A wind turbine blade life intelligent monitoring method and system

Abstract

The application discloses a kind of wind turbine blade life intelligent monitoring method and system, method includes steps: extracting blade low-frequency damage model feature, constructs blade fatigue damage database;High-frequency load analysis model is established, and then based on high-frequency load analysis model, blade load is calculated, and blade fatigue load is obtained;Based on blade fatigue damage database and blade fatigue load, blade fatigue life is obtained, and blade fatigue life is evaluated.The application avoids using complex nonlinear fatigue damage model, while ensuring simulation accuracy, simplifies the order of magnitude of calculation;SCADA data and wind resource data are read online, and the blade load is evaluated in real time by high-frequency load analysis model, which improves the accuracy and timeliness of online monitoring of blade cross-section fatigue load.

Description

TECHNICAL FIELD

[0001] The present application mainly relates to the field of wind power technology, and in particular to a wind turbine blade life intelligent monitoring method and system. BACKGROUND

[0002] Due to the complex structure of the wind turbine, it is inevitable to bear the effects of turbulence, wind shear, tower shadow effect, wake effect, etc., and it often operates in harsh external environments such as low temperature, wind sand, icing, etc. Therefore, the unit load, vibration fluctuation is large, the operation and maintenance is difficult and the cost is high. With the increase of the operation life of the wind turbine, the remaining service life of the unit is increasingly concerned. As the core component of wind turbine to capture wind energy, the blade is subjected to the coupling effect of bending, torsion, shear and other loads for a long time; with the large-scale of wind turbine, the size of the blade is getting larger and larger, and the damage of the blade is more complex, so it is of great significance to master the blade life in real time.

[0003] According to the retrieved patents and papers, the blade fatigue damage evaluation method is usually to calculate the cumulative fatigue value and the change characteristics of the cumulative fatigue value to represent the degree of blade fatigue damage, and linear damage accumulation theory is usually used to process the single-axis fatigue test results to obtain the fatigue performance of the blade under the combined action of flapwise and edgewise loads; or a complex nonlinear fatigue damage model is used to calculate the equivalent stress of the blade under multi-axial load to evaluate the fatigue performance of the blade. If the linear fatigue damage accumulation model is used, the estimation accuracy is greatly reduced, which is difficult to meet the accuracy requirements of the online monitoring system; if the nonlinear fatigue damage model is used, it is difficult to deploy in the online monitoring system and predict the life due to the large order of calculation. SUMMARY

[0004] The technical problem to be solved by the present application is that in view of the technical problems existing in the prior art, the present application provides a wind turbine blade life intelligent monitoring method and system for improving the calculation accuracy and efficiency.

[0005] To solve the above technical problems, the technical solution provided by the present application is:

[0006] A wind turbine blade life intelligent monitoring method, comprising the steps of:

[0007] Extracting blade low-frequency damage model features to construct a blade fatigue damage database;

[0008] Establishing a high-frequency load analysis model, and then calculating the blade load based on the high-frequency load analysis model to obtain the blade fatigue load;

[0009] Based on the blade fatigue damage database and the blade fatigue load, the blade fatigue life is obtained, and the blade fatigue life is evaluated.

[0010] Preferably, the specific process of constructing the blade fatigue damage database is as follows: extract the low-frequency damage model features of the blade through aeroelastic model and finite element model, and then obtain the blade fatigue damage database through aeroelastic model calibration, load time-domain simulation, rainflow calculation, unit load calculation, fatigue stress spectrum and SN curve calculation.

[0011] Preferably, in the load time-domain simulation, the load time-domain sequence of the wind turbine blades under the certification design conditions is calculated; in the rainflow calculation, rainflow counting is performed on the load time-domain sequence of the wind turbine blades to obtain the Markov matrix A for each section of the blades. design,i .

[0012] Preferably, in the unit load calculation, a full-size finite element model of the blade is established based on the blade parameters, and the maximum stress σ generated in each section of the blade under the unit load is calculated. i In the fatigue stress spectrum, calculate the Markov matrix A of blade section i. design,i Mean load F design-mean,i,j The corresponding average blade stress σ design-mean,i,j Where j = 1, 2, ..., m, a total of m load averages; and the Markov matrix A for calculating blade section i. design,i Medium load amplitude F design-amp,i,k The corresponding blade stress amplitude σ design-amp,i,k , where k = 1, 2, ..., p, for a total of p load amplitudes.

[0013] Preferably, in the SN curve calculation, the average blade stress σ is used as the basis. design-mean,i,j and blade stress amplitude σ design-amp,i,k Calculate the allowable load cycle number N for the blade. design,i,j,k .

[0014] Preferably, the specific process for establishing a high-frequency load analysis model is as follows:

[0015] A high-frequency load analysis model integrating the wind turbine subsystem and multiple systems is constructed.

[0016] The monitoring data of the blade root bolts of the wind turbine are obtained within a predetermined time period. The monitoring data of the blade root bolts is compared with the simulated blade root load of the high-frequency load analysis model, and the high-frequency load analysis model is then corrected based on the comparison results.

[0017] Preferably, the specific process for evaluating fatigue life is as follows:

[0018] Calculate the lifetime damage caused by load on the blade And assess the reliability R(T) of the blades. R ), and based on lifespan damage and reliability R(T) R Divide the area into different danger zones;

[0019] The blade stress average σ site-mean,i,j is compared with the design allowable stress σ design-mean,i,m ; when the stress average σ site-mean,i,j is less than the design allowable stress σ design-mean,i,m , the fatigue life state of the wind turbine blade is determined according to the life damage D and the reliability R(T R ); when the stress average σ site-mean,i,j is greater than the design allowable stress σ design-mean,i,m , the fatigue life state of the wind turbine blade is in a high-risk interval.

[0020] Preferably, the specific process of determining the interval in which the fatigue life state of the wind turbine blade is located according to the life damage and the reliability is as follows:

[0021] When D ≤ the first preset value and R(T R ) ≥ the second preset value, the wind turbine blade is in a normal working state, and the fatigue life prediction system tolerates this state;

[0022] When D i Life ≤ the first preset value and R(T R ) < the second preset value, the fatigue life state of the wind turbine blade is in a low-risk interval, and a yellow warning signal is sent out;

[0023] When D i Life > the first preset value and R(T R ) ≥ the second preset value, the fatigue life state of the wind turbine blade is in a medium-risk interval, and a yellow warning signal is sent out;

[0024] When D i Life > the first preset value and R(T R ) < the second preset value, the fatigue life state of the wind turbine blade is in a high-risk interval, and a red warning signal is sent out.

[0025] The application further discloses a wind turbine blade life intelligent monitoring system, which comprises a blade fatigue damage database, a high-frequency load analysis model and a data acquisition module, the data acquisition module is connected with the edge computing module, the data acquisition module acquires wind turbine SCADA data, wind resource data and blade root intelligent bolt data, analyzes the data through the high-frequency load analysis model to obtain blade fatigue load, obtains blade fatigue life based on the blade fatigue load, and evaluates the blade fatigue life.

[0026] Preferably, an edge computing module is further included, the blade fatigue damage database is deployed in the edge computing module, and the data acquisition module acquires wind turbine SCADA data, wind resource data and blade root intelligent bolt data and transmits the data to the edge computing module for analysis and processing.

[0027] Compared with the prior art, the wind turbine blade life intelligent monitoring method has the following advantages:

[0028] The present application considers the maximum stress generated by each cross section of the blade under unit load, constructs a blade fatigue damage database by extracting the characteristics of the low-frequency damage model of the blade, avoids using a complex nonlinear fatigue damage model, simplifies the order of calculation while ensuring simulation accuracy, and improves calculation accuracy and efficiency.

[0029] After the SCADA data and wind resource data are read online, the load of the blade is evaluated in real time by the high-frequency load analysis model, without the need to add a large number of sensors, so that the effect of online monitoring of the service life of the blade can be achieved, the precision and timeliness of online monitoring of the fatigue load of each cross section of the blade are improved, and the risk control capability of the operation and maintenance of the wind farm is improved.

[0030] The blade fatigue damage database is deployed in the edge computing module, and the data is processed in the edge node by edge computing of wind turbine SCADA data, wind resource data, blade root intelligent bolt data and other data, so that the transmission and processing of the data can be effectively reduced, and the speed and timeliness of data transmission are improved. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 The monitoring system of the present application is shown in the structure diagram of the embodiment.

[0032] Figure 2 The monitoring method of the present application is shown in the flowchart of the embodiment. DETAILED DESCRIPTION

[0033] The present application will be further described below in conjunction with the drawings and specific embodiments of the present application.

[0034] As shown in the drawings, Figure 2 the wind turbine blade life intelligent monitoring method of the embodiment of the present application comprises the steps of:

[0035] extracting the characteristics of the low-frequency damage model of the blade to construct a blade fatigue damage database;

[0036] establishing a high-frequency load analysis model, and then calculating the load of the blade based on the high-frequency load analysis model to obtain the fatigue load of the blade;

[0037] based on the blade fatigue damage database and the fatigue load of the blade, obtaining the fatigue life of the blade, and evaluating the fatigue life of the blade.

[0038] In one specific embodiment, the process of constructing the blade fatigue damage database is as follows: Low-frequency damage model features of the blade are extracted using an aeroelastic model and a finite element model. Then, through aeroelastic model calibration, load time-domain simulation, rainflow calculation, unit load calculation, fatigue stress spectrum calculation, and SN curve calculation, the blade fatigue damage database is finally obtained. Specifically, in the load time-domain simulation, the load time-domain sequence of the wind turbine blade under certification design conditions is calculated; in the rainflow calculation, rainflow counting is performed on the load time-domain sequence of the wind turbine blade to obtain the Markov matrix A for each section of the blade. design,i In the unit load calculation, a full-size finite element model of the blade is established based on the blade parameters to calculate the maximum stress σ generated in each section of the blade under a unit load. i In the fatigue stress spectrum, calculate the Markov matrix A of blade section i. design,i Mean load F design-mean,i,j The corresponding average blade stress σ design-mean,i,j Where j = 1, 2, ..., m, a total of m load averages; and the Markov matrix A for calculating blade section i. design,i Medium load amplitude F design-amp,i,k The corresponding blade stress amplitude σ design-amp,i,k Where k = 1, 2, ..., p, a total of p load amplitudes; in the SN curve calculation, based on the blade stress mean σ design-mean,i,j and blade stress amplitude σ design-amp,i,k Calculate the allowable load cycle number N for the blade. design,i,j,k .

[0039] This invention considers the maximum stress generated in each cross section of the blade under unit load, and constructs a blade fatigue damage database by extracting the features of the low-frequency damage model of the blade. This avoids the use of complex nonlinear fatigue damage models, simplifies the computational load while ensuring simulation accuracy, and improves computational accuracy and efficiency.

[0040] In one specific embodiment, the process of establishing a high-frequency load analysis model is as follows: construct a high-frequency load analysis model integrating the wind turbine subsystem and multiple systems, obtain the blade root bolt monitoring data within a predetermined time period of the wind turbine, compare the blade root bolt monitoring data with the simulated blade root load of the high-frequency load analysis model, and then correct the high-frequency load analysis model based on the comparison results, thereby ensuring the accuracy of subsequent data and improving the accuracy of life prediction.

[0041] This invention reads SCADA data and wind resource data online, and then uses a high-frequency load analysis model to evaluate the blade load in real time. Without the need to add a large number of sensors, it can achieve the effect of online monitoring of blade life, improve the accuracy and timeliness of online monitoring of fatigue loads on various sections of the blade, and enhance the risk management capabilities of wind farm operation and maintenance.

[0042] In one specific embodiment, the process of evaluating fatigue life is as follows: calculating the life damage caused by the load on the blade. And assess the reliability R(T) of the blades. R ), and based on lifespan damage and reliability R(T) R Divide the blades into different hazardous zones; calculate the average blade stress σ. site-mean,i,j With the design allowable stress σ design-mean,i,m Compare; when the mean stress σ site-mean,i,j Less than the design allowable stress σ design-mean,i,m Then, based on lifespan damage and reliability R(T) R Determine the range in which the fatigue life state of the wind turbine blades lies; when the mean stress σ site-mean,i,j Greater than the design allowable stress σ design-mean,i,m At that time, the fatigue life state of the wind turbine blades was in the high-risk range.

[0043] like Figure 1 As shown in the figure, this invention also discloses an intelligent monitoring system for the life of wind turbine blades, including a blade fatigue damage database, a high-frequency load analysis model, an edge computing module, and a data acquisition module. The blade fatigue damage database is deployed in the edge computing module, and the data acquisition module is connected to the edge computing module. The data acquisition module collects SCADA data of the wind turbine, wind resource data, and blade root smart bolt data, transmits them to the edge computing module for analysis and processing, and then analyzes them through the high-frequency load analysis model to obtain the blade fatigue load. Based on the blade fatigue load, the blade fatigue life is obtained, and the blade fatigue life is evaluated.

[0044] This invention deploys the blade fatigue damage database on the edge computing module, and processes the data at the edge node through edge computing, including wind turbine SCADA data, wind resource data, and blade root smart bolt data. This effectively reduces data transmission and processing, and improves the speed and timeliness of data transmission.

[0045] To better understand the above technical solution, the following will provide a detailed description of the technical solution in conjunction with the accompanying drawings and specific implementation methods:

[0046] 1) Feature extraction of low-frequency damage model and establishment of fatigue damage database for blade cross section

[0047] Low-frequency damage model features of the blade are extracted using aeroelastic and finite element models. Then, through aeroelastic model calibration, time-domain load simulation, unit load calculation, fatigue stress spectrum, and S-N curve calculation, a blade fatigue damage database is finally obtained and deployed in the edge module. The specific process is as follows:

[0048] 1.1) Calibration of simulation model: Based on the field test data, the simulation model of the wind turbine blade is calibrated;

[0049] 1.2) Load time domain simulation: For the normal power generation condition of wind speed of 4m / s-20m / s, the load time domain sequence of the wind turbine blade under the certification design condition is calculated;

[0050] 1.3) Rainflow counting: Rainflow counting is performed on the load time domain sequence of the wind turbine blade to obtain the Markov matrix A of each section of the blade design,i , wherein i=1, 2, …, n, a total of n blade sections;

[0051] 1.4) Unit load calculation: A full-size blade finite element model is established according to the blade parameters, and the maximum stress σ i of each section of the blade under the action of unit load is calculated;

[0052] 1.5) Fatigue stress spectrum:

[0053] A, the Markov matrix A of the blade section i is calculated design,i , wherein F design-mean,i,j is the load mean value corresponding to the blade stress mean value σ design-mean,i,j , wherein j=1, 2, …, m, a total of m load mean values:

[0054] wherein σ design-mean,i,j =F design-mean,i,j σ i

[0055] B, the Markov matrix A of the blade section i is calculated design,i , wherein F design-amp,i,k is the load amplitude value corresponding to the blade stress amplitude value σ design-amp,i,k , wherein k=1, 2, …, p, a total of p load amplitude values:

[0056] σ design-amp,i,k =F design-amp,i,k σ i

[0057] 1.6) S-N curve database: Based on the blade stress mean value σ design-mean,i,j and the blade stress amplitude value σ design-amp,i,k , the allowable load cycle number N design,i,j,k of the blade is calculated, and a fatigue damage database of each section of the blade is obtained;

[0058] wherein

[0059] m is the Wohler index; γ Ma is the static strength safety factor; γ Mb is the fatigue safety factor; σt σ is the characteristic tensile stress of the material. c This refers to the compressive characteristic stress of the material.

[0060] 2) Online evaluation of fatigue load using high-frequency load analysis model

[0061] 2.1) Establish a high-frequency load analysis model: Using the FAST open-source aeroelastic code as the core, construct a high-frequency load analysis model integrating the wind turbine subsystem and multiple systems;

[0062] 2.2) Collect on-site test data: The test unit needs to be tested for at least three months in order to collect more comprehensive blade root load data of the unit within the operating wind speed range;

[0063] 2.3) Calibrate the high-frequency load analysis model: Based on the field test data, calibrate the simulation model to output more accurate loads on each section of the blade. The main focus is on identifying the natural frequency and damping of the structure, the performance of the subsystem, the load levels of major components such as the blade, and the general dynamic behavior.

[0064] 2.4) Online simulation: The high-frequency load analysis model reads the unit's SCADA data and wind resource data online, calculates the load on the blades, and obtains the time-domain sequence of the load on each section of the blades during this operating phase;

[0065] 2.5) Model Correction: Compare the monitoring data of the blade root smart bolts with the simulated blade root load of the high-frequency load analysis model, and correct the high-frequency load analysis model online;

[0066] 2.6) Rainflow calculation: Obtain the Markov matrix A of fatigue load on each section of the blade during this operating phase. site,i , where i = 1, 2, ..., n, for a total of n blade sections;

[0067] 2.7) Fatigue stress at each section:

[0068] A. Calculate the Markov matrix A for blade section i. site,i Mean load F site-mean,i,j The corresponding average blade stress σ site-mean,i,j Where j = 1, 2, ..., m, there are a total of m load averages:

[0069] Where σ site-mean,i,j =F site-mean,i,j σ i

[0070] B. Calculate the Markov matrix A for blade section i. site,i Medium load amplitude F site-amp,i,k The corresponding blade stress amplitude σ site-amp,i,k Where k = 1, 2, ..., p, there are a total of p load amplitudes:

[0071] Where σ site-amp,i,k =F site-amp,i,k σ i

[0072] 2.8) Load Cycle Correction: Based on the mean blade stress σ site-mean,i,j and blade stress amplitude σ site-amp,i,k Calculate the number of load cycles n for the blade. site,i,j,k :

[0073]

[0074] 2.9) Damage assessment: Calculate the damage to the blade caused by the load based on the allowable number of load cycles.

[0075]

[0076] 2.10) Reliability Analysis: The reliability of the blades is assessed based on their fatigue reliability.

[0077]

[0078] Where f(t) is the failure density function of the blade material life t, T R Given a lifespan of 20 years.

[0079] 3) Wind turbine blade life assessment

[0080] Based on the average blade stress σ of the wind turbine blades under actual operating conditions site-mean,i,j and design allowable stress σ design-mean,i,m The size of the service life and allowable service life are divided into comparison and judgment processes. Based on the judgment of service life conditions and in combination with reliability, the dangerous range is divided into: normal, low dangerous range, medium dangerous range and high dangerous range. Then, different responses are made according to different dangerous ranges.

[0081] See details Figure 1 First, fatigue life and reliability are analyzed, and then the analysis results are judged.

[0082] The average stress σ of the blade site-mean,i,j With the design allowable stress σ design-mean,i,m Compare; when the mean stress σ site-mean,i,j Less than the design allowable stress σ design-mean,i,m Then, the comparison logic between service life and allowable service life is entered; when the average stress σ site-mean,i,j Greater than the design allowable stress σ design-mean,i,m The fatigue life of the wind turbine blades is in a high-risk range, requiring further inspection of their health. Specifically, the comparison logic is as follows:

[0083] when R(T R When the fatigue life is ≥0.9 (service life), the wind turbine blades are in normal working condition, and the fatigue life prediction system can withstand this condition.

[0084] when R(T R When the fatigue life of the wind turbine blade is less than 0.9 (service life), the fatigue life of the wind turbine blade is in the low danger range, and a yellow alarm signal is issued.

[0085] when R(T R When the fatigue life of the wind turbine blade is ≥0.9 (service life), it is in the medium danger range and a yellow alarm signal is issued.

[0086] when R(T R When the fatigue life of the wind turbine blade is less than 0.9 (service life), the fatigue life of the wind turbine blade is in a high-risk range, and a red alarm signal is issued to remind the user to stop the machine for maintenance in time, such as replacing the blade root bolts and checking the health status of the blade.

[0087] When the fatigue life of the wind turbine hub is in the low or medium danger range, a yellow alarm signal is issued. While continuing the next round of fatigue life status determination, the operating status of the wind turbine blades will be further determined based on information such as SCADA data and blade root smart bolt data.

[0088] 4) Construction of an intelligent blade life detection system based on edge computing

[0089] Based on the high-frequency load analysis model, and on the basis of the wind power Internet of Things architecture and big data platform, a collaborative and interactive intelligent monitoring system for wind turbine blade life is created, providing a digital model for the actual operating wind turbine units.

[0090] The system includes a blade data acquisition and detection device, a wireless receiving module (serial server) in the nacelle control cabinet, an edge computing module, and application modules required for intelligent analysis. It uses edge computing to carry out the wind turbine blade life detection business, and transmits data such as group SCADA data, wind resource data, and blade root smart bolts from the data acquisition terminal to the edge computing node at high speed through the network. It adopts an open platform that integrates network, computing, storage, and application core capabilities to provide the nearest end service.

[0091] Edge computing module: An edge computing module is built on the wind turbine with existing wireless network coverage. This enables terminal devices to transmit online monitoring data such as wind turbine SCADA data, wind resource data, and blade root smart bolt data to the edge computing module for analysis and processing via the network. At the same time, the blade section fatigue damage database is deployed on the edge computing module.

[0092] The construction of the edge computing module includes the following: access devices connect to base stations and aggregate to core network MEC equipment; provide rack-mountable edge computing modules; connect to data flow analysis platforms; and connect to IPRAN to provide unified management and customer authentication in 5G CPE.

[0093] The online assessment of blade life includes:

[0094] The homepage of the intelligent wind turbine blade life detection system showcases its main functions and provides a brief introduction to the key points of wind turbine blade life prediction and the significance of the project.

[0095] Wind Farm Module: The "Wind Farm Module" of the intelligent wind turbine blade life detection system contains multiple wind farms. Click to select a specific wind farm.

[0096] Blade Structure Module: After selecting a specific wind farm, click the "Blade Structure Module" icon to bring up sub-menus such as "Blade" and "Blade Root Bolts," where you can set and correct the material SN curve.

[0097] Online stress data: After selecting a specific wind farm, click the "Online Stress Data" icon to display the sub-menus "#1 Blade", "#2 Blade", and "#3 Blade". Each sub-menu displays the blade root bolt monitoring data for that blade. After clicking on a specific blade, you can view the real-time simulated stress data for each section of that blade.

[0098] Blade life prediction: After selecting a specific wind farm, click the "Blade Life Prediction" icon to view the remaining life prediction of the blades of each wind turbine in that wind farm.

[0099] As shown in this disclosure and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. The terms "first," "second," and similar terms used in this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms "comprising" or "including" mean that the element or object preceding the word covers the element or object listed after the word and its equivalents, without excluding other elements or objects. The terms "connected" or "linked" are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect.

[0100] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent monitoring of wind turbine blade life, characterized in that, Including the following steps: Extract features from low-frequency damage models of blades to construct a blade fatigue damage database; A high-frequency load analysis model is established, and the blade load is calculated based on the high-frequency load analysis model to obtain the blade fatigue load. Based on the blade fatigue damage database and blade fatigue load, the blade fatigue life is obtained and evaluated. Based on the average blade stress and blade stress amplitude Calculate the allowable load cycle number for the blade. To obtain a fatigue damage database for each section of the blade; in m The Wohler index; This is the static strength safety factor; The fatigue safety factor; The characteristic tensile stress of the material; The characteristic compressive stress of the material; i This is the blade cross-section; j =1, 2, ..., n, a total of n load averages; k =1, 2, ..., p, a total of p load amplitudes; Load cycle correction: based on blade stress mean and blade stress amplitude Calculate the number of load cycles for the blades. : Damage assessment: Calculate the damage to the blade caused by the load based on the allowable number of load cycles. ; Reliability Analysis: Evaluating Blade Reliability Based on Blade Fatigue Reliability : in For blade material life t The fault density function, For a given lifespan; Based on lifespan damage and reliability Divide the area into different danger zones; Average blade stress With the mean of allowable design stress Compare; when the stress mean Less than the average allowable stress Then, based on lifespan damage and reliability Determine the range in which the fatigue life state of the wind turbine blades lies; when the average stress... Greater than the average allowable stress At that time, the fatigue life state of the wind turbine blades was in the high-risk range.

2. The intelligent monitoring method for wind turbine blade life according to claim 1, characterized in that, The specific process of constructing the blade fatigue damage database is as follows: the low-frequency damage model features of the blade are extracted through aeroelastic model and finite element model, and then the blade fatigue damage database is finally obtained through aeroelastic model calibration, load time-domain simulation, rainflow calculation, unit load calculation, fatigue stress spectrum and SN curve calculation.

3. The intelligent monitoring method for wind turbine blade life according to claim 2, characterized in that, In the time-domain load simulation, the time-domain load sequence of the wind turbine blades under the certification design conditions is calculated; in the rainflow calculation, rainflow counting is performed on the time-domain load sequence of the wind turbine blades to obtain the Markov matrix of each blade section. .

4. The intelligent monitoring method for wind turbine blade life according to claim 3, characterized in that, In the unit load calculation, a full-size finite element model of the blade is established based on the blade parameters to calculate the maximum stress generated in each section of the blade under a unit load. In the fatigue stress spectrum, calculate the Markov matrix of blade section i. Mean of medium load Corresponding average blade stress , where j = 1, 2, ..., m, for a total of m load averages; And calculate the Markov matrix of blade section i. Medium load amplitude Corresponding blade stress amplitude , where k=1,2,…,p, a total of p load amplitudes.

5. The intelligent monitoring method for wind turbine blade life according to any one of claims 1-4, characterized in that, The specific process of establishing a high-frequency load analysis model is as follows: Construct a high-frequency load analysis model for wind turbine subsystems and multi-system integration; The monitoring data of the blade root bolts of the wind turbine are obtained within a predetermined time period. The monitoring data of the blade root bolts is compared with the simulated blade root load of the high-frequency load analysis model, and the high-frequency load analysis model is then corrected based on the comparison results.

6. The intelligent monitoring method for wind turbine blade life according to claim 1, characterized in that, The specific process for determining the fatigue life state of wind turbine blades based on life damage and reliability is as follows: when ≤ First preset value and ≥ The second preset value indicates that the wind turbine blades are in normal working condition and the fatigue life prediction system can tolerate this condition. when ≤ First preset value and <The second preset value indicates that the fatigue life of the wind turbine blades is in the low-risk range, triggering a yellow alarm signal; when >First preset value and If the fatigue life of the wind turbine blades is within the medium-danger range and the value is greater than or equal to the second preset value, a yellow alarm signal will be issued. when >First preset value and The second preset value indicates that the fatigue life of the wind turbine blades is in a high-risk range, triggering a red alarm signal.

7. A wind turbine blade life intelligent monitoring system, used to perform the steps of the wind turbine blade life intelligent monitoring method as described in any one of claims 1-6, characterized in that, It includes a blade fatigue damage database, a high-frequency load analysis model, and a data acquisition module. The data acquisition module collects SCADA data, wind resource data, and blade root smart bolt data from the wind turbine. It then analyzes these data using the high-frequency load analysis model to obtain the blade fatigue load, and finally calculates and evaluates the blade fatigue life based on the load.

8. The intelligent monitoring system for wind turbine blade life according to claim 7, characterized in that, It also includes an edge computing module, in which the blade fatigue damage database is deployed. The data acquisition module is connected to the edge computing module, and the data acquisition module collects wind turbine SCADA data, wind resource data and blade root smart bolt data and transmits them to the edge computing module for analysis and processing.

Images