A method, device and system for monitoring damage of fan blades

Through time-frequency conversion and synergistic index analysis, the problem of signal complexity and confusion in fan blade damage monitoring is solved, more accurate damage monitoring is achieved, and the effectiveness of fan blade damage detection is improved.

CN118836126BActive Publication Date: 2024-12-03国投检测科技(山东)有限公司 +1
View PDF 11 Cites 0 Cited by

Patent Information

Application Number
CN202411319546.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-23
Publication Date
2024-12-03
Estimated Expiration
2044-09-23

AI Technical Summary

Technical Problem

Fan blades are susceptible to damage in complex natural environments, and existing monitoring technologies based on vibration signals are difficult to accurately identify damage signals, resulting in reduced monitoring effectiveness and accuracy.

Method used

The time-frequency conversion algorithm is used to process the vibration signals of the fan blades, calculate the time-frequency entropy change and coordination coefficient, build the time-frequency coordination index, and monitor the fan blade damage with the abnormal data detection algorithm.

Benefits of technology

Through time-frequency characteristic analysis and coordination index, the accuracy and effectiveness of fan blade damage monitoring are improved, the blade damage can be better reflected and the error detection rate can be reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118836126B_ABST
    Figure CN118836126B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of fan blade damage monitoring, and specifically to a fan blade damage monitoring method, device and system, the method comprising: collecting vibration signals of each fan blade in each cycle; obtaining the time-frequency entropy matrix of each fan blade in each cycle; obtaining the time-frequency entropy change of each fan blade at each frequency in each cycle; obtaining the time-frequency entropy mutation index of each fan blade in each cycle; obtaining the coordination coefficient between each fan blade and other fan blades in each cycle; obtaining the time-frequency coordination index of each fan blade in each cycle; obtaining the fan blade damage sensitivity of each fan blade in each cycle; and using an abnormal data detection algorithm combined with the fan blade damage sensitivity to monitor the fan blade damage. The present application improves the effectiveness and accuracy of fan blade damage monitoring.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of wind turbine blade damage monitoring, and in particular to a wind turbine blade damage monitoring method, device and system. Background Art

[0002] As the global demand for renewable energy continues to grow, wind power generation, as a clean and renewable energy form, has been widely developed and applied. As the core equipment of wind power generation, the stability and reliability of wind turbines directly affect the power generation efficiency and economic benefits of the entire wind farm. However, wind turbine blades, as key components that capture wind energy and convert it into mechanical energy, work in a complex and changeable natural environment for a long time, and are subject to the influence of many harsh factors such as strong wind loads, sand scouring, atmospheric oxidation, lightning strikes, and humid air corrosion, and are very prone to damage.

[0003] Damage to fan blades will not only reduce the mechanical properties of the blades and affect their wind-collecting efficiency, but in severe cases may even cause the blades to break, which in turn causes the unit to shut down, resulting in significant economic losses and safety hazards. Damage monitoring technology based on vibration signals has been widely used in the field of fan blade damage monitoring. However, in actual applications, the mechanical noise and environmental noise generated during the operation of the fan are relatively complex and easily confused with the damage signal, making it difficult to accurately identify the operating status of the fan blades, reducing the effectiveness and accuracy of fan blade damage monitoring. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a method, device and system for monitoring wind turbine blade damage. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for monitoring damage of a wind turbine blade, the method comprising the following steps:

[0006] Collect the vibration signal of each fan blade in each cycle;

[0007] The vibration signal is processed using a time-frequency conversion algorithm to obtain the time-frequency matrix and time-frequency entropy matrix of each cycle of each fan blade; based on the column distribution of the differences between the elements at the same position in the time-frequency entropy matrix of each fan blade in adjacent cycles, the time-frequency entropy change of each fan blade at each frequency in each cycle is obtained;

[0008] Based on the size and uncertainty of the change range of the time-frequency entropy change of all frequencies of each fan blade in each cycle, the time-frequency entropy mutation index of each fan blade in each cycle is obtained;

[0009] Based on the similarity between the time-frequency matrix of each fan blade in each period and the time-frequency matrix of other fan blades, the coordination coefficient between each fan blade and other fan blades in each period is obtained; based on the value of the coordination coefficient and the time-frequency entropy mutation index, the time-frequency coordination index of each fan blade in each period is obtained;

[0010] The damage sensitivity of each fan blade in each cycle is obtained based on the time-frequency synergy index; the damage of the fan blades is monitored using the abnormal data detection algorithm combined with the damage sensitivity of the fan blades.

[0011] Furthermore, the method for obtaining the time-frequency entropy matrix is:

[0012] The vibration signal of each fan blade in each cycle is converted into the frequency domain by using the time-frequency conversion algorithm to obtain the time-frequency matrix of each fan blade in each cycle.

[0013] For the time-frequency matrix of each fan blade in each period, a sliding window of a preset size is slid on the time-frequency matrix. After each sliding, the time-frequency entropy of all elements in the sliding window is calculated as the time-frequency entropy of each element in the time-frequency matrix. The time-frequency entropy matrix of each fan blade in each period is constructed according to the position of each element in the time-frequency matrix.

[0014] Furthermore, the method for obtaining the time-frequency entropy change is:

[0015] For the time-frequency entropy matrix of each fan blade in each cycle, the absolute value of the difference between the elements at the same position in the time-frequency entropy matrix and the time-frequency entropy matrix of the previous adjacent cycle is calculated;

[0016] For all elements in the time-frequency entropy matrix corresponding to the same frequencies in the time-frequency matrix of each fan blade in each cycle, the sum of the absolute values ​​of all the differences of the elements is calculated as the change in time-frequency entropy for each frequency of each fan blade in each cycle.

[0017] Furthermore, the formula of the time-frequency entropy mutation index is: ; In the formula, Indicates The fan blades are The time-frequency entropy mutation index of the vibration signal of a cycle; Indicates The vibration signal of the fan blade is The range of all time-frequency entropy changes in the period is Indicates The vibration signal of the fan blade is The information entropy of all time-frequency entropy changes in a period; It is the preset parameter factor.

[0018] Furthermore, the method for obtaining the synergy coefficient is:

[0019] For the time-frequency matrix of each fan blade in each period, the similarity between the time-frequency matrix of the fan blade and the time-frequency matrices of other fan blades is calculated, and the sum of the similarities between each fan blade in each period and all other fan blades is taken as the coordination coefficient between each fan blade in each period and other fan blades.

[0020] Furthermore, the formula of the time-frequency synergy index is: ; In the formula, Indicates The fan blades are The time-frequency coherence index of the cycle; Indicates The fan blades are The coordination coefficient between the vibration signal of a cycle and other fan blades; Indicates The fan blades are The time-frequency entropy mutation index of the vibration signal of a cycle; K is the total number of fan blades.

[0021] Furthermore, the formula for the damage sensitivity of the fan blade is: , where Indicates The fan blades are The time-frequency coherence index of the cycle; Indicates The fan blades are The damage sensitivity of the fan blades in a cycle, e is a natural constant.

[0022] Further, the use of an abnormal monitoring algorithm in combination with the damage sensitivity of the wind turbine blades to monitor the damage of the wind turbine blades includes:

[0023] For each fan blade, the fan blade damage sensitivity of each preset number of cycles is divided into each set;

[0024] An abnormal data detection algorithm is used to detect each set. If abnormal data appears in the set, the fan blade is damaged; otherwise, the fan blade is not damaged.

[0025] In a second aspect, an embodiment of the present application provides a wind blade damage monitoring device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned wind blade damage monitoring methods when executing the computer program.

[0026] In a third aspect, an embodiment of the present application provides a wind turbine blade damage monitoring system, the system comprising:

[0027] A signal acquisition module is used to collect the vibration signal of each fan blade in each cycle;

[0028] A fan blade damage sensitivity acquisition module is used to process the vibration signal using a time-frequency conversion algorithm to obtain the time-frequency matrix and time-frequency entropy matrix of each cycle of each fan blade; based on the column distribution of the differences between the elements at the same position in the time-frequency entropy matrix of each fan blade in adjacent cycles, obtain the time-frequency entropy change of each frequency of each fan blade in each cycle; based on the size and uncertainty of the change range of the time-frequency entropy change of all frequencies of each fan blade in each cycle, obtain the time-frequency entropy mutation index of each fan blade in each cycle; based on the similarity between the time-frequency matrix of each fan blade in each cycle and the time-frequency matrix of each other fan blade, obtain the coordination coefficient between each fan blade and other fan blades in each cycle; based on the size of the coordination coefficient and the value of the time-frequency entropy mutation index, obtain the time-frequency coordination index of each fan blade in each cycle; based on the time-frequency coordination index, obtain the fan blade damage sensitivity of each fan blade in each cycle;

[0029] The fan blade damage monitoring module is used to monitor the fan blade damage using an abnormal data detection algorithm combined with the fan blade damage sensitivity.

[0030] This application has at least the following beneficial effects:

[0031] This application collects vibration signals generated during the operation of fan blades, analyzes the time-frequency characteristics of vibration signals in normal and damaged states of the fan, and constructs a time-frequency entropy mutation index based on the changes in adjacent periodic vibration signals in the time-frequency domain. The time-frequency entropy mutation index takes into account that when blades are damaged, the time-frequency entropy mutation index can better reflect the mutation of the vibration signals of fan blades in two adjacent periods; secondly, a time-frequency synergy index is constructed based on the synergy coefficient between different fan blades, and the damage sensitivity of fan blades is calculated based on the time-frequency synergy index. The beneficial effect is that the time-frequency synergy index takes into account the synergy coefficient between different fan blades, and the damage sensitivity of fan blades calculated based on the time-frequency synergy index can more accurately reflect the damage of fan blades, thereby improving the effectiveness and accuracy of fan blade damage monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 A flowchart of a method for monitoring wind turbine blade damage provided by an embodiment of the present application;

[0034] Figure 2 A schematic diagram of filling edge points of a time-frequency matrix provided by an embodiment of the present application;

[0035] Figure 3 A schematic diagram of vibration signal energy provided by an embodiment of the present application;

[0036] Figure 4 A schematic diagram of anomaly scores provided for one embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, describes in detail a method, device and system for monitoring damage of wind turbine blades proposed in the present application, its specific implementation, structure, features and effects. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0038] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0039] The following is a detailed description of a specific scheme of a wind turbine blade damage monitoring method, device and system provided by the present application in conjunction with the accompanying drawings.

[0040] See also Figure 1 , which shows a flow chart of a method for monitoring wind turbine blade damage provided by an embodiment of the present application, the method comprising the following steps:

[0041] Step S001, obtaining vibration signal data of the fan blade during operation and performing preprocessing.

[0042] This application is aimed at detecting blade damage of wind turbines. During the operation of wind turbines, the healthy state of wind turbine blades is a key link to ensure the efficient operation of wind turbines. In order to accurately assess the overall damage of wind turbine blades, this application detects the damage of three wind turbine blades at a distance from the root of the blade. Arrange the vibration sensor at the position of the total length of the blade. Make sure the sensor is in good contact with the blade surface to obtain accurate vibration signals.

[0043] In order to ensure that the collected vibration signal is closer to the real signal, this application takes into account the Nyquist sampling theorem while taking into account the needs of practical applications. The sampling frequency of the vibration signal is set to 48kHz. For the vibration signal of each period collected by each sensor, the median filter is used to filter the acquired data to reduce the impact of environmental noise on the original signal. Median filtering is a well-known technology, and its specific implementation process will not be repeated here.

[0044] Furthermore, the time taken for each blade to return to its original position after rotating one circle is regarded as a cycle.

[0045] At this point, the vibration signal of each fan blade in each cycle is obtained.

[0046] Step S002: Analyze the collected vibration signal data and construct a damage sensitivity index based on the signal characteristics of the wind turbine blade when it is damaged.

[0047] Since wind turbine blades work in a complex and changeable natural environment for a long time, they are constantly subjected to strong wind loads, sand scouring, atmospheric oxidation, lightning strikes, humid air corrosion and other adverse factors. Various damages such as wear, epidermal shedding, and cracks often occur. Among them, crack damage is particularly worthy of attention because of its serious threat to the structural integrity of the blades. Therefore, this application monitors crack damage in wind turbine blades.

[0048] Since the fan blade signal has significant time-varying characteristics, and the blade's natural frequency of each order changes slightly before and after damage, it is difficult to accurately analyze whether the fan blade has crack damage only from the time domain or frequency domain. Therefore, wavelet transform is used to convert the vibration signal of each fan blade in each cycle into the time-frequency domain, and the time-frequency matrix of the vibration signal of each fan blade in each cycle is obtained. ,in Indicates time, Indicates frequency, Indicates the location of the vibration sensor , where the behavior time of the time-frequency matrix is ​​listed as frequency, and the change characteristics of the vibration signal before and after the fan blade is damaged are further analyzed. Wavelet transform is a well-known technology, and its specific implementation process will not be repeated here.

[0049] When the fan is operating normally, the time-frequency characteristics of the adjacent period vibration signal are relatively stable. However, when crack damage occurs on the fan blade, the local structure of the fan blade changes, causing the natural frequency of the collected vibration signal to change, generating more high-frequency information components, which in turn causes a sudden change in the time-frequency characteristics of the vibration signal in the time-frequency domain. Therefore, by monitoring the change rate of the time-frequency characteristic value of the vibration signal in adjacent periods and then identifying whether the fan blade is damaged, information entropy can measure the complexity and information content of the vibration signal in the time-frequency domain.

[0050] According to the above analysis, for the obtained time-frequency matrix ,by The sliding window slides point by point on the matrix, and the sliding direction is from top to bottom, that is, along the row direction, and from left to right, that is, along the column direction. Each sliding moves an element distance in the row direction or column direction. For each time point in the matrix and frequency points , centered on it The elements in the sliding window calculate the time-frequency entropy of the time point and frequency point, and construct the first The time-frequency entropy matrix of the vibration signal of a fan blade , where the time-frequency matrix has insufficient edge points The matrix is ​​filled with the mean value, and the schematic diagram of the edge point filling of the time-frequency matrix is ​​as follows Figure 2 shown.

[0051] Under normal circumstances, the vibration signal exhibits a stable characteristic. This characteristic is reflected in the time-frequency domain, that is, the energy distribution of the signal is relatively uniform, and the changes in different frequency bands are relatively smooth. Therefore, the calculated time-frequency entropy matrix The overall trend in the continuous time period is relatively stable, indicating that the wind turbine blades are in good operating condition and have no significant signs of damage. However, once the wind turbine blades are damaged, the vibration characteristics of the monitored vibration signal will change significantly, making the energy distribution of the vibration signal no longer uniform, and the energy of certain frequency bands will be significantly enhanced. The vibration signal energy diagram is shown in the figure below. Figure 3 As shown. Curve 1 shows the change of energy with frequency distribution when the fan blade is damaged, and curve 2 shows the change of energy with frequency distribution under normal conditions. Therefore, based on the above analysis, the time-frequency entropy change is constructed, and the formula is: ; In the formula, Indicates The fan blades are The first cycle The change of time-frequency entropy of the frequency; and Respectively represent The fan blades are The cycle and The time-frequency entropy matrix of the vibration signal of cycles, and Resolution represents the time dimension and frequency dimension in the time-frequency entropy matrix; T is the total number of acquisition time points.

[0052] It should be noted that when crack damage occurs in the monitored fan blade, the energy distribution of the vibration signal is no longer uniform, which significantly enhances the energy of the damage frequency band. This causes the entropy value of the calculated time-frequency entropy matrix of the vibration signal of two adjacent cycles to suddenly increase in the local area, that is, Increases, then the value of the time-frequency entropy change increases; conversely, the value of the time-frequency entropy change decreases.

[0053] As the blade damage intensifies, the mutation of the vibration signal will become more obvious, which is manifested as a sudden increase in the entropy value of the local area in the time-frequency entropy matrix. Therefore, the time-frequency entropy mutation index is obtained according to the change of time-frequency entropy. , the formula is: ; In the formula, Indicates The fan blades are The time-frequency entropy mutation index of the vibration signal of a cycle; Indicates The vibration signal of the fan blade is The range of all time-frequency entropy changes in the period is Indicates The vibration signal of the fan blade is The information entropy of all time-frequency entropy changes in a period; It is the preset parameter adjustment factor. In order to prevent the denominator from being 0, the value of 𝛽 is 0.01.

[0054] It should be noted that when the calculated vibration signal changes in different frequency bands The difference increases, even if the calculated as well as becomes larger, and finally the calculated The fan blades are The time-frequency entropy mutation index of the vibration signal of cycles becomes smaller; indicating that the possibility of crack damage on the fan blade is greater at this time; conversely, the larger the time-frequency entropy mutation index is, the smaller the possibility of crack damage on the fan blade is.

[0055] Although the real-time monitoring of the health status of wind turbine blades can be achieved by analyzing the time-frequency entropy of vibration signals of adjacent cycles at the same position, the change of vibration signals is relatively weak in the early stage of blade damage. The fluctuation of time-frequency entropy alone is not enough to capture the early phenomenon of blade damage, and false detection may occur when the wind turbine blades are subjected to harsh environments. Therefore, the damage of wind turbine blades is further analyzed based on the correlation between the monitored vibration signals of different wind turbine blades.

[0056] The fan blades follow strict standards in design, manufacturing and installation to ensure that the fan blades can be evenly stressed and rotate smoothly during operation. Therefore, under normal circumstances, the vibration signals monitored by the three fan blades will show a certain synergistic relationship. Based on the above analysis, the synergy coefficient is constructed, and the formula is: ; In the formula, Indicates The fan blades are The coordination coefficient between the vibration signal of a cycle and other fan blades; and Respectively represent and The fan blades are The time-frequency matrix of the vibration signal of cycles, Indicates the calculation and The fan blades are The cosine similarity between the time-frequency matrices of the vibration signals of the period, the cosine similarity is a well-known technology, and the specific calculation process thereof will not be described in detail in this application; Indicates the number of fan blades. In this embodiment .

[0057] When When a fan blade is cracked, The frequency components and vibration modes of the vibration signals detected by the fan blades change, causing the The time-frequency mutation matrix of a fan blade has poor coordination with other fan blades, which makes the calculated coordination coefficient decreases; on the contrary, the synergy coefficient Increase.

[0058] Specifically, under normal conditions, different wind turbine blades will be subjected to the same external wind load, thus generating the same or similar frequency components or vibration modes. Specifically, the frequency distribution and amplitude of the vibration signals of different blades monitored at the same time should maintain the same or similar characteristics. However, when one of the wind turbine blades is damaged by a local crack, the frequency distribution and amplitude of the vibration signal of the damaged blade will change, resulting in a decrease in the coordination relationship between the vibration signals monitored by the three wind turbine blades. Therefore, based on the above analysis and combined with the time-frequency entropy mutation index, the time-frequency coordination index is constructed. , the formula is: ; In the formula, Indicates The fan blades are The time-frequency coherence index of the cycle; Indicates The fan blades are The coordination coefficient between the vibration signal of a cycle and other fan blades; Indicates The fan blades are The time-frequency entropy mutation index of a vibration signal with a period.

[0059] It should be noted that when When a fan blade is cracked, The frequency components and vibration modes of the vibration signals detected by the fan blades change, causing the The coordination coefficient between the time-frequency mutation matrix of the first fan blade and other fan blades is reduced. The crack damage of the fan blade causes the calculated time-frequency entropy mutation index becomes smaller, thus making the final calculated The fan blades are The time-frequency coherence index of the vibration signal with a period Decrease, indicating that The greater the possibility of crack damage on the fan blade, the greater the time-frequency synergy index. Increase, indicating The smaller the possibility of crack damage on the fan blades.

[0060] According to the above calculation, the time-frequency coordination index of each fan blade in each vibration cycle is finally obtained. , however, the calculated time-frequency synergy index There is a negative correlation between the damage of the fan blade and the crack damage. Therefore, in order to better analyze and evaluate the health status of the fan blade, the damage sensitivity of the fan blade is constructed based on the time-frequency synergy index. , the correlation between the time-frequency synergy index and the fan crack damage is corrected, and the formula is: , where Indicates The fan blades are The time-frequency coherence index of the cycle; Indicates The fan blades are The damage sensitivity of the fan blade in a cycle, e is a natural constant. The time-frequency coordination index calculated when the fan blade is damaged The smaller the damage sensitivity of the fan blades is, the smaller the damage sensitivity of the fan blades is. The bigger.

[0061] At this point, the damage sensitivity of each fan blade in each cycle is obtained.

[0062] Step S003: monitoring the damage condition of the wind turbine blades according to the calculated damage sensitivity index.

[0063] Record damage sensitivity of each wind turbine blade for each cycle , the damage sensitivity of the fan blades for every 100 cycles is a set, denoted as set . Use HBOS algorithm to The data in the collection is used for abnormal monitoring. The data in the set is arranged from large to small, and the set The data in is divided into the number of intervals Set as , Representing a collection The number of elements in this application , Indicates that the calculation result is rounded off.

[0064] According to the divided interval, the collection As input, count the number of elements in each interval and generate a histogram, and calculate the set based on the histogram The anomaly score of each data in is shown in the figure below. Figure 4 As shown, Figure 4 In the figure, the triangle above the dotted line represents an abnormal point, and the origin below the dotted line represents a normal point. For the specific calculation process, please refer to The specific calculation process of the algorithm is not described here. Setting the abnormal threshold Finally, we can get the set The abnormal data set in The abnormal data in the figure indicates that the fan blades are damaged.

[0065] Based on the same inventive concept as the above method, an embodiment of the present application also provides a wind blade damage monitoring device, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned wind blade damage monitoring methods are implemented.

[0066] Similarly, based on the same inventive concept as the above method, an embodiment of the present application provides a wind turbine blade damage monitoring system, the system comprising:

[0067] A signal acquisition module is used to collect the vibration signal of each fan blade in each cycle;

[0068] A fan blade damage sensitivity acquisition module is used to process the vibration signal using a time-frequency conversion algorithm to obtain the time-frequency matrix and time-frequency entropy matrix of each cycle of each fan blade; based on the column distribution of the differences between the elements at the same position in the time-frequency entropy matrix of each fan blade in adjacent cycles, obtain the time-frequency entropy change of each frequency of each fan blade in each cycle; based on the size and uncertainty of the change range of the time-frequency entropy change of all frequencies of each fan blade in each cycle, obtain the time-frequency entropy mutation index of each fan blade in each cycle; based on the similarity between the time-frequency matrix of each fan blade in each cycle and the time-frequency matrix of each other fan blade, obtain the coordination coefficient between each fan blade and other fan blades in each cycle; based on the size of the coordination coefficient and the value of the time-frequency entropy mutation index, obtain the time-frequency coordination index of each fan blade in each cycle; based on the time-frequency coordination index, obtain the fan blade damage sensitivity of each fan blade in each cycle;

[0069] The fan blade damage monitoring module is used to monitor the fan blade damage using an abnormal data detection algorithm combined with the fan blade damage sensitivity.

[0070] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0072] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for monitoring fan blade damage, characterized in that: The method comprises the following steps: Collect the vibration signal of each fan blade in each cycle; The vibration signal is processed using a time-frequency conversion algorithm to obtain the time-frequency matrix and time-frequency entropy matrix of each cycle of each fan blade; based on the column distribution of the differences between the elements at the same position in the time-frequency entropy matrix of each fan blade in adjacent cycles, the time-frequency entropy change of each fan blade at each frequency in each cycle is obtained; Based on the size and uncertainty of the change range of the time-frequency entropy change of all frequencies of each fan blade in each cycle, the time-frequency entropy mutation index of each fan blade in each cycle is obtained; Based on the similarity between the time-frequency matrix of each fan blade in each period and the time-frequency matrix of other fan blades, the coordination coefficient between each fan blade and other fan blades in each period is obtained; based on the value of the coordination coefficient and the time-frequency entropy mutation index, the time-frequency coordination index of each fan blade in each period is obtained; Based on the time-frequency synergy index, the damage sensitivity of each fan blade in each cycle is obtained; the damage of the fan blade is monitored using the abnormal data detection algorithm combined with the damage sensitivity of the fan blade; The method for obtaining the time-frequency entropy matrix is: The vibration signal of each fan blade in each cycle is converted into the frequency domain by using the time-frequency conversion algorithm to obtain the time-frequency matrix of each fan blade in each cycle. For the time-frequency matrix of each fan blade in each period, a sliding window of a preset size is slid on the time-frequency matrix, and the time-frequency entropy of all elements in the sliding window is calculated after each sliding as the time-frequency entropy of each element in the time-frequency matrix. The time-frequency entropy of each element in the time-frequency matrix is ​​constructed according to the position of each element in the time-frequency matrix to construct the time-frequency entropy matrix of each fan blade in each period; The method for obtaining the time-frequency entropy change is: For the time-frequency entropy matrix of each fan blade in each cycle, the absolute value of the difference between the elements at the same position in the time-frequency entropy matrix and the time-frequency entropy matrix of the previous adjacent cycle is calculated; For all elements in the time-frequency entropy matrix corresponding to the same frequency in the time-frequency matrix of each fan blade in each cycle, the sum of the absolute values ​​of all the differences of the elements is calculated as the change in the time-frequency entropy of each frequency of each fan blade in each cycle; The formula of the time-frequency entropy mutation index is: ; In the formula, Indicates The fan blades are The time-frequency entropy mutation index of the vibration signal of a cycle; Indicates The vibration signal of the fan blade is The range of all time-frequency entropy changes in the period is Indicates The vibration signal of the fan blade is The information entropy of all time-frequency entropy changes in a period; is the preset parameter adjustment factor; The method for obtaining the synergy coefficient is: For the time-frequency matrix of each fan blade in each period, calculate the similarity between the time-frequency matrix of the fan blade and the time-frequency matrix of other fan blades, and take the sum of the similarities between each fan blade in each period and all other fan blades as the coordination coefficient between each fan blade in each period and other fan blades; The formula of the time-frequency synergy index is: ; In the formula, Indicates The fan blades are The time-frequency coherence index of the period; Indicates The fan blades are The coordination coefficient between the vibration signal of a cycle and other fan blades; Indicates The fan blades are The time-frequency entropy mutation index of the vibration signal of a cycle; K is the total number of fan blades; The formula for the fan blade damage sensitivity is: , where Indicates The fan blades are The time-frequency coherence index of the cycle; Indicates The fan blades are Damage sensitivity of fan blades in cycles, e is a natural constant; The use of an abnormal monitoring algorithm combined with the fan blade damage sensitivity to monitor the fan blade damage includes: For each fan blade, the fan blade damage sensitivity of each preset number of cycles is divided into each set; An abnormal data detection algorithm is used to detect each set. If abnormal data appears in the set, the fan blade is damaged; otherwise, the fan blade is not damaged.

2. A wind turbine blade damage monitoring device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a wind blade damage monitoring method as claimed in claim 1 are implemented.

3. A wind turbine blade damage monitoring system, implementing a wind turbine blade damage monitoring method as claimed in claim 1, characterized in that: The monitoring system comprises: A signal acquisition module is used to collect the vibration signal of each fan blade in each cycle; A fan blade damage sensitivity acquisition module is used to process the vibration signal using a time-frequency conversion algorithm to obtain the time-frequency matrix and time-frequency entropy matrix of each cycle of each fan blade; based on the column distribution of the differences between the elements at the same position in the time-frequency entropy matrix of each fan blade in adjacent cycles, obtain the time-frequency entropy change of each frequency of each fan blade in each cycle; based on the size and uncertainty of the change range of the time-frequency entropy change of all frequencies of each fan blade in each cycle, obtain the time-frequency entropy mutation index of each fan blade in each cycle; based on the similarity between the time-frequency matrix of each fan blade in each cycle and the time-frequency matrix of each other fan blade, obtain the coordination coefficient between each fan blade and other fan blades in each cycle; based on the size of the coordination coefficient and the value of the time-frequency entropy mutation index, obtain the time-frequency coordination index of each fan blade in each cycle; based on the time-frequency coordination index, obtain the fan blade damage sensitivity of each fan blade in each cycle; The fan blade damage monitoring module is used to monitor the fan blade damage using an abnormal data detection algorithm combined with the fan blade damage sensitivity.

Citation Information

Patent Citations

  • Wind turbine blade state monitoring method and system based on vibration detection and wireless signal transmission

    CN107061185A

  • Method and device for recognizing variable pitch abnormities of wind generating set

    CN107327367A

  • Status monitoring for mechanical plants, in particular wind turbines

    CN109477464A

  • Wind driven generator broken blade detecting method and device based on rectangular microphone array

    CN109505741A

  • Monitoring method and device of wind driven generator

    CN109653961A