A wind turbine blade monitoring method and system based on blade reaching time monitoring

Abstract

The application discloses a kind of wind power blade monitoring method and system based on blade reach time monitoring, belong to wind power blade monitoring technical field, this method needs to collect the torque value, waving vibration value, oscillation vibration value and clearance value of three blades when wind driven generator is connected to grid, when blade is in the position directly below rotation plane, record starting time, and calculate blade through period according to wind wheel rotating speed, in each period, whether the safety setting value is exceeded by judging torque change rate and vibration frequency deviation degree, realize blade fracture, tower scanning early warning, and whether effective value appears to clearance value, to judge whether blade is broken, simultaneously judge whether clearance value is less than safety value to determine whether blade is scanned tower;It also includes establishing blade fracture and tower scanning feature library to correct safety setting value;The application can effectively monitor wind power blade fracture and tower scanning condition, take pitch or shutdown measures in advance, reduce blade damage risk, improve wind driven generator set operation safety.

Description

Technical Field

[0001] This invention belongs to the field of wind turbine blade monitoring technology, specifically relating to a wind turbine blade monitoring method and system based on blade arrival time monitoring. Background Technology

[0002] As a clean energy source, the safe and stable operation of wind turbine generators is crucial for ensuring grid security and extending equipment lifespan. Wind turbine blades, as key components, directly impact the safe operation of the entire generator. With the increasing capacity of individual wind turbine units, blade length also increases. During operation, blades may break or strike towers, posing serious safety hazards and economic losses to wind farms.

[0003] Currently, wind turbine blade monitoring technologies mainly include vibration monitoring and strain monitoring methods. For example, Chinese patent document CN107829885A, published on March 23, 2018, discloses a wind turbine blade vibration monitoring and system that considers environmental parameter correction. This system arranges dual-axis accelerometers at corresponding locations on the wind turbine blades and hub to measure temperature and vibration data at the blades and hub, and transmits the data to a WindBVM data acquisition unit. An industrial-grade router is fixed inside the wind turbine nacelle and connected to the main control cabinet of the nacelle via a network cable. The WindBVM data acquisition unit obtains SCADA data from the main control cabinet of the wind turbine, obtaining environmental parameters such as wind speed and pitch angle. The WindBVM data acquisition unit uniformly saves the vibration data, temperature data, and SCADA operating condition data according to the corresponding clock, and enters the wind farm ring network through the industrial-grade router, achieving data transmission through internal and external network isolation. Combining the collected blade vibration data and wind turbine environmental parameters, the system performs real-time monitoring and diagnosis of whether the blade has tip cracks, lightning damage, or blade icing.

[0004] However, existing technologies still have shortcomings: 1. There is a lack of effective direct monitoring methods for blade fracture or tower sweep. Existing technologies mainly rely on abnormal vibration or strain signals to indirectly determine the blade condition, failing to combine multiple parameters such as blade torque, vibration, and clearance value for comprehensive analysis. This makes it difficult to accurately determine the blade condition and results in insufficient precision in judging whether a blade has fractured.

[0005] 2. Existing monitoring systems cannot simultaneously monitor blade breakage, blade sweeping on the tower, and blade vortex-induced vibration when the wind turbine is shut down. The monitoring function is relatively limited and cannot fully guarantee the safe operation of the wind turbine.

[0006] 3. The lack of a feature library for blade breakage or tower sweeping makes it impossible to establish effective early warnings through historical data analysis, resulting in imperfect early warning functions and difficulty in detecting potential risks in advance.

[0007] Therefore, there is an urgent need for a wind turbine blade monitoring method that can accurately monitor abnormal conditions such as wind turbine blade breakage and tower sweeping, and has an early warning function, in order to improve the safety and reliability of wind turbine generator sets. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a wind turbine blade monitoring method and system based on blade arrival time monitoring. By collecting blade torque, flapping vibration, swaying vibration, and clearance values, it overcomes the limitations of single-parameter monitoring. Furthermore, it sets safety thresholds for blade torque, flapping vibration, and swaying vibration values ​​for fracture risk and tower sweep risk, respectively, enabling early warning of blade fracture and tower sweep risks. Simultaneously, by monitoring the effective value of the clearance value, it further determines whether blade fracture or tower sweep failure has occurred. Through risk warning and fault determination, it achieves accurate judgment of blade fracture and tower sweep status, improving the operational safety of wind turbine units. Independent storage partitions are used for fracture and tower sweep data, ensuring orderly data management and providing data support for subsequent feature database construction based on historical data, optimizing the safety thresholds for early warning, and improving the accuracy of blade tower sweep and fracture risk warnings.

[0009] This invention is achieved through the following technical solution: The first aspect of this invention provides a wind turbine blade monitoring method based on blade arrival time monitoring, comprising the following steps: Step 1: After the wind turbine is connected to the grid, collect the torque values ​​M1, M2 and M3, flapping vibration values ​​V1, V2 and V3, swaying vibration values ​​W1, W2 and W3, and the corresponding blade clearance value J for the three blades. Step 2: When any of M1, M2, and M3 is 0 and the corresponding net clearance value J is a valid value, the blade is considered to be directly below the blade rotation plane, and the start time t1 is recorded. Step 3: Obtain the current wind turbine speed and calculate the angle A that the wind turbine has rotated through by integrating the current wind turbine speed over time. When the angle A satisfies one period T, record that moment as t2. Step 4: Simultaneously calculate the blade torque change rate, the vibration frequency deviation in the blade oscillation direction, and the vibration frequency deviation in the blade flapping direction at times t1-t2. If the rate of change of torque is greater than the safety setting value S2 and the deviation of the vibration frequency in the blade flapping direction is greater than the safety setting value S3 or the deviation of the vibration frequency in the blade flapping direction is greater than the safety setting value S4, then it is determined that there is a risk of blade breakage, and the wind turbine generator set is controlled to execute the pitch protection strategy in advance. If the torque change rate is less than the safety setting value S2 and the vibration frequency deviation in the blade flapping direction is greater than the safety setting value S5 or the vibration frequency deviation in the blade flapping direction is greater than the safety setting value S6, then it is determined that there is a risk of the blade sweeping the tower, and the wind turbine generator set is controlled to execute the pitch protection strategy in advance. Step 5: Check if the net clearance value J has a valid value within the set time range Δt of t2 ± Δt; If the clearance value J is valid, it is determined that the blade has not broken and whether the clearance value J under the corresponding blade is less than the safety value S1. If the clearance value J is less than the safety value S1, it is determined that the blade has swept the tower and the swaying vibration value during this period is stored in the fourth storage partition Y4 and the flapping vibration value is stored in the fifth storage partition Y5. If the clearance value J is not less than the safety value S1, it is determined that the blade has not swept the tower. If the clearance value J does not have a valid value, it is determined that the blade has broken. The wind turbine generator is controlled to stop quickly and the blade torque value from that moment to the previous two rotation cycles is stored in the first storage partition Y1, the swaying vibration value is stored in the second storage partition Y2, and the flapping vibration value is stored in the third storage partition Y3. Step 6: Update blade identifier B after each rotation cycle. i And repeat the above steps.

[0010] More preferably, the rotation cycle T is one rotation period of the wind turbine every 120°.

[0011] More preferably, when the data in Y1-Y3 reach the set number of samples, the torque change rate within a fixed period is calculated for the data in Y1 using a sliding window; for the data in Y2 and Y3, the vibration frequencies of the blade in the flapping direction and the oscillation direction are extracted respectively by fast Fourier transform, and the deviation of the vibration frequencies in the flapping direction and the oscillation direction is calculated, thereby establishing a feature library for blade breakage and completing the correction of safety settings S2, S3, and S4.

[0012] More preferably, when the data in Y4 and Y5 reach the set number of samples, the vibration frequencies of the blades in the flapping direction and the oscillation direction are extracted from the data in Y4 and Y5 by fast Fourier transform, and the deviation of the vibration frequencies in the flapping direction and the oscillation direction is calculated. In this way, a feature library of blades sweeping the tower is established, and the correction of the safety set values ​​S5 and S6 is completed.

[0013] More preferably, the torque change rate is as follows: Torque change rate = Torque change value / Fixed period.

[0014] More preferably, the deviation of the vibration frequency from its inherent characteristic is calculated as follows: Vibration frequency deviation = |Vibration frequency - Natural frequency| / Natural frequency.

[0015] More preferably, when the blades are continuously sweeping the tower, a rapid shutdown strategy is implemented.

[0016] More preferably, when the wind turbine is shut down, if any of V1, V2, or V3 exceeds the set protection value, the wind turbine will execute a yaw or pitch strategy.

[0017] More preferably, when the wind turbine is shut down, if any of W1, W2, or W3 exceeds the set protection value, the wind turbine will execute a yaw or pitch strategy.

[0018] The second aspect of this invention provides a monitoring system applicable to the wind turbine blade monitoring method based on blade arrival time monitoring described in the first aspect, characterized in that it includes a PLC controller, blade strain gauges, blade vibration sensors, and blade clearance radar; the blade strain gauges and blade vibration sensors are installed at the blade root, the blade clearance radar is installed inside the nacelle, the PLC controller is communicatively connected to the blade strain gauges, blade vibration sensors, and blade clearance radar respectively, and interacts with the main controller of the wind turbine generator set to obtain wind turbine rotation speed information and complete real-time data storage, analysis, and alarm, and establish a data feature database for blade anomalies. The beneficial effects of this invention are as follows: 1. This invention overcomes the limitations of single-parameter monitoring by collecting blade torque, flapping vibration, swaying vibration, and clearance values. Furthermore, it sets safety thresholds for blade torque, flapping vibration, and swaying vibration values ​​for fracture risk and tower sweep risk, respectively, to achieve early warning of blade fracture and tower sweep risk. At the same time, by monitoring the effective value of clearance value, it further determines whether blade fracture or tower sweep failure has occurred. Through risk warning and fault determination, it achieves the judgment of blade fracture and tower sweep status, thereby improving the safety of wind turbine generator operation.

[0019] 2. This invention divides the data into independent storage partitions according to the fracture and tower sweep, and stores the data of blade fracture and tower sweep in categories to ensure the orderliness of data management. This provides data support for the subsequent establishment of a feature library based on historical data, and for optimizing the safety setpoints for early warning, thereby improving the accuracy of early warning of blade tower sweep and fracture risks.

[0020] 3. In this invention, a rapid shutdown strategy is added for situations where blades continuously sweep the tower, to avoid cumulative damage to the blades and tower due to continuous collisions and to prevent the accident from escalating.

[0021] 4. In this invention, in response to the possibility that the flapping or swaying vibration values ​​may exceed the set protection value during shutdown, the wind turbine generator set reduces the damage to the blades caused by vortex-induced vibration by executing yaw or pitch strategies, thereby ensuring the equipment safety of the unit in the shutdown state. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0023] Example 1 like Figure 1 As shown, the first aspect of the present invention provides a wind turbine blade monitoring method based on blade arrival time monitoring, comprising the following steps: Step 1: After the wind turbine is connected to the grid, collect the torque values ​​M1, M2 and M3, flapping vibration values ​​V1, V2 and V3, swaying vibration values ​​W1, W2 and W3, and the corresponding blade clearance value J for the three blades. Step 2: When any of M1, M2, and M3 is 0 and the corresponding net clearance value J is a valid value, the blade is considered to be directly below the blade rotation plane, and the start time t1 is recorded. Step 3: Obtain the current wind turbine speed and calculate the angle A that the wind turbine has rotated through by integrating the current wind turbine speed over time. When the angle A satisfies one period T, record that moment as t2. Step 4: Simultaneously calculate the blade torque change rate, the vibration frequency deviation in the blade oscillation direction, and the vibration frequency deviation in the blade flapping direction at times t1-t2. If the rate of change of torque is greater than the safety setpoint S2 and the deviation of the vibration frequency in the blade flapping direction is greater than the safety setpoint S3 or the deviation of the vibration frequency in the blade flapping direction is greater than the safety setpoint S4, then it is determined that the blade is at risk of breakage, and the wind turbine generator set is controlled to execute the pitch protection strategy in advance to reduce power operation. If the torque change rate is less than the safety setting value S2 and the vibration frequency deviation in the blade flapping direction is greater than the safety setting value S5 or the vibration frequency deviation in the blade flapping direction is greater than the safety setting value S6, then it is determined that there is a risk of the blade sweeping the tower, and the wind turbine generator set is controlled to execute the pitch protection strategy in advance to reduce the power operation. Step 5: Check if the net clearance value J has a valid value within the set time range Δt of t2 ± Δt; The effective value of the clearance value J needs to be preset according to parameters such as the wind turbine model, blade length, and tower structure; If the clearance value J is valid, it is determined that the blade has not broken and whether the clearance value J under the corresponding blade is less than the safety value S1. If the clearance value J is less than the safety value S1, it is determined that the blade has swept the tower and the swaying vibration value during this period is stored in the fourth storage partition Y4 and the flapping vibration value is stored in the fifth storage partition Y5. If the clearance value J is not less than the safety value S1, it is determined that the blade has not swept the tower. If the clearance value J does not have a valid value, it is determined that the blade has broken. The wind turbine generator is controlled to stop quickly and the blade torque value from that moment to the previous two rotation cycles is stored in the first storage partition Y1, the swaying vibration value is stored in the second storage partition Y2, and the flapping vibration value is stored in the third storage partition Y3. Step 6: Update blade identifier B after each rotation cycle. i And repeat the above steps.

[0024] More preferably, each 120° rotation of the wind turbine constitutes one rotation cycle T. Defining each 120° rotation of the wind turbine as one cycle matches the evenly distributed structure of the three blades, ensuring that the complete status monitoring of a single blade can be completed independently within each cycle.

[0025] Furthermore, for each subsequent rotation cycle (i.e., when A is a multiple of 120), the blade indicator is incremented by 1, and when it exceeds 3 cycles, the indicator is reset to 1.

[0026] This invention calculates the blade passage period based on the wind turbine rotation speed, and simultaneously calculates the blade torque change rate, the vibration frequency deviation in the blade oscillation direction, and the vibration frequency deviation in the blade flapping direction at times t1-t2. By checking whether the blade torque change rate and vibration frequency deviation exceed the safety set value, it determines whether there is a risk of blade breakage or tower sweep, thus providing an early warning. If there is a risk of breakage or tower sweep, it controls the wind turbine to execute the pitch protection strategy in advance.

[0027] While providing early warning, the system monitors whether the net air value J has an effective value within the set time range Δt of t2±, thereby further determining whether the blade has broken or the tower has been swept. If the clearance value J is valid, it is determined that the blade has not broken, and it is determined whether the clearance value J under the corresponding blade is less than the safety value S1. If the clearance value J is less than the safety value S1, it is determined that the blade has swept the tower and the swaying vibration value during this period is stored in the fourth storage partition Y4 and the flapping vibration value is stored in the fifth storage partition Y5. If the clearance value J is not less than the safety value S1, it is determined that the blade has not swept the tower, and the wind turbine generator is controlled to operate normally. If the clearance value J does not have a valid value, it is determined that the blade has broken. The wind turbine generator is controlled to stop quickly and the blade torque value from that moment to the previous two rotation cycles is stored in the first storage partition Y1, the sway vibration value in the second storage partition Y2, and the flapping vibration value in the third storage partition Y3.

[0028] Blade fracture does not occur instantaneously; it typically involves crack initiation, propagation, and fracture. By storing the torque, flapping vibration, and oscillation vibration values ​​from the moment of blade fracture to the two preceding rotational cycles, data support can be provided for the subsequent establishment of a blade fracture feature library.

[0029] The data stored in the first storage partition Y1, the second storage partition Y2, the third storage partition Y3, the fourth storage partition Y4, and the fifth storage partition Y5 provide data support for the subsequent establishment of a feature library for blade breakage and tower sweeping. By establishing the feature library, the safety setting values ​​can be corrected, making the early warning of blade breakage and tower sweeping more accurate.

[0030] Furthermore, the data from blade breakage and tower sweeping are stored separately. When analyzing the data from blade breakage and tower sweeping later, the data in the corresponding partition can be quickly located and operated on, which is less likely to cause data confusion and facilitates the rapid extraction of the corresponding data later.

[0031] Example 2 This embodiment further elaborates and supplements the implementation of the present invention based on Embodiment 1.

[0032] When the data in Y1-Y3 reach the set number of samples, the torque change rate within a fixed period is calculated for the data in Y1 using a sliding window; for the data in Y2 and Y3, the vibration frequencies of the blade in the flapping direction and the oscillation direction are extracted respectively by fast Fourier transform, and the deviation of the vibration frequencies in the flapping direction and the oscillation direction is calculated. In this way, a feature library for blade fracture is established, and the correction of safety setpoints S2, S3, and S4 is completed.

[0033] Specifically, when the data in Y1-Y3 reaches the set number of samples, the data in Y1-Y3 is processed. For the data in Y1, the torque change rate within a fixed period is calculated using a sliding window. For the data in Y2 and Y3, the vibration frequencies of the blade in the flapping and swaying directions are extracted using Fast Fourier Transform. The natural frequencies are calculated during the blade design phase. Then, the deviation of the vibration frequencies in the flapping and swaying directions is calculated using the vibration frequencies and natural frequencies. Data samples that confirm blade fracture and have effective warnings are selected, and valid samples after misjudgment are excluded. The torque change rate, vibration frequency deviation in the flapping and swaying directions of the valid samples are sorted to form a feature threshold interval, i.e., the blade fracture feature library. The lower limit of the torque change rate in the blade fracture feature library is taken as the corrected S2, the lower limit of the vibration frequency deviation in the swaying direction is taken as the corrected S3, and the lower limit of the vibration frequency deviation in the flapping direction is taken as the corrected S4.

[0034] When the data in Y4 and Y5 reach the set number of samples, the vibration frequencies of the blades in the flapping and oscillating directions are extracted by fast Fourier transform. The deviation of the vibration frequencies in the flapping and oscillating directions from the natural frequencies is calculated to establish a feature library of blades during tower sweeping, and to complete the correction of safety setpoints S5 and S6.

[0035] Specifically, when the data in Y4 and Y5 reach the set number of samples, the data in Y4 and Y5 are processed. The vibration frequencies of the blades in the flapping direction and the oscillation direction are extracted from the data in Y4 and Y5 by Fast Fourier Transform. The natural frequency is the calculated value in the blade design stage. Then, the vibration frequency deviation in the flapping direction and the oscillation direction is calculated by using the vibration frequency and the natural frequency. Data samples that are confirmed to be sweeping the tower and have effective warnings are selected. Valid samples after false judgment are excluded. The vibration frequency deviation in the flapping direction and the oscillation direction in the valid samples are sorted to form a feature threshold interval, i.e., the blade sweeping tower feature library. The lower limit of the vibration frequency deviation of the blades in the oscillation direction in the blade sweeping tower feature library is taken as S5, and the lower limit of the vibration frequency deviation of the blades in the flapping direction in the blade sweeping tower feature library is taken as the corrected S6.

[0036] As the operating time of the wind turbine increases, when the data in Y1, Y2, Y3, Y4, and Y5 reach the set number of samples again, the safety setpoints are readjusted to make the safety setpoints more accurate and the early warning judgments more accurate.

[0037] More preferably, the torque change rate is as follows: Torque change rate = Torque change value / Fixed period.

[0038] The torque change value is the maximum torque value minus the minimum torque value within a fixed period, reflecting the fluctuation range of torque within a fixed period.

[0039] More preferably, the deviation of the vibration frequency from its inherent characteristic is calculated as follows: Vibration frequency deviation = |Vibration frequency - Natural frequency| / Natural frequency.

[0040] Example 3 This embodiment further elaborates and supplements the implementation of the present invention based on Embodiment 1 or Embodiment 2.

[0041] When blades continuously swipe across the tower, a rapid shutdown strategy is implemented. Continuous blade swiping means that the collision between the blades and the tower continues. Rapid shutdown can directly terminate the dangerous operating state, minimize the cumulative damage to the blades and tower, prevent the accident from escalating, and further improve the safety of wind turbine operation.

[0042] When the wind turbine generator set is shut down, if any of V1, V2, V3 or W1, W2, W3 exceeds the set protection value, the wind turbine generator set will execute a yaw or pitch strategy until these six values ​​are lower than the set protection value. This yaw or pitch protection strategy reduces the damage of vortex-induced vibration to the blade structure, extends the blade's service life, and ensures equipment safety during turbine shutdown.

[0043] Example 4 This embodiment further elaborates and supplements the implementation of the present invention based on Embodiment 1, Embodiment 2 or Embodiment 3.

[0044] The second aspect of this invention provides a system applicable to the wind turbine blade monitoring method based on blade arrival time monitoring described in the above embodiments, comprising a PLC controller, blade strain gauges, blade vibration sensors, and blade clearance radar; the blade strain gauges and blade vibration sensors are installed at the blade root and are used to measure the blade root torque value and the blade vibration value in the flapping and swaying directions in real time, respectively; the blade clearance radar is installed in the nacelle and is used to measure the horizontal distance between the blade rotation plane and the tower, i.e., the clearance value; the PLC controller is communicatively connected to the blade strain gauges, blade vibration sensors, and blade clearance radar, and interacts with the wind turbine main controller to obtain wind turbine speed information and complete real-time data storage, analysis, and alarm, and establish a data feature database for blade anomalies.

[0045] Specifically, after the wind turbine is connected to the grid, the blade strain gauges, blade vibration sensors and blade clearance radar of the corresponding blades monitor the torque value, flapping vibration value, oscillation vibration value and clearance value of the blades in real time, and upload them to the PLC controller for analysis. When any of the three blade torque values ​​M1, M2 and M3 is 0 and the corresponding clearance value J is a valid value, the blade is identified as being directly below the blade rotation plane, and the start time t1 is recorded. The PLC controller interacts with the main controller of the wind turbine to obtain the current wind turbine speed and calculates the angle A through which the wind turbine rotates by integrating the current wind turbine speed over time. When the angle A satisfies one period T, the moment is recorded as t2. At the time t1-t2, the blade torque change rate, the vibration frequency deviation in the blade swaying direction, and the vibration frequency deviation in the blade flapping direction are calculated simultaneously. When the torque change rate is greater than the safety setpoint S2 and the vibration frequency deviation in the blade flapping direction is greater than the safety setpoint S3 or the vibration frequency deviation in the blade flapping direction is greater than the safety setpoint S4, it is determined that there is a risk of blade breakage. The PLC controller sends an early warning signal to the wind turbine main controller. The wind turbine main controller receives the early warning signal and controls the wind turbine to execute the pitch protection strategy in advance. When the torque change rate is less than the safety setpoint S2 and the vibration frequency deviation in the blade flapping direction is greater than the safety setpoint S5 or the vibration frequency deviation in the blade flapping direction is greater than the safety setpoint S6, it is determined that there is a risk of the blade sweeping the tower. The PLC controller sends an early warning signal to the wind turbine main controller. The wind turbine main controller receives the early warning signal and controls the wind turbine to execute the pitch protection strategy in advance. The PLC controller obtains whether the clearance value J monitored by the blade clearance radar within the set time range Δt of t2±1000 has a valid value; If the clearance value J is valid, it is determined that the blade has not broken and whether the clearance value J under the corresponding blade is less than the safety value S1. If the clearance value J is less than the safety value S1, it is determined that the blade has swept the tower and the swaying vibration value during this period is stored in the fourth storage partition Y4 and the flapping vibration value is stored in the fifth storage partition Y5. If the clearance value J is not less than the safety value S1, it is determined that the blade has not swept the tower, and the PLC controller sends a signal to the wind turbine generator main controller to control the wind turbine generator to resume normal operation.

[0046] If the clearance value J does not have a valid value, it is determined that the blade has broken. The PLC controller sends a signal to the wind turbine generator set to control the wind turbine generator set to stop quickly and store the blade torque value from that moment to the previous two rotation cycles in the first storage partition Y1, the swaying vibration value in the second storage partition Y2, and the flapping vibration value in the third storage partition Y3.

[0047] In summary, this monitoring system overcomes the limitations of single-parameter monitoring by collecting blade torque, flapping vibration, swaying vibration, and clearance values. Furthermore, it sets safety thresholds for blade torque, flapping vibration, and swaying vibration values ​​separately for fracture risk and tower sweep risk, enabling early warning of blade fracture and tower sweep risks. Simultaneously, by monitoring the effective value of the clearance value, it further determines whether blade fracture or tower sweep failure has occurred. Through risk warnings and fault determination, it achieves accurate judgment of blade fracture and tower sweep status, improving the operational safety of wind turbine units. Independent storage partitions are used for fracture and tower sweep data, ensuring orderly data management and providing data support for establishing a feature library based on historical data to optimize the safety thresholds for early warning, thereby improving the accuracy of blade tower sweep and fracture risk warnings.

Claims

1. A wind turbine blade monitoring method based on blade arrival time monitoring, characterized in that: Includes the following steps: Step 1: After the wind turbine is connected to the grid, collect the torque values ​​M1, M2 and M3, flapping vibration values ​​V1, V2 and V3, swaying vibration values ​​W1, W2 and W3, and the corresponding blade clearance value J for the three blades. Step 2: When any of M1, M2, and M3 is 0 and the corresponding net clearance value J is a valid value, the blade is considered to be directly below the blade rotation plane, and the start time t1 is recorded. Step 3: Obtain the current wind turbine speed and calculate the angle A that the wind turbine has rotated through by integrating the current wind turbine speed over time. When the angle A satisfies one period T, record that moment as t2. Step 4: Simultaneously calculate the blade torque change rate, the vibration frequency deviation in the blade oscillation direction, and the vibration frequency deviation in the blade flapping direction at times t1-t2. If the rate of change of torque is greater than the safety setting value S2 and the deviation of the vibration frequency in the blade flapping direction is greater than the safety setting value S3 or the deviation of the vibration frequency in the blade flapping direction is greater than the safety setting value S4, then it is determined that there is a risk of blade breakage, and the wind turbine generator set is controlled to execute the pitch protection strategy in advance. If the torque change rate is less than the safety setting value S2 and the vibration frequency deviation in the blade flapping direction is greater than the safety setting value S5 or the vibration frequency deviation in the blade flapping direction is greater than the safety setting value S6, then it is determined that there is a risk of the blade sweeping the tower, and the wind turbine generator set is controlled to execute the pitch protection strategy in advance. Step 5: Check if the net clearance value J has a valid value within the set time range Δt of t2 ± Δt; If the clearance value J is valid, it is determined that the blade has not broken and whether the clearance value J under the corresponding blade is less than the safety value S1. If the clearance value J is less than the safety value S1, it is determined that the blade has swept the tower and the swaying vibration value during this period is stored in the fourth storage partition Y4 and the flapping vibration value is stored in the fifth storage partition Y5. If the clearance value J is not less than the safety value S1, it is determined that the blade has not swept the tower. If the clearance value J does not have a valid value, it is determined that the blade has broken. The wind turbine generator is controlled to stop quickly and the blade torque value from that moment to the previous two rotation cycles is stored in the first storage partition Y1, the swaying vibration value is stored in the second storage partition Y2, and the flapping vibration value is stored in the third storage partition Y3. Step 6: Update blade identifier B after each rotation cycle. i And repeat the above steps.

2. The wind turbine blade monitoring method based on blade arrival time monitoring as described in claim 1, characterized in that: The rotation cycle T is one rotation period when the wind turbine rotates 120°.

3. A wind turbine blade monitoring method based on blade arrival time monitoring as described in claim 1 or 2, characterized in that: When the data in Y1-Y3 reach the set number of samples, the torque change rate within a fixed period is calculated for the data in Y1 using a sliding window; for the data in Y2 and Y3, the vibration frequencies of the blade in the flapping direction and the oscillation direction are extracted respectively by fast Fourier transform, and the deviation of the vibration frequencies in the flapping direction and the oscillation direction is calculated. In this way, a feature library for blade fracture is established, and the correction of safety setpoints S2, S3, and S4 is completed.

4. The wind turbine blade monitoring method based on blade arrival time monitoring as described in claim 3, characterized in that: When the data in Y4 and Y5 reach the set number of samples, the vibration frequencies of the blades in the flapping and oscillating directions are extracted by fast Fourier transform, respectively. The deviation of the vibration frequencies in the flapping and oscillating directions is calculated to establish a feature library of blades sweeping the tower, and to complete the correction of the safety setpoints S5 and S6.

5. The wind turbine blade monitoring method based on blade arrival time monitoring as described in claim 3, characterized in that: The torque change rate is as follows: Torque change rate = Torque change value / Fixed period.

6. The wind turbine blade monitoring method based on blade arrival time monitoring as described in claim 3, characterized in that: The deviation of the vibration frequency from its natural frequency is calculated as follows: Vibration frequency deviation = |Vibration frequency - Natural frequency| / Natural frequency.

7. A wind turbine blade monitoring method based on blade arrival time monitoring as described in claim 3, characterized in that: When the blades continuously sweep the tower, a rapid shutdown strategy is executed.

8. The wind turbine blade monitoring method based on blade arrival time monitoring as described in claim 7, characterized in that: When the wind turbine is shut down, if any of V1, V2, or V3 exceeds the set protection value, the wind turbine will execute a yaw or pitch strategy.

9. A wind turbine blade monitoring method based on blade arrival time monitoring as described in claim 8, characterized in that: When the wind turbine generator set is shut down, if any of W1, W2, or W3 exceeds the set protection value, the wind turbine generator set will execute a yaw or pitch strategy.

10. A monitoring system applicable to the wind turbine blade monitoring method based on blade arrival time monitoring as described in any one of claims 1-9, characterized in that: It includes a PLC controller, blade strain gauges, blade vibration sensors, and blade clearance radar. The blade strain gauges and blade vibration sensors are installed at the blade root, and the blade clearance radar is installed in the nacelle. The PLC controller is communicatively connected to the blade strain gauges, blade vibration sensors, and blade clearance radar, and interacts with the main controller of the wind turbine to obtain the wind turbine speed information and complete the real-time storage, analysis, and alarm of the data, and establish a data feature database for blade anomalies.

Images