Wind driven generator control method based on vibration signals

Abstract

The invention relates to the technical field of wind driven generator control, and discloses a wind driven generator control method based on a vibration signal, which comprises the following steps: setting a monitoring point based on the type of a wind driven generator component, and acquiring monitoring point data of a current monitoring time node; constructing a wind driven generator monitoring model based on the historical data; generating running state information of the current wind driven generator in combination with the monitoring point data and the wind driven generator monitoring model; generating a wind driven generator control instruction based on the current operation state information of the wind driven generator; the monitoring point data comprises vibration data and environment data. Based on the constructed wind driven generator monitoring model, factors such as correction of various operation states and weights thereof, vibration data and environment data are comprehensively considered; the small change of the running state of the generator can be more sensitively captured, so that the control instruction can respond in time, and the power generation efficiency reduction caused by deviation of the running state from the optimal point is avoided.

Description

Technical Field

[0001] The invention relates to the technical field of wind turbine generator control, and in particular to a wind turbine generator control method based on vibration signals. Background Art

[0002] As the global demand for clean energy continues to increase, wind power generation has developed rapidly as an important renewable energy generation method. The capacity and scale of wind turbines are increasing, and the number and scale of wind farms are also expanding. In this case, how to improve the operating efficiency of wind turbines, reduce the failure rate, and improve reliability and safety have become important issues facing the wind power industry.

[0003] The wind turbine control method based on vibration signals can monitor the operating status of the wind turbine in real time, discover potential faults in a timely manner, give early warnings and take corresponding control measures, thereby improving the operating efficiency and reliability of the wind turbine, reducing operation and maintenance costs, and meeting the development needs of the wind power industry. Summary of the invention

[0004] The purpose of the present invention is to make the state adjustment of the wind turbine generator during operation more stable through the operation state monitoring based on vibration signals and the control instruction generation mechanism based on the operation state information.

[0005] In order to achieve the above object, the present invention provides a wind turbine generator control method based on vibration signals, comprising: Set monitoring points based on the type of wind turbine components and obtain monitoring point data at the current monitoring time node; Building a wind turbine monitoring model based on historical data; Generate the current wind turbine operating status information by combining the monitoring point data and the wind turbine monitoring model; Generate wind turbine generator control instructions based on current wind turbine generator operation status information; The monitoring point data includes: vibration data and environmental data.

[0006] In some embodiments of the present invention, the construction of the wind turbine monitoring model includes: Generate multiple types of operating states based on historical operating state data of wind turbines, set weights for each operating state, and generate an operating state sample set based on the weight of each operating state; Combine vibration data and environmental data to generate a correction model; A wind turbine monitoring model is constructed based on the revised model and the operating status sample set.

[0007] In some embodiments of the present invention, the step of setting a weight for each operating state includes: Based on the historical operating status data of wind turbines, the operating status of wind turbines is classified to generate multiple types of operating status sets Q, where Q={q1,q2…q i …q n}; Among them, q i represents the i-th operating status category, and n represents the total number of operating status categories; Get the i-th running status category q i The vibration data set E from the occurrence to the end time, E={e1,e2…e j …e w}; Among them, 1 represents the occurrence period of the current operating status category, e j represents the vibration data of the jth period of the current operating status category, and w represents the end period of the current operating status category; Get the i-th running status category q i The state correction period j1 after the occurrence of the state correction is segmented into the vibration data set E based on the state correction period j1 to generate the first vibration data set E1, E1={e1,e2…e j1} and the second rotation data set E2, E2={e j1 …e w}; The weights are set for the first vibration data set E1 in order, and the weights are set for the second vibration data set E2 in reverse order; Generate the first vibration data weight set F1, F1={f1,f2…f j1} and the second vibration data weight set F2, F2{f j1 …f w}; Combine the first vibration data weight set F1 and the second vibration data weight set F2 to generate a vibration data weight set F{f1, f2…f j …f w}; f j is the vibration data weight of the jth period.

[0008] In some embodiments of the present invention, the step of generating the operating state sample set by combining the weight of each operating state includes: Generate an operating state sample by comprehensively considering the current operating state of the wind turbine, the vibration data set corresponding to the current operating state, and the weight set corresponding to the vibration data set; All operating status samples of the wind turbine are acquired to generate an operating status sample set.

[0009] In some embodiments of the present invention, the generating of the correction model includes: Generate an environmental data reference value set R based on the corresponding environmental data obtained under the current operating state, R = {R1, R2…R k …R m}; Among them, R k represents the reference value of the kth environmental data, and m represents the total number of environmental data; Obtaining in sequence the first correction value of the vibration data weight of each type of environmental data reference value; Calculate the environmental type correlation between the reference values ​​of each type of environmental data; generating a second correction value of the vibration data weight by combining the environment type association and the first correction value of the vibration data weight; A correction model is generated by combining the second correction value of the vibration data weight and the acquired historical vibration data.

[0010] In some embodiments of the present invention, the step of obtaining the first correction value d1 of the reference value of each type of environmental data to the vibration data weight includes: Combine the kth environmental data reference value and the vibration data weight set F to obtain the i-th operating state category q of the current wind turbine i The corresponding vibration data weight deviation value Fa(R k ,q i ); Fa(R k ,q i )= ; Among them, z1 is the first fixed coefficient, The weight of the operating state of the i-th wind turbine generator corresponding to the k-th environmental data reference value; Obtaining the vibration data weight deviation values ​​of all operating status categories to generate a first correction value d1; d1= ; Wherein, z2 is the second fixed coefficient.

[0011] In some embodiments of the present invention, the calculation of the environmental type association between each type of environmental data reference value includes: Environmental data reference value R i and R j (i≠j and i,j=1…V) conduct correlation analysis; Get environmental data reference value R i The sample values ​​{x1,x2…x V}; And based on the environmental data reference value R i The sample values ​​{x1,x2…x V}Get the environmental data reference value Ri Sample mean E(x); Get environmental data reference value R j The sample values ​​{y1,y2…y V}; And based on the environmental data reference value R j The sample values ​​{y1,y2…y V}Get the sample mean E(y); Calculate the environmental data reference value R i and R j Pearson correlation coefficient r ij ; Based on the overall Pearson correlation coefficient r ij The composition correlation matrix C = {r ij}(i,j=1,⋯,V).

[0012] In some embodiments of the present invention, the step of combining the environment type association and the first correction value of the vibration data weight to generate the second correction value of the vibration data weight includes: d2=d1 ; Among them, w ij is the weight coefficient of the second correction value, and =1.

[0013] In some embodiments of the present invention, the generating of the current operating status information of the wind turbine includes: Generate a vibration data reference value and an environmental data reference value based on the acquired monitoring point data of the current wind turbine; Generate the current generator operating state and operating state weight fa by combining the vibration data reference value and the operating state sample set; Acquire a second correction value d2 of the current wind turbine by combining the environmental data reference value and the correction model; Generate a corrected operating state weight fb based on the current operating state weight fa of the generator and the second correction value d2; fb=fa-h*d2; Wherein, h is the fixed coefficient of the second correction value.

[0014] In some embodiments of the present invention, the generating of wind turbine control instructions; The current operation status information of the wind turbine generator includes the current operation status type of the wind turbine generator and the corrected operation status weight fb; Combine historical data to set multiple preset values ​​for the operating status weight corresponding to each operating status type; and setting corresponding wind turbine generator control instructions based on each preset value; The corresponding wind turbine generator control instruction is generated based on the current corrected operating state weight fb.

[0015] Compared with the prior art, the wind turbine generator control method based on vibration signals provided by the embodiment of the present invention has the following beneficial effects: By setting monitoring points based on the types of wind turbine components, vibration data and environmental data related to the operating status of the generator can be obtained comprehensively and accurately; this helps to accurately identify different operating status types and their weights, so that control instructions can be more in line with actual operating needs.

[0016] Combined with historical data, multiple preset values ​​are set for the operating state weight corresponding to each operating state type, and control instructions are generated based on the corrected operating state weight fb; based on the long-term operating experience of wind turbines, the most suitable control strategy can be formulated for different operating conditions.

[0017] Setting segmented weights for the vibration data set helps to detect potential faults earlier. When a component begins to have a minor fault, changes in its vibration data will be reflected in the vibration data set E. Since different weights are set for vibration data in different time periods, this change can be captured more clearly, thereby issuing fault warnings in advance, reducing downtime caused by faults, and indirectly improving power generation efficiency.

[0018] Based on accurate operating status information and fault warnings, more targeted maintenance plans can be formulated, and key maintenance can be performed on components that may fail based on specific operating status types, weights, and environmental factors.

[0019] By fully considering environmental data, such as wind speed, temperature, humidity and other environmental factors, and correcting the weight of vibration data, wind turbines can better adapt to environmental changes; whether under complex and changeable meteorological conditions or in different geographical environments, wind turbines can adjust their operating status in time according to environmental changes and maintain stable operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of a wind turbine generator control method based on vibration signals provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0022] In the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0023] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0024] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0025] Embodiment 1: An embodiment of the present invention provides a wind turbine generator control method based on a vibration signal, such as Figure 1 As shown, including: Set monitoring points based on the type of wind turbine components and obtain monitoring point data at the current monitoring time node; Building a wind turbine monitoring model based on historical data; Generate the current wind turbine operating status information by combining the monitoring point data and the wind turbine monitoring model; Generate wind turbine generator control instructions based on current wind turbine generator operation status information; The monitoring point data includes: vibration data and environmental data.

[0026] In this embodiment, for the blade part of the wind turbine: According to the length, material and structural characteristics of the blade, monitoring points are set at key locations such as the root, middle and tip of the blade. This is because these locations are subjected to different forces during operation, with the root bearing greater torsional forces, the middle bearing bending moments, and the tip being more sensitive to vibration.

[0027] Taking into account the possible fatigue damage of the blades, a small number of monitoring points can also be set on the leading and trailing edges of the blades to promptly detect damage caused by airflow scouring and other reasons.

[0028] For the tower part: Monitoring points are set at the bottom, middle and top of the tower. The bottom monitoring point can detect the stability of the tower foundation, the middle monitoring point helps to detect the bending vibration of the tower as a whole, and the top monitoring point is more sensitive to the vibration transmitted by the wind rotor.

[0029] Multiple monitoring points (such as 4 to 8) are evenly arranged along the circumference of the tower to comprehensively monitor the vibration of the tower in different directions.

[0030] For the cabin part: Monitoring points are set up near key equipment such as gearboxes and generators inside the engine room. Gearbox monitoring points can be set on the input shaft, output shaft and the surface of the box to detect gear meshing vibration, bearing vibration, etc.

[0031] Generator monitoring points are set near the stator and rotor to monitor electromagnetic vibration and other conditions.

[0032] Use high-precision accelerometers to obtain vibration data. For blade monitoring points, the sensor range should be selected based on the expected vibration amplitude of the blade under different operating conditions, with an accuracy of at least ±0.1m / s².

[0033] At the tower monitoring point, the sensor should have good low-frequency response characteristics to accurately measure the low-frequency vibration of the tower, and the sampling frequency should be set above 100 Hz to ensure that possible resonance and other phenomena can be captured.

[0034] The vibration sensor in the cabin should have the ability to resist electromagnetic interference, and the sampling frequency is determined based on factors such as the rotational speed of the equipment. For example, for a generator with a rotational speed of 1500r / min, the sampling frequency can be set to 500Hz.

[0035] Meteorological stations are set up around wind turbines to obtain environmental data such as wind speed, wind direction, temperature, humidity, etc. Wind speed and wind direction sensors should have high precision, with wind speed measurement accuracy reaching ±0.5m / s and wind direction measurement accuracy reaching ±5°.

[0036] The measurement range of the temperature sensor should cover the local extreme temperatures, and the accuracy of the humidity sensor should be at least ±3%RH.

[0037] Embodiment 2: The construction of the wind turbine monitoring model includes: Generate multiple types of operating states based on historical operating state data of wind turbines, set weights for each operating state, and generate an operating state sample set based on the weight of each operating state; Combine vibration data and environmental data to generate a correction model; A wind turbine monitoring model is constructed based on the revised model and the operating status sample set.

[0038] Embodiment 3: The step of setting a weight for each operating state includes: Based on the historical operating status data of wind turbines, the operating status of wind turbines is classified to generate multiple types of operating status sets Q, where Q={q1,q2…q i …q n}; Among them, q i represents the i-th operating status category, and n represents the total number of operating status categories; Get the i-th running status category q i The vibration data set E from the occurrence to the end time, E={e1,e2…e j …e w}; Among them, 1 represents the occurrence period of the current operating status category, e j represents the vibration data of the jth period of the current operating status category, and w represents the end period of the current operating status category; Get the i-th running status category q i The state correction period j1 after the occurrence of the state correction is segmented into the vibration data set E based on the state correction period j1 to generate the first vibration data set E1, E1={e1,e2…e j1} and the second rotation data set E2, E2={e j1 …e w}; The weights are set for the first vibration data set E1 in order, and the weights are set for the second vibration data set E2 in reverse order; Generate the first vibration data weight set F1, F1={f1,f2…f j1} and the second vibration data weight set F2, F2{f j1 …f w}; Combine the first vibration data weight set F1 and the second vibration data weight set F2 to generate a vibration data weight set F{f1, f2…f j …f w}; f j is the vibration data weight of the jth period.

[0039] Embodiment 4: The step of generating the operating state sample set by combining the weight of each operating state includes: Generate an operating state sample by comprehensively considering the current operating state of the wind turbine, a vibration data set corresponding to the current operating state, and a weight set corresponding to the vibration data set; All operating status samples of the wind turbine are acquired to generate an operating status sample set.

[0040] In this embodiment, in addition to the conventional classification method based on historical operating status data, the operating status in different seasons and different time periods (such as day and night, because factors such as wind conditions and temperature may be different) can also be considered for classification. For example, in winter, due to the low temperature, the thermal expansion and contraction characteristics of the components may lead to different operating states than in summer.

[0041] The classification of fault conditions can be further subdivided, such as dividing blade failures into different types of operating conditions such as surface damage and internal structure damage.

[0042] When setting weights for the vibration data set E segments, for the first vibration data set E1, a linearly increasing weight setting method can be used, such as f1 = 0.1, f2 = 0.2, and so on, where the weight increment is determined according to the actual operating conditions and data statistical results.

[0043] For the second vibration data set E2, a linearly decreasing weight setting method is adopted, such as f w = 0.1, f w - 1 = 0.2 etc.

[0044] When determining the weight of each operating state, a risk assessment factor can be introduced. For example, if a certain operating state is associated with a high risk of failure, its weight can be appropriately increased.

[0045] Embodiment 5: The generating of the correction model comprises: Generate an environmental data reference value set R based on the corresponding environmental data obtained under the current operating state, R = {R1, R2…R k …R m}; Among them, R k represents the reference value of the kth environmental data, and m represents the total number of environmental data; Obtaining in sequence the first correction value of the vibration data weight of each type of environmental data reference value; Calculate the environmental type correlation between the reference values ​​of each type of environmental data; generating a second correction value of the vibration data weight by combining the environment type association and the first correction value of the vibration data weight; A correction model is generated by combining the second correction value of the vibration data weight and the acquired historical vibration data.

[0046] Embodiment 6: The step of obtaining the first correction value d1 of the reference value of each type of environmental data to the vibration data weight includes: Combine the kth environmental data reference value and the vibration data weight set F to obtain the i-th operating state category q of the current wind turbine i The corresponding vibration data weight deviation value Fa(R k ,q i ); Fa(R k ,q i )= ; Among them, z1 is the first fixed coefficient, The weight of the operating state of the i-th wind turbine generator corresponding to the k-th environmental data reference value; Obtaining the vibration data weight deviation values ​​of all operating status categories to generate a first correction value d1; d1= (R k ,q i ); Wherein, z2 is the second fixed coefficient.

[0047] Embodiment 7: The calculation of the environmental type association between the reference values ​​of each type of environmental data includes: Environmental data reference value R i and R j (i≠j and i,j=1…V) conduct correlation analysis; Get environmental data reference value R i The sample values ​​{x1,x2…x V}; And based on the environmental data reference value R i The sample values ​​{x1,x2…x V}Get the environmental data reference value R i Sample mean E(x); Get environmental data reference value R j The sample values ​​{y1,y2…y V}; And based on the environmental data reference value R j The sample values ​​{y1,y2…y V}Get the sample mean E(y); Calculate the environmental data reference value R i and R j Pearson correlation coefficient rij ; Based on the overall Pearson correlation coefficient r ij The composition correlation matrix C = {r ij}(i,j=1,⋯,V).

[0048] Embodiment 8: The step of combining the environment type association and the first correction value of the vibration data weight to generate the second correction value of the vibration data weight includes: d2=d1 ; Among them, w ij is the weight coefficient of the second correction value, and =1.

[0049] In this embodiment, the construction of the environmental data reference value set R is: For wind speed data, in addition to obtaining the current wind speed value, you can also calculate the average wind speed within a certain period of time (such as 10 minutes), the wind speed fluctuation range, etc. as reference values.

[0050] The temperature data can calculate the temperature change rate as a reference value, and the humidity data can consider the humidity change range, etc.

[0051] First correction value calculation: Calculate the vibration data weight deviation value Fa(R k ,q i ), the method for determining the first fixed coefficient z1 is further refined. For example, through a large amount of experimental data and simulation analysis, the value of z1 is obtained by fitting according to the operating data of different wind turbine models and under different environmental conditions.

[0052] When calculating the first correction value d1, in addition to the simple weighted summation method, it is possible to consider weighted averaging the vibration data weight deviation values ​​according to the importance of different operating states.

[0053] Environment type association calculation: In calculating the Pearson correlation coefficient r ij When the data is processed, you can add data preprocessing steps, such as standardizing the sample values ​​to make their mean 0 and standard deviation 1, to improve the accuracy of the calculation results.

[0054] For the correlation matrix C, in addition to the Pearson correlation coefficient, other correlation measurement methods (such as the Spearman rank correlation coefficient) can also be considered for comparative analysis to ensure the reliability of the correlation calculation.

[0055] The second correction value is generated: In determining the weight coefficient w of the second correction value ijWhen the wind speed is large, the wind speed-related weight coefficient can be appropriately increased.

[0056] Embodiment 9: The generating of the current operating status information of the wind turbine generator includes: Generate a vibration data reference value and an environmental data reference value based on the acquired monitoring point data of the current wind turbine; Generate the current generator operating state and operating state weight fa by combining the vibration data reference value and the operating state sample set; Acquire a second correction value d2 of the current wind turbine by combining the environmental data reference value and the correction model; Generate a corrected operating state weight fb based on the current operating state weight fa of the generator and the second correction value d2; fb=fa-h*d2; Wherein, h is the fixed coefficient of the second correction value.

[0057] In this embodiment, the vibration data reference value is generated: The acquired vibration data is filtered to remove high-frequency noise, and then a reference value is generated using statistical methods (such as mean, median, standard deviation, etc.). For example, the mean value of vibration data within a certain time window (such as 5 minutes) can be calculated as a reference value.

[0058] Vibration data reference value models are established according to different operating modes of wind turbines (such as startup, normal operation, shutdown, etc.).

[0059] Environmental data reference value generation: For wind speed data, in addition to the current wind speed value, a wind speed prediction model can also be established based on historical wind speed data, and the predicted value can be used as part of the reference value.

[0060] The temperature and humidity reference values ​​can be adjusted based on local meteorological historical data and seasonal variation patterns. For example, when seasons change, the range of reference values ​​can be appropriately relaxed.

[0061] When calculating the operating state weight fa, a time decay factor may be introduced. For example, the closer the operating state sample is to the current time, the greater its contribution to fa is. An exponential decay function is used to implement this time weighting.

[0062] When calculating the corrected operating state weight fb, the value range of the fixed coefficient h of the second correction value is further studied. According to different wind turbine models, operating environments and other factors, the optimal value range of h is determined through experiments and data analysis.

[0063] Embodiment 10: generating a wind turbine generator control instruction; The current operation status information of the wind turbine generator includes the current operation status type of the wind turbine generator and the corrected operation status weight fb; Combine historical data to set multiple preset values ​​for the operating status weight corresponding to each operating status type; and setting corresponding wind turbine generator control instructions based on each preset value; The corresponding wind turbine generator control instruction is generated based on the current corrected operating state weight fb.

[0064] In this embodiment, different preset values ​​are set for the operating state weight corresponding to each operating state type according to different load conditions of the wind turbine (such as light load, full load, overload, etc.). For example, in the case of full load, the preset value of the operating state weight related to power generation efficiency is set higher.

[0065] Taking into account different grid access requirements, such as requirements for power quality (frequency, voltage, etc.), the preset values ​​of the operating state weights are adjusted. For example, in areas where the grid has higher requirements for frequency stability, the preset values ​​of the operating state weights related to frequency regulation are increased accordingly.

[0066] According to the different operating state weight fb values, the control instructions are refined. For example, when the fb value is in a certain range, the pitch angle of the blades is adjusted to optimize the wind energy capture efficiency; when the fb value is in another range, the excitation current of the generator is adjusted to stabilize the output voltage.

[0067] A feedback mechanism for control instructions is established. After executing the control instructions, the control instructions are adjusted in time according to the changes in the monitoring point data to achieve more accurate wind turbine control.

[0068] Finally, it should be noted that it is obvious that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technology, the present invention is also intended to include these modifications and variations.

[0069] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention.

Claims

1. A wind turbine generator control method based on vibration signals, characterized in that: include: Set monitoring points based on the type of wind turbine components and obtain monitoring point data at the current monitoring time node; Building a wind turbine monitoring model based on historical data; Generate the current wind turbine operating status information by combining the monitoring point data and the wind turbine monitoring model; Generate wind turbine generator control instructions based on current wind turbine generator operation status information; The monitoring point data includes: vibration data and environmental data.

2. The wind turbine generator control method based on vibration signal according to claim 1, characterized in that: The construction of the wind turbine monitoring model includes: Generate multiple types of operating states based on historical operating state data of wind turbines, set weights for each operating state, and generate an operating state sample set based on the weight of each operating state; Combine vibration data and environmental data to generate a correction model; A wind turbine monitoring model is constructed based on the revised model and the operating status sample set.

3. The wind turbine generator control method based on vibration signal according to claim 2, characterized in that: The step of setting a weight for each operating state includes: Based on the historical operating status data of wind turbines, the operating status of wind turbines is classified to generate multiple types of operating status sets Q, where Q={q1,q2…q i …q n }; Among them, q i represents the i-th operating status category, and n represents the total number of operating status categories; Get the i-th running status category q i The vibration data set E from the occurrence to the end time, E={e1,e2…e j …e w }; Among them, 1 represents the occurrence period of the current operating status category, e j represents the vibration data of the jth period of the current operating status category, and w represents the end period of the current operating status category; Get the i-th running status category q i The state correction period j1 after the occurrence of the state correction is segmented into the vibration data set E based on the state correction period j1 to generate the first vibration data set E1, E1={e1,e2…e j1 } and the second rotation data set E2, E2={e j1 …e w }; The weights are set for the first vibration data set E1 in order, and the weights are set for the second vibration data set E2 in reverse order; Generate the first vibration data weight set F1, F1={f1,f2…f j1 } and the second vibration data weight set F2, F2{f j1 …f w }; Combine the first vibration data weight set F1 and the second vibration data weight set F2 to generate a vibration data weight set F{f1, f2…f j …f w }; f j is the vibration data weight of the jth period.

4. The wind turbine generator control method based on vibration signal according to claim 3, characterized in that: The step of generating the operating state sample set by combining the weight of each operating state includes: Generate an operating state sample by comprehensively considering the current operating state of the wind turbine, the vibration data set corresponding to the current operating state, and the weight set corresponding to the vibration data set; All operating status samples of the wind turbine are acquired to generate an operating status sample set.

5. The wind turbine generator control method based on vibration signal according to claim 4, characterized in that: The generating of the correction model comprises: Generate an environmental data reference value set R based on the corresponding environmental data obtained under the current operating state, R = {R1, R2…R k …R m }; Among them, R k represents the reference value of the kth environmental data, and m represents the total number of environmental data; Obtaining in sequence the first correction value of the vibration data weight of each type of environmental data reference value; Calculate the environmental type correlation between the reference values ​​of each type of environmental data; generating a second correction value of the vibration data weight by combining the environment type association and the first correction value of the vibration data weight; A correction model is generated by combining the second correction value of the vibration data weight and the acquired historical vibration data.

6. The wind turbine generator control method based on vibration signals according to claim 5, characterized in that: The step of obtaining the first correction value d1 of the weight of the vibration data for the reference value of each type of environmental data includes: Combine the kth environmental data reference value and the vibration data weight set F to obtain the i-th operating state category q of the current wind turbine i The corresponding vibration data weight deviation value Fa(R k ,q i ); Fa(R k ,q i )= ; Among them, z1 is the first fixed coefficient, The weight of the operating state of the i-th wind turbine generator corresponding to the k-th environmental data reference value; Obtaining the vibration data weight deviation values ​​of all operating status categories to generate a first correction value d1; d1= (R k ,q i ); Wherein, z2 is the second fixed coefficient.

7. The wind turbine generator control method based on vibration signals according to claim 6, characterized in that: The calculation of the environmental type association between the reference values ​​of each type of environmental data includes: Environmental data reference value R i and R j (i≠j and i,j=1…V) conduct correlation analysis; Get environmental data reference value R i The sample values ​​{x1,x2…x V }; And based on the environmental data reference value R i The sample values ​​{x1,x2…x V }Get the environmental data reference value R i Sample mean E(x); Get environmental data reference value R j The sample values ​​{y1,y2…y V }; And based on the environmental data reference value R j The sample values ​​{y1,y2…y V }Get the sample mean E(y); Calculate the environmental data reference value R i and R j Pearson correlation coefficient r ij ; Based on the overall Pearson correlation coefficient r ij The composition correlation matrix C = {r ij }(i,j=1,⋯,V).

8. The wind turbine generator control method based on vibration signals according to claim 7, characterized in that: The step of combining the environment type association and the first correction value of the vibration data weight to generate the second correction value of the vibration data weight includes: d2=d1 , Among them, w ij is the weight coefficient of the second correction value, and =1.

9. The wind turbine generator control method based on vibration signals according to claim 8, characterized in that: The generating of the current wind turbine operating status information includes: Generate a vibration data reference value and an environmental data reference value based on the acquired monitoring point data of the current wind turbine; Generate the current generator operating state and operating state weight fa by combining the vibration data reference value and the operating state sample set; Acquire a second correction value d2 of the current wind turbine by combining the environmental data reference value and the correction model; Generate a corrected operating state weight fb based on the current operating state weight fa of the generator and the second correction value d2; fb=fa-h*d2; Wherein, h is the fixed coefficient of the second correction value.

10. The wind turbine generator control method based on vibration signals according to claim 9, characterized in that: generating a wind turbine generator control instruction; The current operation state information of the wind turbine generator includes the current operation state type of the wind turbine generator and the corrected operation state weight fb; Combine historical data to set multiple preset values ​​for the operating status weight corresponding to each operating status type; and setting corresponding wind turbine generator control instructions based on each preset value; The corresponding wind turbine generator control instruction is generated based on the current corrected operating state weight fb.

Images