spindle, the spindle 1 and the first-stage planet carrier 3 are connected as one by shrink fitting, and the first-stage planet carrier bearing 4 is a cylindrical roller bearing that provides rear support for the spindle system. The schematic diagram is as follows: Figure 4 shown.

[0053] like Figure 1 As shown, the present invention provides an alarm and early warning method for a wind turbine gearbox failure, comprising:

[0054] S1, obtain the historical operating data of multiple indicators of the gearbox within a preset time period, and generate corresponding standard monitoring data based on the historical operating data; in this step, determine the standard value of the detection data (standard monitoring data) under different ambient temperatures and wind speeds based on the historical operating data.

[0055] S2, collect the real-time operation data of multiple indicators of the gearbox, and use the pulse interference prevention average filtering method to process the multiple real-time data to obtain the corresponding target operation data; in this step, the multiple indicators include: vibration monitoring data of the gearbox, oil cleanliness data, temperature of bearings and gear oil of various components of the gearbox, and pressure of the lubrication system of the gearbox. According to the detection requirements, the vibration data, oil cleanliness data, gearbox temperature data, and gearbox lubrication system pressure data of the wind turbine gearbox are obtained in real time, and filtered to obtain the target operation data of multiple indicators.

[0056] S3, based on the plurality of target operating data, establishing a monitoring data matrix for the gearbox, monitoring and analyzing the operating status of the wind turbine gearbox by establishing the monitoring data matrix, and determining whether there is any abnormality in the target operating data of the gearbox.

[0057] S4, compare and analyze the target operating data in the monitoring data matrix with multiple preset thresholds of the corresponding indicators to determine whether to start the fault alarm warning function, where the multiple preset thresholds include alarm preset thresholds, similarity thresholds and evaluation parameter thresholds. By monitoring the vibration of the gearbox, the cleanliness of the oil, the temperature of the bearings and gear oil of each component of the gearbox, and the lubrication system pressure of the gearbox, timely warning of the occurrence of unit failures can be carried out, maintenance preparations can be made in advance, and serious failures of the unit can be prevented and cannot be handled in time, so as to achieve long-term unattended operation of the unit, ensure the safe and stable operation of the wind turbine gearbox, and improve the operation quality of the wind turbine. This step specifically includes:

[0058] S41, first compare the target operation data with the alarm preset threshold, and determine whether to start the alarm function according to the comparison result. This step includes two situations:

[0059] S411, when the indicators are the vibration monitoring data of the gearbox and the oil cleanliness data, the determination is as follows:

[0060] (1) If the target operating data is less than the first alarm threshold, the alarm function is not activated;

[0061] (2) If the first alarm threshold value is less than the target operating data and less than the second alarm threshold value, the alarm function is activated and the unit does not shut down;

[0062] (3) If the target operating data is greater than the second alarm threshold, the alarm function is activated and the unit is shut down; wherein the alarm preset threshold includes a first alarm threshold and a second alarm threshold, and the second alarm threshold is greater than the first alarm threshold.

[0063] S412, when the indicators are the temperature of the bearings of the gearbox components and the gear oil and the lubrication system pressure of the gearbox, the determination is as follows:

[0064] (1) If the first alarm threshold ≤ the target operating data ≤ the second alarm threshold, the alarm function is not activated;

[0065] (2) If the target operation data is greater than the second alarm threshold or the target operation data is less than the first alarm threshold, the alarm function is activated and the unit is shut down.

[0066] S42, if the alarm function is not activated, the target operation data is subtracted from the corresponding standard monitoring data to obtain a similarity analysis result (similarity analysis process), and whether to activate the early warning function is determined according to the similarity analysis result. This step specifically includes:

[0067] S421, if the similarity result exceeds the similarity threshold, a comprehensive evaluation is performed on the operating state of the gearbox and it is determined whether to start the early warning function; this step specifically includes:

[0068] S4211, calculating a comprehensive evaluation parameter of the gearbox based on the similarity analysis result, and if the comprehensive evaluation parameter exceeds the evaluation parameter threshold, starting a warning function;

[0069] S4212: If the comprehensive evaluation parameter does not exceed the evaluation parameter threshold, it is determined that the gearbox operating state is normal.

[0070] S422: If the similarity analysis result does not exceed the similarity threshold, it is determined that the gearbox is operating normally.

[0071] The following is a detailed description with reference to specific embodiments. Figure 2 As shown:

[0072] Step 1: Obtain and store several groups of historical operating data of multiple indicators of gearboxes within a preset time period, and generate standard values ​​of operating monitoring data (standard monitoring data) of wind turbine gearboxes under different ambient temperatures and wind speeds based on the historical operating data. At the same time, real-time gearbox operating monitoring data (real-time operating data) within a preset time range (the same as the previous preset time period) is collected;

[0073] Step 2: The collected real-time operation data is filtered by the pulse interference-proof average filtering method to obtain the corresponding target operation data, and then the gearbox operation monitoring data matrix is ​​established according to the target operation data. The calculation process of the pulse interference-proof average filtering method includes: sorting the collected real-time operation data in ascending order, removing the maximum and minimum values, and calculating the average value of the remaining real-time operation data to obtain the corresponding target operation data; the target operation data of multiple indicators include: the vibration monitoring data of the gearbox, the oil cleanliness data, the temperature of the bearings and gear oil of each component of the gearbox, and the lubrication system pressure of the gearbox. The calculation of the pulse interference-proof average filtering method is explained as follows using the radial vibration data of the main bearing of the gearbox as an example.

[0074] Z 1i =[Z 11 ,Z 12 ,Z 13 ···Z 1N ]; (1)

[0075]

[0076] Where Z1 is the filtered value of the radial vibration data of the gearbox main bearing within the preset time range (target operation data), Z 11 <Z 12<Z 13 < <Z 1N , other data are processed in the same way. The target operation data is the data within the preset time range, and N is the number of samples within the preset time range. For example, the preset time range is 10 minutes, and the sampling period is 100 ms, that is, there are 100 data within the preset time range.

[0077] The gearbox operation monitoring data matrix in step 2 is specifically:

[0078] A i =[Z1,Z2,Z3,Z4,Z5,Z6,C,T1,T2,T3,T4,T5,T6,Q1,Q2] i ; (3)

[0079] The data monitored by the embodiment of the present invention includes the vibration data Z of each vibration monitoring point of the gear box i , gearbox online oil monitoring system oil cleanliness data C, gearbox component bearing and gear oil temperature T i , Lubrication system pressure data Q i . Among them, Z1 is the radial vibration of the main bearing, Z2 is the axial vibration of the gearbox mounting flange, Z3 is the radial vibration of the first-stage inner gear ring, Z4 is the radial vibration of the second-stage inner gear ring, Z5 is the axial vibration of the high-speed shaft bearing, and Z6 is the radial vibration of the high-speed shaft bearing. T1 is the bearing temperature of the main bearing, T2 is the downwind bearing temperature of the ratio wheel, T3 is the upwind bearing temperature of the high-speed shaft, T4 is the downwind bearing temperature of the high-speed shaft, T5 is the gearbox oil pool temperature, and T6 is the gearbox oil inlet temperature. Q1 is the gearbox oil pump pressure, and Q2 is the gearbox oil inlet pressure.

[0080] Step 3: Determine whether there is any abnormality in the target operating data of the gearbox and whether to activate the alarm function. The determination method is:

[0081] (1) For the gearbox vibration monitoring data, set the alarm preset thresholds, including Y1 (first alarm threshold) and Y2 (second alarm threshold), and compare the filtered vibration data Zi obtained in real time with the set thresholds Y1 and Y2.

[0082] If the vibration data Z i If (vibration monitoring target operation data) is less than Y1, it is determined that the vibration condition of the equipment is normal at present, and the alarm function is not activated. Go to step 4 to perform similarity analysis with the standard monitoring data.

[0083] If Y1<vibration data Z i ≤Y2, the equipment is judged to be abnormal, the alarm function is activated, the unit does not stop, and an inspection is required at an appropriate time.

[0084] If the vibration data Zi If it is greater than Y2, the equipment is judged to be faulty, the alarm function is activated, and the equipment needs to be stopped immediately for inspection.

[0085] (2) For the gear oil cleanliness data, set thresholds Y3 (first alarm threshold) and Y4 (second alarm threshold), filter the oil cleanliness data obtained by the online oil monitoring system, and compare the filtered oil cleanliness data (oil cleanliness target operating data) with the set thresholds Y3 and Y4.

[0086] If the oil cleanliness data C is less than the first set threshold value Y3, the oil cleanliness is determined to be normal, the alarm function is not activated, and the process continues to step 4 to perform similarity analysis with the standard monitoring data;

[0087] If Y3<Oil cleanliness data C i ≤Y4, it is determined that the oil cleanliness is abnormal, the alarm function is activated, the filtration needs to be strengthened, and the unit does not stop;

[0088] If the oil cleanliness data C is greater than Y4, the oil cleanliness is judged to be poor, the alarm function is activated, and the machine needs to be stopped to take oil samples for oil inspection.

[0089] (3) For the bearing and gear oil temperature data of each gearbox component, set the maximum temperature threshold T max (Second alarm threshold), minimum temperature threshold T min (First alarm threshold), the bearing and gear oil temperature T of each component of the gearbox after filtering obtained in real time i (target operation data) and the maximum temperature threshold T max and the minimum temperature threshold T min For comparison:

[0090] If T i Greater than the maximum temperature threshold T max or T i Less than the minimum temperature threshold T min , it is determined that the gearbox has high or low temperature abnormalities, the alarm function is activated, and the machine needs to be stopped immediately for inspection.

[0091] If T max ≥Ti≥T min , the gearbox temperature is determined to be normal, the alarm function is not activated, and the process goes to step 4 to perform similarity analysis with the standard monitoring data.

[0092] (4) For the gearbox lubrication system pressure data, set the maximum pressure threshold Q max (Second alarm threshold), minimum pressure threshold Q min (first alarm threshold), the gearbox lubrication system pressure Q i(Target operating data) and maximum pressure threshold Q max and the minimum pressure threshold Q min Make a comparison.

[0093] If Q i Greater than the maximum pressure threshold Q max or Q i Less than the minimum pressure threshold Q min , it is determined that the gearbox lubrication system has an abnormal high pressure or low pressure, the alarm function is activated, and the machine needs to be stopped immediately for inspection.

[0094] If Q max ≥Qi≥Q min , it is determined that the gearbox lubrication system pressure is normal, the alarm function is not activated, and go to step 4 to perform similarity analysis with the standard monitoring data.

[0095] Step 4, based on the four situations in step 3 above, if there is no abnormality in the target operating data of the gearbox (the alarm function is not activated), the real-time operating data is subjected to similarity analysis with the corresponding standard monitoring data (compared with the standard value of the operating data under the same ambient temperature and wind speed conditions within the same preset time range), and the real-time operating data is subtracted from the corresponding standard monitoring data to obtain the corresponding similarity value under the current operating state. The similarity value is the similarity analysis result. For example, taking the gearbox main bearing temperature data as an example, the similarity value (similarity result) of the gearbox main bearing temperature data under the current operating state is calculated. The specific process of similarity analysis calculation is as follows:

[0096] X1(i)=|T1(i)-T1(i,K i )| ; (4)

[0097] X=[X1,X2,···,X j ,···,X n ]; (5)

[0098] Where i = 1, 2, ···, N, j = Z1, Z2, Z3, Z4, Z5, Z6, C, T1, T2, T3, T4, T5, T6, Q1, Q2, n = 15, T1(i) is the real-time data of the gearbox main bearing temperature under the current operating state, T1(i, K i ) is the gearbox main bearing temperature data (standard monitoring data) under historical working conditions, X1(i) is the similarity value of the gearbox main bearing temperature data under the current operating state, and X is the similarity matrix of the gearbox operating data under the current operating state.

[0099] Step 5: Determine whether to start the warning function based on the similarity analysis results. In this step, set the similarity threshold δ. If X j >δ(Xj is the similarity value of the gearbox main bearing temperature data under the current operating state), indicating that the current operating state is abnormal, giving an abnormal warning of the target detection parameters, and making a comprehensive evaluation of the gearbox operating state.

[0100] If X j ≤δ, indicating that the current operating status is normal and the gearbox is in good operating condition.

[0101] Step 6: If the similarity result exceeds the similarity threshold (if X j >δ), calculate the comprehensive evaluation parameters of the gearbox, and the specific calculation process of the comprehensive evaluation is:

[0102] According to the similarity values ​​of gearbox operating parameters, the comprehensive evaluation parameter η of the gearbox is calculated, which is:

[0103]

[0104] Where n=15, j=Z1, Z2, Z3, Z4, Z5, Z6, C, T1, T2, T3, T4, T5, T6, Q1, Q2.

[0105] Step 7, determining whether to start the warning function according to the comprehensive evaluation result, including: if the comprehensive evaluation parameter of the gearbox exceeds the evaluation parameter threshold, the gearbox state is judged to be abnormal and the warning function is started; if the comprehensive evaluation parameter of the gearbox does not exceed the evaluation parameter threshold, the gearbox operation state is considered to be normal. Set the comprehensive evaluation parameter threshold ε of the gearbox, that is, η>ε, then the gearbox state is judged to be abnormal and the warning function is started; if η≤ε, it means that the gearbox is in good operation.

[0106] like Figure 3 As shown, another aspect of the present invention provides an alarm and early warning system for a wind turbine gearbox failure, comprising:

[0107] The data acquisition module 21 is used to acquire the historical operation data of multiple indicators of the gearbox within a preset time period, and generate corresponding standard monitoring data according to the historical operation data;

[0108] A target data generation module 22 is used to collect real-time operation data of multiple indicators of the gearbox, and process the multiple real-time data using an anti-pulse interference average filtering method to obtain multiple target operation data;

[0109] A monitoring matrix establishment module 23 is used to establish a monitoring data matrix of the gearbox based on the target operation data;

[0110] The alarm warning module 24 is used to compare and analyze the target operating data in the monitoring data matrix with multiple preset thresholds of corresponding indicators to determine whether to start the fault alarm warning function, where the multiple preset thresholds include alarm preset thresholds, similarity thresholds and evaluation parameter thresholds.

[0111] Furthermore, the alarm and warning module is specifically used for:

[0112] First, the target operation data is compared with the preset alarm threshold, and whether to activate the alarm function is determined according to the comparison result;

[0113] If the alarm function is not activated, the target operation data is subtracted from the corresponding standard monitoring data to obtain a similarity analysis result, and whether to activate the early warning function is determined based on the similarity analysis result.

[0114] Furthermore, the alarm module is also used for:

[0115] If the similarity result exceeds the similarity threshold, a comprehensive evaluation is performed on the operating status of the gearbox and it is determined whether to activate the early warning function;

[0116] If the similarity analysis result does not exceed the similarity threshold, it is determined that the gearbox operating state is normal.

[0117] The present invention proposes an alarm and early warning method and system for a gearbox fault of a wind turbine generator set. The gearbox fault alarm or early warning is realized by processing and analyzing the operation monitoring data through the real-time acquired vibration data of the wind turbine generator set gearbox, oil cleanliness data, the temperature of the bearings and gear oil of each component of the gearbox, and the lubrication system pressure of the gearbox. By analyzing and storing the historical data of the gearbox of the wind turbine generator set, the standard value of the detection data of the gearbox under different environmental temperatures and wind speeds is determined, so as to analyze the gearbox operation data and judge whether there is an abnormality in the gearbox operation state. The present invention can realize dynamic comparison of gearbox data of different power units and different manufacturers, avoid setting a single fault threshold, and misjudgment of faults of gearboxes of different power units and different manufacturers, so as to realize flexible control and dynamic comparison of gearbox operation data monitoring. The present invention can also timely warn the occurrence of unit faults, make maintenance preparations in advance, prevent the unit from having serious faults that cannot be handled in time, realize long-term unattended operation of the unit, ensure the safe and stable operation of the wind turbine generator set gearbox, and improve the operation quality of the wind turbine generator set.

[0118] A person of ordinary skill in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes the following process of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (Read-Only Memory, referred to as ROM) or a random access memory (Random Access Memory, referred to as RAM), etc. The steps in the method of the embodiment of the present invention can be adjusted in order, merged and deleted according to actual needs. The modules [units] in the system [terminal or device] of the embodiment of the present invention can be merged, divided and deleted according to actual needs.

Claims

1. A wind turbine gearbox fault alarm and early warning method, characterized in that: include: Acquire historical operating data of multiple indicators of the gearbox within a preset time period, and generate corresponding standard monitoring data according to the historical operating data; Collecting real-time operating data of multiple indicators of the gearbox, and processing the multiple real-time data using an anti-pulse interference average filtering method to obtain corresponding target operating data; Establishing a monitoring data matrix of the gearbox based on a plurality of the target operating data; The target operation data in the monitoring data matrix is ​​compared and analyzed with a plurality of preset thresholds of corresponding indicators to determine whether to start the fault alarm warning function, wherein the plurality of preset thresholds include an alarm preset threshold, a similarity threshold and an evaluation parameter threshold.

2. The wind turbine gearbox fault early warning method according to claim 1 is characterized in that: The comparing and analyzing the target operation data in the monitoring data matrix with a plurality of preset thresholds of corresponding indicators to determine whether to start the fault alarm and early warning function includes: First, the target operation data is compared with the preset alarm threshold, and whether to activate the alarm function is determined according to the comparison result; If the alarm function is not activated, the target operation data is subtracted from the corresponding standard monitoring data to obtain a similarity analysis result, and whether to activate the early warning function is determined based on the similarity analysis result.

3. The wind turbine gearbox fault early warning method according to claim 2 is characterized in that: Determining whether to start the early warning function according to the similarity analysis result includes: If the similarity result exceeds the similarity threshold, a comprehensive evaluation is performed on the operating status of the gearbox and it is determined whether to activate the early warning function; If the similarity analysis result does not exceed the similarity threshold, it is determined that the gearbox operating state is normal.

4. The wind turbine gearbox fault early warning method according to claim 3 is characterized in that: The comprehensive evaluation of the operating status of the gearbox and determining whether to start the early warning function includes: Calculating a comprehensive evaluation parameter of the gearbox based on the similarity analysis result, and activating an early warning function if the comprehensive evaluation parameter exceeds the evaluation parameter threshold; If the comprehensive evaluation parameter does not exceed the evaluation parameter threshold, it is determined that the gearbox operating state is normal.

5. The wind turbine gearbox fault early warning method according to claim 2, characterized in that: The multiple indicators include: vibration monitoring data of the gearbox, oil cleanliness data, temperature of bearings of various components of the gearbox and gear oil, and lubrication system pressure of the gearbox.

6. The wind turbine gearbox fault early warning method according to claim 5, characterized in that: The step of first comparing the target operation data with the preset alarm threshold and determining whether to start the alarm function according to the comparison result includes: When the indicators are the vibration monitoring data of the gearbox and the oil cleanliness data, the determination is as follows: If the target operating data is less than the first alarm threshold, the alarm function is not activated; If the first alarm threshold value is less than the target operating data and less than the second alarm threshold value, the alarm function is activated and the unit does not stop; If the target operating data is greater than the second alarm threshold, the alarm function is activated and the unit is shut down; wherein the alarm preset threshold includes a first alarm threshold and a second alarm threshold, and the second alarm threshold is greater than the first alarm threshold.

7. The wind turbine gearbox fault early warning method according to claim 5, characterized in that: The step of first comparing the target operation data with the preset alarm threshold and determining whether to activate the alarm function according to the comparison result further includes: When the indicators are the temperature of the bearings of the gearbox components and the gear oil and the lubrication system pressure of the gearbox, the determination is as follows: If the first alarm threshold ≤ the target operating data ≤ the second alarm threshold, the alarm function is not activated; If the target operating data is greater than the second alarm threshold or the target operating data is less than the first alarm threshold, the alarm function is activated and the unit is shut down.

8. An alarm and early warning system for wind turbine gearbox failure, characterized in that: include: A data acquisition module is used to acquire historical operating data of multiple indicators of the gearbox within a preset time period, and generate corresponding standard monitoring data according to the historical operating data; Generate a target data module, which is used to collect real-time operating data of multiple indicators of the gearbox, and process the multiple real-time data using an anti-pulse interference average filtering method to obtain multiple target operating data; Establishing a monitoring matrix module, for establishing a monitoring data matrix of the gearbox based on the target operation data; The alarm and warning module is used to compare and analyze the target operating data in the monitoring data matrix with multiple preset thresholds of corresponding indicators to determine whether to start the fault alarm and warning function, where the multiple preset thresholds include alarm preset thresholds, similarity thresholds and evaluation parameter thresholds.

9. The wind turbine gearbox failure alarm and early warning system according to claim 8, characterized in that: The alarm module is specifically used for: First, the target operation data is compared with the preset alarm threshold, and whether to activate the alarm function is determined according to the comparison result; If the alarm function is not activated, the target operation data is subtracted from the corresponding standard monitoring data to obtain a similarity analysis result, and whether to activate the early warning function is determined based on the similarity analysis result.

10. The wind turbine gearbox failure alarm and early warning system according to claim 8, characterized in that: The alarm module is also used for: If the similarity result exceeds the similarity threshold, a comprehensive evaluation is performed on the operating status of the gearbox and it is determined whether to activate the early warning function; If the similarity analysis result does not exceed the similarity threshold, it is determined that the gearbox operating state is normal.

Images