CN118013827A - A method, system and device for online monitoring of fatigue life of wind turbines in the entire field - Google Patents

Abstract

The invention provides a method, a system and equipment for online monitoring of fatigue life of a full-field wind turbine generator, which comprise the following steps: step 1, selecting a plurality of wind turbines from wind turbines of the same type in a wind power plant to be tested as an actual measurement set; step 2, performing discretization processing on the fatigue load measurement data of each actual measurement unit to obtain an actual measurement fatigue load database; step 3, converting the actually measured fatigue load database into a structural component fatigue damage database corresponding to each actually measured unit; step 4, establishing a multi-parameter correlation model of the fatigue damage of the structural component and the state parameters of the wind turbine generator; step 5, extrapolating structural fatigue damage of the rest units in the wind turbines of the same type of wind power plant; step 6, analyzing the fatigue life of structural components of the full-farm wind turbine generator; the invention can effectively control the system cost, and the fatigue damage and life calculation precision can be continuously improved.

Description

A method, system and device for online monitoring of fatigue life of wind turbines in the entire field

Technical Field

The invention belongs to the technical field of wind turbines, and relates to a method, system and equipment for online monitoring of fatigue life of wind turbines in the entire field.

Background technique

Whether wind turbines can operate safely during their design life has always been a hot topic in the industry. Wind turbines operate in harsh environments for a long time, and their components are subjected to loads caused by various complex factors. Fatigue damage is the main form of damage to the load-bearing components of wind turbines. Fatigue damage usually first produces tiny cracks, which then gradually expand. When they are discovered, it is often in the late stage of the fatigue damage process. Its secrecy poses a huge risk to the safe operation of wind turbines. If the fatigue damage status of wind turbines in wind farms can be mastered in real time, fatigue safety countermeasures for wind turbines can be formulated in a targeted manner to ensure the safe operation of wind turbines.

At present, there is no online monitoring system for the fatigue life of wind turbines in wind farms, and there is no such system or equipment on the market. In order to ensure the fatigue safety of each wind turbine in a wind farm and taking into account the cost factor, the present invention proposes a method for real-time monitoring of the fatigue life of all wind turbines in the wind farm.

In the technical research related to the present invention, patent CN113821979B (a method for fatigue damage and life assessment of wind turbines, computer equipment and storage medium) mentioned the training of a correlation model between fatigue damage of wind turbines and SCADA parameters of wind turbines. However, the patent application is for a single unit and has not yet evaluated the life of the same type of wind turbines in the entire field.

Summary of the invention

The purpose of the present invention is to provide a method, system and equipment for online monitoring of fatigue life of wind turbines in the whole field, which solves the above-mentioned deficiencies in the existing fatigue life monitoring of wind turbines in the whole field.

In order to achieve the above object, the technical solution adopted by the present invention is:

The present invention provides a method for online monitoring of fatigue life of wind turbines in the entire field, comprising the following steps:

Step 1, selecting multiple wind turbines from the same type of wind turbines in the wind farm to be tested as the actual test turbines;

Step 2, discretizing the fatigue load measurement data of each measured unit to obtain a measured fatigue load database;

Step 3, converting the measured fatigue load database into a structural component fatigue damage database corresponding to each measured unit;

Step 4, establishing a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbine generator sets;

Step 5, extrapolating the structural fatigue damage of the remaining wind turbines of the same type in the wind farm;

Step 6: Fatigue life analysis of structural components of all wind turbines in the field.

Preferably, in step 1, a wind turbine set that satisfies at least the following two conditions is selected as the actual measurement set from the wind turbine sets of the same type in the wind farm to be measured, and the conditions are:

The first condition: wind turbines with average wind speed and instantaneous wind speed greater than the preset threshold;

The second condition: the two wind turbines with the largest turbulence difference;

The third condition: Conduct a correlation analysis on the wind parameter data of all wind turbines of the same model in the field and select the wind turbine with the highest correlation.

Preferably, in step 2, the fatigue load measurement data of each measured unit is discretized to obtain a measured fatigue load database, and the specific method is:

Perform fatigue load measurement on each measured unit determined in step 1 to obtain fatigue load data corresponding to each wind turbine unit;

Extract the time series from the obtained fatigue load data, and combine the time series with the operating parameters and wind data of the corresponding wind turbine to form the fatigue load time series data corresponding to each measured turbine;

The obtained fatigue load time series data are cleaned and divided into bins to obtain the measured fatigue load database corresponding to each measured unit.

Preferably, in step 3, the measured fatigue load database is converted into a component fatigue damage database corresponding to each measured unit, and the specific method is:

Establish the load-stress quantitative relationship of the structural components corresponding to each measured unit;

According to the load-stress quantification relationship of the structural component, the measured fatigue time series load is converted into the time series stress of the structural component;

The obtained structural component time-series stress is subjected to rain flow statistical analysis to obtain the fatigue damage value corresponding to the component time-series stress, and then the component fatigue damage database corresponding to each measured unit is obtained.

Preferably, in step 4, a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbine generator sets is established, and the specific method is:

The fatigue damage database of structural components corresponding to each measured unit is used as training samples, and a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbine units based on measured fatigue load data is established using a neural network method.

Preferably, in step 5, the structural fatigue damage of the remaining wind turbines of the same type in the wind farm is extrapolated, and the specific method is:

Obtain fatigue load simulation data corresponding to each wind turbine of the same type in the wind farm;

Each group of fatigue load simulation data is cleaned and discretely divided into bins to obtain a simulation fatigue load database corresponding to each wind turbine, and the obtained simulation fatigue load database is converted into a structural component fatigue damage simulation database corresponding to each wind turbine;

Using the multi-parameter correlation analysis method, combined with the fatigue damage simulation database of the structural components corresponding to each wind turbine, a component fatigue damage correlation model between each other wind turbine and the measured unit is established;

The structural component fatigue damage database corresponding to each measured unit is used as the input of the component fatigue damage correlation model between each other wind turbine unit and the measured unit, so as to obtain the structural component fatigue damage database corresponding to each other wind turbine unit.

Preferably, in step 6, fatigue life analysis of structural components of all wind turbines is performed by:

For the measured units, the fatigue life of each component of each measured unit is calculated using the fatigue load data obtained from the actual measurement;

For other wind turbines, the time series of wind turbine state parameters of each other wind turbine since its operation is obtained, and the time series of wind turbine state parameters is cleaned and discretely divided into bins to obtain the characteristic value of the wind turbine state parameter corresponding to each other wind turbine;

The structural component fatigue damage database corresponding to each other wind turbine obtained in step 5 is used as the input of the correlation model between structural component fatigue damage and wind turbine state parameters established in step 4 to optimize, so as to obtain a personalized correlation model between wind turbine state parameters and structural component fatigue damage corresponding to each wind turbine;

The obtained characteristic values of the wind turbine state parameters corresponding to each other wind turbine are combined with the personalized model of association between the wind turbine state parameters and fatigue damage of structural components to obtain the fatigue life damage value corresponding to each other wind turbine.

A system for online monitoring of fatigue life of wind turbines in the entire field, comprising:

A unit selection unit is used to select multiple wind turbines from the same type of wind turbines in the wind farm to be tested as the actual test units;

A fatigue load database acquisition unit is used to discretize the fatigue load measurement data of each measured unit to obtain a measured fatigue load database;

A structural component fatigue damage database acquisition unit is used to convert the measured fatigue load database into a structural component fatigue damage database corresponding to each measured unit;

A model building unit, used to build a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbines;

Fatigue damage extrapolation unit, used to extrapolate the structural fatigue damage of the remaining wind turbines of the same model in the wind farm;

Fatigue life analysis unit is used for fatigue life analysis of structural components of wind turbines in the entire field.

A computer device/apparatus/system comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.

A computer-readable storage medium stores a computer program/instruction thereon, which implements the steps of the method when executed by a processor.

Compared with the prior art, the present invention has the following beneficial effects:

The present invention provides a method for online monitoring of fatigue life of wind turbines in the whole field. First, the load is discretized (to facilitate subsequent model training), and the continuously measured fatigue load is discretized into short-time time series data of 600s or 180s according to the state of the wind turbine, and then converted into a structural fatigue damage database. Then, through a neural network algorithm, a correlation model between the state parameters of the wind turbine (including operating parameters and wind parameters) and the fatigue damage of structural components is established, and the state parameters of the unit are directly and organically matched with the fatigue damage, thereby improving the calculation efficiency of the fatigue life of the whole field unit structure; then a fatigue damage extrapolation model is established, and the measured wind turbine fatigue damage library is extrapolated to all wind turbines in the whole field; finally, the structural fatigue damage database of each wind turbine at each position is used as a sample, and the measured wind turbine state parameter-structural fatigue damage model is trained twice to obtain a personalized correlation model for each position, and then the fatigue damage life of each wind turbine is obtained. Since the present application only needs to monitor the fatigue load of the representative position, the system cost can be effectively controlled; at the same time, for all models of the present invention, as the measured fatigue load online data continues to increase, the model algorithm is continuously optimized, and the fatigue damage and life calculation accuracy is continuously improved.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram of the process of the present invention.

Detailed ways

In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

Example 1

As shown in FIG1 , this embodiment provides a method for online monitoring of fatigue life of wind turbines in a whole field, comprising the following steps:

Step 1: Selection of wind turbines to be measured

Several wind turbines of the same type are selected from the wind farm for continuous fatigue load measurement. In order to ensure that the selected positions can better represent the wind conditions of the entire field, it is necessary to comprehensively analyze the feasibility study report, the wind data of the wind farm wind tower, and the wind data of each position. Among them, wind turbines that meet at least the following two conditions are selected as the actual measurement units:

The first condition: wind turbines with average wind speed and instantaneous wind speed greater than the preset threshold, so as to collect more load data in high wind areas;

The second condition: the two wind turbines with the largest turbulence difference, in order to facilitate the calculation of loads;

The third condition: Conduct a correlation analysis on the wind parameter data (wind speed, wind direction, turbulence, etc.) of all wind turbines of the same model in the entire field, and finally select the wind turbine with the highest correlation to ensure that the selected wind turbine can effectively represent the wind characteristics of each wind turbine in the wind farm, and can extrapolate the fatigue loads of other wind turbines by selecting the measured fatigue loads of the wind turbine.

Step 2: Fatigue load measurement and data discretization to obtain the measured fatigue load database corresponding to each measured unit

S21, measuring fatigue load of each measured unit determined in step 1 to obtain fatigue load data, wherein the fatigue load data at least includes:

Blade root: flapping bending moment and shimmy bending moment;

Hub center loads: pitching moment, lateral bending moment and torque;

Tower: pitch bending moment at tower top, lateral bending moment at tower top, torque at tower top, pitch bending moment at tower bottom and lateral bending moment at tower bottom.

S22, cleaning and binning the fatigue load data obtained from each measured unit

S221, extracting a time series representing the measured fatigue load data of the wind turbine, and extracting synchronous wind turbine operating parameters such as power, speed, pitch angle and impeller phase angle, and extracting synchronous wind data such as wind speed, turbulence and wind direction.

S222, cleaning the obtained fatigue load time series data to obtain cleaned fatigue load time series data, the specific cleaning method is:

Eliminate erroneous or invalid data.

S223, discretizing and classifying the cleaned fatigue load time series data to obtain a plurality of fatigue load time series data sets corresponding to each measured unit, and forming a measured fatigue load time series database corresponding to each measured unit from the plurality of fatigue load time series data sets.

The multiple fatigue load time series data sets are respectively a fatigue load time series data set of the wind turbine startup process, a fatigue load time series data set of the wind turbine shutdown process, a fatigue load time series data set of the wind turbine normal power generation process, and a fatigue load time series data set of the wind turbine idling or standby state, where:

The fatigue load time series data set of the wind turbine startup process is the fatigue load data corresponding to 120s before and 60s after the grid connection of multiple wind turbines.

The fatigue load time series data set of the wind turbine shutdown process is the fatigue load data corresponding to 60 seconds before the shutdown of multiple wind turbines and 120 seconds after the wind turbines tow the net.

The fatigue load time series data set of the normal power generation process of the wind turbine is the fatigue load data corresponding to multiple discrete monomers with a duration of 600s (corresponding to the duration of the turbulent load condition in IEC61400-1) under the normal power generation state of the wind turbine;

The wind turbine idling or standby fatigue load time series data set is the fatigue load data corresponding to multiple discrete units with a duration of 600s (corresponding to the duration of the turbulent load condition in IEC61400-1) in the wind turbine idling or standby state.

S224, calculate the eigenvalues corresponding to the four fatigue load time series data sets corresponding to each measured unit, among which: the power takes the average value of the discrete time period as the eigenvalue; the speed takes the average value as the eigenvalue; the pitch takes the average value as the eigenvalue; the impeller phase angle takes the angle at the beginning as the eigenvalue; the wind speed takes the average value as the eigenvalue; the turbulence degree is the turbulence degree of the period; and the wind direction takes the average value as the eigenvalue.

S225, all measured fatigue load time series are divided into four categories according to the state of the wind turbine, and further subdivided into fatigue load sequence monomers with a duration of 180s or 600s, and correspond to the operating parameters and wind parameters of the wind turbine. Therefore, the characteristic values corresponding to the four fatigue load time series data sets are combined to form a measured fatigue load database corresponding to each measured unit. As the load measurement time increases, the database continues to expand.

Step 3: Convert the measured fatigue load database into the structural component fatigue damage database corresponding to each measured unit

S31, establish the load-stress quantitative relationship of the structural components corresponding to each measured unit through finite element modeling analysis or other mechanical analysis methods.

S32, according to the load-stress quantification relationship of the structural components corresponding to each measured unit, the corresponding measured fatigue load time series database is converted into the structural component time series stress.

Node stress of finite element model of structural components:

σi＝Fx×fi_Fx+Fy×fi_Fy+Fz×fi_Fz+Mx×fi_Mx+My×fi_My+Mz×fi_Mz

Among them, σi is the stress of the node of the finite element model, fi_Fx is the stress value of the node under the unit load of Fx, fi_Fy is the stress value of the node under the unit load, and so on.

S33, performing rain flow statistical analysis on the time-series stress of the structural components in S32, referring to the fatigue S-N curve of the structural components, and according to the linear accumulation theory of fatigue damage, obtaining the fatigue damage value corresponding to the time-series stress of the components, and the fatigue damage value has a corresponding relationship with the operating parameters of the wind turbine.

S34, combining all the obtained fatigue damage values to form a structural component fatigue damage database corresponding to each measured unit.

Step 4: Establish a correlation model between fatigue damage of structural components and state parameters of wind turbines

The fatigue damage database of wind turbine components of representative on-site wind turbines is used as a sample training library, and a neural network method is used for training. Then, a multi-parameter correlation model is established between the structural fatigue damage of the wind turbine based on the measured fatigue load data and the wind turbine operating parameters (such as power, speed, pitch angle, impeller phase angle, etc.) and wind parameters (wind speed, turbulence, wind direction, etc.).

The model continues to evolve as the load measurement data increases, and the model accuracy continues to improve.

Step 5: Extrapolate the structural fatigue damage of the remaining wind turbines of the same type in the wind farm.

Taking into account that the turbulence, wind direction distribution, wind shear, inflow angle, wind frequency distribution and other parameters of the representative position are different from those of other positions in the entire field, in order to make the fatigue damage data of the structural components of the representative group applicable to the fatigue damage calculation of the wind turbine structures in the entire field, extrapolation calculation is required here to make the damage data better applicable to positions with different turbulence, wind direction distribution and wind frequency distribution.

S51, according to the specific parameters of the on-site wind turbines, a corresponding load simulation model is established for each wind turbine of the same model in the wind farm, and according to the wind condition parameters of all wind turbines in the field, fatigue load simulation calculations are performed on wind turbines of different models to obtain fatigue load simulation data corresponding to each model of wind turbine.

S52, cleaning and discrete binning the fatigue load simulation data to obtain a simulation fatigue load database corresponding to each wind turbine generator set, and converting the obtained simulation fatigue load database into a structural component fatigue damage simulation database corresponding to each wind turbine generator set.

S53, compare and analyze the fatigue damage results corresponding to the simulated loads of each wind turbine set, use multi-parameter correlation analysis methods (such as SPSS and other tools) to analyze the impact of changes in parameters such as turbulence, wind direction, wind shear, and inflow angle on the fatigue damage of structural components, and establish a component fatigue damage correlation model between each remaining wind turbine set and the actual measured unit.

S54, using the structural component fatigue damage database corresponding to each measured unit as the input of the component fatigue damage correlation model between each other wind turbine unit and the measured unit, to obtain the structural component fatigue damage database corresponding to each other wind turbine unit.

Step 6: Fatigue life analysis of wind turbine structural components throughout the field

First, for the measured unit

The commonly used method for calculating the fatigue life of structural components in the research and development and design process of wind turbines (Miner's linear damage accumulation theory) is adopted. According to the S-N curve of the components, the fatigue life of each component of each measured unit is calculated using the measured fatigue load.

Second, for other wind turbines

S61, obtaining the wind turbine state parameter time series (wind turbine state parameters: power, speed, pitch angle, etc. and wind parameters: wind speed, turbulence, wind direction, etc.) of each wind turbine in other wind turbines since its operation, and cleaning and discretizing the state parameter time series to obtain the characteristic values of the wind turbine state parameters corresponding to each other wind turbine.

S62, using the structural component fatigue damage database corresponding to each other wind turbine set obtained in step 5 as the input of the association model between structural component fatigue damage and wind turbine state parameters established in step 4 to optimize, and obtain a personalized association model between wind turbine state parameters and structural component fatigue damage corresponding to each wind turbine set.

S63, taking the characteristic value of the wind turbine state parameter corresponding to each other wind turbine obtained in S61 as input, using the wind turbine state parameter corresponding to each wind turbine in S62 and the personalized model for associating fatigue damage of structural components to calculate the corresponding fatigue damage value, and then accumulating the damage values corresponding to each discrete state parameter to obtain the fatigue damage of the wind turbine in the period.

According to Miner's linear accumulation theory of fatigue damage, the fatigue damage values of different components can be obtained by superimposing all the damages since the operation of the wind turbine, and then the fatigue life value can be obtained.

By performing the above analysis and calculation on all wind turbines in the field, the key fatigue life results of each wind turbine can be obtained.

Step 7: Online monitoring of fatigue life and real-time update of algorithms

1) Online monitoring of fatigue life

Through the previous calculation and analysis, the fatigue damage and fatigue life results of the structural components of the wind turbines in the whole field can be obtained. The above steps can be implemented by programming, and then the program is arranged on the wind farm monitoring computer to form a fatigue monitoring system with real-time fatigue load data and wind turbine state parameter data as input and fatigue damage and life of the wind turbines in the whole field as the result.

2) Real-time update of models and algorithms

Three models were established previously: 1) a model correlating the state parameters of wind turbines under measured loads and fatigue damage of structural components, 2) a model extrapolating fatigue damage of wind turbines under load measurement and wind turbines in the entire field, and 3) a model correlating the state parameters of each position obtained by optimizing the previous two models and fatigue damage of structural components.

These models are based on field measured fatigue load data. With the continuous accumulation of load measurement data of wind turbines, the samples for model training continue to grow, and the accuracy of the model will continue to improve. In this way, the accuracy of the fatigue life calculation results of wind turbines will continue to improve, and accurate monitoring of the fatigue life of wind turbines in the entire field will be achieved, providing fatigue safety protection for wind turbines.

Example 2

This embodiment provides a system for online monitoring of fatigue life of wind turbines in the entire field, including:

A unit selection unit is used to select multiple wind turbines from the same type of wind turbines in the wind farm to be tested as the actual test units;

A fatigue load database acquisition unit is used to discretize the fatigue load measurement data of each measured unit to obtain a measured fatigue load database;

A structural component fatigue damage database acquisition unit is used to convert the measured fatigue load database into a structural component fatigue damage database corresponding to each measured unit;

A model building unit, used to build a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbines;

Fatigue damage extrapolation unit, used to extrapolate the structural fatigue damage of the remaining wind turbines of the same model in the wind farm;

Fatigue life analysis unit is used for fatigue life analysis of structural components of wind turbines in the entire field.

Example 3

The present embodiment provides a terminal device, which includes a processor and a memory, wherein the memory is used to store a computer program, wherein the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding functions; the processor described in the embodiment of the present invention can be used for the operation of the method described in embodiment 1.

Example 4

The computer device provided in this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the method for calculating the composition of fluid in a reservoir reformed wellbore in the embodiment is implemented. To avoid repetition, it is not described one by one here. Alternatively, when the computer program is executed by the processor, the functions of each model/unit in the system of embodiment 2 are implemented. To avoid repetition, it is not described one by one here.

The computer device may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device may include, but is not limited to, a processor and a memory. The computer device may include more or fewer components than shown in the figure, or may combine certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

The memory can be an internal storage unit of a computer device, such as a hard disk or memory of a computer device. The memory can also be an external storage device of a computer device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device.

Furthermore, the memory may include both an internal storage unit of a computer device and an external storage device. The memory is used to store computer programs and other programs and data required by the computer device. The memory may also be used to temporarily store data that has been output or is to be output.

In another embodiment, this embodiment further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory (Non-Volatile Memory), such as at least one disk memory.

One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the method described in Example 1 in the above embodiments; one or more instructions in a computer-readable storage medium are loaded and executed by a processor to perform the steps of the method described in Example 1.

The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims (10)

1. A method for online monitoring of fatigue life of wind turbines in the entire field, characterized by comprising the following steps: Step 1, selecting multiple wind turbines from the same type of wind turbines in the wind farm to be tested as the actual test turbines; Step 2, discretizing the fatigue load measurement data of each measured unit to obtain a measured fatigue load database; Step 3, converting the measured fatigue load database into a structural component fatigue damage database corresponding to each measured unit; Step 4, establishing a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbine generator sets; Step 5, extrapolating the structural fatigue damage of the remaining wind turbines of the same type in the wind farm; Step 6: Fatigue life analysis of structural components of all wind turbines in the field. 2. The method for online monitoring fatigue life of wind turbines in the entire field according to claim 1, characterized in that in step 1, wind turbines that meet at least the following two conditions are selected as the actual measurement units from the wind turbines of the same model in the wind farm to be measured, and the conditions are: The first condition: wind turbines with average wind speed and instantaneous wind speed both greater than the preset threshold; The second condition: the two wind turbines with the largest turbulence difference; The third condition: Conduct a correlation analysis on the wind parameter data of all wind turbines of the same model in the field and select the wind turbine with the highest correlation. 3. The method for online monitoring of fatigue life of wind turbines in a whole field according to claim 1 is characterized in that, in step 2, the fatigue load measurement data of each measured unit is discretized to obtain a measured fatigue load database, and the specific method is: Perform fatigue load measurement on each measured unit determined in step 1 to obtain fatigue load data corresponding to each wind turbine unit; Extract the time series from the obtained fatigue load data, and combine the time series with the operating parameters and wind data of the corresponding wind turbine to form the fatigue load time series data corresponding to each measured turbine; The obtained fatigue load time series data are cleaned and divided into bins to obtain the measured fatigue load database corresponding to each measured unit. 4. The method for online monitoring of fatigue life of wind turbines in a whole field according to claim 1 is characterized in that in step 3, the measured fatigue load database is converted into a component fatigue damage database corresponding to each measured unit, and the specific method is: Establish the load-stress quantitative relationship of the structural components corresponding to each measured unit; According to the load-stress quantification relationship of the structural component, the measured fatigue time series load is converted into the time series stress of the structural component; The obtained structural component time-series stress is subjected to rain flow statistical analysis to obtain the fatigue damage value corresponding to the component time-series stress, and then the component fatigue damage database corresponding to each measured unit is obtained. 5. The method for online monitoring of fatigue life of wind turbines in the whole field according to claim 1 is characterized in that in step 4, a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbines is established, and the specific method is: The fatigue damage database of structural components corresponding to each measured unit is used as training samples, and a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbine units based on measured fatigue load data is established using a neural network method. 6. The method for online monitoring fatigue life of wind turbines in the entire field according to claim 1, characterized in that in step 5, the structural fatigue damage of the remaining wind turbines of the same type in the wind farm is extrapolated by: Obtain fatigue load simulation data corresponding to each wind turbine of the same type in the wind farm; Each group of fatigue load simulation data is cleaned and discretely divided into bins to obtain a simulation fatigue load database corresponding to each wind turbine, and the obtained simulation fatigue load database is converted into a structural component fatigue damage simulation database corresponding to each wind turbine; Using the multi-parameter correlation analysis method, combined with the fatigue damage simulation database of the structural components corresponding to each wind turbine, a component fatigue damage correlation model between each other wind turbine and the measured unit is established; The structural component fatigue damage database corresponding to each measured unit is used as the input of the component fatigue damage correlation model between each other wind turbine unit and the measured unit, so as to obtain the structural component fatigue damage database corresponding to each other wind turbine unit. 7. The method for online monitoring of fatigue life of wind turbines in the whole field according to claim 1, characterized in that in step 6, fatigue life analysis of structural components of wind turbines in the whole field is specifically performed by: For the measured units, the fatigue life of each component of each measured unit is calculated using the fatigue load data obtained from the actual measurement; For other wind turbines, the time series of wind turbine state parameters of each other wind turbine since its operation is obtained, and the time series of wind turbine state parameters is cleaned and discretely divided into bins to obtain the characteristic value of the wind turbine state parameter corresponding to each other wind turbine; The structural component fatigue damage database corresponding to each other wind turbine obtained in step 5 is used as the input of the correlation model between structural component fatigue damage and wind turbine state parameters established in step 4 to optimize the correlation model, so as to obtain a personalized correlation model between wind turbine state parameters and structural component fatigue damage corresponding to each wind turbine; The obtained characteristic values of the wind turbine state parameters corresponding to each other wind turbine are combined with the personalized model of association between the wind turbine state parameters and fatigue damage of structural components to obtain the fatigue life damage value corresponding to each other wind turbine. 8. A system for online monitoring of fatigue life of wind turbines in the entire field, characterized by comprising: A unit selection unit is used to select multiple wind turbines from the same type of wind turbines in the wind farm to be tested as the actual test units; A fatigue load database acquisition unit is used to discretize the fatigue load measurement data of each measured unit to obtain a measured fatigue load database; A structural component fatigue damage database acquisition unit is used to convert the measured fatigue load database into a structural component fatigue damage database corresponding to each measured unit; A model building unit, used to build a multi-parameter correlation model between fatigue damage of structural components and state parameters of wind turbines; Fatigue damage extrapolation unit, used to extrapolate the structural fatigue damage of the remaining wind turbines of the same model in the wind farm; Fatigue life analysis unit is used for fatigue life analysis of structural components of wind turbines in the entire field. 9. A computer device/equipment/system, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to claim 1. 10. A computer-readable storage medium having a computer program/instruction stored thereon, wherein the computer program/instruction implements the steps of the method according to claim 1 when executed by a processor.