Wind turbine pin connection stability assessment method and system

Abstract

The invention relates to the technical field of pin shaft stability assessment, and provides a pin shaft connection stability assessment method and system for a wind turbine generator set, comprising: obtaining basic information of a pin shaft to be monitored, deploying a micro positioning chip to monitor its motion characteristics, parsing a reference motion characteristic of the pin shaft based on a wind turbine generator set model and parameters, verifying the monitoring characteristics and the reference characteristics to generate an abnormality coefficient, and when the abnormality coefficient is greater than a threshold value, generating a pin shaft connection loosening alarm label and marking the pin shaft to be monitored, thereby solving the technical problems that the stability of the pin shaft connection cannot be monitored and warned in real time, and it is difficult to timely discover and handle the abnormal situation of the pin shaft connection, and realizing the technical effect of real-time monitoring of the stability of the pin shaft connection without interfering with the normal operation of the wind turbine generator set, generating a pin shaft connection loosening alarm label, marking a pin shaft that may have an abnormality, timely discovering and handling the abnormal situation of the pin shaft connection, avoiding damage to the wind turbine generator set, and improving the operating efficiency and safety of the wind turbine generator set.

Description

Technical Field

[0001] The present invention relates to the technical field related to pin shaft stability assessment, and in particular to a method and system for assessing the stability of a pin shaft connection of a wind turbine generator set. Background Art

[0002] A wind turbine is a device that uses wind energy to generate electricity, in which the pin connection is an important component of the wind turbine and is used to connect the various components of the wind turbine. However, the stability of the pin connection will be affected by many factors, such as wind speed, wind direction, and unit operation time. If the pin connection is loose or abnormal, it will affect the normal operation of the wind turbine and even cause damage to the wind turbine. Therefore, it is very necessary to conduct a stability assessment of the pin connection of the wind turbine.

[0003] Typically, monitoring the stability of a pin connection requires shutdown monitoring, which will affect the normal operation of the wind turbine and has a high detection cost. Alternatively, the acoustic emission signal of the pin connection is monitored by an acoustic emission sensor installed on the wind turbine: when the pin connection becomes loose or abnormal, a specific acoustic emission signal is emitted, which is detected by the acoustic emission sensor and an alarm is sounded. However, this method is difficult to distinguish between loose pin connections and other acoustic emission signals, and cannot monitor the stability of the pin connection in real time.

[0004] In summary, the prior art has the technical problem that it is impossible to monitor and warn the stability of the pin connection in real time, and it is difficult to detect and handle abnormal conditions of the pin connection in a timely manner. Summary of the invention

[0005] The present application provides a method and system for evaluating the stability of the pin connection of a wind turbine generator set, aiming to solve the technical problems in the prior art of being unable to perform real-time monitoring and early warning of the stability of the pin connection, and being difficult to promptly detect and handle abnormal conditions of the pin connection.

[0006] In view of the above problems, the present application provides a method and system for evaluating the stability of a pin connection of a wind turbine generator set.

[0007] The first aspect disclosed in the present application provides a method for evaluating the stability of a pin connection of a wind turbine, wherein the method is applied to a pin connection stability evaluation system of a wind turbine, the system comprising a motion feature monitoring component, a motion reference parsing component and a motion feature verification component, and the method comprises: obtaining basic information of a pin to be monitored, wherein the basic information of the pin to be monitored comprises a pin connection position and a pin connection type; wherein a micro positioning chip is deployed on the surface of the pin to be monitored; activating the motion feature monitoring component, starting the wind turbine, and monitoring the pin monitoring motion feature of the pin to be monitored based on the micro positioning chip; activating the motion reference parsing component, performing motion feature parsing on the pin connection position and the pin connection type based on the wind turbine model and the wind turbine operating parameters, and generating a pin reference motion feature; activating the motion feature verification component to verify the pin monitoring motion feature and the pin reference motion feature, and generate a first motion anomaly coefficient; when the first motion anomaly coefficient is greater than a first motion anomaly coefficient threshold, generating a pin connection loosening alarm label, and marking the pin to be monitored.

[0008] Another aspect disclosed in the present application provides a pin connection stability assessment system for a wind turbine, wherein the system includes a motion feature monitoring component, a motion reference parsing component and a motion feature verification component, and the system includes: a basic information acquisition module, used to obtain basic information of the pin to be monitored, wherein the basic information of the pin to be monitored includes the pin connection position and the pin connection type; a positioning chip deployment module, wherein a micro positioning chip is deployed on the surface of the pin to be monitored; a positioning chip monitoring module, used to activate the motion feature monitoring component, start the wind turbine, and monitor the pin to be monitored based on the micro positioning chip. A pin shaft monitoring motion characteristic; a motion characteristic analysis module, used to activate the motion reference analysis component, based on the wind turbine model and the wind turbine operating parameters, perform motion characteristic analysis on the pin shaft connection position and the pin shaft connection type, and generate a pin shaft reference motion characteristic; a motion characteristic verification module, used to activate the motion characteristic verification component to verify the pin shaft monitoring motion characteristic and the pin shaft reference motion characteristic, and generate a first motion abnormality coefficient; a loose connection monitoring module, used to generate a pin shaft connection loose alarm label when the first motion abnormality coefficient is greater than the first motion abnormality coefficient threshold, and identify the pin shaft to be monitored.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The basic information of the pin shaft to be monitored, including the connection position and connection type, is obtained, and a micro positioning chip is deployed on the surface of the pin shaft to be monitored to monitor its motion characteristics; the motion characteristic monitoring component is activated, the wind turbine is started, and the pin shaft monitoring motion characteristics of the pin shaft to be monitored are monitored based on the micro positioning chip; the motion reference parsing component is activated, and the pin shaft connection position and connection type are parsed based on the wind turbine model and operating parameters to generate the pin shaft reference motion characteristics; the motion characteristic verification component is activated to verify the monitoring characteristics and the reference characteristics to generate a first motion anomaly coefficient; when the first motion anomaly coefficient is greater than a preset threshold, a pin shaft connection loose alarm label is generated, and the pin shaft to be monitored is marked, so that the stability of the pin shaft connection is monitored in real time without interfering with the normal operation of the wind turbine, and a pin shaft connection loose alarm label is generated, and the pin shaft that may have abnormalities is marked, and the abnormal conditions of the pin shaft connection are discovered and handled in time, so as to avoid damage to the wind turbine and improve the operating efficiency and safety of the wind turbine.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A possible flow chart of a method for evaluating the stability of a pin connection of a wind turbine generator system is provided for the embodiment of the present application;

[0013] Figure 2 A possible flow diagram of deploying a micro positioning chip in a method for evaluating the stability of a pin connection of a wind turbine generator system is provided for an embodiment of the present application;

[0014] Figure 3 A possible structural schematic diagram of a pin connection stability assessment system for a wind turbine generator system is provided for an embodiment of the present application.

[0015] Description of the accompanying drawings: basic information acquisition module 100, positioning chip deployment module 200, positioning chip monitoring module 300, motion feature analysis module 400, motion feature verification module 500, loose connection monitoring module 600. DETAILED DESCRIPTION

[0016] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description. Example 1

[0017] like Figure 1 As shown, an embodiment of the present application provides a method for evaluating the stability of a pin connection of a wind turbine generator set, wherein the method is applied to a pin connection stability evaluation system of a wind turbine generator set, the system comprising a motion feature monitoring component, a motion reference parsing component and a motion feature verification component, and the method comprises:

[0018] Step-1: Obtain basic information of the pin to be monitored, wherein the basic information of the pin to be monitored includes the pin connection position and the pin connection type;

[0019] Step-2: A micro positioning chip is deployed on the surface of the pin to be monitored;

[0020] Step-3: Activate the motion feature monitoring component, start the wind turbine, and monitor the pin shaft monitoring motion feature of the pin shaft to be monitored based on the micro positioning chip;

[0021] The commonly used traditional method for monitoring the stability of the pin connection is the vibration monitoring method, which uses a vibration sensor installed on the wind turbine to monitor the vibration of the pin connection. When the pin connection is loose or abnormal, it will cause a large vibration, which will be detected by the vibration sensor and an alarm will be issued. However, this method cannot monitor and warn the stability of the pin connection in real time, and it is also difficult to distinguish between loose and normal vibration of the pin connection.

[0022] Another traditional method is the acoustic emission method, which uses an acoustic emission sensor installed on the wind turbine to monitor the acoustic emission signal of the pin connection. When the pin connection is loose or abnormal, a specific acoustic emission signal will be emitted, which will be detected by the acoustic emission sensor and an alarm will be issued. However, this method cannot monitor the stability of the pin connection in real time, and it is also difficult to distinguish between the looseness of the pin connection and other acoustic emission signals.

[0023] It has been verified that the embodiments of the present application can overcome the shortcomings of the traditional method. The method can monitor the stability of the pin connection in real time without interfering with the normal operation of the wind turbine, and generate a pin connection loosening alarm label to mark the pin that may have abnormalities. It can not only detect and handle the abnormal situation of the pin connection in time to avoid damage to the wind turbine, but also reduce the detection cost and improve the operating efficiency and safety of the wind turbine.

[0024] The future abnormal trend can also be predicted based on the historical motion abnormal coefficient. By analyzing the historical motion abnormal coefficient of the pin shaft, the motion abnormal coefficient historical change curve and the rising rate change curve are constructed, and the average rising rate and the maximum rising rate are calculated to predict the future abnormal trend. It can also warn of possible looseness or abnormal conditions in advance, so that timely measures can be taken to deal with them, avoid the occurrence of abnormal conditions or reduce their impact.

[0025] Specifically, first, determine the connection position and connection type of the pin to be monitored, which can be obtained through the design drawings, operation records or other relevant documents of the wind turbine; deploy the micro positioning chip on the surface of the pin to be monitored, and special tools or equipment, such as a robotic arm, are required to correctly install the chip on the pin;

[0026] Then, the motion characteristic monitoring component is activated and the wind turbine is started, which can be done manually or automatically, depending on the type of wind turbine and the operating requirements; this is achieved by reading the data of the micro positioning chip and analyzing it. As the wind turbine is running, the motion characteristics of the pin shaft will change, and the changes in the motion characteristics can be monitored and recorded in real time by the micro positioning chip. By real-time monitoring of the monitoring motion characteristics of the pin shaft to be monitored, the subsequent analysis provides the necessary data basis, and the monitoring data can be used to evaluate the stability of the pin shaft connection and determine whether any abnormal conditions have occurred.

[0027] Step-4: Activate the motion benchmark analysis component, perform motion feature analysis on the pin connection position and the pin connection type based on the wind turbine model and wind turbine operation parameters, and generate the pin benchmark motion feature;

[0028] Step-5: Activate the motion feature verification component to verify the pin shaft monitoring motion feature and the pin shaft reference motion feature to generate a first motion anomaly coefficient;

[0029] Step-6: When the first motion abnormality coefficient is greater than the first motion abnormality coefficient threshold, a pin shaft loose connection alarm label is generated to mark the pin shaft to be monitored.

[0030] Then, activating the motion benchmark analysis component can be achieved through program code or hardware equipment. The motion benchmark analysis component performs motion feature analysis on the connection position and connection type of the monitored pin shaft based on the wind turbine model and the wind turbine operating parameters, and generates a pin shaft benchmark motion feature. The pin shaft benchmark motion feature includes the pin shaft motion trajectory (the expected motion path or trajectory of the pin shaft under the normal operating conditions of the wind turbine), the pin shaft motion speed (the expected motion speed of the pin shaft under the normal operating conditions of the wind turbine), the pin shaft motion acceleration (the expected acceleration of the pin shaft under the normal operating conditions of the wind turbine), the pin shaft motion angular displacement (the expected angular displacement of the pin shaft under the normal operating conditions of the wind turbine), and the pin shaft motion frequency (the expected motion frequency of the pin shaft under the normal operating conditions of the wind turbine), which is mainly used to compare and evaluate whether the pin shaft monitoring motion features are normal and stable;

[0031] Then, activating the motion feature verification component can also be implemented through program code or hardware equipment. The motion feature verification component verifies the pin monitoring motion feature and the pin reference motion feature to generate a first motion anomaly coefficient. The first motion anomaly coefficient refers to the difference or degree of deviation between the pin monitoring motion feature and the pin reference motion feature. The difference or degree of deviation can reflect the stability and reliability of the pin connection. When the motion feature verification component is implemented through program code or hardware equipment, it can automatically compare the pin monitoring motion feature and the pin reference motion feature, and calculate the difference or degree of deviation between the two. The degree of deviation can be expressed by a numerical value, and the resulting numerical value is the first motion anomaly coefficient. In general, the larger the value of the first motion anomaly coefficient, the greater the difference or degree of deviation between the pin monitoring motion feature and the pin reference motion feature, which means that the pin connection may have instability or reliability problems.

[0032] Finally, the generated first motion abnormality coefficient is compared with the first motion abnormality coefficient threshold: if the first motion abnormality coefficient is not greater than the first motion abnormality coefficient threshold, the monitoring is continued; if the first motion abnormality coefficient is greater than the first motion abnormality coefficient threshold, it indicates that the pin shaft to be monitored may have an abnormality, and a pin shaft connection loose alarm label is generated. At the same time, the pin shaft connection loose alarm label is associated with the corresponding pin shaft to be monitored, and the pin shaft to be monitored is marked. The alarm label can be a warning signal or an error report to remind maintenance personnel to perform timely inspection and processing. Timely discover and handle abnormal conditions of pin shaft connections, realize monitoring and early warning of the motion state of the pin shaft to be monitored in the wind turbine, and timely discover potential abnormal conditions and issue alarms so that maintenance personnel can take corresponding measures to handle them, thereby ensuring the normal operation of the wind turbine. At the same time, it can improve the efficiency and accuracy of maintenance, reduce maintenance costs, and improve the operating efficiency and safety of wind turbines.

[0033] Furthermore, the present application method also includes:

[0034] When the first motion abnormality coefficient is less than or equal to the first motion abnormality coefficient threshold, generating a pin-connected quasi-stable label;

[0035] When generating the pin connection quasi-stable label, based on the pin connection position, the pin connection type and the wind turbine operating parameters, the pin historical motion anomaly coefficient is collected;

[0036] Calculating a second motion abnormality coefficient according to the pin shaft historical motion abnormality coefficient;

[0037] When the second motion abnormality coefficient is greater than the second motion abnormality coefficient threshold, a pin shaft connection loosening warning label is generated to mark the pin shaft to be monitored;

[0038] When the second motion abnormality coefficient is less than or equal to the second motion abnormality coefficient threshold, a pin connection stability label is generated to identify the pin to be monitored.

[0039] Specifically, when the current vibration is normal, the development trend of the pin shaft is analyzed to see whether it is close to abnormality, so that loosening warning can be given in advance, thereby avoiding the occurrence of loosening or reducing its impact, including: when the first motion abnormality coefficient is less than or equal to the first motion abnormality coefficient threshold, it means that the connection state of the pin shaft to be monitored is basically stable. In this case, a pin shaft connection quasi-stable label is generated to identify the current state;

[0040] After generating the quasi-stable label of the pin connection, the historical motion anomaly coefficient of the pin can be collected based on the pin connection position, the pin connection type and the wind turbine operating parameters. This can be achieved by analyzing historical data and operating parameters, specifically including collecting historical motion anomaly data related to the pin to be monitored based on data such as the pin connection position, the pin connection type and the wind turbine operating parameters, and the historical motion anomaly data comes from the operation record of the wind turbine; performing pre-processing work such as data cleaning and format conversion on the collected historical motion anomaly data to ensure the accuracy and consistency of the data; extracting useful features from the historical motion anomaly data, such as the pin motion trajectory, speed, acceleration, etc., and extracting useful features can reflect the motion state and development trend of the pin; using data mining techniques, such as cluster analysis and trend prediction, to deeply analyze the useful features extracted from the historical motion anomaly data, the potential patterns, trends and abnormal signs of the pin motion can be discovered, and the pin historical motion anomaly percentage is output as a percentage in the form of the number of pin motion anomalies: the total number of times, and the pin historical motion anomaly percentage is used as the pin historical motion anomaly coefficient;

[0041] According to the collected historical motion abnormality coefficient of the pin shaft, the second motion abnormality coefficient is calculated: if the second motion abnormality coefficient is greater than the second motion abnormality coefficient threshold, a pin shaft connection loosening warning label can be generated to give an early warning and avoid loosening or other abnormal conditions of the pin shaft connection, so as to remind maintenance personnel to take timely prevention and intervention; if the second motion abnormality coefficient is less than or equal to the second motion abnormality coefficient threshold, a pin shaft connection stable label can be generated to identify the state, and the above steps also include: based on the collected historical motion abnormality coefficient of the pin shaft, the second motion abnormality coefficient is calculated, involving time series analysis; if the second motion abnormality coefficient is greater than the second motion abnormality coefficient threshold, it means that the pin shaft to be monitored may have looseness and other abnormal conditions. In this case, a pin shaft connection loosening warning label can be generated to remind maintenance personnel to take timely prevention and intervention; if the second motion abnormality coefficient is less than or equal to the second motion abnormality coefficient threshold, it means that the connection state of the pin shaft to be monitored is stable. In this case, a pin shaft connection stable label can be generated to identify the state. It helps maintenance personnel better understand the connection status of the pin shaft to be monitored and take appropriate measures in time to prevent and solve potential problems. At the same time, it can also improve the efficiency and accuracy of maintenance, reduce maintenance costs, and improve the operating efficiency and safety of wind turbines.

[0042] Furthermore, if Figure 2 As shown, a micro positioning chip is deployed on the surface of the pin to be monitored, and the application method includes:

[0043] Extracting a connection area according to the pin to be monitored, wherein the connection area includes a pin area, a first connection part area, and a second connection area;

[0044] The micro positioning chip is disposed in the pin region, the first connector region and the second connector region.

[0045] Deployment of the micro positioning chip specifically includes: extracting relevant information and data of the pin shaft to be monitored to determine the range of the connection area, which may include the pin shaft area, the first connecting piece area and the second connecting area, which can be divided according to factors such as the operating principle and structural design of the wind turbine set.

[0046] The micro positioning chip is deployed in the pin shaft area, the first connection part area and the second connection part area: this is achieved by scanning the surface of the pin shaft to be monitored, selecting appropriate positions in the pin shaft area, the first connection part area and the second connection part area, respectively, and deploying the micro positioning chip. The micro positioning chip can record and monitor the position, speed, acceleration and other motion characteristics of the pin shaft, and transmit the data in real time to the pin shaft connection stability assessment system for analysis and processing. By deploying the micro positioning chip on the surface of the pin shaft to be monitored, the motion characteristics and state of the pin shaft can be monitored in real time, and potential abnormal conditions can be discovered in time and alarms can be issued so that maintenance personnel can take corresponding measures to deal with them.

[0047] Further, the motion reference analysis component is activated, and based on the wind turbine model and the wind turbine operating parameters, the motion feature analysis of the pin connection position and the pin connection type is performed to generate the pin reference motion feature. The method of the present application includes:

[0048] Activate the motion reference analysis component, and retrieve the motion feature analysis node based on the wind turbine model, wherein the motion feature analysis node includes a pin motion trajectory analysis channel, a first connector motion trajectory analysis channel, and a second connector motion trajectory analysis channel;

[0049] Retrieving the pin shaft motion trajectory analysis channel, mapping the wind turbine operating parameters, the pin shaft connection position and the pin shaft connection type, and generating a pin shaft motion reference trajectory;

[0050] Retrieving the first connector motion trajectory analysis channel, mapping the wind turbine operating parameters, the pin connection position and the pin connection type, and generating a first connector motion reference trajectory;

[0051] Retrieving the motion trajectory analysis channel of the second connector, mapping the wind turbine operating parameters, the pin connection position and the pin connection type, and generating a second connector motion reference trajectory;

[0052] The pin shaft motion reference trajectory, the first connecting member motion reference trajectory and the second connecting member motion reference trajectory are added to the pin shaft reference motion feature.

[0053] Activating the motion benchmark analysis component can be implemented by program code or hardware device. The motion benchmark analysis component retrieves the motion feature analysis node based on the wind turbine model. The motion feature analysis node is usually pre-stored in the database or knowledge base of the pin connection stability assessment system for subsequent calling and analysis.

[0054] The pin motion trajectory parsing channel is called through the motion feature parsing node, and the pin motion trajectory parsing channel is used to map the position and type of the pin to be monitored to generate a pin motion reference trajectory; the first connecting member motion trajectory parsing channel is called through the motion feature parsing node, and the first connecting member motion trajectory parsing channel is used to map the position and type of the pin to be monitored to generate a first connecting member motion reference trajectory; similarly, the second connecting member motion trajectory parsing channel can be called through the motion feature parsing node, and the second connecting member motion trajectory parsing channel is used to map the position and type of the pin to be monitored to generate a second connecting member motion reference trajectory.

[0055] The pin motion reference track, the first connector motion reference track and the second connector motion reference track are added to the pin motion reference track, which constitute a part of the pin motion reference track, and can reflect the motion states of the pin to be monitored, such as the position and speed at different time points. The motion states of the pin to be monitored, such as the position and speed at different time points, are understood, and potential abnormalities in the pin motion state are discovered in time and an alarm is issued.

[0056] Further, the motion feature verification component is activated to verify the pin shaft monitoring motion feature and the pin shaft reference motion feature to generate a first motion abnormality coefficient. The method of the present application includes:

[0057] Counting a first relative distance interval according to the pin shaft motion reference trajectory and the first connecting member motion reference trajectory;

[0058] Counting a second relative distance interval according to the reference trajectory of the pin movement and the reference trajectory of the second connecting member movement;

[0059] Calculating a first frequency proportional coefficient that does not belong to the first relative distance interval according to the pin shaft monitoring motion characteristics;

[0060] Calculating a second frequency proportional coefficient that does not belong to the second relative distance interval according to the pin shaft monitoring motion characteristics;

[0061] The first frequency proportional coefficient and the second frequency proportional coefficient are added to the first motion abnormality coefficient.

[0062] By comparing the position information of the reference trajectory of the pin movement and the reference trajectory of the first connecting member movement, the relative distance interval of the corresponding positions of the two positioning chips is obtained, and the first relative distance interval can reflect the position relationship between the pin and the first connecting member; similarly, by comparing the position information of the reference trajectory of the pin movement and the reference trajectory of the second connecting member movement, the relative distance interval of the corresponding positions of the two positioning chips can be obtained, and the second relative distance interval can reflect the position relationship between the pin and the second connecting member.

[0063] By analyzing the movement characteristics of the monitored pin shaft, the frequency proportional coefficient that does not belong to the first relative distance interval is calculated, and the first frequency proportional coefficient can reflect whether the relative distance between the pin shaft and the first connecting member is normal; similarly, by analyzing the movement characteristics of the monitored pin shaft, the frequency proportional coefficient that does not belong to the second relative distance interval is calculated, and the second frequency proportional coefficient can reflect whether the relative distance between the pin shaft and the second connecting member is normal.

[0064] The first frequency proportional coefficient and the second frequency proportional coefficient can comprehensively reflect whether the relative position relationship between the pin shaft to be monitored and the surrounding components is normal. Adding the first frequency proportional coefficient and the second frequency proportional coefficient to the first motion abnormality coefficient can more comprehensively evaluate the motion state of the pin shaft. This helps wind turbine maintenance personnel better understand whether the relative position relationship between the pin shaft to be monitored and the surrounding components is normal, and timely discover potential abnormal situations and issue alarms so that maintenance personnel can take corresponding measures to deal with them.

[0065] Further, before activating the motion benchmark analysis component and retrieving the motion feature analysis node based on the wind turbine model, the present application method includes:

[0066] Based on the preset wind turbine model, collect the experimental data of pin connection stability;

[0067] The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped by the LSTM neural network topology, and the pin motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the pin motion trajectory analysis channel;

[0068] The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped by the LSTM neural network topology, and the first connection part motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the first connection part motion trajectory analysis channel;

[0069] The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped through the LSTM neural network topology, and the second connecting part motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the second connecting part motion trajectory analysis channel.

[0070] Collecting experimental data by means of wind turbine experimental equipment or historical data, the pin connection stability experimental data includes wind turbine operating parameters, pin connection positions and pin connection types, and is classified and sorted based on pin connection stability as an indicator;

[0071] The LSTM (Long Short Term Memory) neural network topology is used for training, the wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are used as input, and the pin motion reference trajectory experimental data of the pin connection stability experimental data are used as output supervision data of the model to train the pin motion trajectory analysis channel;

[0072] Referring to the above process, the LSTM neural network topology can also be used for training, taking the wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data as input, and taking the first connection member motion reference trajectory experimental data of the pin connection stability experimental data as the output supervision data of the model, to train the first connection member motion trajectory analysis channel;

[0073] It can also be trained through the LSTM neural network topology, taking the wind turbine operating parameter experimental data of the stability experimental data, the pin connection position experimental data and the pin connection type experimental data as input, and the second connecting member motion reference trajectory experimental data of the pin connection stability experimental data as the output supervision data of the model, to train the second connecting member motion trajectory analysis channel.

[0074] The LSTM neural network topology is trained using historical experimental data to obtain the pin shaft motion trajectory analysis channel, the first connecting part motion trajectory analysis channel and the second connecting part motion trajectory analysis channel for different wind turbine models. In this way, in the subsequent monitoring process, the monitoring data can be quickly analyzed by calling these trained channels to obtain the motion characteristics of the pin shaft, which can improve the efficiency and accuracy of the system and can perform adaptive adjustments according to different wind turbine models.

[0075] Further, according to the historical motion abnormality coefficient of the pin shaft, a second motion abnormality coefficient is calculated, and the method of the present application includes:

[0076] According to the historical motion abnormality coefficient of the pin shaft, a motion abnormality coefficient historical change curve is constructed;

[0077] When the motion anomaly coefficient historical change curve is in an upward trend, constructing an upward rate change curve;

[0078] When the rising rate change curve is in a non-rising trend, the average rising rate is calculated, the trend of the motion abnormality coefficient historical change curve is predicted, and the abnormal triggering duration is generated and set as the second motion abnormality coefficient;

[0079] When the rising rate change curve is in an upward trend, the maximum rising rate is counted, and a trend prediction is performed on the motion abnormality coefficient historical change curve to generate an abnormal trigger duration, which is set as the second motion abnormality coefficient.

[0080] The time information corresponding to the historical motion abnormality coefficient of the pin is taken as the X-axis, and the historical motion abnormality coefficient of the pin is taken as the Y-axis to construct a historical change curve of the motion abnormality coefficient. The historical change curve of the motion abnormality coefficient can reflect the historical changes of the motion abnormality coefficient of the pin; this is achieved by calculating the change rate of the historical motion abnormality coefficient of the pin. Specifically, the difference between the historical motion abnormality coefficient of the pin at two adjacent time points can be calculated, and this difference is used as the change in the rising rate.

[0081] When the motion abnormality coefficient historical change curve is in an upward trend, a rising rate change curve is constructed: on the motion abnormality coefficient historical change curve, for each time point, its rising rate compared with the previous time point is calculated, and a curve of this rising rate is drawn. The rising rate change curve can reflect the severity of the rising change of the abnormality coefficient.

[0082] Based on the collected historical motion anomaly coefficient of the pin shaft, the second motion anomaly coefficient is calculated, which involves time series analysis. Specifically, in the first case, when the rising rate change curve is in a non-rising trend (i.e., the rising speed slows down or tends to be stable), all rising rates are averaged to obtain an average rising rate. The average value of the rising rate change curve can reflect the average rising rate of the historical motion anomaly coefficient of the pin shaft. Then, the average rising rate is used to predict the trend of the motion anomaly coefficient historical change curve to predict possible abnormal coefficients in the future. The moment information corresponding to the predicted value and the current moment information are calculated to obtain the abnormal triggering duration, which is used as the second motion anomaly coefficient.

[0083] Based on the collected historical motion anomaly coefficient of the pin shaft, the second motion anomaly coefficient is calculated, which involves time series analysis. Specifically, in the second case, when the rising rate change curve is in an upward trend (that is, the rising speed is still increasing), the maximum rising rate is counted, and the maximum value of all rising rates is found. The maximum value found is the maximum rising rate. The maximum rising rate can reflect the maximum rising rate of the historical motion anomaly coefficient of the pin shaft. The maximum rising rate is used to perform trend prediction on the motion anomaly coefficient historical change curve, predict possible abnormal coefficients in the future, and calculate the moment information corresponding to the predicted abnormal coefficient and the current moment information, so as to obtain the abnormal triggering duration, and set it as the second motion anomaly coefficient.

[0084] It helps wind turbine maintenance personnel to better understand the changes in the abnormal coefficient of the pin shaft's historical motion, evaluate the stability of the pin shaft connection, and provide early warning of possible looseness or abnormal conditions so that maintenance personnel can take appropriate measures to deal with them and improve the operating efficiency and safety of the wind turbine.

[0085] In summary, the pin connection stability assessment method and system for wind turbines provided in the embodiments of the present application have the following technical effects:

[0086] Without interfering with the normal operation of the wind turbine, the stability of the pin connection is monitored in real time, and a loose pin connection alarm label is generated to mark the pins that may have abnormalities. This can promptly detect and handle abnormal pin connections and avoid damage to the wind turbine.

[0087] The motion feature monitoring component, the motion reference parsing component and the motion feature verification component are used to automatically perform monitoring and parsing, thereby reducing the cost of manual inspection and measurement, and also improving the accuracy and reliability of detection.

[0088] Real-time monitoring and early warning of the stability of the pin connection can promptly detect and handle abnormal situations, thereby avoiding damage to the wind turbine and improving the operating efficiency and safety of the wind turbine.

[0089] It is suitable for different types of wind turbines and pin connections, and is also easy to upgrade and maintain, with good scalability.

[0090] The method uses the historical motion abnormality coefficient of the pin shaft to construct a motion abnormality coefficient historical change curve; when the motion abnormality coefficient historical change curve is in an upward trend, a rising rate change curve is constructed; when the rising rate change curve is in a non-rising trend, the average rising rate is counted, the motion abnormality coefficient historical change curve is trend predicted, and the abnormal trigger duration is generated and set as the second motion abnormality coefficient; when the rising rate change curve is in an upward trend, the maximum rising rate is counted, the motion abnormality coefficient historical change curve is trend predicted, and the abnormal trigger duration is generated and set as the second motion abnormality coefficient. It helps wind turbine maintenance personnel better understand the changes in the historical motion abnormality coefficient of the pin shaft, evaluate the stability of the pin shaft connection, and warn of possible looseness or abnormal conditions in advance, so that maintenance personnel can take corresponding measures to deal with them, thereby improving the operating efficiency and safety of wind turbines. Example 2

[0091] Based on the same inventive concept as the method for evaluating the stability of the pin connection of a wind turbine generator set in the aforementioned embodiment, Figure 3 As shown, the embodiment of the present application provides a pin connection stability assessment system for a wind turbine generator set, wherein the system includes a motion feature monitoring component, a motion reference parsing component and a motion feature verification component, and the system includes:

[0092] A basic information acquisition module 100 is used to obtain basic information of the pin to be monitored, wherein the basic information of the pin to be monitored includes a pin connection position and a pin connection type;

[0093] A positioning chip deployment module 200, in which a micro positioning chip is deployed on the surface of the pin to be monitored;

[0094] The positioning chip monitoring module 300 is used to activate the motion characteristic monitoring component, start the wind turbine, and monitor the pin shaft monitoring motion characteristics of the pin shaft to be monitored based on the micro positioning chip;

[0095] The motion feature analysis module 400 is used to activate the motion reference analysis component, perform motion feature analysis on the pin connection position and the pin connection type based on the wind turbine model and the wind turbine operation parameters, and generate a pin reference motion feature;

[0096] A motion feature verification module 500 is used to activate a motion feature verification component to verify the pin shaft monitoring motion feature and the pin shaft reference motion feature to generate a first motion abnormality coefficient;

[0097] The loose connection monitoring module 600 is used to generate a pin shaft loose connection alarm label and mark the pin shaft to be monitored when the first motion abnormality coefficient is greater than the first motion abnormality coefficient threshold.

[0098] Furthermore, the system also includes:

[0099] When the first motion abnormality coefficient is less than or equal to the first motion abnormality coefficient threshold, generating a pin-connected quasi-stable label;

[0100] When generating the pin connection quasi-stable label, based on the pin connection position, the pin connection type and the wind turbine operating parameters, the pin historical motion anomaly coefficient is collected;

[0101] Calculating a second motion abnormality coefficient according to the pin shaft historical motion abnormality coefficient;

[0102] When the second motion abnormality coefficient is greater than the second motion abnormality coefficient threshold, a pin shaft connection loosening warning label is generated to mark the pin shaft to be monitored;

[0103] When the second motion abnormality coefficient is less than or equal to the second motion abnormality coefficient threshold, a pin connection stability label is generated to identify the pin to be monitored.

[0104] Furthermore, the system also includes:

[0105] Extracting a connection area according to the pin to be monitored, wherein the connection area includes a pin area, a first connection part area, and a second connection area;

[0106] The micro positioning chip is disposed in the pin region, the first connector region and the second connector region.

[0107] Furthermore, the system also includes:

[0108] Activate the motion reference analysis component, and retrieve the motion feature analysis node based on the wind turbine model, wherein the motion feature analysis node includes a pin motion trajectory analysis channel, a first connector motion trajectory analysis channel, and a second connector motion trajectory analysis channel;

[0109] Retrieving the pin shaft motion trajectory analysis channel, mapping the wind turbine operating parameters, the pin shaft connection position and the pin shaft connection type, and generating a pin shaft motion reference trajectory;

[0110] Retrieving the first connector motion trajectory analysis channel, mapping the wind turbine operating parameters, the pin connection position and the pin connection type, and generating a first connector motion reference trajectory;

[0111] Retrieving the motion trajectory analysis channel of the second connector, mapping the wind turbine operating parameters, the pin connection position and the pin connection type, and generating a second connector motion reference trajectory;

[0112] The pin shaft motion reference trajectory, the first connecting member motion reference trajectory and the second connecting member motion reference trajectory are added to the pin shaft reference motion feature.

[0113] Furthermore, the system also includes:

[0114] Counting a first relative distance interval according to the pin shaft motion reference trajectory and the first connecting member motion reference trajectory;

[0115] Counting a second relative distance interval according to the reference trajectory of the pin movement and the reference trajectory of the second connecting member movement;

[0116] Calculating a first frequency proportional coefficient that does not belong to the first relative distance interval according to the pin shaft monitoring motion characteristics;

[0117] Calculating a second frequency proportional coefficient that does not belong to the second relative distance interval according to the pin shaft monitoring motion characteristics;

[0118] The first frequency proportional coefficient and the second frequency proportional coefficient are added to the first motion abnormality coefficient.

[0119] Furthermore, the system also includes:

[0120] Based on the preset wind turbine model, collect the experimental data of pin connection stability;

[0121] The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped by the LSTM neural network topology, and the pin motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the pin motion trajectory analysis channel;

[0122] The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped by the LSTM neural network topology, and the first connection part motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the first connection part motion trajectory analysis channel;

[0123] The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped through the LSTM neural network topology, and the second connecting part motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the second connecting part motion trajectory analysis channel.

[0124] Furthermore, the system also includes:

[0125] According to the historical motion abnormality coefficient of the pin shaft, a motion abnormality coefficient historical change curve is constructed;

[0126] When the motion anomaly coefficient historical change curve is in an upward trend, constructing an upward rate change curve;

[0127] When the rising rate change curve is in a non-rising trend, the average rising rate is calculated, the trend of the motion abnormality coefficient historical change curve is predicted, and the abnormal triggering duration is generated and set as the second motion abnormality coefficient;

[0128] When the rising rate change curve is in an upward trend, the maximum rising rate is counted, and a trend prediction is performed on the motion abnormality coefficient historical change curve to generate an abnormal trigger duration, which is set as the second motion abnormality coefficient.

[0129] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0130] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for evaluating the stability of a pin connection of a wind turbine generator system, characterized in that: The method is applied to a pin connection stability assessment system for a wind turbine generator set, the system comprising a motion feature monitoring component, a motion reference analysis component and a motion feature verification component, including: Obtaining basic information of the pin to be monitored, wherein the basic information of the pin to be monitored includes a pin connection position and a pin connection type; Among them, a micro positioning chip is deployed on the surface of the pin to be monitored; Activate the motion characteristic monitoring component, start the wind turbine, and monitor the pin shaft monitoring motion characteristics of the pin shaft to be monitored based on the micro positioning chip; Activate the motion reference analysis component, perform motion feature analysis on the pin connection position and the pin connection type based on the wind turbine model and the wind turbine operation parameters, and generate the pin reference motion feature, including: Activate the motion reference analysis component, and retrieve the motion feature analysis node based on the wind turbine model, wherein the motion feature analysis node includes a pin motion trajectory analysis channel, a first connector motion trajectory analysis channel, and a second connector motion trajectory analysis channel; Retrieving the pin shaft motion trajectory analysis channel, mapping the wind turbine operating parameters, the pin shaft connection position and the pin shaft connection type, and generating a pin shaft motion reference trajectory; Retrieving the first connector motion trajectory analysis channel, mapping the wind turbine operating parameters, the pin connection position and the pin connection type, and generating a first connector motion reference trajectory; Retrieving the motion trajectory analysis channel of the second connector, mapping the wind turbine operating parameters, the pin connection position and the pin connection type, and generating a second connector motion reference trajectory; Adding the pin shaft motion reference trajectory, the first connecting member motion reference trajectory and the second connecting member motion reference trajectory into the pin shaft reference motion feature; Activate the motion feature verification component to verify the pin shaft monitoring motion feature and the pin shaft reference motion feature to generate a first motion anomaly coefficient; When the first motion abnormality coefficient is greater than the first motion abnormality coefficient threshold, a pin shaft connection loosening alarm label is generated to mark the pin shaft to be monitored; When the first motion abnormality coefficient is less than or equal to the first motion abnormality coefficient threshold, generating a pin-connected quasi-stable label; When generating the pin connection quasi-stable label, based on the pin connection position, the pin connection type and the wind turbine operating parameters, the pin historical motion anomaly coefficient is collected; Calculating a second motion abnormality coefficient according to the pin shaft historical motion abnormality coefficient; When the second motion abnormality coefficient is greater than the second motion abnormality coefficient threshold, a pin shaft connection loosening warning label is generated to mark the pin shaft to be monitored; When the second motion abnormality coefficient is less than or equal to the second motion abnormality coefficient threshold, a pin connection stability label is generated to identify the pin to be monitored.

2. The method according to claim 1, characterized in that The pin surface to be monitored is equipped with a micro positioning chip, including: Extracting a connection area according to the pin to be monitored, wherein the connection area includes a pin area, a first connection part area, and a second connection area; The micro positioning chip is disposed in the pin region, the first connector region and the second connector region.

3. The method according to claim 1, characterized in that Activating the motion feature verification component to verify the pin shaft monitoring motion feature and the pin shaft reference motion feature to generate a first motion abnormality coefficient, including: Counting a first relative distance interval according to the pin shaft motion reference trajectory and the first connecting member motion reference trajectory; Counting a second relative distance interval according to the reference trajectory of the pin movement and the reference trajectory of the second connecting member movement; Calculating a first frequency proportional coefficient that does not belong to the first relative distance interval according to the pin shaft monitoring motion characteristics; Calculating a second frequency proportional coefficient that does not belong to the second relative distance interval according to the pin shaft monitoring motion characteristics; The first frequency proportional coefficient and the second frequency proportional coefficient are added to the first motion abnormality coefficient.

4. The method according to claim 1, characterized in that Activate the motion benchmark analysis component, call the motion feature analysis node based on the wind turbine model, which previously included: Based on the preset wind turbine model, collect the experimental data of pin connection stability; The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped by the LSTM neural network topology, and the pin motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the pin motion trajectory analysis channel; The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped by the LSTM neural network topology, and the first connection part motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the first connection part motion trajectory analysis channel; The wind turbine operating parameter experimental data, the pin connection position experimental data and the pin connection type experimental data of the pin connection stability experimental data are mapped through the LSTM neural network topology, and the second connecting part motion reference trajectory experimental data of the pin connection stability experimental data is used as supervision to train the second connecting part motion trajectory analysis channel.

5. The method according to claim 1, characterized in that Calculating a second motion abnormality coefficient according to the pin shaft historical motion abnormality coefficient includes: According to the historical motion abnormality coefficient of the pin shaft, a motion abnormality coefficient historical change curve is constructed; When the motion anomaly coefficient historical change curve is in an upward trend, constructing an upward rate change curve; When the rising rate change curve is in a non-rising trend, the average rising rate is calculated, the trend of the motion abnormality coefficient historical change curve is predicted, and the abnormal triggering duration is generated and set as the second motion abnormality coefficient; When the rising rate change curve is in an upward trend, the maximum rising rate is counted, and a trend prediction is performed on the motion abnormality coefficient historical change curve to generate an abnormal trigger duration, which is set as the second motion abnormality coefficient.

6. A wind turbine pin connection stability assessment system, characterized in that: The method for evaluating the stability of the pin connection of a wind turbine set according to any one of claims 1 to 5 is used to implement the system, which includes a motion feature monitoring component, a motion reference analysis component and a motion feature verification component, including: A basic information acquisition module, used to obtain basic information of the pin to be monitored, wherein the basic information of the pin to be monitored includes a pin connection position and a pin connection type; A positioning chip deployment module, in which a micro positioning chip is deployed on the surface of the pin to be monitored; A positioning chip monitoring module is used to activate the motion characteristic monitoring component, start the wind turbine, and monitor the pin shaft monitoring motion characteristics of the pin shaft to be monitored based on the micro positioning chip; A motion feature analysis module is used to activate the motion reference analysis component, perform motion feature analysis on the pin connection position and the pin connection type based on the wind turbine model and the wind turbine operation parameters, and generate a pin reference motion feature; A motion feature verification module, used for activating a motion feature verification component to verify the pin shaft monitoring motion feature and the pin shaft reference motion feature, and generate a first motion abnormality coefficient; The loose connection monitoring module is used to generate a pin shaft loose connection alarm label and mark the pin shaft to be monitored when the first motion abnormality coefficient is greater than the first motion abnormality coefficient threshold.

Images