Wind turbine generator variable pitch band-type brake fault diagnosis method, system and device and storage medium

Abstract

The invention relates to the technical field of equipment fault diagnosis, in particular to a wind turbine generator variable pitch band-type brake fault diagnosis method, system and equipment and a storage medium. A state machine instruction, a motor three-phase current, a main shaft rotating speed and a band-type brake switch signal of a variable pitch system of the wind turbine generator are collected in real time, and when the state machine instruction is switched to a shutdown state, a motor three-phase current zero returning state and a main shaft rotating speed fluctuation state are synchronously monitored; based on the time sequence correlation characteristics that the motor current returns to zero and the main shaft rotating speed continuously fluctuates, band-type brake braking force insufficiency judgment conditions are constructed; confirming by combining a locking state of a band-type brake switch signal, and analyzing paddle angle displacement change to obtain band-type brake mechanical failure characteristic parameters; and based on multi-source signal time sequence collaborative analysis, establishing a coupling judgment logic of a state machine instruction and a physical signal, and outputting a band-type brake fault diagnosis result and fault type classification according to the coupling judgment logic. Precise identification and early warning of band-type brake mechanical failure are realized, and the problems of high fault reproduction dependence and high misjudgment rate are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of equipment fault diagnosis, and in particular to a wind turbine generator unit variable pitch brake fault diagnosis method, system, device and storage medium. BACKGROUND

[0002] The variable pitch system of a wind turbine generator unit is one of the core control systems of the wind turbine generator unit, which controls the rotational speed and output power of the wind wheel by adjusting the pitch angle. The motor brake mechanism in the variable pitch system undertakes the important function of locking the pitch position in the shutdown state, preventing accidental rotation of the pitch under the action of wind, and ensuring the safe operation of the unit.

[0003] Traditional variable pitch brake fault detection mainly relies on fault reproduction and single signal monitoring methods. When the brake mechanism has mechanical failure such as insufficient braking force, technicians usually need to wait for the fault to occur again to diagnose, and this passive detection method has obvious lag. At the same time, the existing detection method mainly relies on single signals such as encoder angle to make judgments, which is easy to misjudge the brake mechanical failure as a limit switch fault or calibration problem, leading to maintenance personnel using the trial-and-error method to replace hardware step by step to troubleshoot.

[0004] Although the existing state machine monitoring logic can identify the switching of the running state of the variable pitch system, it lacks deep linkage analysis capability with physical signals. When the state machine shows a shutdown state but the actual pitch is still in a micro-motion state, the traditional monitoring system cannot accurately identify this abnormal condition where the state machine instruction does not match the physical response. In addition, the brake fault has the characteristics of strong concealment, and under the influence of environmental interference factors, the detection method relying solely on threshold comparison is easy to produce false alarms or omissions.

[0005] At present, the wind power industry lacks effective multi-source signal collaborative analysis technology, and cannot realize the time sequence correlation diagnosis of state machine instructions, motor current, rotational speed signals and brake control signals. This makes it difficult to accurately identify key faults such as brake mechanical failure at an early stage, affecting the safe and stable operation of the wind turbine generator unit. SUMMARY

[0006] In view of the problems existing in the prior art, the present application is proposed.

[0007] Therefore, the present application mainly solves the technical problem: the existing wind turbine variable pitch brake fault detection method has the problems of strong diagnosis lag, high misjudgment rate and high maintenance cost. The traditional detection method relies on fault recurrence and cannot realize early warning of brake mechanical failure. Single signal monitoring is easily disturbed by the environment, leading to inaccurate fault positioning. Lack of multi-source signal timing collaborative analysis capability, unable to effectively identify abnormal conditions where state machine instructions and physical signals do not match. Therefore, it is necessary to provide a wind turbine variable pitch brake fault diagnosis method based on multi-source signal timing collaborative analysis, to realize accurate fault positioning without fault recurrence, improve diagnosis accuracy and reduce maintenance cost.

[0008] To solve the above technical problems, the present application provides the following technical solutions:

[0009] In a first aspect, the present application provides a wind turbine variable pitch brake fault diagnosis method, which comprises: collecting state machine instructions, motor three-phase current, main shaft speed and brake switch signals of a wind turbine variable pitch system in real time; when the state machine instructions switch to a shutdown state, synchronously monitoring the motor three-phase current zero state and the main shaft speed fluctuation state;

[0010] Based on the timing correlation characteristics of the motor current zero and the main shaft speed continuous fluctuation, a brake braking force deficiency determination condition is constructed.

[0011] Combined with the lockup state confirmation of the brake switch signal, the blade angle displacement change is analyzed to obtain the brake mechanical failure characteristic parameters.

[0012] Based on multi-source signal timing collaborative analysis, a coupling determination logic of state machine instructions and physical signals is established, and a brake fault diagnosis result and fault type classification are output according to the coupling determination logic.

[0013] As a preferred scheme of the wind turbine variable pitch brake fault diagnosis method, the synchronous monitoring of the motor three-phase current zero state and the main shaft speed fluctuation state comprises: detecting that the motor three-phase current value drops below a preset current threshold, and confirming that the motor drive stops; detecting that the main shaft speed fluctuation exceeds a preset speed threshold and the duration exceeds a preset time threshold.

[0014] As a preferred scheme of the wind turbine variable pitch brake fault diagnosis method, the construction of the brake braking force deficiency determination condition comprises: setting a timing delay compensation of the state machine shutdown instruction and the motor current response.

[0015] The dynamic threshold range of the main shaft speed fluctuation detection is set; the timing correlation determination rule of the current zero and the speed fluctuation is established.

[0016] As a preferred scheme of the wind turbine variable pitch brake fault diagnosis method, the analysis of the blade angle displacement change to obtain the brake mechanical failure characteristic parameter comprises collecting the blade angle displacement when the brake switch signal is in the locked state; calculating the blade angle displacement rate and the displacement cumulative amount; and comparing the blade angle displacement with a preset displacement threshold to determine the brake mechanical failure degree.

[0017] As a preferred scheme of the wind turbine variable pitch brake fault diagnosis method, the coupling determination logic of the state machine instruction and the physical signal comprises a four-dimensional state matrix of the state machine shutdown instruction, the motor current, the main shaft speed and the brake switch signal; setting the identification rule of the abnormal coupling mode in the four-dimensional state matrix; and calculating the brake failure confidence according to the abnormal coupling mode.

[0018] As a preferred scheme of the wind turbine variable pitch brake fault diagnosis method, the output of the brake fault diagnosis result and the fault type classification comprises setting the determination threshold of the warning level and the emergency stop level according to the brake failure confidence; generating a fault report containing the fault time, the fault type and the associated data; and outputting the brake replacement suggestion and the maintenance guidance information.

[0019] As a preferred scheme of the wind turbine variable pitch brake fault diagnosis method, the method further comprises comparing the current operation data with the historical normal operation data to identify the signal abnormal fluctuation mode; excluding the electrical interference factors based on the historical data back analysis to lock the mechanical failure root cause; generating the brake degradation warning according to the signal abnormal fluctuation mode before the main fault alarm is triggered; establishing a brake health degree evaluation system to output the brake remaining service life prediction.

[0020] In a second aspect, the embodiments of the present application provide a wind turbine variable pitch brake fault diagnosis system, which comprises a data acquisition module, a state machine instruction, a motor three-phase current, a main shaft speed and a brake switch signal of a wind turbine variable pitch system are acquired in real time, when the state machine instruction is switched to the shutdown state, the motor three-phase current zero state and the main shaft speed fluctuation state are monitored synchronously;

[0021] A determination module constructs a brake braking force deficiency determination condition based on the time sequence correlation characteristics of the motor current zero and the main shaft speed continuous fluctuation;

[0022] A parameter output module combines the locked state confirmation of the brake switch signal, analyzes the blade angle displacement change, and obtains the brake mechanical failure characteristic parameter;

[0023] A result output module establishes the coupling determination logic of the state machine instruction and the physical signal based on the time sequence cooperative analysis of the multiple source signals, and outputs the brake fault diagnosis result and the fault type classification according to the coupling determination logic.

[0024] Thirdly, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, they implement the steps of the wind turbine pitch brake fault diagnosis method as described in the first aspect of the present invention.

[0025] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, they implement the steps of the wind turbine pitch brake fault diagnosis method as described in the first aspect of the present invention.

[0026] The beneficial effects of this invention are as follows: This invention achieves accurate identification and early warning of pitch brake faults in wind turbines through real-time acquisition of multi-source signals and time-series collaborative analysis technology. The initial steps construct a precise correlation model between state machine commands and physical signals, overcoming the technical bottlenecks of low accuracy and slow response in traditional single-signal monitoring. By synchronously monitoring the motor current at zero state and the main shaft speed fluctuation state, it accurately captures the key characteristics of insufficient braking force, providing a reliable data foundation for intelligent diagnostic decisions. The intermediate steps establish time-series correlation judgment conditions and blade angle displacement analysis to achieve dynamic parameter identification based on physical response characteristics, solving the problem of traditional reliance on fault reproduction for timely diagnosis and effectively avoiding the risk of misdiagnosing mechanical brake failure as other types of faults. The final step, with its four-dimensional state matrix and coupled judgment logic, comprehensively considers the interaction of multi-source signals, achieving system-level fault identification and classification, avoiding overall diagnostic errors caused by local judgments. The overall effect is a significant improvement in fault diagnosis accuracy, reduced maintenance costs and downtime, reduced manual troubleshooting workload, and a reduction in equipment safety accidents through preventative diagnosis, providing a complete technical solution for intelligent operation and maintenance and safe and stable operation of wind turbines. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of the present invention, 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 A flowchart for diagnosing pitch brake faults in wind turbine units;

[0029] Figure 2 A diagram of computer equipment for diagnosing pitch brake faults in wind turbine units;

[0030] Figure 3 A three-state line graph of the blade for the fault diagnosis method of wind turbine pitch brake;

[0031] Figure 4 The motor current, speed change comparison chart for the wind turbine variable pitch brake fault diagnosis method;

[0032] Figure 5 The blade 3 motor brake and blade change broken line chart for the wind turbine variable pitch brake fault diagnosis method. DETAILED DESCRIPTION

[0033] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0034] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the present application, but the present application can also be implemented in other ways different from the description, and those skilled in the art can make similar generalizations without departing from the concept of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0035] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an embodiment that is separate or selectively excluded from other embodiments.

[0036] Embodiment 1

[0037] Reference Figure 1 - Figure 2 For the first embodiment of the present application, the embodiment provides a wind turbine variable pitch brake fault diagnosis method, comprising,

[0038] S100: Real-time acquisition of the state machine instruction of the wind turbine variable pitch system, the three-phase current of the motor, the speed of the main shaft and the brake switch signal, when the state machine instruction is switched to the shutdown state, the three-phase current zero state and the main shaft speed fluctuation state are monitored synchronously;

[0039] S200: Based on the time sequence correlation characteristics of the motor current zero and the continuous fluctuation of the main shaft speed, the brake braking force insufficient judgment condition is constructed;

[0040] S300: Combined with the brake lock state confirmation of the brake switch signal, the blade angle displacement change is analyzed to obtain the brake mechanical failure characteristic parameter;

[0041] S400: Based on the time sequence cooperative analysis of multiple source signals, the coupling judgment logic of the state machine instruction and the physical signal is established, and the brake fault diagnosis result and the fault type classification are output according to the coupling judgment logic.

[0042] It should be noted that during the operation of the wind turbine variable pitch system, the state machine instruction, motor current, main shaft speed, brake switch signal and other parameters change greatly, causing dynamic changes in the brake braking state and the blade position, and the brake braking force will also change due to the influence of wind load and environmental factors. The running variable pitch system is in a complex working condition environment, making it difficult to accurately identify the mechanical failure characteristics of the brake, and the fault diagnosis is often delayed; at the same time, due to the strong concealment of the brake fault, the mechanical failure develops continuously, thereby endangering the system safety, and the traditional single signal monitoring has great limitations for accurate diagnosis, and also increases the maintenance cost due to misjudgment; therefore, the fault diagnosis and early warning of the variable pitch brake are also very important.

[0043] Therefore, in view of the above-mentioned fault identification and diagnosis control problems, through the steps of S100-S400, a multi-source signal cooperative detection model is constructed to simulate the running state of the system under typical working conditions, to obtain the time sequence correlation between the state machine instruction and the physical signal under the influence of brake failure, to realize accurate determination of the brake braking force deficiency; to realize real-time monitoring of the blade angle displacement and the brake switch signal, to dynamically calculate the mechanical failure characteristic parameters of the brake, to realize early warning of the abnormal situation of the brake fault; at the same time, based on the four-dimensional state matrix and the confidence calculation model, accurate diagnosis and type classification of the variable pitch brake fault are realized.

[0044] Embodiment 2

[0045] Reference Figure 1 - Figure 5 This is the second embodiment of the present application.

[0046] In this embodiment, the state machine instruction, motor three-phase current, main shaft speed and brake switch signal of the wind turbine variable pitch system are collected in real time in step S100, and when the state machine instruction switches to the shutdown state, the motor three-phase current zero state and the main shaft speed fluctuation state are monitored synchronously, including the following steps A1-A2:

[0047] A1: detecting that the motor three-phase current value drops below the preset current threshold value, and confirming that the motor drive stops;

[0048] A2: detecting that the main shaft speed fluctuation exceeds the preset speed threshold value and the duration exceeds the preset time threshold value. Specifically, the specific construction form of the multi-source signal cooperative detection model in A1-A2 is:

[0049] S detection (t)=α·I motor (t)+β·ω fluctuation (t)+γ·State machine (t)+δ·Brake signal (t)

[0050] In the formula, Sdetection (t) represents the multi-source signal collaborative detection result at time t; I motor (t) represents the effective value (A) of the three-phase current of the motor; ω fluctuation (t) represents the spindle speed fluctuation value (° / s); State machine (t) represents the state value of the state machine instruction; Brake signal (t) represents the status value of the brake switch signal; δ represents the weight coefficient of each signal, which is obtained through training and optimization using historical fault data.

[0051] It should be noted that in the formula, α·I motor (t) is the dominant term for motor drive state, reflecting whether the motor has truly stopped operating; β·ω fluctuation (t) represents the abnormal speed detection item, which captures the blade slippage phenomenon caused by brake failure; γ·State machine (t)+δ·Brake signal (t) is a control command consistency verification item to ensure the logical correctness of state machine commands and brake signals.

[0052] like Figure 3 As shown in the line graph, by monitoring the state changes of blade 3, the response process of the blade state at the moment of state machine command switching can be clearly observed, verifying the effectiveness of motor drive stop confirmation and speed fluctuation detection. Blue lines represent blade angular displacement changes, and yellow lines represent state machine command states.

[0053] In an optional implementation, the method for confirming the motor drive has stopped in step S100 can also be verified by analyzing the spectral characteristics of the three-phase current. Combining current amplitude detection and frequency domain analysis, when the amplitude of the three-phase current decreases and the high-frequency components disappear, it is confirmed that the motor has completely stopped operating. This method can effectively eliminate false judgments caused by electromagnetic interference and improve detection accuracy. Specifically, the three-phase current is analyzed in the frequency domain using a fast Fourier transform. During normal operation, the motor current contains the fundamental frequency and its harmonic components. After stopping, the high-frequency components rapidly decay to below the noise level.

[0054] In another optional implementation, the abnormal slip phenomenon determined in step S100 can also be comprehensively judged in conjunction with the change in blade angle. When the main shaft speed fluctuates while the blade angle continues to deviate, the slip state caused by insufficient braking force of the brake is further confirmed. Through cross-validation of multi-dimensional signals, misjudgments caused by external factors such as wind load fluctuations are avoided. The cross-validation logic is as follows: when the speed fluctuation amplitude is greater than the threshold and the blade angle displacement is greater than 1° and the duration is greater than 5 seconds, a brake failure warning is triggered.

[0055] In an alternative embodiment, the multi-dimensional data acquisition matrix constructed in step S100 can also be optimized by a time sequence alignment algorithm, which uniformly calibrates the acquisition time stamps of the state machine instructions, motor current, rotating speed signal and brake band signal, and eliminates the differences in signal transmission delay. In specific implementation, a reference clock source is set to time-synchronize the signals of each sensor, ensuring the accuracy of subsequent time sequence correlation analysis.

[0056] In another alternative embodiment, the multi-dimensional data acquisition matrix in step S100 can also be dynamically optimized by an adaptive sampling frequency adjustment mechanism, which automatically adjusts the data acquisition frequency according to the operating state of the wind turbine. In the normal operating state, a standard sampling rate is used, and when a state machine switching signal is detected, the sampling frequency is automatically increased to a high-precision mode, ensuring complete capture of the key state change process.

[0057] In the present embodiment, based on the time sequence correlation characteristics of the motor current returning to zero and the continuous fluctuation of the main shaft rotating speed, the brake band braking force deficiency determination condition is constructed in step S200, including the following steps B1-B3:

[0058] B1: Set the time sequence delay compensation of the state machine shutdown instruction and the motor current response; through historical data statistical analysis, it is found that the normal response delay is 0.5-1.0 seconds, and if it exceeds this range, it is marked as abnormal;

[0059] B2: Set the dynamic threshold range of the main shaft rotating speed fluctuation detection; dynamically adjust the threshold according to the wind speed level and the blade load, set the threshold to ±0.3° / s at low wind speed, and set the threshold to ±0.8° / s at high wind speed;

[0060] B3: Establish the time sequence correlation determination rule of current returning to zero and rotating speed fluctuation; when the current drops below the threshold value and the rotating speed still has continuous fluctuation within 3 seconds, and the fluctuation amplitude exceeds the dynamic threshold, the brake band braking force deficiency warning is triggered.

[0061] Specifically, the brake band braking force deficiency determination condition of B1-B3 can be embodied by the following determination logic:

[0062] Fault brake =f(I motor ,ω fluctuation ,t delay ,State machine )

[0063] Wherein, State machine =0 (shutdown state) and I motor <0.1A (motor current returning to zero) and |ω fluctuation |>ω threshold (rotating speed fluctuation exceeding threshold) and t delayIf the duration exceeds 5 seconds, a Fault is determined. brake =1 (Insufficient braking force of the holding brake).

[0064] For example, the dynamic threshold of the above judgment condition can be specifically expressed as:

[0065] ω threshold =ω base ·K wind (v)·K load (F)·K temp (T)

[0066] In the formula, ω base The reference speed fluctuation threshold is typically taken as 0.5° / s; K wind (v) is the wind speed correction factor; K load (F) is the load correction factor; K temp (T) is the temperature correction factor.

[0067] It should be noted that the core of the above judgment logic lies in capturing the mismatch between state machine instructions and physical responses. The drive is stopped by the motor current returning to zero, and the brake failure is identified by the continuous fluctuation of the speed. The dynamic threshold design takes into account the impact of environmental factors on the judgment accuracy and avoids misjudgment caused by changes in wind load.

[0068] like Figure 4 As shown, the comparison chart of motor current and speed changes visually demonstrates the temporal correlation between the motor current returning to zero and the spindle speed fluctuation. Under normal circumstances, when the motor current drops below the preset threshold, the speed should return to zero synchronously. However, when the brake fails, the speed continues to fluctuate, verifying the accuracy of the brake's insufficient braking force determination condition. The blue line represents the change in the motor's three-phase current, and the yellow line represents the spindle speed fluctuation.

[0069] In an optional implementation, the insufficient braking force determination condition in step S200 can also be combined with a temperature compensation mechanism to consider the influence of ambient temperature on the braking force. In low-temperature environments, the brake material becomes more rigid, potentially leading to a slight increase in braking force; in high-temperature environments, the brake material softens, potentially resulting in a decrease in braking force. Real-time monitoring using an ambient temperature sensor dynamically adjusts the determination threshold, improving the environmental adaptability of the diagnosis. The temperature compensation algorithm employs a piecewise linear interpolation method to establish a mapping relationship between temperature and braking force correction coefficients within the range of -20℃ to 60℃.

[0070] In another alternative embodiment, the timing correlation determination rule in step S200 can also introduce a historical failure mode learning mechanism, by analyzing the signal characteristics of historical brake failure cases, establishing a failure mode library, and using a pattern matching algorithm for failure identification. The failure mode library contains the timing signal characteristics of typical failures, such as complete failure mode, progressive degradation mode, intermittent failure mode, etc. When the current detected signal pattern has a high degree of match with the historical failure pattern, the determination threshold is reduced in advance to realize early warning.

[0071] In the present embodiment, the brake failure state confirmation in step S300 in combination with the brake switch signal analyzes the blade angle displacement change to obtain the brake mechanical failure characteristic parameters, including the following steps C1-C3:

[0072] C1: Collect the blade angle displacement when the brake switch signal is in the locked state;

[0073] C2: Calculate the blade angle displacement rate and displacement accumulation;

[0074] C3: Compare the blade angle displacement with the preset displacement threshold to determine the degree of brake mechanical failure.

[0075] Specifically, the brake mechanical failure characteristic parameters of C1-C3 can be embodied by the following calculation model:

[0076]

[0077] In the formula, Δθ cumulative is the cumulative angle displacement; is the maximum displacement rate; t duration is the duration.

[0078] For example, the failure degree determination criteria of the above parameters can be specifically expressed as:

[0079] When Δθ cumulative <1° and , it is determined to be in normal state;

[0080] When 1°≤Δθ cumulative <3° and , it is determined to be in mild failure;

[0081] When 3°≤Δθ cumulative <5° and , it is determined to be in moderate failure;

[0082] When Δθ cumulative ≥5° or , it is determined to be in severe failure.

[0083] It should be noted that the construction form of the above characteristic parameters can accurately reflect the severity of the mechanical failure of the brake by quantifying the dynamic characteristics of the blade angle change, and provide a reliable basis for subsequent maintenance decisions. The displacement rate reflects the sharpness of the failure, the cumulative displacement reflects the overall impact of the failure, and the duration reflects the stability of the failure. The combination of the three can comprehensively evaluate the state of the brake.

[0084] As shown in Figure 5 The blade 3 motor brake and the blade change broken line chart show the actual displacement change process of the blade angle when the brake switch signal is in the locked state. By comparing the timing relationship between the brake signal state and the blade angle change, the effectiveness of the brake mechanical failure characteristic parameter calculation model is verified. Blue line: represents the brake switch signal state, yellow line: represents the cumulative displacement of the blade angle.

[0085] In an optional embodiment, the analysis of the blade angle displacement change in step S300 can also combine a wind load compensation algorithm to monitor the wind load change in real time through a wind speed sensor and a force sensor, and correct the blade angle displacement. Under high wind speed conditions, even if the brake is normal, the blade may also produce a small displacement due to excessive load. Load compensation can avoid misjudgment and improve diagnostic accuracy. The load compensation model is based on the aerodynamic characteristics and structural stiffness of the blade, and considers the influence of factors such as wind direction angle and pitch angle on load distribution.

[0086] In another optional embodiment, the brake mechanical failure characteristic parameters obtained in step S300 can also be verified by vibration analysis. Acceleration sensors are installed on the blade hub and the nacelle to monitor the vibration characteristics when the brake is actuated. Normal brake action should produce a specific vibration mode, including impact vibration at the moment of brake contact and steady-state vibration after locking. When the brake fails, the vibration characteristics will change significantly, such as reduced vibration amplitude due to insufficient contact force, abnormal frequency components due to poor contact, etc. The vibration characteristics are extracted through frequency spectrum analysis and wavelet transform, and compared with the normal vibration mode.

[0087] In this embodiment, based on the multi-source signal timing collaborative analysis in step S400, the coupling determination logic of the state machine instruction and the physical signal is established, and the brake fault diagnosis result and the fault type classification are output according to the coupling determination logic, including the following steps D1-D3:

[0088] D1: Establishing the coupling determination logic of the state machine instruction and the physical signal includes establishing a four-dimensional state matrix of the state machine shutdown instruction, the motor current, the main shaft speed, and the brake switch signal; setting the recognition rules of the abnormal coupling mode in the four-dimensional state matrix; and calculating the brake failure confidence according to the abnormal coupling mode.

[0089] D2: The output clutch failure diagnosis result and failure type classification includes setting the warning level and emergency stop level determination threshold according to the clutch failure confidence; generating a failure report containing the failure time, failure type and associated data; outputting clutch replacement suggestions and maintenance guidance information.

[0090] D3: Further includes, comparing the current running data with the historical normal running data, identifying the signal abnormal fluctuation mode; based on historical data backtracking analysis, excluding electrical interference factors, locking mechanical failure causes; before triggering the main failure alarm, generating clutch degradation warning according to the signal abnormal fluctuation mode; establishing a clutch health degree evaluation system, outputting clutch remaining service life prediction.

[0091] Specifically, the coupling determination logic of D1-D3 can be embodied by the following model:

[0092] Four-dimensional state matrix analysis:

[0093] M normal =[0, <0.1A, <0.1° / s, 1]

[0094] M abnormal =[0, <0.1A, ≥0.5° / s, 1]

[0095] Clutch failure confidence calculation:

[0096]

[0097] In the formula, I expected is the expected current value (0A); ω expected is the expected speed (0° / s); t max is the maximum allowed duration (10s); weight coefficients α=0.2, β=0.6, γ=0.2.

[0098] For example, the failure type classification standard can be set as:

[0099] When 0.3≤Confidence<0.5, output "clutch degradation warning";

[0100] When 0.5≤Confidence<0.7, output "clutch mild failure", and suggest monitoring operation;

[0101] When 0.7≤Confidence<0.9, output "clutch moderate failure", and suggest planned maintenance;

[0102] When Confidence≥0.9, output "clutch severe failure", and suggest immediate shutdown for maintenance.

[0103] It should be noted that the four-dimensional state matrix can accurately identify hidden faults that cannot be found by traditional single signal monitoring through the cooperative analysis of multi-dimensional signals. The confidence calculation comprehensively considers the influence of current abnormalities, speed abnormalities and duration, and provides quantitative fault severity evaluation to provide a scientific basis for maintenance decision-making.

[0104] In an alternative embodiment, the coupling determination logic of the state machine instruction and the physical signal established in step S400 can also use the Bayesian network method to model the causal relationship between signals as a directed graph and calculate the posterior probability of fault occurrence through conditional probability. The Bayesian network model includes a state machine instruction node, a motor current node, a main shaft speed node, a brake signal node and a fault state node, and the conditional probability table between the nodes is obtained by statistical analysis of historical data. The inference process uses the variable elimination algorithm to calculate the probability distribution of the brake fault when some signal abnormalities are observed.

[0105] In another alternative embodiment, the output of the brake fault diagnosis result in step S400 can also be combined with an expert system to establish a fault diagnosis knowledge base containing fault symptoms, cause analysis, treatment suggestions and other expert experience. The knowledge base uses production rule representation in the form of "IF condition THEN conclusion" to derive possible fault causes from observed symptoms through a forward reasoning mechanism. The expert system also integrates uncertainty reasoning capability, using a certainty factor to represent the credibility of the rule, and handling incomplete information-based diagnosis decisions.

[0106] In summary, the construction of the multi-source signal cooperative detection model realizes accurate quantification of motor drive state, speed anomaly detection and control instruction consistency through weight coefficient optimization and historical data training, which has higher stability compared to traditional experience judgment methods. The introduction of spectral feature analysis and cross-validation mechanism eliminates the influence of external factors such as electromagnetic interference and wind load fluctuations, reduces the misjudgment rate and improves the environmental adaptability of the diagnosis system. The time sequence alignment algorithm and adaptive sampling frequency adjustment ensure the complete capture of key state change processes, solving the technical limitations of traditional fixed sampling that cannot adapt to dynamic working conditions. Dynamic threshold design combined with temperature compensation and historical fault mode learning realizes the self-learning and adaptive ability of the diagnosis algorithm, which can intelligently adjust according to the actual operating environment and equipment state. The wind load compensation algorithm and vibration analysis auxiliary verification further enhance the reliability of the diagnosis result, and through the comprehensive analysis of multi-dimensional physical signals, the limitations of single-dimensional judgment are avoided. The four-dimensional state matrix and the confidence calculation model provide quantitative fault severity evaluation to support accurate maintenance decision-making. The integrated application of Bayesian networks and expert systems realizes uncertainty reasoning capability, which can provide reliable diagnosis conclusions in the case of incomplete information.

[0107] Example 3

[0108] The above is a schematic scheme of a wind turbine variable pitch brake failure diagnosis method. It should be noted that the technical scheme of the wind turbine variable pitch brake failure diagnosis system is the same as the technical scheme of the wind turbine variable pitch brake failure diagnosis method described above. The technical details of the wind turbine variable pitch brake failure diagnosis system in this embodiment are not described in detail, and can be referred to the description of the technical scheme of the wind turbine variable pitch brake failure diagnosis method described above.

[0109] The embodiment also provides a wind turbine variable pitch brake failure diagnosis system, comprising:

[0110] The data acquisition module acquires the state machine instruction, the motor three-phase current, the main shaft speed and the brake switch signal of the wind turbine variable pitch system in real time, and synchronously monitors the motor three-phase current zero state and the main shaft speed fluctuation state when the state machine instruction switches to the shutdown state.

[0111] The determination module constructs a brake braking force deficiency determination condition based on the timing correlation characteristics of the motor current zero and the main shaft speed continuous fluctuation.

[0112] The parameter output module analyzes the blade angle displacement change to obtain the brake mechanical failure characteristic parameter in combination with the lockup state confirmation of the brake switch signal.

[0113] The result output module establishes the coupling determination logic of the state machine instruction and the physical signal based on the multi-source signal timing collaborative analysis, and outputs the brake failure diagnosis result and the fault type classification according to the coupling determination logic.

[0114] The embodiment also provides an electronic device suitable for wind turbine variable pitch brake failure diagnosis, comprising: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the wind turbine variable pitch brake failure diagnosis method proposed in the above embodiment.

[0115] The embodiment also provides a storage medium having a computer program stored thereon, which is executed by a processor to realize the wind turbine variable pitch brake failure diagnosis method proposed in the above embodiment.

[0116] The storage medium proposed in the embodiment belongs to the same inventive concept as the wind turbine variable pitch brake failure diagnosis method proposed in the above embodiment. The technical details not described in detail in this embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0117] Those skilled in the art can clearly understand the present application by the above description of the embodiments, and the present application can be realized by software and necessary general hardware, and of course, can also be realized by hardware. Based on such understanding, the technical solutions of the present application or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk, or an optical disc, and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.

[0118] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method for diagnosing pitch brake faults in wind turbine generators, characterized in that: This includes real-time acquisition of state machine commands, three-phase motor current, spindle speed, and brake switch signals of the wind turbine pitch system. When the state machine command switches to the stop state, the system simultaneously monitors the zero-state of the three-phase motor current and the fluctuation state of the spindle speed. Based on the time-series correlation characteristics of motor current returning to zero and spindle speed continuously fluctuating, a condition for determining insufficient braking force of the holding brake is constructed. By confirming the locked state of the brake switch signal and analyzing the changes in blade angle displacement, the characteristic parameters of brake mechanical failure are obtained. Based on multi-source signal timing collaborative analysis, a coupling judgment logic between state machine instructions and physical signals is established, and the brake fault diagnosis result and fault type classification are output according to the coupling judgment logic.

2. The method for diagnosing wind turbine pitch brake faults as described in claim 1, characterized in that: The synchronous monitoring of the motor's three-phase current returning to zero and the spindle speed fluctuation includes detecting when the motor's three-phase current value drops below a preset current threshold and confirming that the motor drive has stopped; and detecting when the spindle speed fluctuation exceeds a preset speed threshold and the duration exceeds a preset time threshold.

3. The method for diagnosing wind turbine pitch brake faults as described in claim 2, characterized in that: The criteria for determining insufficient braking force of the brake include setting time delay compensation between the state machine stop command and the motor current response; Set the dynamic threshold range for spindle speed fluctuation detection; Establish a timing correlation judgment rule between current returning to zero and speed fluctuation.

4. The method for diagnosing wind turbine pitch brake faults as described in claim 3, characterized in that: The analysis of blade angle displacement changes to obtain brake mechanical failure characteristic parameters includes: collecting the blade angle change when the brake switch signal is in a locked state; calculating the blade angle displacement rate and cumulative displacement; and comparing the blade angle displacement with a preset displacement threshold to determine the degree of brake mechanical failure.

5. The method for diagnosing wind turbine pitch brake faults as described in claim 4, characterized in that: The logic for establishing the coupling between state machine commands and physical signals includes establishing a four-dimensional state matrix of state machine stop commands, motor current, spindle speed, and brake switch signals. Define the identification rules for abnormal coupling patterns in the four-dimensional state matrix; calculate the confidence level of brake failure based on the abnormal coupling patterns.

6. The method for diagnosing wind turbine pitch brake faults as described in claim 5, characterized in that: The output brake fault diagnosis results and fault type classification include setting judgment thresholds for warning level and emergency stop level based on the brake failure confidence level; Generates a fault report containing fault time, fault type, and related data; outputs brake replacement recommendations and maintenance guidance information.

7. The method for diagnosing wind turbine pitch brake faults as described in claim 6, characterized in that: It also includes comparing current operating data with historical normal operating data to identify abnormal signal fluctuation patterns; Based on historical data retrospective analysis, electrical interference factors were eliminated, and the root cause of mechanical failure was identified. Before the main fault alarm is triggered, a brake deterioration warning is generated based on the abnormal signal fluctuation pattern; Establish a brake health assessment system to generate a prediction of the remaining service life of the brake.

8. A wind turbine pitch brake fault diagnosis system, based on the wind turbine pitch brake fault diagnosis method according to any one of claims 1 to 7, characterized in that: It also includes a data acquisition module, which collects the state machine commands, three-phase current of the motor, spindle speed and brake switch signals of the wind turbine pitch system in real time. When the state machine command switches to the stop state, it synchronously monitors the three-phase current of the motor to zero and the spindle speed fluctuation. The judgment module constructs the judgment condition for insufficient braking force of the holding brake based on the time-series correlation characteristics of the motor current returning to zero and the spindle speed continuously fluctuating. The parameter output module, combined with the confirmation of the brake switch signal's locked state, analyzes the changes in blade angle displacement to obtain the characteristic parameters of brake mechanical failure. The output module, based on multi-source signal timing collaborative analysis, establishes a coupling judgment logic between state machine instructions and physical signals, and outputs the brake fault diagnosis results and fault type classification according to the coupling judgment logic.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the wind turbine pitch brake fault diagnosis method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the wind turbine pitch brake fault diagnosis method according to any one of claims 1 to 7.

Images