Whole machine behavior diagnosis method, device, controller, wind farm, system and medium

By receiving and diagnosing the whole machine behavior data of the wind turbine unit in real time in the wind farm, the problems of delay and inaccurate diagnosis in the wind farm are solved, and efficient and accurate whole machine behavior diagnosis in the wind farm is achieved. .

CN114647223BActive Publication Date: 2024-12-06GOLDWIND SCI & TECH CO LTD
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202011501648.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-12-18
Publication Date
2024-12-06
Estimated Expiration
2040-12-18

AI Technical Summary

Technical Problem

Due to the disconnection of the network from the server in the wind farm, the entire behavior data of the wind turbine unit cannot be uploaded in real time, resulting in inaccurate and delayed diagnosis results.

Method used

Set up a wind farm controller in the wind farm to receive the entire machine behavior data of the wind turbine through a network protocol and perform real-time diagnosis. There is no need to upload data to the server, and store and process finer granular data to improve diagnostic accuracy.

Benefits of technology

Real-time diagnosis of wind turbines is realized, the accuracy and response speed of diagnosis results are improved, and the real-time and intelligent needs of wind farms are met.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114647223B_ABST
    Figure CN114647223B_ABST
Patent Text Reader

Abstract

The present application discloses a whole machine behavior diagnosis method, device, controller, wind farm, system and medium. The method is applied to a wind farm controller set in a wind farm. The method includes: receiving whole machine behavior data sent by a target wind turbine generator set in the wind farm through a network protocol; determining whole machine behavior characteristic information of the target wind turbine generator set in each operation stage according to the whole machine behavior data of the target wind turbine generator set in each operation stage; for each operation stage, diagnosing the whole machine behavior of the target wind turbine generator set in the operation stage according to the whole machine behavior characteristic information in the operation stage and the first reference characteristic information, and obtaining the whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage. The method disclosed in the present application can perform whole machine behavior diagnosis on the target wind turbine generator set in real time, and the diagnosis result is more accurate.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of wind power generation technology, and in particular relates to a whole machine behavior diagnosis method, device, controller, wind farm, system and medium. Background Art

[0002] While the wind power industry is booming, it is also facing the situation of wind farm failures. In order to reduce the economic losses caused by wind farm failures, it is very important to timely diagnose the behavior of wind turbines in wind farms.

[0003] In the related art, the whole machine behavior data of the wind turbine generator set in the wind farm is uploaded to a server, and the server analyzes the whole machine behavior data to perform whole machine behavior diagnosis on the wind turbine generator set.

[0004] However, since there is no network connection between the wind farm and the server, the wind farm cannot upload the whole machine behavior data of the wind turbine to the server in real time, which in turn makes it impossible for the server to perform whole machine behavior diagnosis on the wind turbine in a timely manner. In addition, the server performs diagnosis based on the coarse-grained whole machine behavior data of the wind turbine, which results in inaccurate results for the whole machine behavior diagnosis of the wind turbine. Summary of the invention

[0005] The embodiments of the present application provide a whole-machine behavior diagnosis method, device, controller, wind farm, system and medium, which can solve the technical problems of being unable to timely perform whole-machine behavior diagnosis on a wind turbine generator set and inaccurate diagnosis results.

[0006] On the one hand, an embodiment of the present application provides a method for diagnosing the behavior of a wind turbine generator set, which is applied to a wind farm controller provided in a wind farm, and the method includes:

[0007] Receiving whole-machine behavior data sent by a target wind turbine generator set in the wind farm through a network protocol, wherein the whole-machine behavior data is whole-machine behavior data of the target wind turbine generator set in at least one operation stage;

[0008] Determining whole machine behavior characteristic information of the target wind turbine generator set at each of the operation stages according to the whole machine behavior data of the target wind turbine generator set at each of the operation stages;

[0009] For each of the operating stages, the whole machine behavior of the target wind turbine generator set in the operating stage is diagnosed based on the whole machine behavior characteristic information and the first reference characteristic information in the operating stage to obtain a whole machine behavior diagnosis result of the target wind turbine generator set in the operating stage.

[0010] On the other hand, an embodiment of the present application provides a device for diagnosing the behavior of a wind turbine generator set, which is applied to a wind farm controller arranged in a wind farm, and the device includes:

[0011] A first receiving module is used to receive the whole machine behavior data sent by the target wind turbine generator set in the wind farm through a network protocol, wherein the whole machine behavior data is the whole machine behavior data of the target wind turbine generator set in at least one operation stage;

[0012] A first determination module is used to determine the whole machine behavior characteristic information of the target wind turbine generator set at each of the operation stages according to the whole machine behavior data of the target wind turbine generator set at each of the operation stages;

[0013] The unit diagnosis module is used to diagnose the whole machine behavior of the target wind turbine generator set in the operation stage for each of the operation stages according to the whole machine behavior characteristic information and the first reference characteristic information in the operation stage, so as to obtain the whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage.

[0014] In yet another aspect, an embodiment of the present application provides a wind farm controller, comprising: a processor and a memory storing computer program instructions;

[0015] When the processor executes the computer program instructions, the whole machine behavior diagnosis method of the wind turbine generator set described in any one of the above items is implemented.

[0016] In another aspect, an embodiment of the present application provides a wind farm, including:

[0017] Target wind turbines;

[0018] A wind farm controller as described in any one of the above items, used for diagnosing the overall behavior of the target wind turbine generator set.

[0019] On the other hand, the embodiment of the present application provides a whole machine behavior diagnosis system, including:

[0020] At least two of the above-mentioned wind farms, each of which is used to determine the whole machine behavior characteristic information of the target wind turbine generator set according to the whole machine behavior data of the target wind turbine generator set in the wind farm, and send the whole machine behavior characteristic information to the server;

[0021] The server is configured to determine a second reference whole machine behavior feature based on whole machine behavior feature information of the target wind turbine generator sets having the target configuration in at least two of the wind farms, and send the second reference whole machine behavior feature to each of the wind farms;

[0022] Each of the wind farms is further configured to update the first reference characteristic information corresponding to the target wind turbine generator set having the target configuration in the wind farm according to the second reference whole-machine behavior characteristic.

[0023] On the other hand, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, any one of the above-mentioned methods for diagnosing the overall behavior of a wind turbine generator set is implemented.

[0024] The whole machine behavior diagnosis method, device, controller, wind farm, system and medium of the embodiment of the present application diagnoses the whole machine behavior of the target wind turbine generator set through the wind farm controller in the wind farm, and there is no need to upload the whole machine behavior data of the target wind turbine generator set to the server. Since the wind farm controller in the wind farm can obtain the whole machine behavior data of the target wind turbine generator set in real time, the whole machine behavior diagnosis of the target wind turbine generator set can be performed in real time. In addition, the wind farm only needs to store and process the data of the wind turbine generator set in the wind farm, without the need to store and process the data of the wind turbine generator sets in other wind farms. In this way, the pressure of storing and processing data on the wind farm is smaller than that on the server. Therefore, the wind farm can store and process finer granularity data of the wind turbine generator set, so that the wind farm controller performs whole machine behavior diagnosis based on finer granularity data, thereby making the result of whole machine behavior diagnosis more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A flow chart of an embodiment of the whole machine behavior diagnosis method provided in the present application is shown.

[0027] Figure 2 A flow chart of an embodiment of the whole machine behavior diagnosis process provided by the present application is shown.

[0028] Figure 3 A flow chart of an embodiment of a pitch behavior analysis process during the startup phase provided by the present application is shown.

[0029] Figure 4 A flow chart of an embodiment of a pitch behavior analysis process during the startup phase provided by the present application is shown.

[0030] Figure 5A flow chart showing an embodiment of a characteristic curve analysis process provided by the present application.

[0031] Figure 6 A flow chart of another embodiment of the whole machine behavior diagnosis method of a wind turbine generator set provided in the present application is shown.

[0032] Figure 7 A flow chart of an embodiment of a whole-machine behavior diagnosis system for a wind turbine generator set provided by the present application is shown.

[0033] Figure 8 A flow chart of another embodiment of the whole machine behavior diagnosis system of the wind turbine generator set provided by the present application is shown.

[0034] Fig. 9 A flow chart of another embodiment of the whole machine behavior diagnosis method of a wind turbine generator set provided by the present application is shown.

[0035] Fig.10 A schematic diagram of an embodiment of a source of abnormal whole machine behavior provided by the present application is shown.

[0036] Fig.11 A structural schematic diagram of an embodiment of a whole-machine behavior diagnosis device for a wind turbine generator set provided in the present application is shown.

[0037] Fig.12 A schematic diagram of the hardware structure of an embodiment of a wind farm controller provided by the present application is shown. DETAILED DESCRIPTION

[0038] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0039] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0040] Before describing the embodiments of the present application, the professional terms involved in the embodiments of the present application are first described.

[0041] Wind Farm Controller (WFC) is a hardware carrier that implements cluster control decisions for wind turbines on the wind farm side. It consists of two parts: a real-time core part and a non-real-time core part.

[0042] Decision Module (DM) refers to a data operation decision module that complies with the development guideline. Its function is to output decision information such as control and data display of wind turbines through data operation, so as to achieve functions such as increasing the power generation of wind farms, health warning of wind turbines, grid friendliness, and safety protection. It can be deployed in real-time cores, non-real-time cores or other peripheral devices.

[0043] Turbine Behavior Diagnosis: Turbine behavior can be any execution action of the wind turbine in a broad sense. For the control strategy, the turbine behavior diagnosis is mainly for the wind turbine pitch control, torque setting related actions and execution result evaluation. The turbine behavior diagnosis content can include at least one of the following: pitch control and torque execution during startup and shutdown; control signal execution boundary under power generation state; control signal fluctuation under power generation state; control signal response under power generation state; basic characteristic curve of the unit under power generation state.

[0044] After explaining the above technical terms, the embodiments of the present application are described.

[0045] The whole-machine behavior diagnosis of a wind turbine can reflect the operating characteristics of the unit, such as the control execution state, control stability, and characteristic curve quality. In addition, if the pitch and torque behavior of a wind turbine is abnormal, the overall control performance, control stability, and characteristic curve of the wind turbine will be affected, which will cause the actual load and design load deviation of the wind turbine to increase, the power generation performance to decrease, and the long-term life of the unit to decrease. Therefore, the diagnosis of the whole-machine behavior of a wind turbine is very important. The whole-machine behavior diagnosis of a wind turbine has always been a difficult point in the operation of wind farm units.

[0046] A large amount of millisecond-level whole-machine behavior data is generated in the daily operation of wind turbines in wind farms, and these data often contain important information. Therefore, the whole-machine behavior of the wind turbine can be diagnosed based on the whole-machine behavior data. In related technologies, the whole-machine behavior diagnosis process of wind turbines is generally performed manually. Specifically, engineers manually collect the whole-machine behavior data of wind turbines on site and send it to the server. The server analyzes the whole-machine behavior data and determines the state and operating characteristics of the wind turbine to diagnose the whole-machine behavior of the wind turbine.

[0047] However, for the sake of network security of wind farms, there is no network connection between wind farms and servers, so engineers need to copy the whole machine behavior data of wind turbines from wind farms first, and then upload it to the server. However, uploading the whole machine behavior data to the server in the above offline way will result in the inability to upload the whole machine behavior data of wind turbines in real time, which in turn makes it impossible for the server to perform abnormal diagnosis of wind turbines in a timely manner.

[0048] In addition, since the server stores and processes the data of wind turbines in multiple wind farms, and considering the limited storage and processing resources of the server, before the wind farm uploads the whole behavior data of the wind turbine to the server, it is necessary to convert the fine-grained data (such as millisecond-level data) into coarse-grained data (such as second-level data), so as to reduce the storage and processing pressure of the server. However, this will result in inaccurate results of the abnormal diagnosis of the wind turbine by the server.

[0049] In order to solve the above technical problems, the embodiments of the present application provide a whole machine behavior diagnosis method for a wind turbine generator set (hereinafter referred to as the whole machine behavior diagnosis method), a device, a controller, a wind farm, a system and a medium. The whole machine behavior diagnosis method provided by the embodiments of the present application is first introduced below.

[0050] Figure 1 The flowchart of an embodiment of the whole machine behavior diagnosis method provided by the present application is shown. The whole machine behavior diagnosis method is applied to a wind farm controller arranged in a wind farm, such as Figure 1As shown, the whole machine behavior diagnosis method includes:

[0051] S102, receiving whole-machine behavior data sent by a target wind turbine generator set in a wind farm through a network protocol, where the whole-machine behavior data is whole-machine behavior data of the target wind turbine generator set in at least one operation stage.

[0052] In S102, when the target wind turbine generator set in the wind farm generates whole machine behavior data during operation, the target wind turbine generator set can send the whole machine behavior data generated in real time to the wind farm controller in the wind farm through the network protocol. In this way, the wind farm controller can receive the whole machine behavior data generated in real time by the target wind turbine generator set through the network protocol, so as to diagnose the whole machine behavior of the target wind turbine generator set in real time.

[0053] As an example, the whole machine behavior data may include at least one of the following: instantaneous data such as wind turbine status, wind speed, wind deviation, power, feedback blade angle, given blade angle, feedback torque, given torque, air density, turbulence, generator speed, etc. of the target wind turbine generator set within a predetermined time period.

[0054] After S102, the whole machine behavior diagnosis method further includes:

[0055] S104, determining whole machine behavior characteristic information of the target wind turbine generator set at each operation stage according to the whole machine behavior data of the target wind turbine generator set at each operation stage.

[0056] In S104, as an example, the whole machine behavior characteristic information of the target wind turbine generator set in each operation stage can be determined in sequence. For example, the whole machine behavior characteristic information of the startup stage is first determined based on the whole machine behavior data of the target wind turbine generator set in the startup stage; then, the whole machine behavior characteristic information of the shutdown stage is determined based on the whole machine behavior data of the target wind turbine generator set in the shutdown stage; finally, the whole machine behavior characteristic information of the power generation stage is determined based on the whole machine behavior data of the target wind turbine generator set in the power generation stage.

[0057] The whole machine behavior characteristic information of each operation stage can be determined after the whole machine behavior data of the target wind turbine generator set in each operation stage is obtained. Alternatively, the whole machine behavior characteristic information of an operation stage can be determined after the whole machine behavior data in an operation stage is obtained.

[0058] As another example, after obtaining the whole machine behavior data of the target wind turbine generator set in each operating stage, the whole machine behavior data in each operating stage can be processed concurrently, so that the whole machine behavior characteristic information of the target wind turbine generator set in each operating stage can be determined simultaneously.

[0059] In addition, if the whole machine behavior data of the target wind turbine generator set in multiple operating stages is obtained, for example, the whole machine behavior data of the target wind turbine generator set in the standby stage, maintenance stage, startup stage, shutdown stage, and power generation stage is obtained, but the whole machine behavior data is not grouped according to the operating stage, that is, the whole machine behavior data of the target wind turbine generator set in multiple operating stages are mixed together. Therefore, in one example, the whole machine behavior data of the target wind turbine generator set is grouped according to the operating stage of the target wind turbine generator set to distinguish the whole machine behavior data of the target wind turbine generator set in each operating stage, so as to determine the whole machine behavior characteristic information in the corresponding operating stage according to the whole machine behavior data in each operating stage.

[0060] The whole machine behavior diagnosis method also includes:

[0061] S106, for each operation stage, diagnose the whole machine behavior of the target wind turbine generator set in the operation stage according to the whole machine behavior characteristic information in the operation stage and the first reference characteristic information, and obtain the whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage.

[0062] In S106, the wind farm controller performs a correlation analysis on the whole machine behavior characteristic information of the target wind turbine generator set at each operation stage to obtain the whole machine behavior diagnosis result of the target wind turbine generator set at each operation stage.

[0063] Specifically, the wind farm controller can compare the whole machine behavior characteristic information in each operation stage with the first reference characteristic information in the corresponding operation stage, and compare the difference between the whole machine behavior characteristic information and the first reference characteristic information, to obtain the whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage. If the difference between the whole machine behavior characteristic information and the first reference characteristic information is relatively large, it can be diagnosed that the target wind turbine generator set is abnormal in the operation stage; if the difference between the whole machine behavior characteristic information and the first reference characteristic information is relatively small, it can be diagnosed that the target wind turbine generator set is not abnormal in the operation stage.

[0064] In an embodiment of the present application, the whole machine behavior of the target wind turbine generator set is diagnosed by the wind farm controller in the wind farm, and there is no need to upload the whole machine behavior data of the target wind turbine generator set to the server. Since the wind farm controller in the wind farm can obtain the whole machine behavior data of the target wind turbine generator set in real time, the whole machine behavior of the target wind turbine generator set can be diagnosed in real time. In addition, the wind farm only needs to store and process the data of the wind turbine generator sets in the wind farm, without the need to store and process the data of the wind turbine generator sets in other wind farms. In this way, the pressure of storing and processing data on the wind farm is smaller than that on the server. Therefore, the wind farm can store and process finer granularity data of the wind turbine generator set, so that the wind farm controller can perform whole machine behavior diagnosis based on finer granularity data, thereby making the result of whole machine behavior diagnosis more accurate.

[0065] Secondly, the wind farm controller in the wind farm provides the whole machine behavior diagnosis service of the wind turbine generator set nearby. Compared with the diagnosis on the server side, the diagnosis on the wind farm side can produce faster business response and meet the needs of the wind farm and wind turbine generator set in terms of real-time, intelligence, network security, etc. At the same time, the server side can still access the wind farm controller's diagnosis results of the whole machine behavior of the wind turbine generator set.

[0066] In one or more embodiments of the present application, Figure 2 As shown, S104 may include:

[0067] S1042, behavior analysis during the startup phase, specifically, determining a whole-machine behavior characteristic value of the target wind turbine generator set during the startup phase according to the whole-machine behavior data of the target wind turbine generator set during the startup phase;

[0068] S1044, after determining the whole machine behavior characteristic value of the target wind turbine generator set in the startup phase, analyzing the behavior in the shutdown phase, specifically, determining the whole machine behavior characteristic value of the target wind turbine generator set in the shutdown phase according to the whole machine behavior data of the target wind turbine generator set in the shutdown phase;

[0069] S1046, after determining the whole-machine behavior characteristic values ​​of the target wind turbine generator set in the shutdown stage, conduct behavior analysis and characteristic curve analysis in the power generation stage. Specifically, based on the whole-machine behavior data of the target wind turbine generator set in the power generation stage, determine the whole-machine behavior characteristic values ​​and whole-machine behavior curve of the target wind turbine generator set in the power generation stage.

[0070] In an embodiment of the present application, the whole machine behavior data of the target wind turbine generator set at each operating stage is analyzed in turn to obtain the whole machine behavior characteristic information of the target wind turbine generator set at this stage, so as to facilitate the diagnosis of the whole machine behavior of the target wind turbine generator set at this stage.

[0071] The whole machine behavior characteristic values ​​at each stage may include pitch execution dynamic deviation, pitch system response, pitch execution static deviation, torque execution dynamic deviation, torque execution static deviation, and variable current system response. The whole machine behavior curve may include at least one of the following: a characteristic curve and a power curve.

[0072] The whole machine behavior characteristic information of the target wind turbine generator set in each operation stage may include the whole machine behavior characteristic values ​​of the three blades of the target wind turbine generator set. For example, the whole machine behavior characteristic information of the target wind turbine generator set in the startup stage includes the dynamic deviation of the pitch execution of the three blades of the target wind turbine generator set.

[0073] In one or more embodiments of the present application, S106 may include:

[0074] For each operation stage, obtaining a characteristic extreme value among a maximum value and a minimum value from a plurality of whole machine behavior characteristic values ​​of the target wind turbine generator set in the operation stage, wherein obtaining the maximum value or the minimum value is determined by the operation stage;

[0075] A diagnosis result of the target wind turbine generator set in the operation stage is generated according to the characteristic extreme value and the reference characteristic value in the first reference characteristic information.

[0076] As an example, generating a diagnosis result of the target wind turbine generator set at the operation stage according to the characteristic extreme value and the reference characteristic value in the first reference characteristic information may include:

[0077] Calculate the difference between the characteristic extreme value and the reference characteristic value in the first reference characteristic information;

[0078] According to the magnitude relationship between the difference value and the first preset difference value, a diagnosis result of the target wind turbine generator set in the operation stage is generated.

[0079] In the embodiment of the present application, by calculating the difference between the characteristic extreme value of the target wind turbine generator set and the reference characteristic value, a diagnosis result of the target wind turbine generator set in the operating stage is generated, thereby achieving a diagnosis of the overall machine behavior of the target wind turbine generator set in the operating stage.

[0080] The following is an exemplary description of the embodiments of the present application.

[0081] In one example, the above-mentioned multiple whole-machine behavior characteristic values ​​may be the whole-machine behavior characteristic values ​​of the target wind turbine generator set at a certain moment. For example, the above-mentioned multiple whole-machine behavior characteristic values ​​may be the pitch change rate of the three blades of the target wind turbine generator set. Then, the characteristic extreme value is obtained from the multiple whole-machine behavior characteristic values, and the difference between the characteristic extreme value and the reference characteristic value is calculated. When the difference is greater than the first preset difference, an abnormal diagnosis result of the target wind turbine generator set in the operation stage can be generated; when the difference is less than the first preset difference, a normal diagnosis result of the target wind turbine generator set in the operation stage can be generated.

[0082] The following uses the startup phase as an example to illustrate the analysis process of the pitch change behavior of the target wind turbine generator set during the startup phase. The target wind turbine generator set generally limits the pitch change rate, considering that the target wind turbine generator set generally does not evaluate the pitch change rate during the startup phase. Moreover, considering the data transmission and storage costs on the server side, the server side obtains second-level data, which is a coarse-grained evaluation. Some instantaneous pitch change rate spikes may be difficult to detect and record. The pitch change behavior analysis in the whole machine behavior diagnosis on the wind farm side provides a new problem solution.

[0083] Figure 3 FIG. 1 is a flow chart of an embodiment of the pitch behavior analysis process during the startup phase provided by the present application. Figure 3 As shown, the process of analyzing the pitch behavior during the startup phase may include:

[0084] S202, obtaining whole machine behavior data of the target wind turbine generator set;

[0085] S204, filtering out the whole machine behavior data in the startup phase from the acquired whole machine behavior data;

[0086] S206, verifying and cleaning the whole machine behavior data in the startup phase;

[0087] S208, obtaining the pitch rates of three blades of the target wind turbine generator set from the whole machine behavior data in the startup phase, and then obtaining the pitch rate extreme value from the pitch rates of the three blades;

[0088] S210, based on the difference between the pitch rate extreme value and the reference characteristic value, generate a diagnostic result of the pitch behavior of the target wind turbine generator set during the startup phase. Specifically, when the difference is less than a first preset difference, a normal diagnostic result of the target wind turbine generator set during the startup phase is generated. When the difference is greater than the first preset difference, an abnormal diagnostic result of the target wind turbine generator set during the startup phase is generated, and a first warning information may also be generated to warn that the target wind turbine generator set is abnormal during the startup phase.

[0089] In this example, by evaluating the pitch rate during the startup phase, it is possible to diagnose abnormal pitch behavior caused by reasons such as mismatch of control parameters during the startup phase of the target wind turbine, abnormal rate control of the pitch system, abnormal execution of the pitch mechanism, and design defects of the unit. In addition, in addition to being used to diagnose the pitch behavior during the startup phase, the extreme value of the pitch rate can also be used to statistically classify data and provide data samples for correlation analysis and trend analysis. Moreover, this type of statistical data can not only be used as a feature extraction of real-time data of the unit, but can also be transmitted to the server through an authorized network or file for further analysis by the server. The solution of this example can form an efficient flow of information among wind turbines, wind farms, and servers, and provide a complete solution for optimizing and protecting the functions related to the overall behavior of wind turbines. In addition, the analysis process of the torque rate during the startup phase, the pitch change and torque rate during the shutdown phase is similar to this example, and will not be repeated here.

[0090] The following uses the power generation stage as an example to illustrate the analysis process of the pitch behavior of the target wind turbine in the power generation stage. Considering resource constraints, warnings are generally given for the maximum following error of the blade angle and the maximum difference between the blade angles. The relevant wind speed, turbulence, power, speed and other recorded information are scattered and difficult to extract effectively, making it difficult to form a closed-loop optimization of the threshold parameters. Moreover, considering the data transmission and storage costs on the server side, the server side obtains second-level data, which is a coarse-grained assessment. Some instantaneous pitch rate spikes may be difficult to detect and record. The pitch behavior analysis in the whole machine behavior diagnosis on the wind farm side provides a new solution to this problem.

[0091] Figure 4 FIG. 1 is a flow chart showing an embodiment of the pitch behavior analysis process in the power generation phase provided by the present application. Figure 4 As shown, the process of analyzing pitch behavior during power generation can include:

[0092] S302, obtaining whole machine behavior data of the target wind turbine generator set;

[0093] S304, filtering out the whole machine behavior data of the target wind turbine generator set in the power generation stage from the acquired whole machine behavior data;

[0094] S306, verifying and cleaning the whole machine behavior data in the power generation stage;

[0095] S308, obtaining the dynamic deviations of the blade angles of three blades of the target wind turbine generator set in a preset time period from the whole machine behavior data in the power generation stage, such as the deviation between the set blade angle of blade 1 and the actual feedback blade angle of blade 1, the deviation between the set blade angle of blade 2 and the actual feedback blade angle of blade 2, and the deviation between the set blade angle of blade 3 and the actual feedback blade angle of blade 3;

[0096] S310, obtaining a blade angle dynamic deviation extreme value from the blade angle dynamic deviations of the three blades, where the blade angle dynamic deviation extreme value is a maximum value or a minimum value;

[0097] S312, calculating a first difference between the extreme value of the blade angle dynamic deviation obtained in S310 and a reference blade angle dynamic deviation (the reference blade angle dynamic deviation is the reference characteristic value mentioned above);

[0098] S314, obtaining the difference between the blade angles of the three blades of the target wind turbine generator set in a preset time period from the whole machine behavior data in the power generation stage, such as the difference between the actual feedback blade angle of blade 1 and the actual feedback blade angle of blade 2, the difference between the actual feedback blade angle of blade 2 and the actual feedback blade angle of blade 3, and the difference between the actual feedback blade angle of blade 3 and the actual feedback blade angle of blade 1;

[0099] S316, obtaining an extreme value of the difference between the two blade angles from the two differences between the blade angles obtained in S314;

[0100] S318, calculating a second difference between the extreme value of the difference between the two blade angles obtained in S316 and a reference difference (the reference difference is the reference characteristic value mentioned above);

[0101] S320, based on the first difference and the second difference, generate a diagnostic result of the pitch change behavior in the power generation stage; specifically, if the first difference and the second difference are respectively smaller than the first preset difference, then generate a diagnostic result of the normal operation of the target wind turbine in the power generation stage; if at least one of the first difference and the second difference is larger than the first preset difference, then generate a first warning information of the target wind turbine in the power generation stage to warn of the abnormality of the target wind turbine in the power generation stage.

[0102] In this example, according to the dynamic deviations of the three blade pitch angles of the target wind turbine in the power generation stage and the difference between the two blade angles, the diagnosis results of the pitch behavior in the power generation stage are generated. In this way, the abnormal pitch behavior caused by abnormal control of the pitch system, abnormal execution of the pitch mechanism, and design defects of the unit can be diagnosed from multiple dimensions. In addition to being used to calculate the difference between the reference characteristic value, the extreme value of the dynamic deviation of the pitch angle can also be used as statistical classification data to provide data samples for correlation analysis and trend analysis. At the same time, this type of statistical data can be used as a feature extraction of the real-time data of the unit, and can also be transmitted to the server (the server can be a physical server or a cloud server) through an authorized network or file form for further analysis by the server. Through the scheme of this example, an efficient flow of information between the wind turbine, wind farm and server related to the whole machine behavior is formed, providing a complete solution for the optimization and protection of the functions related to the whole machine behavior of the wind turbine. In addition, the analysis process of the torque behavior in the power generation stage is similar to this example, and will not be repeated here.

[0103] In one or more embodiments of the present application, when the whole machine behavior characteristic information includes a whole machine behavior curve, S106 may include:

[0104] Obtain multiple discrete points on the whole machine behavior curve;

[0105] According to the discrete degree of multiple discrete points, the diagnostic result of the whole machine behavior curve is generated.

[0106] As an example, generating a diagnostic result of the whole machine behavior curve according to the discreteness of multiple discrete points may include:

[0107] Perform clustering on multiple discrete points to obtain at least one cluster;

[0108] The diagnostic result of the whole machine behavior curve is generated according to the size relationship between the number of clusters and the reference number in the first reference feature information (for example, the reference number is 1).

[0109] In the embodiment of the present application, if the target wind turbine generator set is normal, then when the discrete points on the whole machine behavior curve are clustered, the number of clusters is equal to or close to the reference number, and when the number of clusters is greater than the reference number, the curve can be considered abnormal. Therefore, the whole machine behavior of the target wind turbine generator set can be diagnosed according to the number of clusters during clustering.

[0110] The following is an exemplary description of how to generate a diagnostic result of the whole machine behavior curve according to the number of clusters and the reference number.

[0111] As an example, when the above-mentioned cluster number is greater than the reference number, the second warning information of the whole machine behavior curve is generated to warn that the target wind turbine generator set has an abnormality; when the above-mentioned cluster number is equal to the reference number, a diagnosis result that the whole machine behavior curve is normal is generated.

[0112] As another example, clustering is performed on multiple discrete points to obtain at least one cluster, including:

[0113] Divide the multiple discrete points into multiple groups of discrete points, the number of the multiple groups of discrete points is less than the number of the multiple discrete points, and the values ​​of the same group of discrete points on the predetermined dimension are within the same interval;

[0114] For at least one group of discrete points in the multiple groups of discrete points, each group of discrete points in the at least one group of discrete points is clustered respectively to obtain a cluster corresponding to each group of discrete points.

[0115] Based on the size relationship between the number of clusters and the reference number, the diagnostic results of the whole machine behavior curve are generated, including:

[0116] According to the number of clusters corresponding to each group of discrete points, a target group of discrete points is obtained from at least one group of discrete points, where the target group of discrete points is a group whose number of corresponding clusters is greater than a reference number;

[0117] When the number of discrete points in the target group is greater than the predetermined number of groups, generating second warning information of the whole machine behavior curve;

[0118] When the number of discrete points in the target group is not greater than the predetermined number of groups, a diagnosis result is generated that the behavior curve of the whole machine is normal.

[0119] In this example, multiple discrete points are divided into multiple groups of discrete points, and the discrete points of each group are clustered separately. If the target wind turbine generator set is normal, the number of clusters obtained by clustering the discrete points of each group is relatively small. Therefore, the overall behavior of the target wind turbine generator set can be diagnosed according to the number of clusters when the discrete points of each group are clustered.

[0120] In one or more examples, before clustering each group of discrete points in at least one group of discrete points to obtain a cluster corresponding to each group of discrete points, the whole machine behavior diagnosis method may further include:

[0121] Among the multiple groups of discrete points, at least one group of discrete points having a number of discrete points greater than a predetermined number is screened.

[0122] In this example, since the number of discrete points involved in clustering is relatively small, the clustering result may be inaccurate. Therefore, at least one group of discrete points with a relatively large number of discrete points is selected, and each group of selected discrete points is clustered separately. In this way, it can be ensured that the clustering result more accurately reflects the diagnosis result of the whole machine behavior curve.

[0123] The following uses the whole machine behavior curve as a characteristic curve of wind speed-power as an example to illustrate the analysis process of obtaining the whole machine behavior curve.

[0124] In the related art, the second-level data display of the whole machine behavior curve is manually performed, and the timeliness of the data on the server side is limited due to data transmission. At the same time, the wind turbine generator set is limited by the unit resources and cannot perform such functional analysis. Based on this, the present application can analyze the characteristic curve on the wind farm side.

[0125] Figure 5 FIG. 1 is a flow chart showing an embodiment of the characteristic curve analysis process provided by the present application. Figure 5 As shown, the process of characteristic curve analysis may include:

[0126] S402, obtaining whole machine behavior data;

[0127] S404, filtering out the whole machine behavior data in the power generation stage;

[0128] S406, verifying and cleaning the whole machine behavior data in the power generation stage;

[0129] S408, extracting multiple discrete points of wind speed and power from the whole machine behavior data, where the multiple discrete points may be part or all of all discrete points forming characteristic curves of wind speed and power;

[0130] S410, dividing the plurality of discrete points into bins according to the wind speed to form a plurality of groups of discrete points, for example, the discrete points with wind speeds in the interval from v1 to v2 are one group, and the discrete points with wind speeds in the interval from v2 to v3 are another group;

[0131] S412, screening out at least one group of discrete points whose number of discrete points meets a predetermined number requirement;

[0132] S414, clustering each group of discrete points respectively to obtain the number of clusters corresponding to each group of discrete points;

[0133] S416, comparing the number of clusters in the clustering results of each group of discrete points with the reference number, and evaluating the characteristic curve.

[0134] S416 may specifically include: based on the number of clusters corresponding to each group of discrete points, selecting target group discrete points whose number of clusters is greater than the reference number from at least one group of discrete points; then, when the number of groups of discrete points in the target group is greater than the predetermined number of groups, it means that the number of groups whose number of clusters is greater than the reference number is relatively large, that is, the discrete points are relatively scattered, and the characteristic curve is likely to be abnormal, and a second warning information of abnormal characteristic curve is generated. When the number of groups of discrete points in the target group is not greater than the predetermined number of groups, it means that the discrete points of each group are relatively concentrated, and further, it means that the characteristic curve is normal, and a diagnosis result of normal characteristic curve can be generated.

[0135] Through the above process from S402 to S416, timely and automatic identification of wind speed-power characteristic curve classification can be achieved. After receiving the warning information of the characteristic curve, the operation and maintenance personnel can promptly feedback the curve abnormality to the relevant departments and quickly check and eliminate similar operation hazards.

[0136] In one or more embodiments of the present application, after determining the whole machine behavior characteristic information of the target wind turbine generator set at each operation stage, the whole machine behavior diagnosis method may further include:

[0137] For the whole machine behavior characteristic information of the target wind turbine generator set at each operation stage, respectively: decomposing the whole machine behavior characteristic information at the operation stage to obtain decomposed characteristic values;

[0138] When the difference between the decomposed characteristic value and the preset characteristic value is greater than the second preset difference, third warning information indicating that the decomposed characteristic value is abnormal is generated.

[0139] In an embodiment of the present application, by decomposing the whole machine behavior characteristic information of the target wind turbine generator set at each operation stage, a decomposed characteristic value is obtained, and each decomposed characteristic value can characterize the operation characteristics of the target wind turbine generator set at the corresponding operation stage. As an example, the decomposed characteristic value can be wind speed, rotation speed, power, etc. When the difference between the decomposed characteristic value and the preset characteristic value is greater than the second preset difference, it means that the target wind turbine generator set is abnormal, and the cause of the abnormality of the target wind turbine generator set is the decomposed characteristic value. Then, a third warning information of abnormal decomposed characteristic value is generated, so that it is convenient for the staff to solve the abnormality of the target wind turbine generator set according to the decomposed characteristic value.

[0140] In one or more embodiments of the present application, after S104, the whole machine behavior diagnosis method may further include:

[0141] For the whole machine behavior characteristic information of the target wind turbine generator set in each operation stage, respectively: according to the whole machine behavior characteristic information in the operation stage, the whole machine behavior characteristic information of the target wind turbine generator set in the operation stage is predicted to obtain the predicted whole machine behavior characteristic information.

[0142] In the embodiment of the present application, by predicting the whole machine behavior characteristic information of the target wind turbine generator set in the operation stage, it is possible to discover in advance whether the target wind turbine generator set is abnormal, so that it is convenient for the staff to handle the abnormality of the target wind turbine generator set in advance.

[0143] It should be noted that at least two of the above embodiments can be combined to implement a method for diagnosing the behavior of a wind turbine generator set. The method for diagnosing the behavior of a wind turbine generator set is described below by way of example.

[0144] Figure 6 FIG. 2 is a flow chart showing another embodiment of the wind turbine generator set behavior diagnosis method provided by the present application. Figure 6 As shown, the wind turbine generator set overall behavior diagnosis method may include:

[0145] S502, determining the operation phase of the target wind turbine generator set, where the operation phase includes a startup phase, a shutdown phase, and a power generation phase;

[0146] S504, pitch change behavior analysis during the startup phase. The process of pitch change behavior analysis during the startup phase may be: Figure 3 The analytical process shown;

[0147] S506, torque behavior analysis during the startup phase. The torque behavior analysis during the startup phase may be similar to the pitch behavior analysis during the startup phase.

[0148] S508, pitch behavior analysis during the shutdown phase. The pitch behavior analysis during the shutdown phase may be similar to the pitch behavior analysis during the startup phase.

[0149] S510, torque behavior analysis during the shutdown phase. The torque behavior analysis during the shutdown phase may be similar to the torque behavior analysis during the startup phase.

[0150] S512, pitch behavior analysis during power generation. The process of pitch behavior analysis during power generation can be: Figure 4 The analytical process shown;

[0151] S514, torque behavior analysis in the power generation stage. The torque behavior analysis in the power generation stage may be similar to the pitch behavior analysis in the power generation stage.

[0152] S516, characteristic curve evaluation, the characteristic curve evaluation process can be Figure 5 The process shown;

[0153] S518, power curve performance diagnosis;

[0154] It should be noted that S504 to S518 diagnose the whole-machine behavior of the target wind turbine. On the one hand, it integrates the target wind turbine's pitch following detection, blade angle difference detection, and pitch rate and torque rate detection on the server side in the relevant technology into the wind farm side. On the other hand, it utilizes the characteristics of the wind farm end being close to the unit to obtain data in real time and having sufficient CPU and storage resources for on-site analysis and processing, and adds the previously coarse-grained or missing functions to the whole-machine behavior diagnosis function at the wind farm level, forming a multi-dimensional and refined systematic evaluation.

[0155] After S518, the wind turbine behavior diagnosis method also includes:

[0156] S520, statistically classifying the acquired whole machine behavior data to establish a whole machine behavior data sample, including whole machine behavior feature statistics, such as wind speed, wind direction, power, blade angle, air density, turbulence, and other features that may be related to the whole machine behavior;

[0157] S522, performing correlation analysis on the acquired whole machine behavior data. Specifically, the whole machine behavior feature information can be decomposed by a machine learning algorithm to obtain a generalized decomposition feature value, and determine the feature value principal component; if the feature value principal component is found to be abnormal, a relevant warning signal is generated;

[0158] S524, performing trend analysis on the obtained whole machine behavior data, predicting the change of pitch or torque execution results under specific wind parameters in the future, and generating relevant warning signals if the trend predicts that the pitch or torque execution results will exceed a certain threshold.

[0159] In one or more embodiments of the present application, when the wind farm controller obtains the whole machine behavior characteristic information of each target wind turbine generator set in the wind farm, it can send the whole machine behavior characteristic information to the server for the server to analyze according to the whole machine behavior characteristic information sent by each wind farm.

[0160] Specifically, the wind farm controller sends the whole machine behavior characteristic information of each target wind turbine generator set in the wind farm to the server, so that the server determines the second reference whole machine behavior characteristic according to the whole machine behavior characteristic information of the wind turbine generator sets having the target configuration in at least two wind farms;

[0161] The wind farm controller receives the second reference characteristic information sent by the server;

[0162] The wind farm controller updates the first reference characteristic information corresponding to the target wind turbine generator set with the target configuration according to the second reference characteristic information, wherein the updated first reference characteristic information is used for the next whole machine behavior diagnosis of the target wind turbine generator set with the target configuration in the wind farm.

[0163] As an example, the wind farm controller updates the first reference characteristic information corresponding to the target wind turbine generator set with the target configuration according to the second reference characteristic information, which may include:

[0164] Calculating a target value for representing the degree of difference between the second reference feature information and the first reference feature information according to the second reference feature information and the first reference feature information;

[0165] When the target value is greater than the preset value, the second reference feature information is determined as the updated first reference feature information.

[0166] In an embodiment of the present application, the wind farm controller sends the whole-machine behavior characteristic information of each target wind turbine generator set in the wind farm to the server, so that the server updates the first reference characteristic information in combination with the whole-machine behavior characteristic information of the wind turbine generator sets in other wind farms, so that the whole-machine behavior diagnosis results based on the updated first reference characteristic information can be more and more accurate.

[0167] Below through Figure 7 The examples of the present invention are described below.

[0168] like Figure 7 As shown, the wind farm controller of each wind farm includes a decision maker and an information library. The information library stores the whole machine behavior data of each target generator set in the wind farm. The decision maker analyzes the whole machine behavior data to obtain the characteristic information of each target generator set in the wind farm. Then, the decision maker sends the characteristic information to the cloud information library through the information library.

[0169] The cloud information library stores the characteristic information sent by each wind farm. The cloud analyzes the characteristic information of the target generator set with the same configuration sent by each wind farm to obtain the second reference characteristic information of the target generator set with the configuration. Then, the cloud sends the second reference characteristic information to each wind farm, so that each wind farm updates the first reference characteristic information stored in the information library according to the second reference characteristic information sent by the cloud.

[0170] The decision maker can regularly collect the whole machine behavior data of the target wind turbines in the wind farm according to the updated first reference feature information in the information database, conduct whole machine behavior diagnosis and early warning decisions, and push the decision results to each target wind turbine, so that the target wind turbine can perform corresponding actions (such as power limitation, speed limitation, blade angle limitation, shutdown or yaw, etc.) to suppress or eliminate abnormal whole machine behavior.

[0171] In addition, the decision maker can also diagnose the overall behavior status of the wind turbines used in the wind farm, and push warning information to each wind turbine based on the diagnosis results.

[0172] Taking the pitch change behavior analysis in the startup phase as an example, the startup pitch change rate extreme value in the whole machine behavior characteristic information is transmitted to the cloud. After cloud data aggregation and analysis, it is determined whether the startup pitch change rate warning threshold of the whole machine behavior under the same model configuration needs to be optimized. The updated startup pitch change rate warning threshold is transmitted to the information library of the wind farm controller, so that the abnormal startup pitch change rate behavior can be diagnosed more accurately.

[0173] Taking characteristic curve evaluation as an example, the scattered classification information of the characteristic curve is transmitted to the cloud. After the cloud data is summarized and analyzed, it is determined whether there are similar operating risks for the same model configuration. Through big data correlation analysis, since more data samples are collected and human intervention is possible, the cause of the anomaly can be located faster, and a solution to the effectiveness of characteristic curve anomalies can be formulated earlier.

[0174] In order to better illustrate the wind turbine behavior diagnosis method of the embodiment of the present application, a specific example is used below to illustrate. The method of this example can be applied to Figure 8 The wind farm controller in the overall behavior diagnostic system of the wind turbine generator set is shown. Fig. 9 A flow chart of another embodiment of the whole machine behavior diagnosis method of a wind turbine generator set provided by the present application is shown.

[0175] like Figure 8 and Fig. 9 As shown, the wind turbine generator set overall behavior diagnosis method may include:

[0176] S602, the decision maker of the wind farm controller numbers and indexes the target wind turbine generator sets that can be bound in the wind farm. Specifically, the decision maker of the wind farm controller obtains the number of target wind turbine generator sets that can be bound in the wind farm from the data service platform, and numbers each target wind turbine generator set to create an index.

[0177] S604, the decision maker can obtain the whole machine behavior characteristic information from the server through the data service platform of the wind farm controller. Specifically, the decision maker obtains the whole machine behavior characteristic information in the cloud information library based on big data according to the unit configuration type in the wind farm, such as foundation, tower, blades, power, current conversion system, pitch control system, etc. These whole machine behavior characteristic information can be the whole machine behavior characteristic information such as dynamic and static deviations of pitch control and torque execution, system response of pitch control and current conversion, and unit characteristic curves under the characteristics of wind speed, wind deviation, power, blade angle, air density, turbulence, and available flags of characteristic curves of the same unit configuration;

[0178] S606, the decision maker determines first reference characteristic information according to the whole machine behavior characteristic information, where the first reference characteristic information may include at least one of the following: a pitch dynamic deviation threshold, a pitch static deviation threshold, a torque dynamic deviation threshold, a torque static deviation threshold, a pitch system response function, a torque system response function, a standard characteristic curve, etc.;

[0179] S608, the decision maker receives the whole machine behavior data within a preset time period T sent by the target wind turbine generator set through the data service platform according to the index of each target wind turbine generator set, where the whole machine behavior data includes at least one of the following: instantaneous data such as wind turbine state, wind speed, wind deviation, power, feedback blade angle, given blade angle, feedback torque, given torque, air density, turbulence, and generator speed;

[0180] S610, the decision maker matches the whole machine behavior data of the target wind turbine generator set with each whole machine behavior characteristic information to be determined according to the whole machine behavior characteristic information to be determined, so as to obtain the whole machine behavior data matching each whole machine behavior characteristic information to be determined.

[0181] For example, the whole machine behavior characteristic information to be determined includes dimensions such as pitch execution dynamic deviation, torque execution dynamic deviation, pitch execution static deviation, torque execution static deviation, pitch system response, variable current system response and characteristic curve, so it is necessary to match the data required for the evaluation of these dimensions. In order to determine the pitch execution dynamic deviation, the wind speed, wind turbine state, feedback blade angle, given blade angle, power and other data are matched with the pitch execution dynamic deviation to determine the pitch execution dynamic deviation through the wind speed, wind turbine state, feedback blade angle, given blade angle, power and other data of the target wind turbine generator set.

[0182] In order to determine the response of the inverter system, the wind speed, wind turbine state, feedback torque, given torque, power and other data are matched with the response of the inverter system to determine the response of the inverter system through the wind speed, wind turbine state, feedback torque, given torque, power and other data of the target wind turbine generator set.

[0183] In order to determine the characteristic curve, data such as wind speed, wind turbine status, wind deviation, power, air density, and available flags of the characteristic curve are matched with the characteristic curve to determine the characteristic curve through data such as wind speed, wind turbine status, wind deviation, power, air density, and available flags of the characteristic curve of the target wind turbine generator set.

[0184] After S610, the wind turbine generator set behavior diagnosis method may further include:

[0185] S612, the decision maker performs data screening and cleaning to clean out inappropriate data, for example, cleaning out data in which the target wind turbine generator set is in standby state for too short a time or has a large number of time series missing or abnormal values;

[0186] S614, the decision maker performs whole machine behavior data diagnosis to obtain whole machine behavior diagnosis results. Specifically, the decision maker analyzes the whole machine behavior data of the target wind turbine generator set in different operation stages (start-up stage, shutdown stage, power generation stage, etc.) to obtain whole machine behavior diagnosis results; in addition, statistical classification, correlation analysis and trend analysis can also be performed; specific examples of diagnosis can be Figure 7 Example:

[0187] S616, the decision maker determines whether an early warning is needed according to the whole machine behavior diagnosis result obtained in S614, and determines that an early warning is needed when the whole machine behavior diagnosis result includes an abnormal diagnosis result of the target wind turbine generator set, and determines that an early warning is not needed when the whole machine behavior diagnosis result includes a normal diagnosis result of the target wind turbine generator set;

[0188] S618, when an early warning is needed, the decision maker generates a first type of early warning information, which can be displayed on the human machine interface (HMI) of the wind farm controller, and can also be pushed to the target wind turbine generator set through the network protocol, and the target wind turbine generator set records and processes the pushed early warning information accordingly; the first type of early warning information may include at least one of the following: pitch following early warning, excessive deviation between blades early warning, pitch rate over-limit early warning, torque following early warning, torque deviation early warning, torque rate over-limit early warning, pitch system response abnormal early warning, current conversion system response abnormal early warning, characteristic curve abnormal early warning, etc.;

[0189] S620, the decision maker saves the instantaneous whole machine behavior data before and after the warning information is triggered, so as to facilitate the R&D personnel to further analyze the root cause and mechanism behind the abnormal whole machine behavior;

[0190] S622, the decision maker determines whether control protection or control optimization is required. If it is determined that control protection or control optimization is required, S624 is executed; if it is determined that control protection or control optimization is not required, S626 is executed;

[0191] S624, the decision maker pushes the control information to the target wind turbine generator set through the data service platform of the wind farm controller.

[0192] The sources of abnormal behavior of wind turbines can be roughly divided into: Fig.10 Input class, design class, control class and subsystem class as shown. For example, for the torque following warning of the unit caused by turbulence, the decision maker in the wind farm can push the control protection information such as power limit, pitch angle limit, and shutdown to the target wind turbine generator set; for example, for the pitch following warning of the unit caused by the mismatch of control parameters, the decision maker in the wind farm can push the matching parameters to the target wind turbine generator set for control optimization. Similar control optimization methods can be to adjust the control proportional integral (PI) parameters and filters, or to shut down the control function that causes abnormal behavior of the whole machine.

[0193] After S624, the wind turbine generator set behavior diagnosis method may further include:

[0194] S626, the decision maker determines whether to traverse each target wind turbine generator set in the wind farm according to the serial number of each target wind turbine generator set in the wind farm, and executes S628 if each target wind turbine generator set in the wind farm is traversed, and continues to execute S614 if each target wind turbine generator set in the wind farm is not traversed;

[0195] S628, the decision maker evaluates the behavior state of the entire unit in the wind farm, for example, obtains the dynamic following deviation of the pitch of all target wind turbines in the wind farm, and screens the units with abnormal pitch following deviation. The method for detecting the abnormal relative pitch following deviation of the wind farm units can be the quartile method, which is used to calculate the minimum and maximum estimated values ​​in the data set, so that the values ​​beyond the range of the minimum and maximum estimated values ​​may be abnormal values;

[0196] S630, the decision maker determines whether the target wind turbine generator set in the wind farm is abnormal as a whole. If the target wind turbine generator set in the wind farm is abnormal as a whole, S632 is executed; if the target wind turbine generator set in the wind farm is normal as a whole, S634 is executed;

[0197] S632, the decision maker generates warning information of overall abnormality of the units in the wind farm, which can be displayed on the HMI interface, and pushes the warning information of the whole machine behavior to the abnormal target wind turbine generator set through the network protocol, so that the target wind turbine generator set records and processes the pushed information accordingly. The warning information is the second type of warning information, which is generally a warning of the comparison result of the wind farm relative to the whole machine behavior state;

[0198] S634, the decision maker may push the whole machine behavior feature information to the server through the data service platform of the wind farm controller, and generate a wind farm decision maker log.

[0199] As an example, after authorization, the whole machine behavior characteristic information is pushed to the cloud through the network or files. Since the whole machine behavior characteristic information converted by the decision maker is much less than the instantaneous data of the wind turbine generator set, while providing direct and effective whole machine behavior characteristic information to the cloud, it also reduces the cloud data transmission and storage load. In addition, the decision maker can also record the whole machine behavior characteristic information to the information library in the wind farm controller and the information library of the Supervisory Control And Data Acquisition (SCADA) system.

[0200] In the embodiment of the present application, a more comprehensive diagnostic evaluation can be efficiently performed on the whole machine behavior state of the wind turbine generator set in the wind farm, so that problems related to the whole machine behavior of the wind turbine generator set can be better discovered in advance, and the root cause of the abnormal whole machine behavior of the wind turbine generator set can be quickly located with the help of correlation analysis. Through the connection of cloud data and trend analysis of valid data accumulated on the wind farm side, some common whole machine behavior abnormalities of wind turbine generator sets can be discovered in advance. In addition, early warning and control information can be pushed to the wind turbine generator set in real time to suppress or eliminate related anomalies in time. Through the combined application of the above strategies, while reducing labor costs, data storage and transmission costs, and project resource investment, the needs of wind farms and units for whole machine behavior evaluation in terms of real-time, intelligence, and network security are met.

[0201] Corresponding to the wind turbine generator set whole machine behavior diagnosis method provided in the present application, the present application also provides a wind turbine generator set whole machine behavior diagnosis device. The device is applied to a wind farm controller arranged in a wind farm.

[0202] Fig.11 FIG. 1 is a schematic diagram showing a structure of an embodiment of a wind turbine generator set behavior diagnostic device provided by the present application. Fig.11 As shown, the wind turbine generator set whole machine behavior diagnosis device 700 includes:

[0203] A first receiving module 702 is used to receive whole-machine behavior data sent by a target wind turbine generator set in a wind farm through a network protocol, where the whole-machine behavior data is whole-machine behavior data of the target wind turbine generator set in at least one operation stage;

[0204] The first determination module 704 is used to determine the whole machine behavior characteristic information of the target wind turbine generator set at each operation stage according to the whole machine behavior data of the target wind turbine generator set at each operation stage;

[0205] The unit diagnosis module 706 is used to diagnose the whole machine behavior of the target wind turbine generator set in the operation stage according to the whole machine behavior characteristic information in the operation stage and the first reference characteristic information for each operation stage, and obtain the whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage.

[0206] In an embodiment of the present application, the whole machine behavior of the target wind turbine generator set is diagnosed by the wind farm controller in the wind farm, and there is no need to upload the whole machine behavior data of the target wind turbine generator set to the server. Since the wind farm controller in the wind farm can obtain the whole machine behavior data of the target wind turbine generator set in real time, the whole machine behavior of the target wind turbine generator set can be diagnosed in real time. In addition, the wind farm only needs to store and process the data of the wind turbine generator sets in the wind farm, without the need to store and process the data of the wind turbine generator sets in other wind farms. In this way, the pressure of storing and processing data on the wind farm is smaller than that on the server. Therefore, the wind farm can store and process finer granularity data of the wind turbine generator set, so that the wind farm controller can perform whole machine behavior diagnosis based on finer granularity data, thereby making the result of whole machine behavior diagnosis more accurate.

[0207] Secondly, the wind farm controller in the wind farm provides the whole machine behavior diagnosis service of the wind turbine generator set nearby. Compared with the diagnosis on the server side, the diagnosis on the wind farm side can produce faster business response and meet the needs of the wind farm and wind turbine generator set in terms of real-time, intelligence, network security, etc. At the same time, the server side can still access the wind farm controller's diagnosis results of the whole machine behavior of the wind turbine generator set.

[0208] In one or more embodiments of the present application, at least one operation phase includes a startup phase, a shutdown phase, and a power generation phase. The first determination module 704 may include:

[0209] A first determination unit is used to determine the whole machine behavior characteristic value of the target wind turbine generator set in the startup phase according to the whole machine behavior data of the target wind turbine generator set in the startup phase;

[0210] a second determination unit, configured to determine the whole machine behavior characteristic value of the target wind turbine generator set in the shutdown phase according to the whole machine behavior data of the target wind turbine generator set in the shutdown phase after determining the whole machine behavior characteristic value of the target wind turbine generator set in the startup phase;

[0211] The third determination unit is used to determine the whole machine behavior characteristic value and the whole machine behavior curve of the target wind turbine generator set in the power generation stage according to the whole machine behavior data of the target wind turbine generator set in the power generation stage after determining the whole machine behavior characteristic value of the target wind turbine generator set in the shutdown stage.

[0212] In one or more embodiments of the present application, the whole-machine behavior characteristic information of the target wind turbine generator set in each operation stage includes: a plurality of whole-machine behavior characteristic values ​​of the target wind turbine generator set.

[0213] The crew diagnostic module 706 may include:

[0214] A first acquisition unit is used to acquire a characteristic extreme value of a maximum value and a minimum value from a plurality of whole machine behavior characteristic values, wherein the acquisition of the maximum value or the minimum value is determined by the operation stage;

[0215] A first calculation unit, used for calculating the difference between the characteristic extreme value and the reference characteristic value in the first reference characteristic information;

[0216] The first generating unit is used to generate a diagnosis result of the target wind turbine generator set in the operation stage according to the size relationship between the difference value and the first preset difference value.

[0217] In one or more embodiments of the present application, the whole machine behavior characteristic information includes a whole machine behavior curve. The unit diagnosis module 706 may include:

[0218] A second acquisition unit is used to acquire multiple discrete points on the whole machine behavior curve;

[0219] A clustering unit, used for performing clustering processing on a plurality of discrete points to obtain at least one cluster;

[0220] The second generating unit is used to generate a diagnostic result of the whole machine behavior curve according to the size relationship between the number of clusters and the reference number in the first reference feature information.

[0221] In one or more embodiments of the present application, the wind turbine generator set whole machine behavior diagnosis device 700 may further include:

[0222] The decomposition module is used to perform the following operations on the whole machine behavior characteristic information of the target wind turbine generator set in each operation stage: decomposing the whole machine behavior characteristic information in the operation stage to obtain decomposed characteristic values;

[0223] A generating module is used to generate third warning information of abnormal decomposition characteristic value when the difference between the decomposition characteristic value and the preset characteristic value is greater than a second preset difference.

[0224] In an embodiment of the present application, by decomposing the whole machine behavior characteristic information of the target wind turbine generator set at each operation stage, a decomposed characteristic value is obtained, and each decomposed characteristic value can characterize the operation characteristics of the target wind turbine generator set at the corresponding operation stage. As an example, the decomposed characteristic value can be wind speed, rotation speed, power, etc. When the difference between the decomposed characteristic value and the preset characteristic value is greater than the second preset difference, it means that the target wind turbine generator set is abnormal, and the cause of the abnormality of the target wind turbine generator set is the decomposed characteristic value. Then, a third warning information of abnormal decomposed characteristic value is generated, so that it is convenient for the staff to solve the abnormality of the target wind turbine generator set according to the decomposed characteristic value.

[0225] In one or more embodiments of the present application, the wind turbine generator set whole machine behavior diagnosis device 700 may further include:

[0226] The prediction module is used to perform the following operations on the whole machine behavior characteristic information of the target wind turbine generator set in each operation stage: predict the whole machine behavior characteristic information of the target wind turbine generator set in the operation stage according to the whole machine behavior characteristic information in the operation stage, and obtain the predicted whole machine behavior characteristic information.

[0227] In the embodiment of the present application, by predicting the whole machine behavior characteristic information of the target wind turbine generator set in the operation stage, it is possible to discover in advance whether the target wind turbine generator set is abnormal, so that it is convenient for the staff to handle the abnormality of the target wind turbine generator set in advance.

[0228] In one or more embodiments of the present application, the wind turbine generator set whole machine behavior diagnosis device 700 may further include:

[0229] A sending module, configured to send the whole machine behavior characteristic information of each target wind generator set in the wind farm to a server when the whole machine behavior characteristic information of each target wind generator set in the wind farm is obtained, so that the server determines a second reference whole machine behavior characteristic according to the whole machine behavior characteristic information of wind generator sets having the target configuration in at least two wind farms;

[0230] A second receiving module, used for receiving second reference feature information sent by the server;

[0231] The updating module is used to update the first reference characteristic information corresponding to the target wind turbine generator set with the target configuration according to the second reference characteristic information, and the updated first reference characteristic information is used to perform the next whole machine behavior diagnosis on the target wind turbine generator set with the target configuration in the wind farm.

[0232] In an embodiment of the present application, the wind farm controller sends the whole-machine behavior characteristic information of each target wind turbine generator set in the wind farm to the server, so that the server updates the first reference characteristic information in combination with the whole-machine behavior characteristic information of the wind turbine generator sets in other wind farms, so that the whole-machine behavior diagnosis results based on the updated first reference characteristic information can be more and more accurate.

[0233] The embodiment of the present application also provides a wind farm controller. As an example, the wind farm controller includes: a decision maker, a data service platform and a human-machine interface. The data service platform is used to interact with the wind turbines in the wind farm. For example, the data service platform can obtain the whole machine behavior data of the wind turbines and store the obtained whole machine behavior data in the information base of the data service platform. The decision maker diagnoses the whole machine behavior of the target wind turbine according to the whole machine behavior data of the target wind turbine in the information base to obtain the whole machine behavior diagnosis result of the target wind turbine. When the decision maker diagnoses an abnormality, it generates early warning information and displays it through the human-machine interface.

[0234] As another example, the wind farm controller includes: a processor and a memory storing computer program instructions, and the processor implements the whole machine behavior diagnosis method of the wind turbine generator set according to any one of the above embodiments when executing the computer program instructions.

[0235] Fig.12 FIG. 1 is a schematic diagram showing the hardware structure of an embodiment of a wind farm controller provided by the present application. Fig.12 As shown, the wind farm controller may include a processor 801 and a memory 802 storing computer program instructions.

[0236] Specifically, the processor 801 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0237] The memory 802 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 802 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 802 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 802 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 802 is a non-volatile solid-state memory.

[0238] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0239] The processor 801 reads and executes the computer program instructions stored in the memory 802 to implement any one of the wind turbine generator set behavior diagnosis methods in the above embodiments.

[0240] In one example, the wind farm controller may further include a communication interface 803 and a bus 810. Fig.12 As shown, the processor 801, the memory 802, and the communication interface 803 are connected via a bus 810 and communicate with each other.

[0241] The communication interface 803 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0242] Bus 810 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 810 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.

[0243] The present application also provides a wind farm, which includes: a target wind turbine generator set; and a wind farm controller according to any one of the above embodiments, wherein the wind farm controller is used to diagnose the overall behavior of the target wind turbine generator set.

[0244] The present application also provides a whole machine behavior diagnosis system, the system comprising:

[0245] At least two of the above wind farms, each wind farm is used to determine the whole machine behavior characteristic information of the target wind turbine generator set according to the whole machine behavior data of the target wind turbine generator set in the wind farm, and send the whole machine behavior characteristic information to the server;

[0246] The server is used to determine a second reference whole machine behavior characteristic based on whole machine behavior characteristic information of target wind turbines with target configuration in at least two wind farms, and send the second reference whole machine behavior characteristic to each wind farm, wherein each wind farm is further used to: update the first reference characteristic information corresponding to the target wind turbines with target configuration in the wind farm based on the second reference whole machine behavior characteristic.

[0247] As an example, the whole machine behavior diagnosis system can be as follows Figure 7 The system shown.

[0248] In addition, in combination with the wind turbine generator set overall behavior diagnosis method in the above embodiment, the present application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by the processor, any wind turbine generator set overall behavior diagnosis method in the above embodiment is implemented.

[0249] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0250] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0251] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0252] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0253] The above are only specific implementation methods of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A wind turbine behavior diagnosis method, characterized in that: Applied to a wind farm controller arranged in a wind farm, the method comprises: Receiving whole-machine behavior data sent by a target wind turbine generator set in the wind farm through a network protocol, wherein the whole-machine behavior data is whole-machine behavior data of the target wind turbine generator set in at least one operation stage; Determining whole machine behavior characteristic information of the target wind turbine generator set at each of the operation stages according to the whole machine behavior data of the target wind turbine generator set at each of the operation stages; For each of the operation stages, according to the whole machine behavior characteristic information and the first reference characteristic information in the operation stage, diagnose the whole machine behavior of the target wind turbine generator set in the operation stage, and obtain the whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage; The at least one operation phase includes one or more of a standby phase, a maintenance phase, a startup phase, a shutdown phase, and a power generation phase.

2. The method according to claim 1, characterized in that The at least one operation phase includes a startup phase, a shutdown phase and a power generation phase; Determining the whole machine behavior characteristic information of the target wind turbine generator set at each of the operation stages according to the whole machine behavior data of the target wind turbine generator set at each of the operation stages includes: Determining a whole-machine behavior characteristic value of the target wind turbine generator set during the startup phase according to the whole-machine behavior data of the target wind turbine generator set during the startup phase; After determining the whole machine behavior characteristic value of the target wind turbine generator set in the startup phase, determining the whole machine behavior characteristic value of the target wind turbine generator set in the shutdown phase according to the whole machine behavior data of the target wind turbine generator set in the shutdown phase; After determining the whole machine behavior characteristic value of the target wind turbine generator set in the shutdown phase, the whole machine behavior characteristic value and the whole machine behavior curve of the target wind turbine generator set in the power generation phase are determined respectively according to the whole machine behavior data of the target wind turbine generator set in the power generation phase.

3. The method according to claim 1, characterized in that The whole machine behavior characteristic information of the target wind turbine generator set in each operation stage includes: a plurality of whole machine behavior characteristic values ​​of the target wind turbine generator set; The step of diagnosing the whole machine behavior of the target wind turbine generator set in the operation stage according to the whole machine behavior characteristic information and the first reference characteristic information in the operation stage to obtain the whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage includes: From the plurality of whole machine behavior characteristic values, obtaining a characteristic extreme value among a maximum value and a minimum value, wherein obtaining the maximum value or the minimum value is determined by the operation stage; Calculating the difference between the characteristic extreme value and the reference characteristic value in the first reference characteristic information; According to the magnitude relationship between the difference and the first preset difference, a diagnosis result of the target wind turbine generator set in the operation stage is generated.

4. The method according to claim 1, characterized in that: The whole machine behavior characteristic information includes a whole machine behavior curve; The step of diagnosing the whole machine behavior of the target wind turbine generator set in the operation stage according to the whole machine behavior characteristic information and the first reference characteristic information in the operation stage to obtain the whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage includes: Acquire a plurality of discrete points on the whole machine behavior curve; Performing clustering processing on the multiple discrete points to obtain at least one cluster; The diagnostic result of the whole machine behavior curve is generated according to the size relationship between the number of clusters and the reference number in the first reference feature information.

5. The method according to any one of claims 1 to 4, characterized in that: After determining the whole machine behavior characteristic information of the target wind turbine generator set at each of the operation stages according to the whole machine behavior data of the target wind turbine generator set at each of the operation stages, the method further comprises: For the whole machine behavior characteristic information of the target wind turbine generator set in each of the operation stages, respectively: decomposing the whole machine behavior characteristic information in the operation stage to obtain decomposed characteristic values; When the difference between the decomposed characteristic value and the preset characteristic value is greater than a second preset difference, third warning information indicating that the decomposed characteristic value is abnormal is generated.

6. The method according to any one of claims 1 to 4, characterized in that After determining the whole machine behavior characteristic information of the target wind turbine generator set at each of the operation stages according to the whole machine behavior data of the target wind turbine generator set at each of the operation stages, the method further comprises: For the whole machine behavior characteristic information of the target wind turbine generator set in each of the operation stages, respectively perform: predict the whole machine behavior characteristic information of the target wind turbine generator set in the operation stage according to the whole machine behavior characteristic information in the operation stage to obtain predicted whole machine behavior characteristic information.

7. The method according to any one of claims 1 to 4, characterized in that After determining the whole machine behavior characteristic information of the target wind turbine generator set at each of the operation stages according to the whole machine behavior data of the target wind turbine generator set at each of the operation stages, the method further comprises: When the whole machine behavior characteristic information of each of the target wind turbine generator sets in the wind farm is obtained, the whole machine behavior characteristic information of each of the target wind turbine generator sets in the wind farm is sent to the server, so that the server determines the second reference whole machine behavior characteristic according to the whole machine behavior characteristic information of the wind turbine generator sets with the target configuration in at least two of the wind farms; Receiving the second reference whole-machine behavior feature sent by the server; The first reference feature information corresponding to the target wind turbine generator set with the target configuration is updated according to the second reference whole-machine behavior feature, and the updated first reference feature information is used to perform the next whole-machine behavior diagnosis on the target wind turbine generator set with the target configuration in the wind farm.

8. A wind turbine generator system behavior diagnostic device, characterized in that: Applicable to a wind farm controller arranged in a wind farm, the device comprises: A first receiving module is used to receive the whole machine behavior data sent by the target wind turbine generator set in the wind farm through a network protocol, wherein the whole machine behavior data is the whole machine behavior data of the target wind turbine generator set in at least one operation stage; A first determination module is used to determine the whole machine behavior characteristic information of the target wind turbine generator set at each of the operation stages according to the whole machine behavior data of the target wind turbine generator set at each of the operation stages; a unit diagnosis module, configured to diagnose the whole machine behavior of the target wind turbine generator set in each of the operation stages according to the whole machine behavior characteristic information and the first reference characteristic information in the operation stage, and obtain a whole machine behavior diagnosis result of the target wind turbine generator set in the operation stage; The at least one operation phase includes one or more of a standby phase, a maintenance phase, a startup phase, a shutdown phase, and a power generation phase.

9. A wind farm controller, characterized in that: include: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the whole machine behavior diagnosis method of the wind turbine generator set according to any one of claims 1-7 is implemented.

10. A wind farm, characterized in that: include: Target wind turbines; The wind farm controller according to claim 9 is used to diagnose the overall behavior of the target wind turbine generator set.

11. A whole machine behavior diagnosis system, characterized in that: include: At least two wind farms according to claim 10, each of the wind farms being used to determine the whole machine behavior characteristic information of the target wind turbine generator set according to the whole machine behavior data of the target wind turbine generator set in the wind farm, and send the whole machine behavior characteristic information to a server; The server is configured to determine a second reference whole machine behavior feature based on whole machine behavior feature information of the target wind turbine generator sets having the target configuration in at least two of the wind farms, and send the second reference whole machine behavior feature to each of the wind farms; Each of the wind farms is further configured to update the first reference characteristic information corresponding to the target wind turbine generator set having the target configuration in the wind farm according to the second reference whole-machine behavior characteristic.

12. A computer storage medium, characterized in that: The computer storage medium stores computer program instructions, which, when executed by a processor, implement the wind turbine generator set behavior diagnosis method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Wind power generator set as well as state monitoring system thereof and method

    CN101705908A

  • Wind power monitoring and fault diagnosis system

    CN102999025A

  • Wind generating set monitoring system based on physical component model and real-time data

    CN104018988A

  • Wind generating set fault diagnosis system and method

    CN107704933A

  • Wind power generator unit status monitoring device

    CN202600421U