Wind turbine variable pitch bearing crack monitoring and risk warning method based on acoustic emission

Abstract

The present application belongs to the technical field of wind power generation equipment monitoring, and provides a wind power variable pitch bearing crack monitoring and risk early warning method based on acoustic emission, which comprises the following steps: firstly, collecting the acoustic emission cumulative energy, acoustic emission instantaneous energy, operating state data and environmental wind speed of the wind power variable pitch bearing; then, constructing a risk preliminary analysis model based on the acoustic emission energy characteristics to generate a preliminary score of the bearing crack state; at the same time, constructing a comprehensive influence analysis model according to the operating state data and the environmental wind speed to generate a comprehensive influence correction coefficient; finally, correcting the preliminary score by using the correction coefficient to obtain a comprehensive score of the bearing crack state, and monitoring and performing risk early warning according to the comprehensive score; the present application solves the false alarm and missed alarm problems caused by working condition fluctuation and external interference in the prior art, and has the advantages of improving the monitoring accuracy and the early warning reliability.

Description

Technical Field

[0001] This invention belongs to the field of wind power equipment monitoring technology, and particularly relates to a method for monitoring and risk warning of cracks in wind turbine pitch bearings based on acoustic emission. Background Technology

[0002] As a core rotating support component of wind turbine generators, wind turbine pitch bearings are subjected to high dynamic loads and complex weather conditions for extended periods, continuously bearing alternating stresses, which easily induce micro-fatigue cracks in the material. If these cracks are not identified and maintained in time, they will accelerate to a critical size, ultimately leading to bearing fracture and causing major safety accidents such as uncontrolled blade fall and complete turbine collapse.

[0003] Currently, the bearing crack monitoring technologies commonly used in the wind power industry encompass vibration spectrum analysis, oil detection, and ultrasonic imaging. Among these, acoustic emission technology, with its highly sensitive ability to capture elastic waves released during microcrack initiation, demonstrates unique advantages in bearing fault diagnosis. Traditional acoustic emission monitoring often employs a static energy threshold determination mechanism, triggering an alarm when the acoustic emission signal exceeds a preset threshold; some methods introduce time-series trend analysis of accumulated energy to infer crack development by assessing changes in the energy accumulation rate.

[0004] However, the aforementioned technologies face significant limitations in practical applications. On the one hand, acoustic emission signals are highly susceptible to fluctuations in bearing operating conditions (such as sudden changes in speed or load) and changes in ambient wind speed, leading to frequent false alarms or missed alarms under non-steady-state conditions using the fixed threshold method. On the other hand, existing methods are generally limited to a single energy characteristic dimension, failing to effectively integrate the complementary information of accumulated energy and instantaneous energy, resulting in insufficient ability to identify weak signals in the early stages of crack initiation. Furthermore, existing technologies lack a quantitative correction framework for the impact of operating conditions, making it impossible to dynamically compensate for external interference, causing deviations between monitoring results and the actual crack state. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for monitoring and warning of cracks in wind turbine pitch bearings based on acoustic emission, thus solving the aforementioned problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for monitoring and risk warning of cracks in wind turbine pitch bearings based on acoustic emission, the method specifically comprising:

[0007] A preliminary risk analysis model was established based on the acoustic emission energy of the wind turbine pitch bearing, and a preliminary score for the crack state of the wind turbine pitch bearing was generated.

[0008] A comprehensive impact analysis model is established based on the operating status data of the wind turbine pitch bearing and the ambient wind speed, and a comprehensive impact correction coefficient is generated.

[0009] Based on the comprehensive impact correction coefficient and the preliminary score of the crack state of the wind turbine pitch bearing, a comprehensive score of the crack state of the wind turbine pitch bearing is generated.

[0010] Based on the comprehensive score of the crack status of wind turbine pitch bearings, we monitor and provide risk warnings for cracks in wind turbine pitch bearings.

[0011] Based on the above technical solutions, the present invention also provides the following optional technical solutions:

[0012] Further technical solution: The method for generating the preliminary score of the crack state of the wind turbine pitch bearing specifically includes:

[0013] Based on the accumulated acoustic emission energy of the wind turbine pitch bearing, a score for the characteristic of accumulated energy change is generated.

[0014] Based on the instantaneous energy of acoustic emission, generate an instantaneous energy state score for acoustic emission;

[0015] A preliminary analysis model was established based on the cumulative energy change characteristic score and the instantaneous energy state score of acoustic emission, and a preliminary score of the crack state of the wind turbine pitch bearing was generated.

[0016] Further technical solution: The specific method for generating the accumulated energy change feature score includes:

[0017] Through the formula: ;

[0018] Generate cumulative energy change characteristic score ;

[0019] In the formula, This represents the cumulative acoustic emission energy at time t. This represents the cumulative acoustic emission energy at time t-1. This represents the cumulative energy change deviation threshold.

[0020] Further technical solution: The method for generating the instantaneous energy state score of acoustic emission specifically includes:

[0021] Through the formula: ;

[0022] Generate an instantaneous energy state score for acoustic emission. ;

[0023] In the formula, This represents the historical average of the instantaneous energy of acoustic emission. This represents the preset instantaneous acoustic emission energy reference value.

[0024] Further technical solution: The expression of the preliminary analysis model is specifically as follows: ;

[0025] In the expression, This indicates the preliminary score for the crack condition of the wind turbine pitch bearing. This represents the cumulative energy change characteristic score. This represents the instantaneous energy state score of acoustic emission. , All are weighting coefficients, and .

[0026] Further technical solution: The method for generating the comprehensive influence correction coefficient specifically includes:

[0027] Based on the operating status data of the wind turbine pitch bearing, an operating status influence coefficient is generated;

[0028] Based on the ambient wind speed, generate the ambient wind speed influence coefficient;

[0029] A comprehensive impact analysis model is established based on the operational status impact coefficient and the environmental wind speed impact coefficient, and a comprehensive impact correction coefficient is generated.

[0030] The specific expression of the comprehensive impact analysis model is as follows: ;

[0031] In the expression, This represents the comprehensive impact correction coefficient. This represents the influence coefficient of the operating status. This represents the environmental wind speed influence coefficient. This represents the weighting coefficient of the operational status influence coefficient.

[0032] Further technical solution: The specific method for generating the operating state influence coefficient includes:

[0033] Through the formula: ;

[0034] Generate operating status influence coefficient ;

[0035] In the formula, This represents the i-th operating status data of the wind turbine pitch bearing. This represents the safety threshold for the i-th operating state data of the wind turbine pitch bearing. This represents the deviation threshold of the i-th operating state data of the wind turbine pitch bearing. This represents the weighting coefficient for the i-th operating state of the wind turbine pitch bearing, and n represents the number of operating state data for the wind turbine pitch bearing.

[0036] Further technical solution: The weighting coefficient of the i-th operating state of the wind turbine pitch bearing:

[0037] Through the formula: ;

[0038] Weighting coefficients for generating the i-th operating state of the wind turbine pitch bearing ;

[0039] In the formula, This represents the Pearson correlation coefficient between the i-th operating state of the wind turbine pitch bearing and the acoustic emission energy. This represents the Pearson correlation coefficient between the j-th operating state of the wind turbine pitch bearing and the acoustic emission energy, where n represents the number of operating state data for the wind turbine pitch bearing. The Pearson correlation coefficient between the operating state of the wind turbine pitch bearing and the acoustic emission energy is calculated based on the historical healthy operating period of the wind turbine pitch bearing.

[0040] Further technical solution: The method for generating the environmental wind speed influence coefficient specifically includes:

[0041] Through the formula: ;

[0042] Environmental wind speed influence coefficient ;

[0043] In the formula, This represents the ambient wind speed at time t. This represents the ambient wind speed at time t-1. This represents the threshold for changes in ambient wind speed.

[0044] Further technical solution: The method for generating the comprehensive score of the wind turbine pitch bearing crack state specifically includes:

[0045] Through the formula: ;

[0046] Generate a comprehensive score for the crack condition of wind turbine pitch bearings. ;

[0047] In the formula, This indicates the preliminary score for the crack condition of the wind turbine pitch bearing. This represents the comprehensive impact correction coefficient. This represents the corrected strength coefficient.

[0048] This invention provides a method for monitoring and warning of cracks in wind turbine pitch bearings based on acoustic emission, which has the following advantages compared with the prior art:

[0049] This invention collects data on acoustic emission accumulated energy, instantaneous energy, operating status, and ambient wind speed, and establishes a preliminary risk analysis model and a comprehensive impact analysis model to dynamically generate crack status scores for early warning. This solves the problem of false alarms and missed alarms caused by operating condition fluctuations and external interference in existing technologies, and has the advantages of improving monitoring accuracy and early warning reliability. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating the method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission provided by the present invention.

[0051] Figure 2 This is a flowchart illustrating step S20 of the present invention.

[0052] Figure 3 This is a flowchart illustrating step S30 of the present invention.

[0053] Figure 4 This is a schematic diagram of the structure of the wind turbine pitch bearing crack monitoring and risk early warning system based on acoustic emission provided by the present invention.

[0054] Figure 5 This is a schematic diagram of the structure of the preliminary analysis unit provided by the present invention.

[0055] Figure 6 A schematic diagram of the external factor analysis unit provided by the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0057] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0058] Please see Figure 1 The present invention provides a method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission, which specifically includes the following steps:

[0059] Step S10: Collect the cumulative acoustic emission energy, instantaneous acoustic emission energy, operating status data of the wind turbine pitch bearing, and ambient wind speed.

[0060] Step S20: Establish a preliminary risk analysis model based on the acoustic emission energy of the wind turbine pitch bearing, and generate a preliminary score for the crack state of the wind turbine pitch bearing.

[0061] Step S30: Establish a comprehensive impact analysis model based on the operating status data of the wind turbine pitch bearing and the ambient wind speed, and generate comprehensive impact correction coefficients;

[0062] Step S40: Generate a comprehensive score for the crack status of the wind turbine pitch bearing based on the comprehensive influence correction coefficient and the preliminary score for the crack status of the wind turbine pitch bearing.

[0063] Step S50: Based on the comprehensive score of the crack status of the wind turbine pitch bearing, monitor and provide risk warning for the cracks in the wind turbine pitch bearing;

[0064] Acoustic emission refers to the transient elastic wave phenomenon generated when energy is rapidly released locally inside a material. It is often used to monitor material damage and crack propagation.

[0065] The wind turbine pitch bearing is a key bearing used to adjust the blade angle in a wind turbine generator set. Its operating conditions are complex, and it is subjected to alternating loads for a long time, which makes it prone to fatigue cracks.

[0066] Acoustic emission cumulative energy refers to the sum of the energy of acoustic emission events over a period of time, reflecting the overall activity level of crack propagation;

[0067] Acoustic emission instantaneous energy refers to the energy of a single acoustic emission event or an acoustic emission event over a short period of time, reflecting the instantaneous intensity of crack propagation;

[0068] Operating status data refers to data that reflects the working conditions of wind turbine pitch bearings, such as speed, load, and temperature.

[0069] Ambient wind speed refers to the real-time wind speed of the environment in which the wind turbine generator is located. Its changes will affect the bearing load and operational stability.

[0070] Specifically, in step S10, the cumulative acoustic emission energy, instantaneous acoustic emission energy, operating status data of the wind turbine pitch bearing, and ambient wind speed are collected.

[0071] The cumulative acoustic emission energy and the instantaneous acoustic emission energy can be collected by installing acoustic emission sensors on the surface of the wind turbine pitch bearing; the sensors convert the elastic waves generated by the activity of cracks inside the bearing into electrical signals, which are then received and processed by the acquisition system.

[0072] Accumulated energy can be obtained by simply summing the energies of all acoustic emission events over a period of time. Instantaneous energy can be obtained by calculating the energy of a single or short-duration acoustic emission event signal.

[0073] Operating status data for wind turbine pitch bearings can be obtained from the wind turbine's Supervisory Control and Data Acquisition (SCADA) system, including parameters such as bearing temperature, speed, and load. Ambient wind speed can be measured in real time using a wind speed sensor installed at the top of the wind turbine tower.

[0074] In step S20, a preliminary risk analysis model is established based on the acoustic emission energy of the wind turbine pitch bearing, generating a preliminary score for the crack state of the wind turbine pitch bearing. After obtaining the accumulated acoustic emission energy and the instantaneous acoustic emission energy, a preliminary risk analysis model can be constructed. This model can be a simple set of rules; for example, when the accumulated energy or instantaneous energy exceeds a preset fixed threshold, a crack risk is considered to exist. Alternatively, the model can be a statistical model based on historical data, assessing the risk by comparing the current energy value with historical normal values. Through this model, a quantitative preliminary crack state score can be generated, reflecting the preliminary risk assessment results based on the acoustic emission energy characteristics.

[0075] In step S30, a comprehensive influence analysis model is established based on the operating status data of the wind turbine pitch bearing and the ambient wind speed, generating a comprehensive influence correction coefficient. To consider the impact of non-crack factors on the acoustic emission signal, a comprehensive influence analysis model can be established using the collected operating status data and ambient wind speed. This model can be a simple lookup table, with a preset correction value based on different operating states (e.g., high speed, low load) and wind speed ranges (e.g., strong wind, light wind). Alternatively, the model can be based on an empirical formula, using the operating status data and ambient wind speed as input, and calculating a correction coefficient through a simple weighted average or linear combination; this correction coefficient aims to quantify the degree of interference of these external factors on the acoustic emission signal.

[0076] In step S40, a comprehensive score for the crack state of the wind turbine pitch bearing is generated based on the comprehensive influence correction coefficient and the preliminary score of the crack state. After obtaining the preliminary crack state score and the comprehensive influence correction coefficient, the two can be combined to generate a more accurate comprehensive crack state score. This combination can be a simple multiplication or addition operation. For example, the preliminary score can be multiplied by a factor based on the correction coefficient, or the correction coefficient can be directly added to or subtracted from the preliminary score. This comprehensive score aims to eliminate or reduce the interference of operating conditions and ambient wind speed on acoustic emission signals, thereby more realistically reflecting the crack state of the bearing.

[0077] In step S50, the wind turbine pitch bearing crack status comprehensive score is used to monitor and provide risk warnings for the wind turbine pitch bearing cracks. Once the comprehensive score for the wind turbine pitch bearing crack status is generated, it can be used for crack monitoring and risk warnings. This may include comparing the comprehensive score with multiple preset risk level thresholds. For example, when the comprehensive score exceeds a first threshold, the system issues a low-level warning; when it exceeds a second threshold, a high-level warning is issued. Warning information can be output in various forms, such as through audible and visual alarms, SMS notifications, or display on the monitoring interface. The monitoring process can be continuous to track the changing trend of the crack status in real time.

[0078] Through the above technical solution, this application enables the monitoring results to more accurately reflect the crack state, overcomes the shortcomings of the existing technology in lacking a quantitative correction mechanism for the influence of operating conditions and environmental factors, and improves the accuracy and reliability of monitoring.

[0079] For preferred options, please refer to [link / reference]. Figure 2 The present invention further proposes a method for generating a preliminary score for the crack state of the wind turbine pitch bearing, specifically including:

[0080] Step S21: Generate a cumulative energy change characteristic score based on the cumulative acoustic emission energy of the wind turbine pitch bearing;

[0081] Step S22: Generate an acoustic emission instantaneous energy state score based on the acoustic emission instantaneous energy;

[0082] Step S23: Establish a preliminary analysis model based on the cumulative energy change characteristic score and the acoustic emission instantaneous energy state score, and generate a preliminary score for the crack state of the wind turbine pitch bearing;

[0083] In step S21, generating a cumulative energy change characteristic score aims to quantify the trend of acoustic emission cumulative energy over time. Cumulative energy reflects the cumulative effect of the total amount and intensity of acoustic emission events over a period of time; its abnormal growth is usually closely related to the accumulation of internal material damage, such as crack initiation and propagation. By scoring the changes in cumulative energy, the long-term trend of crack development can be effectively captured. This score can be generated in various ways. For example, it can be generated by calculating the growth rate or slope of acoustic emission cumulative energy over a continuous time period; when the growth rate exceeds a preset threshold, the score increases with the increase in the growth rate. Alternatively, a sliding window averaging method can be used to smooth the cumulative energy, calculate the deviation between the current value and the historical average, and generate a score based on the magnitude of the deviation.

[0084] In step S22, generating an acoustic emission instantaneous energy state score aims to assess the intensity and activity of individual acoustic emission events. Instantaneous energy reflects the severity of an acoustic emission event at a specific moment; high instantaneous energy events often indicate more severe local damage or rapid crack propagation. By scoring the instantaneous energy, sudden, high-energy damage events can be detected in a timely manner. This score can be generated, for example, by statistically analyzing the average, peak, or variance of the instantaneous energy and comparing it with a baseline value under healthy conditions, generating a score based on the degree of deviation; alternatively, a threshold discrimination method can be used, directly assigning a higher score when the instantaneous energy exceeds a preset danger threshold, or a score can be generated based on the distribution characteristics of the instantaneous energy (such as the frequency of high-energy events).

[0085] In step S23, a preliminary analysis model is established and a preliminary score for the crack state of the wind turbine pitch bearing is generated. This aims to comprehensively utilize information from two different dimensions—the accumulated energy change characteristic score and the acoustic emission instantaneous energy state score—to obtain a more comprehensive and accurate crack state assessment. The model can employ a weighted summation model, linearly combining the two scores according to preset weights to obtain the preliminary score. The weights can be set based on expert experience or historical data analysis. Alternatively, a machine learning model, such as a support vector machine, neural network, or decision tree, can be used to learn the nonlinear relationship between the two scores and the actual crack state by training on historical data, thereby generating a more accurate preliminary score.

[0086] This application's solution refines the analysis of acoustic emission energy into two independent dimensions: cumulative energy change characteristic score and instantaneous acoustic emission energy state score. Based on these dimensions, a preliminary analysis model is established to generate a preliminary score for the crack state of the wind turbine pitch bearing. The cumulative energy change characteristic score captures the trend of long-term cumulative damage from the crack, while the instantaneous acoustic emission energy state score reflects the instantaneous intensity of crack activity. This dual-dimensional, multi-angle analysis method allows the preliminary score to more comprehensively and sensitively reflect early crack initiation and propagation signs. In this way, this application concretizes and optimizes step S20, "establishing a preliminary risk analysis model based on the acoustic emission energy of the wind turbine pitch bearing and generating a preliminary score for the crack state of the wind turbine pitch bearing," in the aforementioned method. This makes the generation of the preliminary score no longer a single or coarse judgment, but based on a more refined acoustic emission characteristic analysis, thus providing a more solid and accurate foundation for subsequent comprehensive impact correction and final monitoring and risk warning.

[0087] Through the above technical solution, this application can analyze acoustic emission energy in two dimensions: cumulative energy and instantaneous energy, and generate corresponding feature scores for each. This allows for a more comprehensive and precise capture of the early initiation and propagation characteristics of wind turbine pitch bearing cracks. The cumulative energy change feature score reflects the long-term cumulative damage trend of the crack, while the instantaneous energy state score captures the instantaneous intensity of crack activity. This meticulous analysis method makes the generated preliminary crack state score for wind turbine pitch bearings more sensitive and accurate. Compared to relying solely on a single or coarse acoustic emission energy index, the solution in this application can significantly improve the accuracy and timeliness of early warning of wind turbine pitch bearing cracks, thus providing a more reliable basis for subsequent comprehensive impact correction and final risk warning, effectively avoiding equipment failures and economic losses caused by insufficient early crack identification.

[0088] Preferably, the present invention further proposes a method for generating the accumulated energy change feature score, specifically including:

[0089] Through the formula: ;

[0090] Generate cumulative energy change characteristic score ;

[0091] In the formula, This represents the cumulative acoustic emission energy at time t. This represents the cumulative acoustic emission energy at time t-1. This represents the threshold value for the cumulative energy change deviation;

[0092] The cumulative energy change characteristic score is a quantitative indicator used to characterize the degree to which the cumulative acoustic emission energy of a wind turbine pitch bearing changes over time. Its function is to transform the dynamic change of the cumulative acoustic emission energy into a numerical value that can be used for risk assessment, thereby reflecting the trend of crack initiation or propagation. This score is usually a dimensionless value, and its magnitude is directly related to the potential risk of crack development.

[0093] The formula for the cumulative energy change characteristic score provides a standardized and robust method for calculating the cumulative energy change characteristic score. The max function ensures the non-negativity of the score; that is, a positive score is only generated when the cumulative energy increases beyond a certain level, effectively avoiding false alarms caused by normal fluctuations or measurement errors.

[0094] In the above formula The term directly reflects the actual increase in the accumulated energy of acoustic emission between the current time t and the previous time t-1, and is a direct manifestation of the activity level of acoustic emission.

[0095] The cumulative acoustic emission energy at time t refers to the total acoustic emission energy generated by the wind turbine pitch bearing from the start of monitoring to the current time t. This data can be obtained by continuously collecting instantaneous acoustic emission signals through acoustic emission sensors and then integrating or accumulating them.

[0096] The cumulative acoustic emission energy at time t-1 refers to the total acoustic emission energy generated by the wind turbine pitch bearing from the start of monitoring up to the previous time t-1. This data is a historical cumulative energy value, which, together with the cumulative energy at time t, is used to calculate the energy change.

[0097] The cumulative energy change deviation threshold is a preset key parameter used to measure whether the cumulative energy change is abnormal. Its function is to set a benchmark; only when the increment of acoustic emission cumulative energy exceeds this threshold is it considered a significant change that may be related to crack propagation. This threshold can be determined through statistical analysis of historical acoustic emission data of wind turbine pitch bearings under healthy operating conditions (e.g., set as a multiple of the standard deviation of cumulative energy change under healthy conditions), or it can be set based on industry experience or expert knowledge to ensure sensitivity to early cracks.

[0098] This application's solution addresses the problem of accurately quantifying the trend of acoustic emission cumulative energy variation by introducing the aforementioned calculation formula for cumulative energy variation characteristic scoring. Specifically, the method first obtains the cumulative acoustic emission energy at the current time t and the previous time t-1, and calculates the difference between the two to obtain the incremental cumulative energy within a specific time interval. Subsequently, this increment is compared and normalized with a preset cumulative energy variation deviation threshold. When the increment of cumulative energy significantly exceeds this deviation threshold, it indicates that there may be abnormal acoustic emission activity inside the wind turbine pitch bearing, suggesting the initiation or propagation of cracks. The max function ensures that a positive characteristic score is generated only when the increase in cumulative energy significantly exceeds the normal fluctuation range, effectively filtering out background noise and minor fluctuations caused by normal operation. This quantification method allows the trend of cumulative energy variation to be accurately captured and transformed into a value that can be used for risk assessment, providing a more accurate and reliable input for the subsequent preliminary scoring of the crack status of the wind turbine pitch bearing.

[0099] Through the above technical solution, this application provides a method for accurately quantifying the cumulative energy variation trend of acoustic emission in wind turbine pitch bearings. This method, by introducing a cumulative energy variation deviation threshold, can effectively distinguish between normal energy fluctuations and significant energy increases caused by crack propagation, avoiding misjudgments due to minor fluctuations. Simultaneously, the application of the max function ensures the non-negativity of the score, so that a positive risk indication is only generated when the cumulative energy change reaches or exceeds a preset danger threshold, thereby improving the accuracy and reliability of the preliminary assessment of crack conditions in wind turbine pitch bearings. This makes early warning and monitoring of cracks in wind turbine pitch bearings more sensitive and effective, helping to promptly detect potential faults and reduce maintenance costs and accident risks.

[0100] Preferably, the present invention further proposes a method for generating the instantaneous energy state score of acoustic emission, specifically including:

[0101] Through the formula: ;

[0102] Generate an instantaneous energy state score for acoustic emission. ;

[0103] In the formula, This represents the historical average of the instantaneous energy of acoustic emission. This represents the preset instantaneous acoustic emission energy reference value;

[0104] The acoustic emission instantaneous energy state score quantifies the deviation of the instantaneous acoustic emission energy of a wind turbine pitch bearing at a given moment from its normal operating baseline. It is a dimensionless indicator reflecting the severity of any anomalies in instantaneous acoustic emission activity. This score can be calculated by real-time monitoring of the acoustic emission signal, extracting its instantaneous energy characteristics, and combining them with historical data; or by using digital signal processing technology to calculate the instantaneous energy by integrating or calculating the root mean square value of the acquired acoustic emission waveform, thereby obtaining the score.

[0105] The above formula provides a specific, quantitative method for calculating the instantaneous energy state score of acoustic emission. It ensures the validity and comparability of the score by comparing the historical average of the instantaneous acoustic emission energy with a preset benchmark value and performing normalization. This formula can be implemented in a data processing unit through programming, such as using an algorithm written in a high-level programming language; it can also be implemented through hardware logic circuits or a dedicated signal processor to meet real-time requirements.

[0106] The historical mean of acoustic emission instantaneous energy refers to the average instantaneous acoustic emission energy generated by wind turbine pitch bearings over a period of time, reflecting the overall level of acoustic emission activity in the current or recent period. This historical mean can be calculated using a moving average window on continuously collected acoustic emission instantaneous energy data; for example, the average of the instantaneous energy from the most recent N sampling points can be taken.

[0107] The preset acoustic emission instantaneous energy reference value refers to the reference value of the instantaneous acoustic emission energy of the wind turbine pitch bearing when it is in a healthy or normal operating state. It is an important basis for judging whether the current instantaneous energy is abnormal. This reference value can be determined by collecting a large amount of acoustic emission instantaneous energy data in the early stage of healthy operation of the wind turbine pitch bearing and performing statistical analysis; or it can be set and adjusted according to the design specifications of the wind turbine pitch bearing, manufacturer's recommendations, or industry experience, combined with the actual operating environment.

[0108] This application's solution introduces an acoustic emission instantaneous energy state score to accurately quantify the degree of anomaly in the instantaneous acoustic emission activity of wind turbine pitch bearings. Specifically, the solution first obtains the historical average of the instantaneous acoustic emission energy, which represents the activity level of instantaneous acoustic emission in the current or recent period. Then, this historical average is compared with a preset benchmark value for the instantaneous acoustic emission energy. The benchmark value is a reference for the instantaneous energy of the wind turbine pitch bearing under healthy conditions; therefore, when the historical average of the instantaneous acoustic emission energy is significantly higher than the preset benchmark value, it indicates that the instantaneous acoustic emission activity may be abnormal. The formula... The algorithm partially calculates the relative growth rate of historical averages relative to a benchmark value, while the max function ensures that the instantaneous acoustic emission energy state score is always non-negative. That is, a positive score is only generated when the historical average of the instantaneous energy exceeds the benchmark value, avoiding the negative impact of normal fluctuations or values ​​below the benchmark on the score. In this way, the scheme effectively filters out background noise and random fluctuations during normal operation, highlighting the instantaneous energy increase caused by potential crack propagation, thus providing a stable and sensitive input for generating the preliminary score of the crack state of wind turbine pitch bearings. This scoring mechanism based on the comparison of historical averages and benchmark values ​​makes the preliminary judgment of crack state more accurate and reliable, avoiding false alarms or missed alarms that may result from judging based on a single instantaneous value, thereby improving the effectiveness of the entire monitoring and risk warning method.

[0109] The above technical solution enables a more accurate assessment of the anomaly level of transient acoustic emission energy when generating a preliminary score for the crack status of wind turbine pitch bearings. This solution effectively avoids misjudgments caused by random factors affecting a single transient energy value by introducing the historical average of the transient acoustic emission energy and comparing it with a preset benchmark value. The use of the max function ensures that a positive score is generated only when the transient energy activity is indeed higher than the normal benchmark, thereby improving the robustness of the score and its sensitivity to crack initiation and propagation. This quantification method allows the transient acoustic emission energy status score to more stably and reliably reflect the actual operating condition of the wind turbine pitch bearing, providing high-quality input for subsequent preliminary analysis models, thus improving the accuracy and reliability of the entire crack monitoring and risk warning method and reducing the false alarm and false negative rates.

[0110] Preferably, the present invention further proposes the following expression for the preliminary analysis model: ;

[0111] In the expression, This indicates the preliminary score for the crack condition of the wind turbine pitch bearing. This represents the cumulative energy change characteristic score. This represents the instantaneous energy state score of acoustic emission. , All are weighting coefficients, and ;

[0112] The expression of this preliminary analytical model defines how to quantitatively combine two independent acoustic emission energy-related scores to form a unified preliminary assessment result. Its purpose is to provide a structured mathematical framework that ensures the contribution of different feature scores to the final preliminary score is controllable and interpretable.

[0113] The preliminary score for the crack condition of wind turbine pitch bearings is a quantitative indicator based on acoustic emission energy data to preliminarily assess the crack condition of wind turbine pitch bearings. It reflects the crack development trend or degree of anomaly revealed by the acoustic emission signal itself, without considering the influence of external operational and environmental factors. This score can serve as the basis for subsequent more comprehensive risk assessments, and its value is directly related to the potential severity of the cracks.

[0114] Among them, the weighting coefficient , These are used to adjust the relative importance of the accumulated energy change characteristic score and the acoustic emission instantaneous energy state score in the preliminary analysis model. They ensure that, when generating the preliminary score, the contribution of different acoustic emission characteristics to crack state assessment can be flexibly allocated based on the actual application scenario, bearing type, or historical data analysis results. The weighting coefficients can be set through expert experience or determined by training and optimizing the model using historical data.

[0115] This application's solution uses a linear weighted combination method to fuse the accumulated energy change characteristic score and the acoustic emission instantaneous energy state score to generate a preliminary score for the crack state of wind turbine pitch bearings. Specifically, this is achieved by assigning weight coefficients to both the accumulated energy change characteristic score and the acoustic emission instantaneous energy state score. and and ensure This model can flexibly adjust the contribution ratio of different acoustic emission characteristics in the initial scoring based on their sensitivity or importance to crack state indication. For example, when the cumulative energy change is considered to better reflect the long-term development trend of the crack, it can be assigned... Higher values; and when abrupt changes in instantaneous energy are considered a better indicator of sudden crack propagation, then higher values ​​can be assigned. Higher values. This weighted summation mechanism allows the preliminary score to not only integrate two key acoustic emission energy characteristics, but also to differentiate these characteristics based on actual needs or empirical knowledge, thereby forming a more comprehensive and targeted preliminary assessment of the crack state. In this way, the model effectively solves the technical problem of how to quantify and integrate different acoustic emission characteristics to form a reliable preliminary assessment, laying a solid foundation for subsequent comprehensive risk warning.

[0116] Through the above technical solution, this application provides a structured and quantifiable method for effectively integrating the cumulative energy change characteristic score and the instantaneous acoustic emission energy state score of wind turbine pitch bearings into a single preliminary crack state score. This weighted linear combination preliminary analysis model overcomes the problems of information distortion and inaccurate assessment that may result from simple superposition or averaging. By introducing weighting coefficients, this scheme can flexibly adjust the actual contribution of different acoustic emission characteristics to crack indication, enabling the preliminary score to more accurately reflect the true development state of the crack. This not only improves the objectivity and reliability of the preliminary assessment but also provides a more solid and refined foundation for subsequent comprehensive impact correction and final risk warning, thereby significantly improving the overall accuracy and effectiveness of wind turbine pitch bearing crack monitoring and risk warning.

[0117] For preferred options, please refer to [link / reference]. Figure 3 The present invention further proposes a method for generating the comprehensive influence correction coefficient, specifically including:

[0118] Step S31: Generate the operating status influence coefficient based on the operating status data of the wind turbine pitch bearing;

[0119] Step S32: Generate the environmental wind speed influence coefficient based on the environmental wind speed;

[0120] Step S33: Establish a comprehensive impact analysis model based on the operating status impact coefficient and the environmental wind speed impact coefficient, and generate comprehensive impact correction coefficients;

[0121] The specific expression of the comprehensive impact analysis model is as follows: ;

[0122] In the expression, This represents the comprehensive impact correction coefficient. This represents the influence coefficient of the operating status. This represents the environmental wind speed influence coefficient. This represents the weighting coefficient of the operational status influence coefficient;

[0123] In step S31, this application first generates an operating state influence coefficient based on the operating state data of the wind turbine pitch bearing. This step aims to quantify the degree of influence of the wind turbine pitch bearing on crack assessment under different operating states. Specifically, this can be achieved in various ways. For example, specific operating state data (such as bearing temperature, spindle speed, pitch angle, torque, etc.) can be mapped to corresponding influence coefficients using a preset rule base or lookup table; alternatively, a machine learning model, such as a support vector machine or neural network, can be trained to learn the relationship between operating state data and crack development using historical operating data, thereby predicting and generating the operating state influence coefficient.

[0124] Simultaneously, in step S32, an environmental wind speed influence coefficient is generated based on the ambient wind speed. This step is used to quantify the impact of environmental wind speed changes on the assessment of cracks in wind turbine pitch bearings. This can be achieved by: setting wind speed ranges and corresponding influence coefficients; for example, when the rate of wind speed change exceeds a certain range, the environmental wind speed influence coefficient will increase accordingly; or, by establishing a physical model between wind speed changes and bearing stress, the environmental wind speed influence coefficient can be generated based on the stress changes calculated from the model.

[0125] Subsequently, in step S33, a comprehensive impact analysis model is established based on the operational status influence coefficient and the environmental wind speed influence coefficient to generate a comprehensive impact correction coefficient. This step aims to comprehensively consider the influence of operational status and environmental wind speed to generate a comprehensive correction coefficient. Specifically, this can be achieved by using weighted averaging, multiplicative combination, or other methods to fuse the operational status influence coefficient and the environmental wind speed influence coefficient; alternatively, a fuzzy logic system or expert system can be used to comprehensively judge and generate the correction coefficient based on a preset rule base and fuzzy inference mechanism.

[0126] Among them, the operating condition influence coefficient is used to capture the accelerated crack propagation or abnormal acoustic emission signal caused by changes in operating conditions;

[0127] The environmental wind speed influence factor is used to account for the additional loads, vibrations, or fatigue that may be caused by changes in wind speed, thereby affecting the initiation and propagation of cracks.

[0128] The weighting coefficient of the operational status influence coefficient typically ranges from 0 to 1 and is used to adjust the relative importance of operational status influence and environmental wind speed influence in the comprehensive correction coefficient. By adjusting the value of the weighting coefficient of the operational status influence coefficient, the contributions of the two influencing factors can be flexibly balanced according to the actual application scenario and experience.

[0129] This application's solution achieves precise correction of the initial score for crack status in wind turbine pitch bearings by meticulously decomposing and quantifying complex external influencing factors and integrating them using a weighted combination method. Specifically, in step S31, the method first conducts in-depth analysis of the operating status data of the wind turbine pitch bearings to extract operating status influence coefficients that reflect the degree of influence of the bearing's actual operating conditions on cracks. This allows the system to identify and quantify the accelerated crack development or changes in acoustic emission signals that may be caused by abnormal operating parameters such as load, speed, and temperature. Next, in step S32, the system processes environmental wind speed data, paying particular attention to wind speed fluctuations and changes, to generate an environmental wind speed influence coefficient. This coefficient is specifically used to capture the potential impact of additional mechanical stress or vibration caused by changes in environmental wind speed on bearing cracks. Subsequently, in step S33, these two independent but interrelated influence coefficients and the environmental wind speed influence coefficient are input into the comprehensive influence analysis model, organically combining the two to generate the final comprehensive influence correction coefficient. Weighting coefficients The introduction of this method allows the system to flexibly adjust the relative importance of the influence of operating conditions and environmental wind speed in the final correction coefficient based on actual conditions or experiential knowledge. Through this step-by-step quantification and weighted fusion mechanism, the proposed solution can comprehensively and meticulously assess the combined influence of external environment and operating conditions on the crack state of wind turbine pitch bearings, thereby providing a more accurate and reliable correction basis for subsequent comprehensive crack state scoring.

[0130] Through the above technical solution, this application overcomes the coarseness of traditional methods in assessing the impact of operating conditions and environmental wind speed, achieving a refined quantification and comprehensive consideration of the influence of these external factors. By generating operating condition influence coefficients and environmental wind speed influence coefficients separately, and then fusing them using a weighted model, the generated comprehensive influence correction coefficient can more accurately and comprehensively reflect the external influences experienced by the wind turbine pitch bearing under actual operating conditions. This significantly improves the correction accuracy of the preliminary score for the crack state of the wind turbine pitch bearing, making the final comprehensive score for the crack state of the wind turbine pitch bearing closer to the actual health condition of the bearing. This precise assessment capability helps to achieve earlier and more reliable crack risk warnings, effectively avoiding unexpected equipment downtime or major failures caused by inaccurate assessments, thereby ensuring the safe and stable operation of wind turbine units and reducing maintenance costs.

[0131] Preferably, the present invention further proposes a method for generating the operating state influence coefficient, specifically including:

[0132] Through the formula: ;

[0133] Generate operating status influence coefficient ;

[0134] In the formula, This represents the i-th operating status data of the wind turbine pitch bearing. This represents the safety threshold for the i-th operating state data of the wind turbine pitch bearing. This represents the deviation threshold of the i-th operating state data of the wind turbine pitch bearing. This represents the weighting coefficient of the i-th operating state of the wind turbine pitch bearing, and n represents the number of operating state data of the wind turbine pitch bearing.

[0135] The operating condition influence coefficient is used to quantify the comprehensive impact of various operating condition data of wind turbine pitch bearings on the risk of bearing cracking. Its value reflects the severity of the deviation of the current operating condition from the safe range; the larger the value, the greater the negative impact of the operating condition on the crack risk. This coefficient is an important component of the comprehensive influence correction coefficient, providing a crucial basis for subsequent crack risk assessment.

[0136] The i-th operational status data of a wind turbine pitch bearing refers to various real-time or historical data collected during the operation of the wind turbine pitch bearing, such as bearing lubricating oil pressure, lubricating oil level, and motor no-load current. This data directly or indirectly reflects the bearing's health status and operating load. This data can be monitored and collected in real time by various sensors installed on the wind turbine pitch bearing, or obtained from the wind farm's SCADA (Supervisory and Data Acquisition) system.

[0137] The safety threshold for the i-th operating status data of a wind turbine pitch bearing defines the minimum permissible value for that i-th operating status data under normal and safe operating conditions. When the actual operating data falls below this threshold, it indicates that the bearing may have an anomaly or increased risk. For example, for lubricating oil pressure data, the safety threshold could be the minimum permissible pressure recommended by the bearing manufacturer; for lubricating oil level, the safety threshold could be a minimum level set based on experience or standards. This threshold can be set based on the equipment manufacturer's specifications, industry standards, historical operating data analysis, or expert experience.

[0138] The deviation threshold for the i-th operating state data of the wind turbine pitch bearing is used to measure the degree to which the i-th operating state data deviates from the safety threshold. It defines an acceptable deviation range, and deviations exceeding this range are considered significant risk factors. This threshold can be adjusted according to the equipment's sensitivity, the severity of the failure mode, and the risk tolerance.

[0139] The weighting coefficient for the i-th operating state of a wind turbine pitch bearing represents the relative importance or influence of the i-th operating state data on the crack risk of the wind turbine pitch bearing. Different operating state data may have different effects on bearing health; for example, abnormal lubricating oil pressure may be a better predictor of crack risk than a minor abnormality in the motor's no-load current. By assigning different weighting coefficients, the role of each operating state data in the comprehensive assessment can be more accurately reflected. This weighting coefficient can be determined based on expert experience, historical failure data analysis, or through statistical methods (such as correlation analysis).

[0140] This application's solution introduces a refined method for calculating the impact coefficient of operating conditions, enabling accurate quantification of the comprehensive impact of various operating condition data of wind turbine pitch bearings on crack risk. Specifically, for each operating condition data point, the degree of standardization of the data's deviation from the safe state is calculated by first comparing its difference with a safety threshold and then combining this with a deviation threshold. Risk contribution is calculated only when operating data falls below its safety threshold, and this contribution value, after being normalized to a deviation threshold and subtracted by 1, is ensured to be non-negative by an outer max function. This means that a positive risk contribution only occurs when the degree to which operating data falls below the safety threshold exceeds the deviation threshold. Each such standardized risk contribution is then multiplied by its corresponding weighting coefficient, reflecting the relative importance of that operating status data to the overall risk. Finally, the weighted risk contributions of all operating status data are summed to obtain the operating status influence coefficient. In this way, this scheme can comprehensively consider the abnormal situations of multiple operating status parameters and their respective influence weights, thereby generating a comprehensive and quantitative operating status influence coefficient. This coefficient is then used to correct the preliminary score of the wind turbine pitch bearing crack status, so that the final crack risk assessment is not only based on acoustic emission energy, but also fully reflects the actual operating conditions of the bearing, improving the accuracy and reliability of risk warning. This method avoids simply treating all operating parameters equally, but distinguishes them according to their actual impact on crack risk, making the generation of the comprehensive influence correction coefficient more scientific and reasonable.

[0141] The above technical solution enables a refined assessment of various operational status data of wind turbine pitch bearings, accurately quantifying their comprehensive impact on bearing crack risk. By introducing safety thresholds, deviation thresholds, and weighting coefficients, this method makes the calculation of operational status impact coefficients more scientific and reasonable, distinguishing the degree of anomaly of different operational parameters and their contribution to risk. This significantly improves the accuracy of the comprehensive impact correction coefficient, allowing the comprehensive score of wind turbine pitch bearing crack status to more realistically reflect the actual health condition of the bearing. This effectively avoids false alarms or missed alarms caused by inaccurate assessment of operational status data, thereby enhancing the reliability and effectiveness of wind turbine pitch bearing crack monitoring and risk early warning.

[0142] Preferably, the present invention further proposes a weighting coefficient for the i-th operating state of the wind turbine pitch bearing:

[0143] Through the formula: ;

[0144] Weighting coefficients for generating the i-th operating state of the wind turbine pitch bearing ;

[0145] In the formula, This represents the Pearson correlation coefficient between the i-th operating state of the wind turbine pitch bearing and the acoustic emission energy. represents the Pearson correlation coefficient between the j-th operating state of the wind turbine pitch bearing and the acoustic emission energy, and n represents the number of operating state data of the wind turbine pitch bearing.

[0146] This generation method aims to quantify the impact of different operating state data on crack monitoring in wind turbine pitch bearings. Its purpose is to provide a reasonable weight allocation for subsequent calculations of operating state influence coefficients, ensuring a more accurate and objective assessment of crack risk. Besides determining weights through correlation analysis, these weight coefficients can also be generated through expert experience methods, i.e., subjectively assessing and assigning values ​​to the importance of different operating states based on the knowledge and experience of domain experts; or through machine learning methods, such as training regression or classification models using historical data and extracting feature importance scores from the models as weight coefficients.

[0147] The Pearson correlation coefficient is a statistic that measures the degree of linear correlation between two variables. In this scheme, it is used to quantify the strength and direction of the linear relationship between the i-th operating state data of a wind turbine pitch bearing and acoustic emission energy. Its role is to identify which operating state data are more consistent with the changing trend of acoustic emission energy (as a crack indicator), thus providing an objective basis for weight allocation. Besides the Pearson correlation coefficient, the Spearman rank correlation coefficient can also be used, which measures the monotonic relationship between variables and has some ability to capture nonlinear relationships; or mutual information can be used, a more generalized correlation measure that can capture nonlinear dependencies between variables.

[0148] The Pearson correlation coefficient between the operating status of the wind turbine pitch bearing and the acoustic emission energy was calculated based on the historical healthy operating period of the wind turbine pitch bearing.

[0149] It should be noted that the phrase "calculated based on the historical healthy operating period of the wind turbine pitch bearing" in this application refers to the fact that the correlation coefficient is calculated based on data collected during periods when the wind turbine pitch bearing is in normal condition, without cracks, or with cracks not yet developed to a dangerous level. Its purpose is to establish a benchmark, ensuring that the calculated correlation coefficient reflects the inherent relationship between different operating states and acoustic emission energy under normal operating conditions, avoiding data interference when cracks have already developed. Data from the historical healthy operating period can be determined through regular maintenance inspections, fault-free operation records, or expert diagnostic results. Furthermore, the calculation method for the Pearson correlation coefficient is a publicly available technique and will not be elaborated upon here.

[0150] This method calculates the Pearson correlation coefficient between the i-th operating state of the wind turbine pitch bearing and the acoustic emission energy, and generates weighting coefficients based on these correlation coefficients. Specifically, during the historical healthy operation period of the wind turbine pitch bearing, the system can continuously collect various operating status data and corresponding acoustic emission energy data. Then, for each operating status parameter, the Pearson correlation coefficient between it and the acoustic emission energy is calculated. These correlation coefficients reflect the strength of the linear correlation between different operating status parameters and changes in acoustic emission energy under healthy conditions. The stronger the correlation, the greater the indicative effect of the operating status parameter on changes in acoustic emission energy. Next, these correlation coefficients are normalized using the above formula, considering only positive correlations (i.e., when an abnormal increase in an operating status parameter, acoustic emission energy tends to increase, which is usually related to crack development), thus obtaining the weight coefficient for each operating status. This data-driven and correlation analysis-based weight allocation method objectively highlights those operating status parameters with a higher correlation to crack development when calculating the operating status influence coefficient, avoiding biases caused by subjective assumptions or empirical assignments. The weighting coefficients determined in this way enable the operating condition influence coefficient to more accurately reflect the impact of the actual operating condition of the wind turbine pitch bearing on the crack risk, thereby improving the accuracy of the comprehensive influence correction coefficient. Ultimately, this makes the assessment results of the comprehensive score of the crack condition of the wind turbine pitch bearing more reliable, providing more solid data support for crack monitoring and risk warning of the wind turbine pitch bearing.

[0151] Through the above technical solution, the weighting coefficient of the i-th operating state of the wind turbine pitch bearing no longer relies on subjective experience or fixed values, but is objectively determined based on the historical health operation data of the wind turbine pitch bearing by calculating the Pearson correlation coefficient between each operating state and acoustic emission energy. This makes the weighting allocation more scientific and reasonable, and can accurately reflect the actual impact of different operating state parameters on the crack development of the wind turbine pitch bearing. Therefore, when calculating the operating state influence coefficient, it can more effectively highlight those key operating parameters closely related to crack development, thereby significantly improving the accuracy of the operating state influence coefficient. This improvement further enhances the reliability of the comprehensive influence correction coefficient, ultimately effectively enhancing the overall accuracy and sensitivity of wind turbine pitch bearing crack monitoring and risk early warning, and providing a more accurate decision-making basis for the health management of wind turbine pitch bearings.

[0152] Preferably, the present invention further proposes a method for generating the environmental wind speed influence coefficient, specifically including:

[0153] Through the formula: ;

[0154] Environmental wind speed influence coefficient ;

[0155] In the formula, This represents the ambient wind speed at time t. This represents the ambient wind speed at time t-1. This represents the threshold for changes in ambient wind speed;

[0156] The environmental wind speed influence coefficient is an indicator that quantifies the impact of environmental wind speed changes on the monitoring results of wind turbine pitch bearing cracks. Its function is to reflect the potential impact of wind speed fluctuations on bearing loads and vibration modes, thereby correcting the initial assessment results based on acoustic emission energy. This coefficient can be a dimensionless value, used to indicate the severity of wind speed changes.

[0157] The ambient wind speed at time t and the ambient wind speed at time t-1 represent the ambient wind speed values ​​measured at the current time t and the previous sampling time t-1. This data is typically provided by the wind farm's meteorological monitoring system, for example, through wind speed sensors installed on the top of the wind turbine nacelle. Wind speed data can be collected in real time by devices such as ultrasonic anemometers or cup anemometers, and recorded and transmitted at fixed time intervals; alternatively, historical wind speed data can be obtained through the wind farm's SCADA system, and interpolated or smoothed to obtain the wind speed value at a specific moment.

[0158] The environmental wind speed variation threshold is a preset benchmark value used to measure whether changes in environmental wind speed are significant. When the wind speed variation exceeds this threshold, it indicates that wind speed fluctuations may have a non-negligible impact on the wind turbine pitch bearing. This threshold is typically set based on the design characteristics of the wind turbine pitch bearing, operational experience, and the climate characteristics of the wind farm location. The threshold can be determined by statistically analyzing a large amount of historical wind speed data to identify the typical range of wind speed variation under normal operating conditions, and setting values ​​exceeding this range as the threshold; alternatively, it can be determined through simulation or experimental testing to study the impact of different wind speed variation amplitudes on the stress, vibration, and acoustic emission signals of the wind turbine pitch bearing, thereby identifying the critical threshold.

[0159] This application's solution introduces an environmental wind speed influence coefficient to accurately quantify the impact of dynamic changes in environmental wind speed on the monitoring results of wind turbine pitch bearing cracks. Specifically, this method compares the absolute difference between the environmental wind speed at the current time t and the environmental wind speed at the previous time t-1, and normalizes it with a preset environmental wind speed change threshold. When the wind speed change exceeds this threshold, it indicates that wind speed fluctuations may significantly affect the bearing's load distribution, vibration characteristics, and acoustic emission signals. In this case, the calculated environmental wind speed influence coefficient will be greater than 0, reflecting the potential risks brought about by environmental wind speed changes. This coefficient is then incorporated into the comprehensive influence analysis model, working together with the operating state influence coefficient to generate a more accurate comprehensive influence correction coefficient. In this way, the solution of this application can dynamically assess the degree of interference or influence of environmental wind speed changes on crack monitoring, avoiding misjudgments or omissions caused by drastic fluctuations in environmental factors, so that the final comprehensive score of wind turbine pitch bearing crack status can more realistically reflect the health status of the bearing.

[0160] Through the above technical solution, this application can effectively quantify the impact of dynamic changes in ambient wind speed on crack monitoring of wind turbine pitch bearings. Given that wind turbine pitch bearings operate outdoors, the ambient wind speed fluctuates frequently and drastically, directly affecting the bearing's load spectrum and vibration response, thus interfering with the accuracy of acoustic emission signals. By introducing an ambient wind speed influence coefficient and incorporating it into the calculation of the comprehensive influence correction coefficient, this application's solution can specifically correct the preliminary assessment results of acoustic emission energy, thereby reducing the uncertainty caused by changes in ambient wind speed. This allows the comprehensive score of wind turbine pitch bearing crack status to more accurately reflect the true health condition of the bearing, improving the sensitivity and reliability of crack monitoring, avoiding false alarms or missed alarms caused by environmental interference, and providing a more accurate decision-making basis for the maintenance and repair of wind turbine pitch bearings.

[0161] Preferably, the present invention further proposes a method for generating the comprehensive score of the crack state of the wind turbine pitch bearing, specifically including:

[0162] Through the formula: ;

[0163] Generate a comprehensive score for the crack condition of wind turbine pitch bearings. ;

[0164] In the formula, This indicates the preliminary score for the crack condition of the wind turbine pitch bearing. This represents the comprehensive impact correction coefficient. This represents the corrected strength coefficient;

[0165] Among them, the comprehensive score of crack status in wind turbine pitch bearings is the core indicator output by this method for crack monitoring and risk early warning in wind turbine pitch bearings. It comprehensively considers the crack development trend reflected by acoustic emission signals as well as the potential impact of external operating environment and working conditions on crack status, aiming to provide a more comprehensive and accurate risk assessment result.

[0166] in, The term serves as a correction factor, influencing the final score through the comprehensive impact correction coefficient. This mathematical model quantifies the combined effect of different factors on crack risk, resulting in a more refined risk assessment.

[0167] The correction strength coefficient is an adjustable weighting parameter used to control the correction strength of the comprehensive influence correction coefficient on the initial score of the crack state of the wind turbine pitch bearing. By adjusting this coefficient, the weight of external influencing factors in the final risk assessment can be flexibly adjusted according to the actual application scenario and experience. For example, when it is necessary to emphasize the influence of the external environment and operating conditions, the coefficient can be appropriately increased; conversely, it can be decreased. This coefficient is usually a positive number; for example, it can be set to 1.0 for direct correction, or set to other values ​​to amplify or reduce the correction effect.

[0168] This application's solution introduces a clear mathematical model to integrate the preliminary assessment of crack state based on acoustic emission signals with the comprehensive influence of the external operating environment and conditions, thereby generating a more comprehensive and accurate comprehensive score for the crack state of wind turbine pitch bearings. Specifically, the preliminary score for the wind turbine pitch bearing crack state serves as the basis for the assessment, directly reflecting the crack activity information revealed by the acoustic emission signals. Based on this, a comprehensive influence correction coefficient is introduced. This coefficient quantifies the combined influence of the external environment and operating conditions on the crack state of the wind turbine pitch bearing; a higher value generally indicates that the external conditions deviate more from the ideal state or are more unfavorable. By combining the preliminary score with the correction strength coefficient and... Multiplying the correction factors in various forms allows for dynamic adjustment of the initial score. A high overall impact correction coefficient indicates significant deviations or adverse factors in the external environment or operating conditions. The value of this term will be relatively small, thus reducing the final comprehensive score of the wind turbine pitch bearing crack state to some extent. Conversely, when the comprehensive influence correction coefficient is low, it indicates that the external environment or operating conditions are relatively stable or favorable. The value of this factor will be relatively large, making the final score closer to the product of the initial score and the corrected strength coefficient. The corrected strength coefficient provides a flexible adjustment mechanism, allowing the weight of external influencing factors in the final assessment to be adjusted according to actual needs. This integrated mechanism ensures that the final risk assessment not only relies on direct evidence of acoustic emission but also fully considers the actual operating conditions of the wind turbine pitch bearing, thereby providing a more robust and instructive crack monitoring and risk warning result.

[0169] Through the above technical solution, this application provides a method for effectively integrating preliminary acoustic emission crack assessment with the influence of external environment and operating conditions. This method introduces a modified strength coefficient and a comprehensive influence correction coefficient, and uses a specific mathematical formula to correct the preliminary score of the wind turbine pitch bearing crack state, thereby generating a comprehensive score for the wind turbine pitch bearing crack state. This ensures that the final crack risk assessment is not only based on direct evidence from acoustic emission signals, but also fully considers the actual operating environment and conditions of the wind turbine pitch bearing, avoiding the one-sidedness that may result from a single indicator assessment. Therefore, this solution can provide a more comprehensive, accurate, and robust crack state assessment result, significantly improving the reliability and effectiveness of wind turbine pitch bearing crack monitoring and risk early warning, and providing more solid data support for wind farm operation and maintenance decisions.

[0170] Please see Figure 4 In another embodiment of the present invention, a wind turbine pitch bearing crack monitoring and risk warning system based on acoustic emission is proposed. This system is used to execute the above-mentioned wind turbine pitch bearing crack monitoring and risk warning method based on acoustic emission, specifically including:

[0171] The data acquisition unit 10 is used to collect the cumulative acoustic emission energy, instantaneous acoustic emission energy, operating status data of the wind turbine pitch bearing, and ambient wind speed.

[0172] Preliminary analysis unit 20 is used to establish a preliminary risk analysis model based on the acoustic emission energy of the wind turbine pitch bearing and generate a preliminary score for the crack state of the wind turbine pitch bearing.

[0173] External factor analysis unit 30 is used to establish a comprehensive impact analysis model based on the operating status data of the wind turbine pitch bearing and the ambient wind speed, and generate comprehensive impact correction coefficients.

[0174] The comprehensive analysis unit 40 is used to generate a comprehensive score of the crack status of the wind turbine pitch bearing based on the comprehensive influence correction coefficient and the preliminary score of the crack status of the wind turbine pitch bearing.

[0175] The monitoring and risk warning unit 50 is used to monitor and provide risk warnings for wind turbine pitch bearing cracks based on the comprehensive score of the crack status of the wind turbine pitch bearing.

[0176] For preferred options, please refer to [link / reference]. Figure 5 The present invention further proposes that the preliminary analysis unit 20 specifically includes:

[0177] The cumulative energy analysis module 21 is used to generate a cumulative energy change characteristic score based on the acoustic emission cumulative energy of the wind turbine pitch bearing.

[0178] The instantaneous energy analysis module 22 is used to generate an acoustic emission instantaneous energy state score based on the acoustic emission instantaneous energy.

[0179] The comprehensive analysis module 23 is used to establish a preliminary analysis model based on the cumulative energy change characteristic score and the instantaneous energy state score of acoustic emission, and to generate a preliminary score of the crack state of the wind turbine pitch bearing.

[0180] For preferred options, please refer to [link / reference]. Figure 6 The present invention further proposes that the external factor analysis unit 30 specifically includes:

[0181] The operation status analysis module 31 is used to generate the operation status influence coefficient based on the operation status data of the wind turbine pitch bearing.

[0182] The environmental wind speed analysis module 32 is used to generate an environmental wind speed influence coefficient based on the environmental wind speed.

[0183] The comprehensive impact correction coefficient output module 33 is used to establish a comprehensive impact analysis model based on the operating status impact coefficient and the environmental wind speed impact coefficient, and generate the comprehensive impact correction coefficient.

[0184] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring and risk warning of cracks in wind turbine pitch bearings based on acoustic emission, characterized in that, The method specifically includes: Collect the cumulative acoustic emission energy, instantaneous acoustic emission energy, operating status data of the wind turbine pitch bearing, and ambient wind speed. A preliminary risk analysis model was established based on the acoustic emission energy of the wind turbine pitch bearing, and a preliminary score for the crack state of the wind turbine pitch bearing was generated. A comprehensive impact analysis model is established based on the operating status data of the wind turbine pitch bearing and the ambient wind speed, and a comprehensive impact correction coefficient is generated. Based on the comprehensive impact correction coefficient and the preliminary score of the crack state of the wind turbine pitch bearing, a comprehensive score of the crack state of the wind turbine pitch bearing is generated. Based on the comprehensive score of the crack status of wind turbine pitch bearings, we monitor and provide risk warnings for cracks in wind turbine pitch bearings.

2. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 1, characterized in that, The method for generating the preliminary score of the crack condition of the wind turbine pitch bearing specifically includes: Based on the accumulated acoustic emission energy of the wind turbine pitch bearing, a score for the characteristic of accumulated energy change is generated. Based on the instantaneous energy of acoustic emission, generate an instantaneous energy state score for acoustic emission; A preliminary analysis model was established based on the cumulative energy change characteristic score and the instantaneous energy state score of acoustic emission, and a preliminary score of the crack state of the wind turbine pitch bearing was generated.

3. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 2, characterized in that, The specific methods for generating the accumulated energy change characteristic score include: Through the formula: ; Generate cumulative energy change characteristic score ; In the formula, This represents the cumulative acoustic emission energy at time t. This represents the cumulative acoustic emission energy at time t-1. This represents the cumulative energy change deviation threshold.

4. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 2, characterized in that, The method for generating the instantaneous energy state score of acoustic emission specifically includes: Through the formula: ; Generate an instantaneous energy state score for acoustic emission. ; In the formula, This represents the historical average of the instantaneous energy of acoustic emission. This represents the preset instantaneous acoustic emission energy reference value.

5. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 2, characterized in that, The specific expression of the preliminary analysis model is as follows: ; In the expression, This indicates the preliminary score for the crack condition of the wind turbine pitch bearing. This represents the cumulative energy change characteristic score. This represents the instantaneous energy state score of acoustic emission. , All are weighting coefficients, and .

6. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 1, characterized in that, The specific methods for generating the comprehensive impact correction coefficient include: Based on the operating status data of the wind turbine pitch bearing, an operating status influence coefficient is generated; Based on the ambient wind speed, generate the ambient wind speed influence coefficient; A comprehensive impact analysis model is established based on the operational status impact coefficient and the environmental wind speed impact coefficient, and a comprehensive impact correction coefficient is generated. The specific expression of the comprehensive impact analysis model is as follows: ; In the expression, This represents the comprehensive impact correction coefficient. This represents the influence coefficient of the operating status. This represents the environmental wind speed influence coefficient. This represents the weighting coefficient of the operational status influence coefficient.

7. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 6, characterized in that, The specific methods for generating the operational status influence coefficient include: Through the formula: ; Generate operating status influence coefficient ; In the formula, This represents the i-th operating status data of the wind turbine pitch bearing. This represents the safety threshold for the i-th operating state data of the wind turbine pitch bearing. This represents the deviation threshold of the i-th operating state data of the wind turbine pitch bearing. This represents the weighting coefficient for the i-th operating state of the wind turbine pitch bearing, and n represents the number of operating state data for the wind turbine pitch bearing.

8. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 7, characterized in that, The weighting coefficient for the i-th operating state of the wind turbine pitch bearing: Through the formula: ; Weighting coefficients for generating the i-th operating state of the wind turbine pitch bearing ; In the formula, This represents the Pearson correlation coefficient between the i-th operating state of the wind turbine pitch bearing and the acoustic emission energy. This represents the Pearson correlation coefficient between the j-th operating state of the wind turbine pitch bearing and the acoustic emission energy, where n represents the number of operating state data for the wind turbine pitch bearing. The Pearson correlation coefficient between the operating state of the wind turbine pitch bearing and the acoustic emission energy is calculated based on the historical healthy operating period of the wind turbine pitch bearing.

9. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 6, characterized in that, The specific methods for generating the environmental wind speed influence coefficient include: Through the formula: ; Environmental wind speed influence coefficient ; In the formula, This represents the ambient wind speed at time t. This represents the ambient wind speed at time t-1. This represents the threshold for changes in ambient wind speed.

10. The method for monitoring and risk warning of wind turbine pitch bearing cracks based on acoustic emission according to claim 1, characterized in that, The specific methods for generating the comprehensive score for the crack state of the wind turbine pitch bearing include: Through the formula: ; Generate a comprehensive score for the crack condition of wind turbine pitch bearings. ; In the formula, This indicates the preliminary score for the crack condition of the wind turbine pitch bearing. This represents the comprehensive impact correction coefficient. This represents the corrected strength coefficient.

Images