Method and device for determining health state of wind generating set

Abstract

The invention discloses a health state determination method and device of a wind generating set, and relates to the technical field of wind power generation. Comprising the steps of calculating blade state data on a preset time window based on blade monitoring data, and calculating tower barrel state data on the preset time window based on tower barrel monitoring data; inputting the blade state data, the tower barrel state data and the operation state data of the wind generating set into a tower barrel bending moment prediction model to obtain theoretical maximum bending moments of sections corresponding to the height positions of a plurality of tower barrels; calculating the actual maximum bending moment of the corresponding section based on the stress data of the sections corresponding to the height positions of the plurality of tower drums; performing data processing on the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain a tower drum bending moment deviation and a self-adaptive bending moment threshold value; and the health state of the wind generating set is judged based on the tower barrel bending moment deviation and the self-adaptive bending moment threshold value. The accuracy of health state evaluation of the wind generating set can be improved.

Description

Technical Field

[0001] This application relates to the field of wind power generation technology, and in particular to a method and apparatus for determining the health status of a wind turbine generator set. Background Technology

[0002] In recent years, with the rapid development of the wind power industry, wind turbine generators have shown a significant trend towards larger sizes, making continuous monitoring of their health status crucial for ensuring safe operation. In practical applications, mainstream technologies for wind turbine generator structural monitoring primarily focus on monitoring single structural components (such as blades or towers), collecting data on blade loads and tower tilt angles through sensor deployment to achieve structural early warning analysis and assessment. However, existing monitoring methods have significant shortcomings: First, monitoring only a single structural component of the wind turbine generator lacks collaborative analysis of multi-source data, making it difficult to comprehensively reflect the state of the wind turbine generator; furthermore, the limitations of single-component monitoring are highlighted by the increasing size of wind turbine generators and the influence of special weather conditions, resulting in insufficient accuracy of early warnings due to a lack of holistic assessment. Second, load warnings from strain monitoring rely on manually set thresholds, which must be determined based on blade design parameters. However, the design drawings and parameters of many older units are no longer available, making threshold calculation extremely difficult. Therefore, there is an urgent need to design a method for determining the health status of wind turbine generators to improve the accuracy of health status assessments. Summary of the Invention

[0003] This application provides a method and apparatus for determining the health status of wind turbine generator sets, which can improve the accuracy of health status assessment of wind turbine generator sets.

[0004] In a first aspect, this application provides a method for determining the health status of a wind turbine generator set, the method comprising:

[0005] The blade status data for a preset time window is calculated based on the blade monitoring data of the wind turbine generator set, and the tower status data for the preset time window is calculated based on the tower monitoring data of the wind turbine generator set.

[0006] Based on the blade state data, the tower state data, and the wind turbine's operating state data, the theoretical maximum bending moment of the cross-section corresponding to multiple tower height positions is obtained by inputting them into a pre-trained tower bending moment prediction model.

[0007] The actual maximum bending moment of the corresponding section is calculated based on the stress data of the sections at multiple tower height locations;

[0008] Data processing is performed on the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain the tower bending moment deviation and the adaptive bending moment threshold;

[0009] The health status of the wind turbine generator is determined based on the tower bending moment deviation and the adaptive bending moment threshold.

[0010] Furthermore, the step of calculating the actual maximum bending moment of the corresponding cross-section based on the stress data of the cross-sections corresponding to the height positions of multiple towers includes: calculating the bending moment of the corresponding cross-section based on the stress data of the cross-sections corresponding to the height positions of multiple towers; performing linear fitting on the bending moment of each cross-section to obtain the bending moment distribution of the corresponding cross-section; and calculating the actual maximum bending moment of the corresponding cross-section at each time point within a preset time window based on the bending moment distribution.

[0011] Furthermore, the process of processing the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain the tower bending moment deviation includes: calculating the difference between the theoretical maximum bending moment and the actual maximum bending moment of each section at the same time point within a preset time window; taking the maximum value of the difference as the tower bending moment deviation at the corresponding time point; and traversing each time point within the preset time window to obtain the tower bending moment deviation at each time point within the preset time window.

[0012] Furthermore, the step of processing the theoretical maximum bending moment and actual maximum bending moment of each section to obtain an adaptive bending moment threshold includes: obtaining an initial bending moment threshold and a first bending moment deviation threshold, a first local bending moment deviation mean, an extreme value boundary line, and a first extreme value fluctuation vector corresponding to the initial bending moment threshold; the initial bending moment threshold is determined by the modeling dataset, the first bending moment deviation threshold is the tower bending moment deviation threshold corresponding to the last time point of the modeling dataset, the first local bending moment deviation mean is the tower bending moment deviation mean corresponding to a preset number of time points before the last time point of the modeling dataset, and the first extreme value fluctuation vector is the set of extreme value fluctuations at all time points in the modeling dataset; calculating the corresponding local feature quantification index for the tower bending moment deviation at each time point within a preset time window; and iteratively calculating the local feature quantification index corresponding to each time point based on the first bending moment deviation threshold, the first local bending moment deviation mean, and the extreme value boundary line to obtain the adaptive bending moment threshold corresponding to each time point.

[0013] Furthermore, the local feature quantification index includes the mean local bending moment deviation, local fluctuation, and extreme value fluctuation. The step of iteratively calculating the local feature quantification index corresponding to each time point based on the first bending moment deviation threshold, the first mean local bending moment deviation, and the extreme value boundary line to obtain the adaptive bending moment threshold for each time point includes: if the local fluctuation at the current time point is less than the extreme value boundary line, then the adaptive bending moment threshold remains unchanged; if the local fluctuation at the current time point is greater than the extreme value boundary line, and the local fluctuation at the current time point is greater than the adaptive bending moment threshold at the previous time point, then the adaptive bending moment threshold at the previous time point is used as the adaptive bending moment threshold at the current time point; if the local fluctuation at the current time point is not less than the extreme value boundary line, and the local fluctuation at the current time point is less than the adaptive bending moment threshold at the previous time point, then the extreme value fluctuation at the current time point is included in the first extreme value fluctuation vector, and the adaptive bending moment threshold at the current time point is recalculated iteratively based on the first bending moment deviation threshold and the first mean local bending moment deviation to obtain the adaptive bending moment threshold at the current time point.

[0014] Furthermore, the blade monitoring data includes blade load data and blade vibration data; the calculation of blade state data within a preset time window based on the blade monitoring data of the wind turbine includes: calculating the average blade load borne by the blade during one rotation based on the blade load data; and calculating the average blade vibration acceleration during one rotation based on the blade vibration data.

[0015] Furthermore, the tower monitoring data includes tower vibration data and tower tilt angle data; the calculation of tower status data within the preset time window based on the tower monitoring data of the wind turbine includes: calculating the average vibration acceleration of the tower based on the tower vibration data; and calculating the top displacement of the tower based on the tower tilt angle data.

[0016] Furthermore, a first preset number of fiber optic strain sensors and a second preset number of temperature sensors are respectively installed on the tower wall at multiple tower height positions in the wind turbine generator set.

[0017] Furthermore, a third preset number of blade load sensors, a fourth preset number of temperature sensors, and a fifth preset number of acceleration sensors are arranged on the tower wall at preset positions of the blades in the wind turbine generator set.

[0018] Secondly, this application provides a health status determination device for a wind turbine generator set, the device comprising:

[0019] The monitoring data processing module is used to calculate blade status data within a preset time window based on the blade monitoring data of the wind turbine generator set, and to calculate tower status data within the preset time window based on the tower monitoring data of the wind turbine generator set.

[0020] The theoretical bending moment prediction module is used to input the blade state data, the tower state data and the wind turbine's operating state data into a pre-trained tower bending moment prediction model to obtain the theoretical maximum bending moment of multiple tower height positions corresponding to the cross sections.

[0021] The actual bending moment calculation module is used to calculate the actual maximum bending moment of the corresponding section based on the stress data of the section corresponding to multiple tower height positions;

[0022] The bending moment threshold adjustment module is used to process the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain the tower bending moment deviation and the adaptive bending moment threshold.

[0023] A health status assessment module is used to determine the health status of the wind turbine generator set based on the tower bending moment deviation and the adaptive bending moment threshold. To address the shortcomings of existing technologies, this application provides a method for determining the health status of a wind turbine generator set. This method offers the following advantages: It integrates blade monitoring data, tower monitoring data, and generator set operating status data, overcoming the limitations of monitoring single components. Multi-source data is collaboratively input into the tower bending moment prediction model, covering the operating characteristics of key generator set structures and incorporating environmental parameters, enabling a comprehensive assessment of the entire generator set. This effectively avoids interference from special weather conditions on single monitoring points and comprehensively reflects the true state of the wind turbine generator set. Furthermore, this application dynamically adjusts the assessment criteria by calculating theoretical and actual bending moment deviations, combined with an adaptive threshold algorithm. It eliminates the need to rely on missing design parameters from older generator sets, solving the traditional threshold calculation problem. Simultaneously, it ensures the threshold closely matches the real-time operating status of the generator set, improving the accuracy of the wind turbine generator set's health status assessment.

[0024] It should be noted that the aforementioned computer instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the processor of the wind turbine generator health status determination device, or it may be packaged separately from the processor of the wind turbine generator health status determination device; this application does not impose any limitations on this.

[0025] The description of the second aspect in this application can be referred to the detailed description of the first aspect; and the beneficial effects of the description of the second aspect can be referred to the analysis of the beneficial effects of the first aspect, which will not be repeated here.

[0026] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description.

[0027] It is understood that before using the technical solutions disclosed in the various embodiments of this application, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this application in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart illustrating a method for determining the health status of a wind turbine generator set, provided in an embodiment of this application;

[0030] Figure 2 This is a schematic diagram illustrating the configuration of an optical fiber strain sensor on a tower, as provided in an embodiment of this application.

[0031] Figure 3 This is a schematic diagram illustrating the configuration of an acceleration sensor on a tower, as provided in an embodiment of this application.

[0032] Figure 4 This is a schematic diagram illustrating the configuration of a sensor device on a blade, as provided in an embodiment of this application.

[0033] Figure 5 A schematic diagram showing the theoretical and actual bending moments of the first-layer tower section calculated based on multi-source data, provided for embodiments of this application;

[0034] Figure 6 A schematic diagram of a health status determination device for a wind turbine generator set provided in an embodiment of this application;

[0035] Figure 7 This is a block diagram of an electronic device used to implement a method for determining the health status of a wind turbine generator set according to an embodiment of this application. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0037] The terms "first," "second," "target," and "original," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein. Furthermore, the terms "comprising," "having," and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0038] It should be noted that the information collected in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant countries and regions, necessary confidentiality measures have been taken, and it does not violate public order and good morals. Corresponding operation portals are provided for users to choose to authorize or refuse.

[0039] Figure 1 This is a flowchart illustrating a method for determining the health status of a wind turbine generator set according to an embodiment of this application. This embodiment is applicable to scenarios where the health status of a wind turbine generator set is assessed by analyzing blade monitoring data, tower monitoring data, and operational status data, combined with an adaptive bending moment threshold. The method for determining the health status of a wind turbine generator set provided in this embodiment can be executed by the device for determining the health status of a wind turbine generator set provided in this application. This device can be implemented through software and / or hardware and integrated into the electronic device executing this method.

[0040] See Figure 1 The method in this embodiment includes, but is not limited to, the following steps:

[0041] S110. Calculate blade status data within a preset time window based on the blade monitoring data of the wind turbine generator set, and calculate tower status data within a preset time window based on the tower monitoring data of the wind turbine generator set.

[0042] Blade monitoring data refers to raw data collected by sensors deployed on or inside the blade surface. Tower monitoring data refers to raw data collected by sensors deployed on the tower body. The preset time window can be a fixed, manually set duration (e.g., one minute), designed to convert high-frequency real-time data into statistically significant state data. Blade state data refers to data obtained by statistically calculating the blade monitoring data within the preset time window, quantifying the blade's operational status during that period. Tower state data refers to data obtained by statistically calculating the tower monitoring data within the preset time window, quantifying the overall stress and deformation state of the tower during that period.

[0043] In one embodiment, a first preset number of fiber optic strain sensors and a second preset number of temperature sensors are respectively installed on the tower wall at multiple tower height positions of the wind turbine tower. These multiple tower height positions can cover the lower, middle, and upper regions of the tower. For example, if the tower height is 80 meters, eight fiber optic strain sensors and three temperature sensors are installed on the tower wall at corresponding height positions of 20 meters, 50 meters, and 70 meters.

[0044] like Figure 2 This is a schematic diagram of the configuration of fiber optic strain sensors on a tower according to an embodiment of this application. The diagram shows a cross-section of the tower wall at a certain height. Eight fiber optic strain sensors are configured on a ring of the tower wall at the following cross-section. The installation angles of these eight fiber optic strain sensors on the tower wall are 0°, 45°, 90°, 135°, 180°, 225°, 270° and 315°, respectively.

[0045] In one embodiment, dual-axis tilt sensors are installed on the tower walls in both the lower and upper regions of the tower to monitor tower displacement. For example, if the tower height is 80 meters, one dual-axis tilt sensor is installed on the tower walls at corresponding heights of 20 meters and 70 meters.

[0046] In one embodiment, acceleration sensors are respectively installed on the tower wall in the middle and upper regions of the tower to monitor the tower's vibration. For example, if the tower height is 80 meters, one acceleration sensor is installed on the tower wall at the corresponding heights of 50 meters and 70 meters.

[0047] like Figure 3 This is a schematic diagram illustrating the configuration of an acceleration sensor on a tower according to an embodiment of this application. The diagram shows a side of the tower wall at a certain height, on which an acceleration sensor is configured. The line connecting the mounting holes on both sides of the acceleration sensor (as shown by line xy in the diagram) is parallel to the foot platform, and the fiber output direction of the acceleration sensor (as shown by arrow z in the diagram) points towards the ground.

[0048] In one embodiment, a third preset number of blade load sensors, a fourth preset number of temperature sensors, and a fifth preset number of acceleration sensors are configured at preset positions on the blades of the wind turbine generator set. For example, four blade load sensors, four temperature sensors, and one acceleration sensor are configured at one-third of the distance from the blade root on each blade.

[0049] like Figure 4 This is a schematic diagram of a sensor device configured on a blade according to an embodiment of this application. The diagram shows a cross-section of a blade at a certain location, on which four blade load sensors (P1 to P4 in the diagram) and four temperature sensors (T1 to T4 in the diagram) are configured. The diagram also shows the windward side (or pressure side), leeward side (or pressure-raising side), trailing edge, leading edge, and blade web of a blade.

[0050] In one embodiment, the blade monitoring data includes blade load data and blade vibration data. The tower monitoring data includes tower vibration data and tower tilt angle data.

[0051] Specifically, based on the blade monitoring data of the wind turbine generator, the blade status data is calculated within a preset time window, including: calculating the average blade load borne by the blade during one rotation based on the blade load data; and calculating the average blade vibration acceleration during one rotation based on the blade vibration data.

[0052] In this embodiment, the sampling frequency of the blade load signal is 50Hz, that is, there are 3000 time series points per minute. The maximum number of complete cycles is taken, and the average load on the blade during one rotation is calculated using formula (1):

[0053] (1);

[0054] In the formula, It is the average blade load at the i-th minute. is the j-th load signal recorded by the blade load sensor within the i-th minute, n is the sequence number of the first load signal peak within the i-th minute, and N is the sequence number of the last load signal peak within the i-th minute.

[0055] The sampling frequency of the blade acceleration signal is 100Hz, which means there are 6000 time series points per minute. Taking the largest number of complete cycles, the average vibration acceleration when the blade rotates one revolution is calculated using formula (2):

[0056] (2);

[0057] In the formula, It is the average vibration acceleration of the blade at the i-th minute. is the j-th load signal recorded by the blade acceleration sensor within the i-th minute, n is the sequence number of the first load signal peak within the i-th minute, and N is the sequence number of the last load signal peak within the i-th minute.

[0058] Specifically, the tower status data within a preset time window is calculated based on the tower monitoring data of the wind turbine generator set, including: calculating the average vibration acceleration of the tower based on the tower vibration data; and calculating the top displacement of the tower based on the tower tilt angle data.

[0059] In this embodiment, the sampling frequency of the tower acceleration signal is 50Hz, that is, there are 3000 time series points per minute. The maximum number of complete cycles is taken, and the average acceleration of the tower per minute is calculated by formula (3):

[0060] (3);

[0061] In the formula, It is the average vibration acceleration of the tower at the i-th minute. is the j-th load signal recorded by the tower acceleration sensor within the i-th minute, n is the sequence number of the first load signal peak within the i-th minute, and N is the sequence number of the last load signal peak within the i-th minute.

[0062] The displacement (sway) at the top of the tower is calculated from the signals of the dual-axis tilt sensors at the top and bottom of the tower. The average displacement of the tower top per minute under a fixed period is calculated using formula (4):

[0063] (4);

[0064] In the formula, This represents the displacement at the top of the tower, where l is the direction of the dual-axis tilt sensor and H is the tower height. It is the tilt angle of the top of the tower in the l direction. It is the tilt angle of the bottom of the tower in the l direction.

[0065] S120. Based on the blade state data, tower state data, and wind turbine operating state data, the pre-trained tower bending moment prediction model is input to obtain the theoretical maximum bending moment of the cross-section corresponding to multiple tower height positions.

[0066] The wind turbine generator set is equipped with a Supervisory Control and Data Acquisition (SCADA) system. This system collects real-time operational status data of the wind turbine generator set, which may include at least wind speed, active power, blade pitch angle, rotor speed, and outside temperature. The tower bending moment prediction model can be trained using an autoencoder algorithm in unsupervised learning. Alternatively, it can be a prediction model trained on historical multi-source data, taking blade condition, tower condition, and generator set operational status data as input, and outputting theoretical bending moment values ​​for different tower height sections. Its core function is to establish a mapping relationship between input features and tower bending moments. The theoretical maximum bending moment can be the theoretical value of the maximum stress bending moment of the tower.

[0067] In this embodiment, the three types of input data—blade status data, tower status data, and unit operating status data—are first cleaned, normalized, and time-series aligned to form a standardized dataset conforming to the model input format. The standardized data is then input into a pre-trained tower bending moment prediction model. Through forward propagation calculations, the nonlinear mapping relationship between the input data and the tower bending moment is automatically extracted, such as the combined influence of blade vibration characteristics, tower deformation state, and unit power output on the bending moment. The model outputs the theoretical maximum bending moment for multiple preset tower height positions (e.g., key sections at 1 meter, 20 meters, 50 meters, and 70 meters above the ground).

[0068] The theoretical maximum bending moment of each section can be expressed as: Let i represent the time series and j represent the section number. In the calculation, the relevant data and sensor signals need to be normalized so that the value of each parameter is within the range of [-1, 1] before being substituted into the machine learning algorithm. For example... Figure 5 The diagram shows the theoretical and actual bending moments of the first-layer tower section calculated based on multi-source data, which are provided for embodiments of this application. The two are basically linearly correlated.

[0069] S130. Calculate the actual maximum bending moment of the corresponding section based on the stress data of the sections corresponding to multiple tower height positions.

[0070] The stress data can be the stress-deformation data of the tower section material acquired through fiber optic strain sensors, directly reflecting the stress state of the section. The actual maximum bending moment can be the actual value of the maximum bending moment of the tower.

[0071] Specifically, the calculation of the actual maximum bending moment of the corresponding section based on the stress data of the section corresponding to multiple tower height positions includes: calculating the bending moment of the corresponding section based on the stress data of the section corresponding to multiple tower height positions; performing linear fitting on the bending moment of each section to obtain the bending moment distribution of the corresponding section, that is, the bending moment distribution at different tower height positions; and calculating the actual maximum bending moment of the corresponding section at each time point within a preset time window based on the bending moment distribution.

[0072] When a tower is subjected to load in space, it will generate bending moments in two perpendicular directions, x and y. Since the x and y directions are perpendicular to each other, the bending moments in the x and y directions of a certain section can be calculated using the Pythagorean theorem.

[0073] S140. Perform data processing on the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain the tower bending moment deviation and the adaptive bending moment threshold.

[0074] Among them, the tower bending moment deviation can be the maximum value of the difference between the theoretical maximum bending moment and the actual maximum bending moment among all monitored sections of the tower at the same time point.

[0075] Specifically, the theoretical maximum bending moment and the actual maximum bending moment of each section are processed to obtain the tower bending moment deviation, including: calculating the difference between the theoretical maximum bending moment and the actual maximum bending moment of each section at the same time point within a preset time window; taking the maximum value of the difference as the tower bending moment deviation at the corresponding time point; and iterating through each time point within the preset time window to obtain the tower bending moment deviation at each time point within the preset time window.

[0076] In this embodiment, two types of data are first extracted from all time points within a preset time window: the theoretical maximum bending moment of each section output from the tower bending moment prediction model, and the actual maximum bending moment calculated from stress data conversion and fitting. Two sets of data at the same time point and the same section are precisely aligned using timestamps, and time points with missing or abnormal data (such as invalid actual bending moment values ​​caused by sensor malfunctions or theoretical bending moment values ​​with abnormal model inference) are removed to ensure the accuracy and completeness of data pairing. For each time point, all sections corresponding to the tower height are traversed, and the difference between the theoretical maximum bending moment and the actual maximum bending moment of each section is calculated, resulting in a set of bending moment differences for each section at that time point. Extreme value filtering is performed on the set of bending moment differences for each section at each time point, and the maximum value represents the overall deviation level for the current period, which is then determined as the tower bending moment deviation at that time point. Subsequently, all time points within the preset time window are traversed in chronological order, and the above difference calculation and extreme value filtering steps are repeated, finally outputting the tower bending moment deviation corresponding to each time point within the time window, forming continuous deviation time-series data.

[0077] Specifically, the theoretical maximum bending moment and actual maximum bending moment of each section are processed to obtain an adaptive bending moment threshold. This includes: obtaining an initial bending moment threshold and the corresponding first bending moment deviation threshold, first local bending moment deviation mean, extreme value boundary line, and first extreme value fluctuation vector. The initial bending moment threshold is determined by the modeling dataset. The first bending moment deviation threshold is the tower bending moment deviation threshold corresponding to the last time point of the modeling dataset. The first local bending moment deviation mean is the tower bending moment deviation mean corresponding to a preset number of time points before the last time point of the modeling dataset. The first extreme value fluctuation vector is the set of extreme value fluctuations at all time points in the modeling dataset. The corresponding local feature quantification index is calculated for the tower bending moment deviation at each time point within a preset time window. Based on the first bending moment deviation threshold, the first local bending moment deviation mean, and the extreme value boundary line, the local feature quantification index corresponding to each time point is iteratively calculated to obtain the adaptive bending moment threshold corresponding to each time point.

[0078] Optional, local characteristic quantification indicators include the mean local bending moment deviation, local fluctuation, and extreme value fluctuation.

[0079] Furthermore, based on the first bending moment deviation threshold, the first local bending moment deviation mean, and the extreme value boundary line, the local feature quantification index corresponding to each time point is iteratively calculated to obtain the adaptive bending moment threshold corresponding to each time point. This includes: if the local fluctuation amount at the current time point is less than the extreme value boundary line, the adaptive bending moment threshold is kept unchanged, and the process proceeds directly to the next time point; if the local fluctuation amount at the current time point is greater than the extreme value boundary line, and the local fluctuation amount at the current time point is greater than the adaptive bending moment threshold of the previous time point, the adaptive bending moment threshold of the previous time point is used as the adaptive bending moment threshold of the current time point; if the local fluctuation amount at the current time point is not less than the extreme value boundary line and is less than the adaptive bending moment threshold of the previous time point, the extreme value fluctuation amount at the current time point is included in the first extreme value fluctuation amount vector, and the adaptive bending moment threshold at the current time point is iteratively fitted and calculated again based on the first bending moment deviation threshold and the first local bending moment deviation mean to obtain the adaptive bending moment threshold of the current time point. The advantage of doing this is that new extreme value data can be added to the modeling dataset, and γ can be refitted to make the exponential distribution more closely match the current data characteristics.

[0080] In this embodiment, the process of determining the initial bending moment threshold can be as follows: based on the modeling dataset, the tower bending moment deviation for each time series i is obtained, and then the following formula (5) is used to determine the initial bending moment threshold. In turn, they are converted into local fluctuations:

[0081] (5);

[0082] In the formula, It is the local fluctuation at point i in the time series. It is the tower bending moment deviation of time series i. It is before time series point i The mean local bending moment deviation of a time series The value is set manually, and in this embodiment it is set to 6; K is the number of time series data in the modeling dataset.

[0083] Will Arranged from largest to smallest, the t%th element The value of (denoted as T) is used as the extreme value boundary. The q%th The value of is taken as the extreme value among the extreme values, i.e., the threshold line. The extreme value fluctuation and the probability density function of the extreme value fluctuation are expressed by the following formula (6):

[0084] (6);

[0085] In the formula, It is the extreme fluctuation of time series point i; T is the extreme value boundary; according to extreme value theory The distribution of the values ​​is independent of the dataset itself and follows an exponential distribution. It is the probability density function of the extreme value fluctuation, and γ is the formal parameter of the exponential distribution function.

[0086] The initial bending moment threshold of the tower is expressed by the following formula (7):

[0087] (7);

[0088] In the formula, Let be the initial bending moment threshold of the tower at time i, r be the correction term for the tower bending moment deviation threshold, t be the percentage value of the extreme value boundary line in all local fluctuations, and q be the percentage value of the extreme value (threshold line) in all local fluctuations.

[0089] Compared to the traditional fixed bending moment threshold, this solution uses the local bending moment deviation mean, data re-iteration fitting (i.e., incorporating the extreme value fluctuation of the current time point into the first extreme value fluctuation vector), and dynamic threshold correction (i.e., recalculating the adaptive bending moment threshold of the current time point with each update). This not only avoids misjudging normal fluctuations but also provides timely alarms when there are real anomalies, ensuring the structural safety of the tower and the stable operation of the wind turbine.

[0090] S150. Determine the health status of wind turbine generator sets based on tower bending moment deviation and adaptive bending moment threshold.

[0091] In this embodiment, after obtaining the tower bending moment deviation, it is compared with an adaptive threshold. The bending moment health status of the wind turbine generator can be divided into three levels: deviation less than 0.1 is excellent, deviation between 0.2 and the threshold is good, and deviation exceeding the threshold is poor.

[0092] For example, assuming a 2.2MW wind turbine tower with a length of 90 meters, the fiber optic strain sensor signals of the blades and the SCADA monitoring data of the unit are collected. The model is then built according to the steps above. t=5, that is, the value of the 5th largest deviation between the theoretical maximum bending moment and the actual maximum bending moment in the modeling data is taken as the extreme value boundary. q=1, and the threshold for each time point is calculated sequentially using 6 data points as the sliding window. During the 4-hour operation, the tower bending moment deviation did not exceed the adaptive threshold and was in good condition.

[0093] The technical solution provided in this embodiment calculates blade status data within a preset time window based on blade monitoring data of the wind turbine generator set, and calculates tower status data within the preset time window based on tower monitoring data of the wind turbine generator set. The blade status data, tower status data, and wind turbine generator set operation status data are input into a pre-trained tower bending moment prediction model to obtain the theoretical maximum bending moment of multiple tower height positions corresponding to different sections. The actual maximum bending moment of each section is calculated based on stress data from these sections. Data processing is performed on the theoretical and actual maximum bending moments of each section to obtain tower bending moment deviation and adaptive bending moment threshold. The health status of the wind turbine generator set is determined based on the tower bending moment deviation and adaptive bending moment threshold. This application integrates blade monitoring data, tower monitoring data, and generator set operation status data, breaking the limitations of single-component monitoring. The collaborative input of multi-source data into the tower bending moment prediction model covers the operational characteristics of key generator set structures and incorporates environmental parameters, enabling a holistic assessment of the entire generator set. This effectively avoids interference from special weather conditions on single monitoring and comprehensively reflects the true state of the wind turbine generator set. Furthermore, this application dynamically adjusts the judgment criteria by calculating the deviation between theoretical and actual bending moments, combined with an adaptive threshold algorithm. It eliminates the need to rely on missing design parameters from older units, solving the traditional threshold calculation problem. Simultaneously, it ensures the threshold closely aligns with the real-time operating status of the unit, thereby improving the accuracy of health status assessment for wind turbine generators.

[0094] Figure 6 This is a schematic diagram of a health status determination device for a wind turbine generator provided in an embodiment of this application, as shown below. Figure 6 As shown, the device 600 may include:

[0095] The monitoring data processing module 610 is used to calculate blade status data within a preset time window based on the blade monitoring data of the wind turbine generator set, and to calculate tower status data within the preset time window based on the tower monitoring data of the wind turbine generator set.

[0096] The theoretical bending moment prediction module 620 is used to input the blade state data, the tower state data and the wind turbine's operating state data into a pre-trained tower bending moment prediction model to obtain the theoretical maximum bending moment of multiple tower height positions corresponding to the cross sections.

[0097] The actual bending moment calculation module 630 is used to calculate the actual maximum bending moment of the corresponding section based on the stress data of the corresponding sections at multiple tower height positions;

[0098] The bending moment threshold adjustment module 640 is used to process the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain the tower bending moment deviation and the adaptive bending moment threshold.

[0099] The health status judgment module 650 is used to judge the health status of the wind turbine generator based on the tower bending moment deviation and the adaptive bending moment threshold.

[0100] In one embodiment, the above-mentioned actual bending moment calculation module 630 can be specifically used to: calculate the bending moment of the corresponding section based on the stress data of the sections corresponding to the height positions of multiple towers; perform linear fitting on the bending moment of each section to obtain the bending moment distribution of the corresponding section; and calculate the actual maximum bending moment of the corresponding section at each time point within a preset time window based on the bending moment distribution.

[0101] In one embodiment, the bending moment threshold adjustment module 640 can be specifically used to: calculate the difference between the theoretical maximum bending moment and the actual maximum bending moment of each section at the same time point within a preset time window; take the maximum value of the difference as the tower bending moment deviation at the corresponding time point; and traverse each time point within the preset time window to obtain the tower bending moment deviation at each time point within the preset time window.

[0102] In one embodiment, the bending moment threshold adjustment module 640 can be specifically used to: obtain an initial bending moment threshold and a first bending moment deviation threshold, a first local bending moment deviation mean, an extreme value boundary line, and a first extreme value fluctuation vector corresponding to the initial bending moment threshold; the initial bending moment threshold is determined by the modeling dataset, the first bending moment deviation threshold is the tower bending moment deviation threshold corresponding to the last time point of the modeling dataset, the first local bending moment deviation mean is the tower bending moment deviation mean corresponding to a preset number of time points before the last time point of the modeling dataset, and the first extreme value fluctuation vector is the set of extreme value fluctuations at all time points in the modeling dataset; calculate the corresponding local feature quantification index for the tower bending moment deviation at each time point within a preset time window; iteratively calculate the local feature quantification index corresponding to each time point based on the first bending moment deviation threshold, the first local bending moment deviation mean, and the extreme value boundary line to obtain the adaptive bending moment threshold corresponding to each time point.

[0103] In one embodiment, the local feature quantification index includes the mean local bending moment deviation, local fluctuation amount, and extreme value fluctuation amount;

[0104] The bending moment threshold adjustment module 640 described above can also be specifically used for: if the local fluctuation at the current time point is less than the extreme value boundary line, then the adaptive bending moment threshold remains unchanged; if the local fluctuation at the current time point is greater than the extreme value boundary line, and the local fluctuation at the current time point is greater than the adaptive bending moment threshold at the previous time point, then the adaptive bending moment threshold at the previous time point is used as the adaptive bending moment threshold at the current time point; if the local fluctuation at the current time point is not less than the extreme value boundary line, and the local fluctuation at the current time point is less than the adaptive bending moment threshold at the previous time point, then the extreme value fluctuation at the current time point is included in the first extreme value fluctuation vector, and the adaptive bending moment threshold at the current time point is recalculated iteratively based on the first bending moment deviation threshold and the first local bending moment deviation mean to obtain the adaptive bending moment threshold at the current time point.

[0105] In one embodiment, the blade monitoring data includes blade load data and blade vibration data;

[0106] The aforementioned monitoring data processing module 610 can be specifically used to: calculate the average blade load borne by the blade during one rotation based on the blade load data; and calculate the average blade vibration acceleration during one rotation based on the blade vibration data.

[0107] In one embodiment, the tower monitoring data includes tower vibration data and tower tilt angle data;

[0108] The aforementioned monitoring data processing module 610 can be specifically used to: calculate the average vibration acceleration of the tower based on the tower vibration data; and calculate the displacement of the top of the tower based on the tower tilt angle data.

[0109] In one embodiment, a first preset number of fiber optic strain sensors and a second preset number of temperature sensors are respectively disposed on the tower wall at multiple tower height positions in the wind turbine generator.

[0110] In one embodiment, a third preset number of blade load sensors, a fourth preset number of temperature sensors, and a fifth preset number of acceleration sensors are disposed on the tower wall at a preset position of the blade in the wind turbine generator set.

[0111] The health status determination device for wind turbine generator sets provided in this embodiment can be applied to the health status determination method for wind turbine generator sets provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0112] Figure 7 This is a block diagram of an electronic device used to implement a method for determining the health status of a wind turbine generator according to embodiments of this application. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present application described and / or claimed herein.

[0113] like Figure 7 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0114] Multiple components in electronic device 10 are connected to input / output (I / O) interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of monitors, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0115] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for determining the health status of wind turbine generators.

[0116] In some embodiments, the wind turbine health status determination method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded into and / or installed on electronic device 10 via read-only memory (ROM) 12 and / or communication unit 19. When the computer program is loaded into random access memory (RAM) 13 and executed by processor 11, one or more steps of the wind turbine health status determination method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the wind turbine health status determination method by any other suitable means (e.g., by means of firmware).

[0117] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0118] Computer programs used to implement the methods of this application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0119] In the context of this application, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0120] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube (CRT) or liquid crystal display (LCD) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0121] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0122] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to address the shortcomings of traditional physical hosts and virtual private servers, such as high management difficulty and weak business scalability.

[0123] Note that the above are merely preferred embodiments and technical principles applied in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. For example, those skilled in the art can use the various forms of processes shown above to reorder, add, or delete steps; the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solution of this application can be achieved, and no limitations are imposed herein.

[0124] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for determining the health status of a wind turbine generator set, characterized in that, The method includes: The blade status data for a preset time window is calculated based on the blade monitoring data of the wind turbine generator set, and the tower status data for the preset time window is calculated based on the tower monitoring data of the wind turbine generator set. Based on the blade state data, the tower state data, and the wind turbine's operating state data, the theoretical maximum bending moment of the cross-section corresponding to multiple tower height positions is obtained by inputting them into a pre-trained tower bending moment prediction model. The actual maximum bending moment of the corresponding section is calculated based on the stress data of the sections at multiple tower height locations; Data processing is performed on the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain the tower bending moment deviation and the adaptive bending moment threshold; The health status of the wind turbine generator is determined based on the tower bending moment deviation and the adaptive bending moment threshold.

2. The method for determining the health status of a wind turbine generator set according to claim 1, characterized in that, The calculation of the actual maximum bending moment of the corresponding section based on stress data at multiple tower height positions includes: Based on the stress data of the cross sections corresponding to multiple tower height positions, the bending moment of the corresponding cross sections is calculated; Linear fitting is performed on the bending moment of each section to obtain the bending moment distribution of the corresponding section; The actual maximum bending moment of the corresponding section at each time point within the preset time window is calculated based on the bending moment distribution.

3. The method for determining the health status of a wind turbine generator set according to claim 1, characterized in that, The process of processing the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain the tower bending moment deviation includes: Calculate the difference between the theoretical maximum bending moment and the actual maximum bending moment at the same time point within a preset time window for each section; The maximum value of the difference is taken as the tower bending moment deviation at the corresponding time point; Iterate through each time point within the preset time window to obtain the tower bending moment deviation at each time point within the preset time window.

4. The method for determining the health status of a wind turbine generator set according to claim 3, characterized in that, The process of processing the theoretical and actual maximum bending moments of each cross section to obtain an adaptive bending moment threshold includes: Obtain the initial bending moment threshold and the corresponding first bending moment deviation threshold, first local bending moment deviation mean, extreme value boundary line, and first extreme value fluctuation vector; the initial bending moment threshold is determined by the modeling dataset, the first bending moment deviation threshold is the tower bending moment deviation threshold corresponding to the last time point of the modeling dataset, the first local bending moment deviation mean is the tower bending moment deviation mean corresponding to a preset number of time points before the last time point of the modeling dataset, and the first extreme value fluctuation vector is the set of extreme value fluctuations at all time points in the modeling dataset; Calculate the corresponding local characteristic quantification index for the tower bending moment deviation at each time point within the preset time window; Based on the first bending moment deviation threshold, the first local bending moment deviation mean, and the extreme value boundary line, the local feature quantification index corresponding to each time point is iteratively calculated to obtain the adaptive bending moment threshold corresponding to each time point.

5. The method for determining the health status of a wind turbine generator set according to claim 4, characterized in that, The local feature quantification indicators include the mean local bending moment deviation, local fluctuation, and extreme value fluctuation. The step of iteratively calculating the local feature quantification index corresponding to each time point based on the first bending moment deviation threshold, the first local bending moment deviation mean, and the extreme value boundary line to obtain the adaptive bending moment threshold corresponding to each time point includes: If the local fluctuation at the current time point is less than the extreme value boundary, then the adaptive bending moment threshold remains unchanged; If the local fluctuation at the current time point is greater than the extreme value boundary line, and the local fluctuation at the current time point is greater than the adaptive bending moment threshold at the previous time point, then the adaptive bending moment threshold at the previous time point is used as the adaptive bending moment threshold at the current time point. If the local fluctuation at the current time point is not less than the extreme value boundary line, and the local fluctuation at the current time point is less than the adaptive bending moment threshold at the previous time point, then the extreme value fluctuation at the current time point is included in the first extreme value fluctuation vector, and the adaptive bending moment threshold at the current time point is obtained by re-iteratio calculation based on the first bending moment deviation threshold and the first local bending moment deviation mean.

6. The method for determining the health status of a wind turbine generator set according to claim 1, characterized in that, The blade monitoring data includes blade load data and blade vibration data; The calculation of blade status data within a preset time window based on wind turbine blade monitoring data includes: The average load on the blade during one rotation is calculated based on the blade load data. The average vibration acceleration of the blade is calculated based on the blade vibration data for one rotation.

7. The method for determining the health status of a wind turbine generator set according to claim 1, characterized in that, The tower monitoring data includes tower vibration data and tower tilt angle data; the calculation of tower status data within the preset time window based on the tower monitoring data of the wind turbine generator includes: The average vibration acceleration of the tower is calculated based on the tower vibration data. The displacement of the top of the tower is calculated based on the tower tilt angle data.

8. The method for determining the health status of a wind turbine generator set according to claim 1, characterized in that, A first preset number of fiber optic strain sensors and a second preset number of temperature sensors are respectively installed on the tower wall at multiple tower height positions in the wind turbine generator set.

9. The method for determining the health status of a wind turbine generator set according to claim 1, characterized in that, A third preset number of blade load sensors, a fourth preset number of temperature sensors, and a fifth preset number of acceleration sensors are installed on the tower wall at preset positions of the blades in the wind turbine generator set.

10. A device for determining the health status of a wind turbine generator set, characterized in that, The device includes: The monitoring data processing module is used to calculate blade status data within a preset time window based on the blade monitoring data of the wind turbine generator set, and to calculate tower status data within the preset time window based on the tower monitoring data of the wind turbine generator set. The theoretical bending moment prediction module is used to input the blade state data, the tower state data and the wind turbine's operating state data into a pre-trained tower bending moment prediction model to obtain the theoretical maximum bending moment of multiple tower height positions corresponding to the cross sections. The actual bending moment calculation module is used to calculate the actual maximum bending moment of the corresponding section based on the stress data of the section corresponding to multiple tower height positions; The bending moment threshold adjustment module is used to process the theoretical maximum bending moment and the actual maximum bending moment of each section to obtain the tower bending moment deviation and the adaptive bending moment threshold. The health status assessment module is used to determine the health status of the wind turbine generator based on the tower bending moment deviation and the adaptive bending moment threshold.

Images