Generator bearing temperature abnormity identification method and system of wind turbine generator

By calculating the difference between the generator bearing temperature and the cabin in the wind turbine and using outlier point detection algorithm, we can identify the abnormality of the generator bearing temperature, which solves the problems of missing data and many characteristics in the existing technology, and realizes the wind turbine Safe and efficient operation and maintenance.

CN119062527APending Publication Date: 2024-12-03ДУНФАН ЭЛЕКТРИК ВИНД ПАУЭР КО ЛТД
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411283173.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-13
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The prior art requires a large amount of historical data and multiple input features in the identification of bearing temperature abnormalities of wind turbine generators, which makes it difficult to effectively identify abnormalities when some data is missing or when new units are put into operation, increasing safety risks and downtime losses.

Method used

By obtaining historical data for preprocessing, the difference between the generator bearing temperature and the cabin temperature is calculated and different power intervals are divided. The outlier point detection algorithm is used to mark abnormal points, the abnormal factors are calculated and the state is fused, and early warning information is generated to identify abnormalities. state.

Benefits of technology

It realizes accurate identification of generator bearing temperature abnormalities under less data conditions, reduces safety risks and downtime losses, is suitable for onshore and offshore wind farms, and supports planned operation and maintenance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119062527A_ABST
    Figure CN119062527A_ABST
Patent Text Reader

Abstract

The invention discloses a generator bearing temperature anomaly identification method and system of a wind turbine generator. The method comprises the following steps: S1, obtaining historical data of a to-be-detected generator; s2, performing data preprocessing on historical data; s3, the difference between the generator bearing temperature of the to-be-detected unit and the temperature in the cabin is obtained, and the difference is divided into different intervals according to the power at the corresponding moment; s4, outlier detection is carried out on the time sequence data in each power interval; s5, calculating an abnormal factor in each power interval, and judging the temperature state of a generator bearing in the interval; s6, fusing the bearing temperature states in all the power intervals, and judging the final generator bearing temperature state; and S7, early warning information is sent to the wind turbine generator with the abnormal temperature of the generator bearing. According to the invention, the abnormal temperature condition of the generator bearing can be quickly identified, the unit is protected to reliably operate within an allowable temperature range, the power generation loss is effectively reduced, and the utilization rate of the wind generating set is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power generation, and in particular relates to a method and system for identifying abnormal temperature of a generator bearing of a wind turbine set. Background Art

[0002] Generators are key equipment in the transmission chain system of wind turbines. They have a high probability of failure and high replacement costs, and have always been the focus of attention in wind turbines. When the bearing temperature of the generator is abnormal, it often reflects the potential risks of the equipment, including poor lubrication, failure of the cooling system, and even bearing damage. If such abnormal changes can be detected in a timely and accurate manner, and the operation and maintenance personnel are informed in advance to arrange maintenance, it will effectively reduce the safety risks in the operation of the wind turbine and the loss of power generation caused by abnormal shutdown.

[0003] In existing related technologies, the model is first trained through machine learning technology, and then the residual between the model prediction value and the actual collected value is used to identify anomalies. However, this solution often requires a large amount of historical data, and the required input features are usually more, otherwise it is difficult to achieve the desired effect.

[0004] In practical applications, many wind turbines have some input feature data that is not collected or missing, and newly commissioned wind turbines may not have sufficient historical data. Therefore, there is an urgent need for an anomaly identification method that relies on relatively fewer features and data. Summary of the invention

[0005] The purpose of the present invention is to accurately monitor the health status of the generator and provide maintenance personnel with a clear maintenance basis. A method and system for identifying abnormal temperature of a generator bearing of a wind turbine set are disclosed.

[0006] On the one hand, the purpose of the present invention is achieved by the following technical solutions:

[0007] A method for identifying abnormal temperature of a generator bearing of a wind turbine generator set, the method comprising the following steps:

[0008] S1: Obtain historical data of the unit to be tested;

[0009] S2: Data preprocessing for historical data;

[0010] S3: Difference between the bearing temperature of the generator of the unit to be tested and the temperature in the cabin, and divide them into different intervals according to the power at the corresponding moment;

[0011] S4: Perform outlier detection on the time series data in each power interval;

[0012] S5: Calculate the abnormal factor in each power interval and determine the temperature state of the generator bearing in the interval;

[0013] S6: integrating the bearing temperature states in all power ranges to determine the final generator bearing temperature state;

[0014] S7: Send warning information to wind turbines with abnormal generator bearing temperature.

[0015] According to a preferred implementation, step S1 specifically includes obtaining wind farm information, wind turbine information, date and time, active power, generator bearing temperature, and cabin temperature data.

[0016] According to a preferred implementation, step S2 specifically includes: invalidating null value data, invalidating stuck data, invalidating data that is obviously beyond the range, and setting the power screening condition to be greater than 0.

[0017] According to a preferred embodiment, step S3 includes: calculating the difference between the generator bearing temperature and the cabin temperature of the wind turbine to be detected within a period of time, and dividing the data into different power intervals according to the active power at the corresponding moment;

[0018] Assuming that m power intervals are divided, namely p1,...,pm, then m time series data are obtained, and each data point is the difference between the bearing temperature and the cabin temperature.

[0019] According to a preferred embodiment, step S4 includes: for each time series data, using an outlier detection algorithm to mark the data, if the data of the wind turbine to be detected at a certain moment appears as an abnormal point, it is marked as 1; if it appears as a normal point, it is marked as 0.

[0020] According to a preferred embodiment, in step S4, the cluster point detection method includes but is not limited to the n-sigma method, the box-type box method, and the isolation forest or local anomaly detection algorithm LOF.

[0021] According to a preferred embodiment, step S5 includes: performing statistical analysis on the abnormal marking data in each time series data of the wind turbine to be detected, and calculating the abnormal factor of a single day and the overall abnormal factor within the time;

[0022] A threshold is set. If both types of abnormal factors exceed a preset ratio, it is considered that the generator bearing temperature of the wind turbine to be detected is in an abnormal state within the corresponding power range.

[0023] According to a preferred embodiment, in step S5, the abnormal factor is calculated as follows: assuming that the total number of data items is p and the number of abnormal items is q, the abnormal proportion factor A=q / p.

[0024] According to a preferred embodiment, step S7 includes generating early warning information and generator bearing maintenance application by abnormal factor calculation and state fusion; the early warning information includes wind farm, wind turbine, model, early warning components, possible causes, and maintenance suggestions.

[0025] On the other hand, the present invention also discloses:

[0026] A system for identifying abnormal temperature of a generator bearing of a wind turbine generator set, the system comprising:

[0027] The data acquisition module is configured to complete the historical data acquisition of the unit to be tested, including: wind farm information, wind turbine information, date and time, active power, generator bearing temperature, and cabin temperature acquisition;

[0028] The data preprocessing module is configured to invalidate null value data, invalidate stuck data, invalidate data that is obviously beyond the range, and set the power screening condition to be greater than 0;

[0029] The feature generation module is configured to calculate the difference between the generator bearing temperature and the cabin temperature of the wind turbine to be detected within a preset time period, and divide the data into different power intervals according to the active power at the corresponding time; assuming that m power intervals p1,...,pm are divided, m time series data are obtained, and each data point is the difference between the bearing temperature and the cabin temperature;

[0030] The outlier detection module is configured to use the outlier detection algorithm to mark each time series data; if the data of the wind turbine to be detected at a certain moment is an abnormal point, it is marked as 1; if it is a normal point, it is marked as 0;

[0031] The single power interval abnormal factor calculation module is configured to perform statistical analysis on the abnormal marking data in each time series data of the wind turbine to be detected, and calculate the abnormal factor of a single day and the overall abnormal factor within the time;

[0032] And based on the set threshold, if both types of abnormal factors exceed the preset ratio, it is considered that the generator bearing temperature of the wind turbine to be detected is in an abnormal state within the corresponding power range;

[0033] The power interval state fusion module is configured to fuse the state of the generator bearing temperature in each power interval, and if there is an abnormality in any power interval, it indicates that the generator bearing temperature is abnormal;

[0034] The fault warning module is configured to generate warning information and generator bearing maintenance application through abnormal factor calculation and state fusion, wherein the warning information includes wind farm, wind turbine, model, warning components, possible causes, and maintenance suggestions.

[0035] The aforementioned main scheme of the present invention and its further options can be freely combined to form multiple schemes, all of which are schemes that can be adopted and claimed for protection by the present invention. After understanding the scheme of the present invention, those skilled in the art can understand that there are multiple combinations based on the prior art and common knowledge, all of which are technical schemes to be protected by the present invention, and they are not exhaustively listed here.

[0036] Beneficial effects of the present invention:

[0037] 1. The method and system for identifying abnormal bearing temperature of a wind turbine generator provided by the present invention can accurately identify abnormal bearing temperature of a generator through outlier detection in different power ranges, and simultaneously push early warning information, which is beneficial to the planned operation and maintenance of the wind farm;

[0038] 2. The method and system for identifying abnormal bearing temperature of a wind turbine generator provided by the present invention will not have any adverse effects or potential dangers on the wind turbine generator set, and are also applicable to wind farms in any terrain on land and offshore. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a schematic diagram of the principle structure of the abnormal temperature identification system for the generator bearing of the wind turbine generator set of the present invention;

[0040] Figure 2 The present invention is a flow chart of a method for identifying abnormal temperature of a generator bearing of a wind turbine generator set. DETAILED DESCRIPTION

[0041] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0042] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. In addition, the present invention should point out that in the present invention, if the specific structure, connection relationship, position relationship, power source relationship, etc. are not specifically written, the structure, connection relationship, position relationship, power source relationship, etc. involved in the present invention are all known to those skilled in the art based on the existing technology without creative work.

[0043] Example 1

[0044] refer to Figure 1 As shown in the figure, a method for identifying abnormal temperature of a generator bearing of a wind turbine generator set is shown, and its characteristic is that the method for identifying abnormal temperature of a generator bearing of a wind turbine generator set includes the following steps.

[0045] Step S1: Obtain historical data of the unit to be detected.

[0046] Preferably, step S1 specifically includes obtaining wind farm information, wind turbine information, date and time, active power, generator bearing temperature, and cabin temperature data.

[0047] Step S2: Perform data preprocessing on historical data.

[0048] Preferably, step S2 specifically includes: invalidating null value data, invalidating stuck data, invalidating data that is obviously beyond the range, and setting the power screening condition to be greater than 0.

[0049] Step S3: subtract the generator bearing temperature of the unit to be tested from the cabin temperature, and divide them into different intervals according to the power at the corresponding moment.

[0050] Preferably, step S3 comprises: calculating the difference between the generator bearing temperature and the cabin temperature of the wind turbine to be detected within a period of time, and dividing the data into different power intervals according to the active power at the corresponding moment;

[0051] Assuming that m power intervals are divided, namely p1,...,pm, then m time series data are obtained, and each data point is the difference between the bearing temperature and the cabin temperature.

[0052] Step S4: Perform outlier detection on the time series data within each power interval.

[0053] Preferably, step S4 includes: for each time series data, using an outlier detection algorithm to mark the data, if the data of the wind turbine to be detected at a certain moment appears as an abnormal point, it is marked as 1; if it appears as a normal point, it is marked as 0.

[0054] Furthermore, in step S4, the outlier detection method can use a simple threshold over-limit method, an n-sigma method, a box-type box method, and some machine learning algorithms, such as isolation forest, local anomaly detection algorithm (LOF), etc., without limitation.

[0055] Step S5: Calculate the abnormal factor in each power interval and determine the temperature state of the generator bearing in the interval.

[0056] Preferably, step S5 includes: performing statistical analysis on the abnormal marking data in each time series data of the wind turbine to be detected, calculating the abnormal factor of a single day and the overall abnormal factor within the time; setting a threshold value, if both types of abnormal factors exceed a preset ratio, it is considered that the generator bearing temperature of the wind turbine to be detected is in an abnormal state within the corresponding power range.

[0057] Furthermore, in step S5, the calculation method of the anomaly factor is as follows: assuming that the total number of data items is p and the number of anomalies is q, then the anomaly proportional factor A=q / p.

[0058] Step S6: The bearing temperature states in all power intervals are integrated to determine the final generator bearing temperature state.

[0059] Specifically, in order to determine whether the final generator bearing temperature is abnormal, the status of the generator bearing temperature in each power interval needs to be fused. For example, the following fusion strategy can be adopted: if there is an abnormality in any power interval, it indicates that the generator bearing temperature is abnormal.

[0060] Step S7: Sending warning information to the wind turbine set with abnormal generator bearing temperature.

[0061] Preferably, step S7 includes generating warning information and generator bearing maintenance application by abnormal factor calculation and state fusion; the warning information includes wind farm, wind turbine, model, warning components, possible causes, and maintenance suggestions.

[0062] Example 2

[0063] On the basis of embodiment 1, this embodiment further discloses: a system for identifying abnormal temperature of a generator bearing of a wind turbine generator set, the system for identifying abnormal temperature of a generator bearing of a wind turbine generator set comprising:

[0064] The data acquisition module is configured to complete the historical data acquisition of the unit to be tested, including: wind farm information, wind turbine information, date and time, active power, generator bearing temperature, and cabin temperature.

[0065] The data preprocessing module is configured to invalidate null value data, invalidate stuck data, invalidate data that is obviously beyond the range, and set the power screening condition to be greater than 0.

[0066] The feature generation module is configured to calculate the difference between the generator bearing temperature and the cabin temperature of the wind turbine to be detected within a preset time period, and divide the data into different power intervals according to the active power at the corresponding moment; assuming that m power intervals p1,...,pm are divided, m time series data are obtained, and each data point is the difference between the bearing temperature and the cabin temperature.

[0067] The outlier detection module is configured to use the outlier detection algorithm to mark each time series data; if the data of the wind turbine to be detected at a certain moment is an abnormal point, it is marked as 1; if it is a normal point, it is marked as 0.

[0068] The single power interval abnormal factor calculation module is configured to perform statistical analysis on the abnormal marking data in each time series data of the wind turbine to be detected, and calculate the abnormal factor of a single day and the overall abnormal factor within the time.

[0069] And based on the set threshold, if both types of abnormal factors exceed the preset ratio, it is considered that the generator bearing temperature of the wind turbine to be detected is in an abnormal state within the corresponding power range.

[0070] The power interval state fusion module is configured to fuse the state of the generator bearing temperature in each power interval, and if there is an abnormality in any power interval, it indicates that the generator bearing temperature is abnormal.

[0071] The fault warning module is configured to generate warning information and generator bearing maintenance application through abnormal factor calculation and state fusion, wherein the warning information includes wind farm, wind turbine, model, warning components, possible causes, and maintenance suggestions.

[0072] The wind turbine generator bearing temperature anomaly identification method and system provided by the present invention can accurately identify the abnormal state of the generator bearing temperature through outlier detection in different power ranges, and simultaneously realize the push of early warning information, which is conducive to the planned operation and maintenance of the wind farm;

[0073] The method and system for identifying abnormal bearing temperature of a wind turbine generator provided by the present invention will not have any adverse effects or potential dangers on the wind turbine generator set, and are also generalizable and applicable to wind farms in any land terrain and offshore wind farms.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying abnormal temperature of a generator bearing of a wind turbine generator set, characterized in that: The method for identifying abnormal temperature of a generator bearing of a wind turbine generator set comprises the following steps: S1: Obtain historical data of the unit to be tested; S2: Data preprocessing for historical data; S3: Calculate the difference between the generator bearing temperature and the cabin temperature of the unit to be tested, and divide them into different intervals according to the power at the corresponding moment; S4: Perform outlier detection on the time series data within each power interval; S5: Calculate the abnormal factor in each power interval and determine the temperature status of the generator bearing in the interval; S6: integrating the bearing temperature states within all power ranges to determine the final generator bearing temperature state; S7: Send warning information to wind turbines with abnormal generator bearing temperature.

2. The method for identifying abnormal temperature of a generator bearing of a wind turbine according to claim 1, wherein: Step S1 specifically includes obtaining wind farm information, wind turbine information, date and time, active power, generator bearing temperature, and cabin temperature data.

3. The method for identifying abnormal temperature of a generator bearing of a wind turbine according to claim 2, wherein: Step S2 specifically includes: invalidating null value data, invalidating stuck data, invalidating data that is obviously out of the range, and setting the power screening condition to be greater than 0.

4. The method for identifying abnormal temperature of a generator bearing of a wind turbine according to claim 1, wherein: Step S3 includes: Calculate the difference between the generator bearing temperature and the cabin temperature of the wind turbine to be tested over a period of time, and divide the data into different power intervals based on the active power at the corresponding moment; Assuming that m power intervals are divided, namely p1,...,pm, then m time series data are obtained, and each data point is the difference between the bearing temperature and the cabin temperature.

5. The method for identifying abnormal temperature of a generator bearing of a wind turbine according to claim 4, characterized in that: Step S4 includes: for each time series data, using an outlier detection algorithm to mark the data, if the data of the wind turbine to be detected at a certain moment appears as an abnormal point, it is marked as 1; if it appears as a normal point, it is marked as 0.

6. The method for identifying abnormal temperature of a generator bearing of a wind turbine according to claim 5, characterized in that: In step S4, the cluster point detection method includes but is not limited to the n-sigma method, the box-shaped method, and the isolation forest or local outlier detection algorithm LOF.

7. The method for identifying abnormal temperature of a generator bearing of a wind turbine according to claim 5, characterized in that: Step S5 includes: performing statistical analysis on the abnormality mark data in each time series data of the wind turbine to be detected, and calculating the abnormality factor of a single day and the overall abnormality factor within the time; A threshold is set. If both types of abnormal factors exceed a preset ratio, it is considered that the generator bearing temperature of the wind turbine to be detected is in an abnormal state within the corresponding power range.

8. The method for identifying abnormal temperature of a generator bearing of a wind turbine according to claim 7, wherein: In step S5, the anomaly factor is calculated as follows: assuming the total number of data items is p and the number of anomalies is q, then the anomaly proportional factor A = q / p.

9. The method for identifying abnormal temperature of a generator bearing of a wind turbine according to claim 7, wherein: Step S7 includes generating warning information and generator bearing maintenance request through abnormal factor calculation and state fusion; the warning information includes wind farm, wind turbine, model, warning components, possible causes, and maintenance suggestions.

10. A system for identifying abnormal temperature of a generator bearing of a wind turbine generator set, characterized in that: The generator bearing temperature abnormality identification system of the wind turbine generator set includes: The data acquisition module is configured to complete the historical data collection of the unit to be tested, including: wind farm information, wind turbine information, date and time, active power, generator bearing temperature, and cabin temperature; The data preprocessing module is configured to invalidate null value data, invalidate stuck data, invalidate data that is obviously out of the range, and set the power screening condition to be greater than 0; The feature generation module is configured to calculate the difference between the generator bearing temperature and the cabin temperature of the wind turbine to be tested within a preset time period and divide the data into different power intervals based on the active power at the corresponding time. Assuming that m power intervals p1,...,pm are divided, m time series data are obtained, and each data point is the difference between the bearing temperature and the cabin temperature. The outlier detection module is configured to use an outlier detection algorithm to mark each time series data; if the data of the wind turbine to be detected at a certain moment is an outlier, it is marked as 1; if it is a normal point, it is marked as 0; The single power interval abnormality factor calculation module is configured to perform statistical analysis on the abnormality mark data in each time series data of the wind turbine to be detected, and calculate the abnormality factor of a single day and the overall abnormality factor within the time period; Based on the set threshold, if both types of abnormal factors exceed the preset ratio, it is considered that the generator bearing temperature of the wind turbine to be tested is in an abnormal state within the corresponding power range; The power interval state fusion module is configured to fuse the state of the generator bearing temperature in each power interval, and if there is an abnormality in any power interval, it indicates that the generator bearing temperature is abnormal; The fault warning module is configured to generate warning information and generator bearing maintenance requests through abnormal factor calculation and state fusion, wherein the warning information includes wind farm, wind turbine, model, warning components, possible causes, and maintenance suggestions.

Citation Information

Patent Citations

  • Wind generating set generator bearing temperature alarm system

    CN112324627A

  • Meteorological prediction method and device for wind driven generator

    CN113902943A

  • Wind turbine generator bearing temperature anomaly detection method and device and wind turbine generator

    CN114412728A

  • Wind driven generator temperature anomaly fault diagnosis method and system based on SCADA data

    CN115342035A

  • Method and system for monitoring the status of a roller bearing of a wind turbine

    EP2927662A1