Wind turbine blade state acquisition system and method based on optical fiber sensor

Abstract

The application discloses a wind turbine blade state acquisition system and method based on an optical fiber sensor, and belongs to the technical field of wind turbine state monitoring. The system acquires real-time strain signals of a wind turbine blade by using an optical fiber sensor through a monitoring module, converts the real-time strain signals into time-frequency characteristics through a data analysis module, and comprehensively analyzes the time-frequency characteristics in combination with environmental information and unit operation state information. When a blade fault occurs, a fault classification module is used to determine the fault type of the blade. The system can improve the real-time monitoring capability of the working state of the blade, is suitable for a wind turbine system with large-size blades, can more comprehensively and objectively understand the health state of the blade, determines the working mode of the blade and each target component, judges the health state of the blade based on the detection standards corresponding to each target component, realizes rapid evaluation of the health state of the blade, enables an operator to rapidly master the fault condition of the blade, takes timely measures, and reduces downtime and maintenance costs.

Description

Technical Field

[0001] This application relates to the field of wind turbine condition monitoring technology, and in particular to a wind turbine blade condition acquisition system based on fiber optic sensors and a wind turbine blade condition acquisition method based on fiber optic sensors. Background Technology

[0002] With the increasing demand for energy, the promotion of environmentally friendly and renewable new energy sources is urgently needed. Compared with traditional energy sources, wind energy, as a green new energy source, has the advantages of large energy storage, no pollution, and sustainability, greatly meeting human needs. Wind power generation mainly relies on tower-type wind turbine units, and the performance of the blade structure is directly related to the safety, stability, and power generation efficiency of wind power generation. As the single-unit capacity of wind turbines continues to increase, the centrifugal force, bending moment, and aerodynamic forces they bear also increase accordingly. Due to the harsh working environment of wind turbine blades, high durability requirements, and their huge size, they are also at risk of corrosion and damage. Therefore, it is necessary to effectively monitor the health status of the blades to prevent serious fatigue damage to the wind turbine's operational safety.

[0003] Currently, wind turbine blade fault diagnosis often employs non-destructive testing technologies such as infrared thermography and ultrasonic testing. These technologies can characterize the internal quality of the blade without damaging its structure or reducing its performance, allowing for the development of appropriate fault diagnosis methods based on the specific structural characteristics and material properties of the blade. However, large blades place certain requirements on the output power and operating distance of infrared lasers, and factors such as ambient temperature, humidity, and background radiation can affect the accuracy of infrared thermography, leading to inaccurate interpretation of the thermal images. Ultrasonic waves gradually attenuate as they propagate through materials; therefore, for large blades, the penetration depth of ultrasonic waves may be limited, making it difficult to completely cover all parts of the blade, especially posing a challenge for detecting defects in complex structures or internal components.

[0004] Therefore, it is necessary to propose a wind turbine blade status acquisition system based on fiber optic sensors, which can solve the technical problems of existing technologies being greatly affected by the environment when monitoring the status of wind turbine blades, not being suitable for large-size wind turbines, and being unable to comprehensively assess the health status of wind turbine blades by integrating real-time monitoring signals and external environmental information. Summary of the Invention

[0005] In view of this, the present invention provides a wind turbine blade status acquisition system and method based on fiber optic sensors to solve the technical problems of large errors in existing wind turbine blade monitoring methods, insufficient utilization of monitoring information, and inability to provide effective technical support for actual production.

[0006] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a wind turbine blade status acquisition system based on fiber optic sensors, including a monitoring module, a parameter acquisition module, a data analysis module, and a fault classification module; wherein the monitoring module and the parameter acquisition module are both connected to the data analysis module, and the data analysis module is also connected to the fault classification module; The monitoring module is used to acquire real-time strain signals of wind turbine blades using fiber optic sensors. The parameter acquisition module is used to acquire environmental information and unit operating status information in real time; The data analysis module is used to perform frequency domain conversion on real-time strain signals to obtain the corresponding time-frequency characteristics, and determine the blade operating mode based on the time-frequency characteristics, environmental information and unit operating status information. Based on the blade operating mode, multiple target components of the time-frequency characteristics are determined, and the health status of the blade is judged based on the detection standard of each target component under the current blade operating mode. The fault classification module is used to input the time-frequency characteristics, environmental information and unit operating status information into a preset fault classification model when a blade fault is determined, to determine the fault type of the blade, and to upload the fault type.

[0007] Furthermore, the data analysis module includes a frequency domain conversion unit, a working condition confirmation unit, a target component confirmation unit, and a status judgment unit; The frequency domain conversion unit is used to perform wavelet transform on the real-time strain signal to obtain the spectrum of the strain signal at various scales. The operating condition confirmation unit is used to determine the frequency variation values ​​of the windward and leeward sides of the blade over time based on the spectrum diagram, and to determine the blade operating mode based on the frequency variation values, environmental information and unit operating status information. The target component confirmation unit is used to determine multiple target components of the strain signal based on the blade operating mode and the natural frequency of the wind turbine. The state judgment unit is used to determine the standard threshold range and influence weight of each target component under the current environmental information and unit operating status based on a preset knowledge graph, and to determine whether the blade has failed based on the standard threshold range and influence weight.

[0008] Furthermore, the real-time environmental information acquired includes wind speed, temperature, and humidity information; the unit operating status information includes power generation and rotational speed.

[0009] Furthermore, the operating condition confirmation unit determines the blade operating mode based on frequency change values, environmental information, and unit operating status information, including: When the amplitude changes of the windward and leeward sides of the blades are the same and opposite in direction on each frequency component, the unit has power output and the wind speed is greater than the preset start-up threshold, the blade working mode is determined to be the start-up operation state. When the amplitude changes of each frequency component on both the windward and leeward sides of the blades exhibit positive and negative periodic variations and the wind speed is less than the preset start-up threshold, the blade operating mode is determined to be in shutdown state.

[0010] Furthermore, the target component confirmation unit determines multiple target components of the strain signal based on the blade operating mode and the natural frequency of the wind turbine blade, including: The phase offset and amplitude adjustment coefficient are determined based on the current blade operating mode of the wind turbine, and the harmonic frequency and resonant frequency are determined based on the natural frequency of the blade. Multiple target analysis frequency bands are determined based on phase offset, natural frequency, harmonic frequency and resonant frequency. The amplitude of each frequency band is adjusted according to the amplitude adjustment coefficient. The absolute value, change value and rate of change of the amplitude in the target analysis frequency band are used as target components.

[0011] Furthermore, the status determination unit includes a graph database unit and a risk analysis unit; The knowledge graph database unit is used to store a pre-constructed knowledge graph containing blade structure, target frequency components, environmental information, unit operating status information, and detection standards; wherein, the knowledge graph includes the frequency band range and influence weight of the target frequency components under different environmental information, blade structures, and unit operating states; The risk analysis unit is used to calculate the fault risk value based on the frequency band range and influence weight of the target component, and to determine whether the blade has failed based on the risk value and the preset risk warning value.

[0012] Furthermore, the fiber optic sensor is a stress grating sensor, which is laid along the root of the blade at preset intervals towards the tip.

[0013] Furthermore, the preset fault classification model is constructed based on the support vector machine model, using time-frequency characteristics, environmental information and unit operating status information as input variables, and blade fault type as output variable, and using one-hot encoding to represent fault categories.

[0014] Furthermore, the monitoring module also includes a temperature compensation sensor, which is placed around the fiber optic sensor to eliminate the influence of temperature changes on the fiber optic sensor.

[0015] On the other hand, the present invention also provides a method for acquiring the state of wind turbine blades based on fiber optic sensors, comprising: The monitoring module uses fiber optic sensors to acquire real-time strain signals of the wind turbine blades. The parameter acquisition module obtains environmental information and unit operating status information in real time. The data analysis module performs frequency domain conversion on the real-time strain signal to obtain the corresponding time-frequency characteristics. Based on the time-frequency characteristics, environmental information and unit operating status information, the blade operating mode is determined. Based on the blade operating mode, multiple target components of the time-frequency characteristics are determined. Based on the detection standard of each target component in the current blade operating mode, the health status of the blade is judged. When a blade failure is determined by the fault classification module, the time-frequency characteristics, environmental information, and unit operating status information are input into a preset fault classification model to determine the fault type of the blade and upload the fault type.

[0016] This invention provides a wind turbine blade status acquisition system based on fiber optic sensors. The system uses a monitoring module to collect real-time strain signals from the blades via fiber optic sensors, facilitating the monitoring of blade status characteristics during operation and improving the real-time monitoring capability of wind turbine blade operating status. The system offers stable monitoring with minimal environmental influence, making it particularly suitable for wind turbine systems with large blades. A data analysis module performs frequency domain conversion on the real-time strain signals to obtain time-frequency characteristics. Combined with environmental information and turbine operating status information, a comprehensive analysis is performed, providing a more complete and objective understanding of the blade's operating status and enabling more timely and accurate detection of potential blade failures. By determining the blade's operating mode and target components, and based on the detection standards for each target component, the system judges the blade's health status, achieving rapid assessment of blade health and providing early warning of possible failures. When a blade failure occurs, a fault classification module determines the fault type and promptly uploads the fault information, helping operators quickly grasp the blade fault situation and take timely measures to reduce downtime and maintenance costs. Attached Figure Description

[0017] Figure 1 A schematic diagram of an embodiment of the wind turbine blade status acquisition system based on fiber optic sensors provided by the present invention; Figure 2 This is a schematic diagram of the structure of an embodiment of the data analysis module provided by the present invention; Figure 3 This is a schematic diagram of the structure of an embodiment of the state determination module provided by the present invention; Figure 4 This is a flowchart illustrating an embodiment of the wind turbine blade status acquisition method based on fiber optic sensors provided by the present invention. Detailed Implementation

[0018] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0019] This invention provides a wind turbine blade status acquisition system and method based on fiber optic sensors, which will be described below.

[0020] Combination Figure 1 As shown, a specific embodiment of the present invention discloses a wind turbine blade status acquisition system 100 based on fiber optic sensors, including a monitoring module 101, a parameter acquisition module 102, a data analysis module 103, and a fault classification module 104. The monitoring module 101 and the parameter acquisition module 102 are both connected to the data analysis module 103, and the data analysis module 103 is also connected to the fault classification module 104. The monitoring module 101 is used to acquire real-time strain signals of wind turbine blades using fiber optic sensors. Parameter acquisition module 102 is used to acquire environmental information and unit operating status information in real time; The data analysis module 103 is used to perform frequency domain conversion on the real-time strain signal to obtain the corresponding time-frequency characteristics, and determine the blade working mode based on the time-frequency characteristics, environmental information and unit operating status information. Based on the blade working mode, it determines multiple target components of the time-frequency characteristics, and judges the health status of the blade based on the detection standard of each target component in the current blade working mode. The fault classification module 104 is used to input the time-frequency characteristics, environmental information and unit operating status information into a preset fault classification model when it is determined that a blade has failed, to determine the fault type of the blade, and to upload the fault type.

[0021] Compared to existing technologies, the system in this embodiment uses fiber optic sensors in the monitoring module to collect real-time strain signals from the blades. This helps monitor the blade's state characteristics during operation, improving the real-time monitoring capability of the wind turbine blades' operating status. The monitoring status is stable and less affected by the environment, making it particularly suitable for wind turbine systems with large-sized blades. The data analysis module performs frequency domain conversion on the real-time strain signals to obtain time-frequency characteristics. Combined with environmental information and unit operating status information, a comprehensive analysis is performed, enabling a more comprehensive and objective understanding of the blade's operating status, thus allowing for more timely and accurate detection of potential blade failures. By determining the blade's operating mode and each target component, and judging the blade's health status based on the detection standards of each target component, rapid assessment of blade health status is achieved, providing early warning of possible failures. When a blade failure occurs, the fault classification module determines the type of blade failure and promptly uploads the fault information, helping operators quickly grasp the blade failure situation and take appropriate measures in a timely manner, reducing downtime and maintenance costs.

[0022] In a preferred embodiment, the fiber optic sensor is a stress grating sensor, which is laid along the root of the blade at preset intervals towards the tip.

[0023] Stress grating sensors can achieve highly sensitive detection of minute strains, capturing subtle changes in blade structure and helping to promptly identify potential problems. Simultaneously, they possess strong anti-interference capabilities, enabling stable and reliable operation in complex environments, ensuring the accuracy and reliability of monitoring. Furthermore, they can achieve distributed monitoring, making them suitable for real-time monitoring of large blades.

[0024] In practical applications, it has been found that blade failures often occur at the blade root. Therefore, the focus of monitoring is on the blade root, with a more densely spaced pre-set spacing at the root, gradually decreasing in density along the blade tip direction. A crisscross arrangement of horizontal and vertical forces is used on each individual blade to monitor both axial and lateral forces acting on it.

[0025] As a preferred embodiment, such as Figure 2 As shown, the data analysis module 103 includes a frequency domain conversion unit 301, a working condition confirmation unit 302, a target component confirmation unit 303, and a status judgment unit 304; The frequency domain conversion unit 301 is used to perform wavelet transform on the real-time strain signal to obtain the spectrum of the strain signal at various scales. The operating condition confirmation unit 302 is used to determine the frequency change values ​​of the windward and leeward sides of the blade over time based on the spectrum diagram, and to determine the blade operating mode based on the frequency change values, environmental information and unit operating status information. The target component confirmation unit 303 is used to determine multiple target components of the strain signal based on the blade operating mode and the natural frequency of the wind turbine. The state judgment unit 304 is used to determine the standard threshold range and influence weight of each target component under the current environmental information and unit operating state based on a preset knowledge graph, and to judge whether the blade has failed based on the standard threshold range and influence weight.

[0026] Specifically, the frequency domain conversion unit obtains the spectrum of the blade vibration signal through wavelet transform, displays the change of amplitude of different frequency components over time, and determines characteristic information such as the windward and leeward sides by the peak values ​​of different frequency components on the spectrum.

[0027] In a preferred embodiment, the real-time environmental information acquired includes wind speed, temperature, and humidity information; the unit operating status information includes power generation and rotational speed.

[0028] In a preferred embodiment, the operating condition confirmation unit determines the blade operating mode based on frequency change values, environmental information, and unit operating status information, including: When the amplitude changes of the windward and leeward sides of the blades are the same and opposite in direction on each frequency component, the unit has power output and the wind speed is greater than the preset start-up threshold, the blade working mode is determined to be the start-up operation state. When the amplitude changes of each frequency component on both the windward and leeward sides of the blades exhibit positive and negative periodic variations and the wind speed is less than the preset start-up threshold, the blade operating mode is determined to be in shutdown state.

[0029] It should be noted that when the wind turbine is in a stopped state, the blades are only subjected to gravity load, and the strain value is a fixed value; when the wind turbine starts to rotate but is not generating electricity, the main load on the blades is gravity load, and the strain value changes periodically due to the rotation of the blades; when the wind turbine is turned on to generate electricity, the blades are subjected to the combined action of air load and gravity load.

[0030] In a preferred embodiment, the target component confirmation unit determines multiple target components of the strain signal based on the blade operating mode and the natural frequency of the wind turbine blade, including: The phase offset and amplitude adjustment coefficient are determined based on the current blade operating mode of the wind turbine, and the harmonic frequency and resonant frequency are determined based on the natural frequency of the blade. Multiple target analysis frequency bands are determined based on phase offset, natural frequency, harmonic frequency and resonant frequency. The amplitude of each frequency band is adjusted according to the amplitude adjustment coefficient. The absolute value, change value and rate of change of the amplitude in the target analysis frequency band are used as target components.

[0031] In one specific embodiment, the target analysis frequency band also includes specific frequencies corresponding to certain faults, such as frequency components corresponding to blade damage, cracks, and other faults. In some cases, nonlinear effects may lead to the emergence of new frequency components, which may also become target components. Resonance frequency points tend to increase vibration amplitude and are one of the key targets that need to be focused on. Under different operating conditions, the weight of each target component is different. For example, when the blade is in the on-state operation, it is necessary to ensure indicators such as power generation, so the influence weight of each analysis component at the resonance frequency point needs to be increased. However, when the blade is off but rotating, the blade balance needs to be considered, so the influence weight of the parameters related to the torque frequency point needs to be increased. By analyzing the blade operating mode and the blade's natural frequency, adaptive adjustments can be made according to the actual situation, which is beneficial for real-time monitoring and analysis of blade vibration.

[0032] As a preferred embodiment, such as Figure 3 As shown, the state determination unit 304 includes a map database unit 311 and a risk analysis unit 312; The graph database unit 311 is used to store a pre-constructed knowledge graph containing blade structure, target frequency components, environmental information, unit operating status information, and detection standards; wherein, the knowledge graph includes the frequency band range and influence weight of the target frequency components under different environmental information, blade structures, and unit operating states. The risk analysis unit 312 is used to calculate the fault risk value based on the frequency band range and influence weight of the target component, and to determine whether the blade has failed based on the risk value and the preset risk warning value.

[0033] By establishing a knowledge graph, various information is stored and managed in a structured manner, improving data query efficiency and enabling rapid and efficient risk assessment and early warning. Once a fault occurs, an early warning is issued first, followed by analysis of the specific fault type through a fault classification module, thus improving the response speed to abnormal situations.

[0034] In one specific embodiment, the risk value is calculated as follows: Risk Value = Σ (Actual Frequency Point Analysis Components) Frequency offset influence coefficient Target frequency band weighting factor The actual frequency point analysis component has multiple components. The influence values ​​of these analysis components are weighted to obtain the overall fault risk value.

[0035] By analyzing the components of actual frequency values, the deviation between actual and target frequency values, and considering frequency band range and influence weights, the failure risk of blade vibration can be assessed more accurately, reducing the possibility of false alarms and missed alarms. Comparing the risk value with preset risk warning values ​​enables intelligent early warning functions, promptly notifying relevant personnel to take appropriate maintenance and repair measures, and reducing losses caused by failures.

[0036] The blades used in wind turbine generators are typically made of composite materials such as glass fiber reinforced plastic or carbon fiber reinforced plastic. These materials possess high strength and lightweight characteristics, enabling them to withstand the dynamic loads of wind turbine operation. They also exhibit corrosion resistance and good weather resistance, making them suitable for long-term operation in outdoor environments. Blade damage may include fatigue cracks in stress concentration areas caused by wind loads and inertial forces, as well as surface skin detachment, cracking and deformation, and pinholes. In severe cases, it can even lead to blade breakage. It is necessary to differentiate between different types of failures to facilitate subsequent assessment of blade life.

[0037] In a preferred embodiment, the preset fault classification model is constructed based on the support vector machine model, using time-frequency characteristics, environmental information and unit operating status information as input variables, blade fault type as output variable, and one-hot encoding to represent fault categories.

[0038] By establishing a pre-defined fault classification model using a support vector machine (SVM) model, multi-dimensional feature space can be effectively processed, comprehensively considering data from various aspects such as time-frequency characteristics, environmental information, and unit operating status, to fully analyze the probability of blade faults. Since SVM models typically have high computational speed, they can achieve real-time monitoring and diagnosis of blade fault types in practical applications, facilitating timely maintenance measures by operators. Furthermore, fault categories are represented and transmitted using one-hot encoding, which significantly reduces the amount of data transmitted compared to transmitting raw data. Because one-hot encoding is a discrete encoding method, it is less susceptible to interference or errors during transmission, further improving transmission efficiency and data processing speed.

[0039] In a preferred embodiment, the monitoring module further includes a temperature compensation sensor, which is arranged around the fiber optic sensor to eliminate the influence of temperature changes on the fiber optic sensor.

[0040] Because the working environment of wind turbine blades is harsh, there are often temperature changes such as icing, rapid heating and cooling. By deploying temperature compensators around the fiber optic sensor, the influence of temperature changes on the sensor measurement data can be eliminated, thereby improving the accuracy of system monitoring.

[0041] This embodiment also provides a method for acquiring the state of wind turbine blades based on fiber optic sensors. Figure 4A flowchart illustrating the method is provided, including: Step S101: The real-time strain signal of the wind turbine blades is acquired by the monitoring module using fiber optic sensors; Step S102: Obtain environmental information and unit operating status information in real time through the parameter acquisition module; Step S103: The real-time strain signal is frequency-domain converted by the data analysis module to obtain the corresponding time-frequency characteristics. The blade working mode is determined according to the time-frequency characteristics, environmental information and unit operating status information. Multiple target components of the time-frequency characteristics are determined according to the blade working mode. The health status of the blade is judged based on the detection standard of each target component in the current blade working mode. Step S104: When a blade failure is determined by the fault classification module, the time-frequency characteristics, environmental information and unit operating status information are input into the preset fault classification model to determine the blade failure type and upload the failure type.

[0042] This invention provides a wind turbine blade status acquisition system based on fiber optic sensors. The system uses a monitoring module to collect real-time strain signals from the blades via fiber optic sensors, facilitating the monitoring of blade status characteristics during operation and improving the real-time monitoring capability of wind turbine blade operating status. The system offers stable monitoring with minimal environmental influence, making it particularly suitable for wind turbine systems with large blades. A data analysis module performs frequency domain conversion on the real-time strain signals to obtain time-frequency characteristics. Combined with environmental information and turbine operating status information, a comprehensive analysis is performed, providing a more complete and objective understanding of the blade's operating status and enabling more timely and accurate detection of potential blade failures. By determining the blade's operating mode and target components, and based on the detection standards for each target component, the system judges the blade's health status, achieving rapid assessment of blade health and providing early warning of possible failures. When a blade failure occurs, a fault classification module determines the fault type and promptly uploads the fault information, helping operators quickly grasp the blade fault situation and take timely measures to reduce downtime and maintenance costs.

[0043] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A wind turbine blade status acquisition system based on fiber optic sensors, characterized in that, It includes a monitoring module, a parameter acquisition module, a data analysis module, and a fault classification module; among which, the monitoring module and the parameter acquisition module are both connected to the data analysis module, and the data analysis module is also connected to the fault classification module; The monitoring module is used to acquire real-time strain signals of wind turbine blades using fiber optic sensors. The parameter acquisition module is used to acquire environmental information and unit operating status information in real time; The data analysis module is used to perform frequency domain conversion on real-time strain signals to obtain the corresponding time-frequency characteristics, and determine the blade operating mode based on the time-frequency characteristics, environmental information and unit operating status information. Based on the blade operating mode, multiple target components of the time-frequency characteristics are determined, and the health status of the blade is judged based on the detection standard of each target component under the current blade operating mode. The fault classification module is used to input the time-frequency characteristics, environmental information and unit operating status information into a preset fault classification model when a blade fault is determined, to determine the fault type of the blade, and to upload the fault type.

2. The wind turbine blade status acquisition system based on fiber optic sensors according to claim 1, characterized in that, The data analysis module includes a frequency domain conversion unit, a working condition confirmation unit, a target component confirmation unit, and a status judgment unit. The frequency domain conversion unit is used to perform wavelet transform on the real-time strain signal to obtain the spectrum of the strain signal at various scales. The operating condition confirmation unit is used to determine the frequency variation values ​​of the windward and leeward sides of the blade over time based on the spectrum diagram, and to determine the blade operating mode based on the frequency variation values, environmental information and unit operating status information. The target component confirmation unit is used to determine multiple target components of the strain signal based on the blade operating mode and the natural frequency of the wind turbine. The state judgment unit is used to determine the standard threshold range and influence weight of each target component under the current environmental information and unit operating status based on a preset knowledge graph, and to determine whether the blade has failed based on the standard threshold range and influence weight.

3. The wind turbine blade status acquisition system based on fiber optic sensors according to claim 2, characterized in that, The real-time environmental information acquired includes wind speed, temperature, and humidity information; the unit operating status information includes power generation and rotational speed.

4. The wind turbine blade status acquisition system based on fiber optic sensors according to claim 3, characterized in that, The operating condition confirmation unit determines the blade operating mode based on frequency change values, environmental information, and unit operating status information, including: When the amplitude changes of the windward and leeward sides of the blades are the same and opposite in direction on each frequency component, the unit has power output and the wind speed is greater than the preset start-up threshold, the blade working mode is determined to be the start-up operation state. When the amplitude changes of each frequency component on both the windward and leeward sides of the blades exhibit positive and negative periodic variations and the wind speed is less than the preset start-up threshold, the blade operating mode is determined to be in shutdown state.

5. The wind turbine blade status acquisition system based on fiber optic sensors according to claim 2, characterized in that, The target component confirmation unit determines multiple target components of the strain signal based on the wind turbine blade operating mode and the blade's natural frequency, including: The phase offset and amplitude adjustment coefficient are determined based on the current blade operating mode of the wind turbine, and the harmonic frequency and resonant frequency are determined based on the natural frequency of the blade. Multiple target analysis frequency bands are determined based on phase offset, natural frequency, harmonic frequency and resonant frequency. The amplitude of each frequency band is adjusted according to the amplitude adjustment coefficient. The absolute value, change value and rate of change of the amplitude in the target analysis frequency band are used as target components.

6. The wind turbine blade status acquisition system based on fiber optic sensors according to claim 2, characterized in that, The status determination unit includes a graph database unit and a risk analysis unit; The knowledge graph database unit is used to store a pre-constructed knowledge graph containing blade structure, target frequency components, environmental information, unit operating status information, and detection standards; wherein, the knowledge graph includes the frequency band range and influence weight of the target frequency components under different environmental information, blade structures, and unit operating states; The risk analysis unit is used to calculate the fault risk value based on the frequency band range and influence weight of the target component, and to determine whether the blade has failed based on the risk value and the preset risk warning value.

7. The wind turbine blade status acquisition system based on fiber optic sensors according to claim 1, characterized in that, The fiber optic sensor is a stress grating sensor, which is laid along the root of the blade at preset intervals towards the tip.

8. The wind turbine blade status acquisition system based on fiber optic sensors according to claim 1, characterized in that, The preset fault classification model is built based on the support vector machine model. It uses time-frequency characteristics, environmental information and unit operating status information as input variables, blade fault type as output variable, and uses one-hot encoding to represent fault categories.

9. The wind turbine blade status acquisition system based on fiber optic sensors according to claim 1, characterized in that, The monitoring module also includes a temperature compensation sensor, which is placed around the fiber optic sensor to eliminate the effect of temperature changes on the fiber optic sensor.

10. A method for acquiring the state of wind turbine blades based on fiber optic sensors, characterized in that, include: The monitoring module uses fiber optic sensors to acquire real-time strain signals of the wind turbine blades. The parameter acquisition module obtains environmental information and unit operating status information in real time. The data analysis module performs frequency domain conversion on the real-time strain signal to obtain the corresponding time-frequency characteristics. Based on the time-frequency characteristics, environmental information and unit operating status information, the blade operating mode is determined. Based on the blade operating mode, multiple target components of the time-frequency characteristics are determined. Based on the detection standard of each target component in the current blade operating mode, the health status of the blade is judged. When a blade failure is determined by the fault classification module, the time-frequency characteristics, environmental information, and unit operating status information are input into a preset fault classification model to determine the fault type of the blade and upload the fault type.

Images