Fan blade current collection line tripping fault recovery method
By building a regulation model library based on historical fault data and a multi-stage testing method, the problems of low efficiency and insufficient accuracy of the traditional fan blade collector line fault recovery method are solved, and efficient and reliable fault repair is achieved.
Patent Information
- Application Number
- CN202510407036.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-01
- Publication Date
- 2025-07-11
AI Technical Summary
The traditional method of blower blade collector line tripping fault recovery depends on manual experience, is inefficient, difficult to effectively deal with complex faults, and lacks dynamic consideration of environmental data, resulting in inaccurate and reliable fault repair.
By setting monitoring points based on historical fault data, building a first-level adjustment model library, combining environmental data and operating status data, generating adjustment instructions, conducting multi-stage testing to judge the fault area and repair effect, and introducing stability and execution evaluation value to evaluate the effectiveness of adjustment instructions.
It improves the accuracy of fault location and identification accuracy, shortens the time for fault repair decisions, reduces the trial and error process, ensures the effectiveness and reliability of fault repair, prevents fault spread, and improves overall repair efficiency and accuracy.
Smart Images
Figure CN120297948A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of collector line fault detection, and particularly to a method for recovering the tripping fault of a collector line of a wind turbine blade. Background Art
[0002] After a tripping fault occurs in the collector line, the traditional approach is for maintenance personnel to manually troubleshoot and repair based on experience. For example, first check whether the fuse is blown, and then check whether the line connection is loose, etc. For some simple faults, this method can solve the problem, but the efficiency is low and it depends on the experience and skill level of the maintenance personnel.
[0003] In some large-scale wind farms, the fault recovery is also carried out according to the pre-established conventional operation procedures. These procedures are usually formulated based on common fault types, but may not be able to effectively handle some special and complex fault situations. Summary of the Invention
[0004] The object of the present invention is to consider environmental data in multiple links, so that the fault recovery method can adapt to different environmental conditions and improve the reliability of the fault recovery method under different working conditions; the dynamic consideration of environmental factors can more accurately handle complex situations in actual operation, ensure the effectiveness of fault repair, and thus improve the operation reliability of the entire collector line of the wind turbine blade.
[0005] To achieve the above object, the present invention provides a method for recovering the tripping fault of a collector line of a wind turbine blade, including: Setting the positions of monitoring points based on historical fault data, and obtaining the operation status data of the collector line based on the monitoring points; judging the operation status of the collector line based on the obtained operation status data of each monitoring point; Constructing a primary adjustment model library based on historical fault data; Combining the current operation status of the extreme line and the primary adjustment model library to select the corresponding fault handling model, and generating the corresponding adjustment instruction based on the primary adjustment model; Running the adjustment instruction to repair the fault, and judging the effectiveness of the adjustment instruction by testing the current collector line of the wind turbine blade based on the obtained operation status data of the collector line after the fault is repaired.
[0006] In some embodiments of the present invention, when establishing the primary adjustment model library, it further includes: Classifying the fault types based on historical fault data to obtain a set A of primary fault types, A = {a1, a2... ai... an}; wherein, ai represents the first type of fault of the i-th type, and n represents the total number of primary fault types; Obtain the operation status data of each monitoring point corresponding to the first type of fault ai of the i-th type; Obtain the operation status evaluation value of each monitoring point based on the operation status data of the monitoring point; Construct a first-level fault identification model by combining the operation status evaluation values of all monitoring points and the environmental data of the current fan blade collector line; Determine the first-level fault type of the current fan blade collector line by combining the operation status evaluation value of each monitoring point at the current monitoring time node and the first-level fault identification model; Set the adjustment instruction for each first-level fault type based on historical fault data to generate a corresponding first-level adjustment model; Generate a first-level adjustment model library by combining all first-level adjustment models.
[0007] In some embodiments of the present invention, when constructing the first-level fault identification model, it further includes: Generate a first-level fault mapping table by combining historical fault data and historical environmental data corresponding to the fault occurrence time; Obtain the first-level fault types that occurred under the same environmental data based on the first-level fault mapping table; Among them, the first-level fault types that occurred under the same environmental data are one or more; Obtain the operation status evaluation value of each monitoring point corresponding to each first-level fault type under the current environmental data; Generate the distribution characteristics of the evaluation values based on the operation status evaluation values of all monitoring points; Construct a first-level fault identification model by combining historical environmental data and the distribution characteristics of the evaluation values.
[0008] In some embodiments of the present invention, when generating the distribution characteristics of the evaluation values, it further includes: Divide the operation status of the fan blade collector line based on historical fault data to generate a normal operation status and an abnormal operation status; Obtain the set of operation status evaluation values of each monitoring point when the fan blade collector line is in the normal operation status corresponding to the current environmental data; And generate a normal operation interval [a, b] based on the set of operation status evaluation values of each monitoring point in the normal operation status; Wherein, a is the left endpoint of the interval of the operation status evaluation value of the current monitoring point in the normal operation status, and b is the right endpoint of the interval of the operation status evaluation value of the current monitoring point in the normal operation status; Screen the operation status evaluation values of each monitoring point obtained based on the normal operation interval [a, b], and obtain the operation status evaluation values of the monitoring points not within the normal operation interval [a, b] to generate abnormal monitoring points; Divide the area of abnormal monitoring points, and generate the geometric distribution characteristics of abnormal monitoring points by judging the density of abnormal monitoring points in the area; Obtain the variance, mean, maximum value and minimum value of the evaluation value of abnormal monitoring points to generate the numerical distribution characteristics of abnormal monitoring points; Integrate the geometric distribution characteristics and numerical distribution characteristics of abnormal monitoring points to generate the distribution characteristics of evaluation values.
[0009] In some embodiments of the present invention, when generating the geometric distribution characteristics of abnormal monitoring points, it further includes: Construct a rectangular area to include all abnormal monitoring points; Divide and screen the rectangular area to obtain the first geometric distribution; Perform division and screening on the first geometric distribution again to obtain the geometric distribution characteristics of abnormal monitoring points; The division and screening include: Divide the current rectangular area into 9 sub-rectangular areas of the same size of 3×3; Obtain the number of abnormal monitoring points in each sub-rectangular area; Eliminate the m sub-rectangular areas with the smallest number of abnormal monitoring points; Among them, the value of m is selected based on actual accuracy requirements; Obtain the positions based on the remaining sub-rectangular areas.
[0010] In some embodiments of the present invention, when judging the effectiveness of the adjustment instruction, it further includes: Determine the fault area based on the adjustment instruction, and determine the warning area based on the fault area; Separate the fault area and the warning area through the adjustment instruction; After repairing the fault area, conduct the first test on the current fault area through a simulation signal to generate the first test result; If the first test result is passed, connect the fault area and the warning area, and conduct the second test on the connected fault area and warning area through a simulation signal to generate the second test result; If the second test result is passed, make the collector line of the fan blades operate normally, and obtain the change curve of the operating state data of each monitoring point of the collector line of the fan blades; Generate the third test result based on the change curve of the operating state data; Judge the effectiveness of the adjustment instruction based on the third test result.
[0011] In some embodiments of the present invention, when determining the warning area based on the fault area, it includes: Obtain the monitoring points connected to the fault area based on the topological structure and electrical characteristics of the collector line; Predict the spread path of the fault based on the current fault type; Modify the spread path based on the environmental data at the current monitoring time node to generate a second spread path; Combine the second spread path and the monitoring points connected to the fault area to construct a warning area; The starting boundary of the warning area is the monitoring point connected to the fault area; The ending boundary of the warning area is delimited by the second spread path.
[0012] In some embodiments of the present invention, when generating the second test result, it further includes: Set the parameters of the analog signal according to the fault type; Start testing from one end of the fault area to the warning area in sequence, use the analog signal to test each node or device, and record the signal response data of each test point; Based on the comparison between the change situation of the analog signal when passing through the fault area and the warning area and the preset transformation situation, judge the connection status between the repaired fault area and the warning area; Among them, the change situation includes: the attenuation degree and distortion condition of the signal; The connection status between the repaired fault area and the warning area generates the second test result.
[0013] In some embodiments of the present invention, when judging the effectiveness of the adjustment instruction based on the third test result, it further includes: Calculate the volatility of the change curve based on the operation status data change curves of each monitoring point, and judge the line stability by comparing the volatility with the first preset value to generate a stability evaluation value c1; Conduct a comparative analysis on the same parameter change curves of different monitoring points, and judge the executability of the adjustment instruction by checking whether there are local differences exceeding the second preset value to generate an executability evaluation value c2; Generate an effectiveness evaluation value c3 of the adjustment instruction based on the line stability evaluation value and the executability evaluation value; C3 = w1*c1 + w2*c2; Among them, w1 is the weight of the stability evaluation value c1, and w2 is the weight of the executability evaluation value c2.
[0014] Compared with the prior art, the beneficial effects of a method for restoring the tripping fault of a fan blade collector line provided by an embodiment of the present invention are as follows: By setting the positions of the monitoring points based on historical fault data, the operation status data of the collector line can be obtained more pertinently, which helps to improve the positioning accuracy of the fault area, reduce the blindness of monitoring, and thus more quickly determine the approximate range where the fault is located.
[0015] When building a primary fault identification model, in the case where multiple fault types may coexist or the fault manifestations are similar, it can distinguish different fault modes and improve the accuracy of fault identification.
[0016] In the process of generating the distribution characteristics of evaluation values, detailed geometric and numerical feature analyses are carried out on abnormal monitoring points. By determining the geometric distribution characteristics (such as regional density, etc.) and numerical distribution characteristics (such as variance, mean, etc.) of abnormal monitoring points, a deeper understanding of the spatial and numerical distributions of faults can be obtained, which helps to accurately judge the severity and influence range of faults.
[0017] Build a primary regulation model library based on historical fault data, so that when facing the current fault, the appropriate fault handling model can be quickly selected in combination with the operating status of the current line and adjustment instructions can be generated. This avoids formulating a repair strategy from scratch when a fault occurs, greatly shortens the decision-making time for fault repair, and improves the repair efficiency.
[0018] Build a regulation model according to historical fault types and corresponding effective solutions, so the generated adjustment instructions are more likely to be effective, reducing the trial-and-error process, thereby accelerating the speed of fault repair.
[0019] When determining the early warning area, consider various factors such as the topological structure, electrical characteristics, fault types, diffusion paths, and environmental data of the collector line. By accurately constructing the early warning area, measures can be taken in advance to prevent the further spread of faults. At the same time, it also helps to conduct more comprehensive tests (such as the second test) after repairing the fault area, improving the overall fault repair efficiency.
[0020] Adopt a phased testing method to gradually verify the effect of fault repair. This multi-stage testing method can comprehensively check the operating status of the fault area, early warning area, and the entire collector line after repair.
[0021] In each test, by reasonably setting the test signal, test process, and result judgment criteria, the repair effect can be more accurately evaluated to ensure that the fault is truly resolved and no new problems are introduced due to the repair process.
[0022] When judging the effectiveness of the adjustment instruction based on the results of the third test, introduce stability evaluation values, executability evaluation values, and comprehensive effectiveness evaluation values, and evaluate the effectiveness of the adjustment instruction through quantitative indicators, making the judgment result more objective and accurate; this evaluation method can timely detect possible problems in the adjustment instruction so as to adjust and optimize the repair strategy. Description of the Drawings
[0023] Figure 1It is a flowchart of a method for recovering from a tripping fault of a collector line of a wind turbine blade provided by an embodiment of the present invention. Detailed implementation manners
[0024] The following further describes in detail the specific implementation manners of the present invention in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.
[0025] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is 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 should not be construed as limiting the present invention.
[0026] The terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more.
[0027] 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 may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected, or indirectly connected through an intermediate medium, and it may be the internal communication of two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0028] Embodiment 1: A method for recovering from a tripping fault of a collector line of a wind turbine blade provided by an embodiment of the present invention, as Figure 1 shown, includes: Setting the positions of monitoring points based on historical fault data, and obtaining the operation state data of the collector line based on the monitoring points; judging the operation state of the collector line based on the obtained operation state data of each monitoring point; Constructing a primary adjustment model library based on historical fault data; Selecting a corresponding fault handling model in combination with the current operation state of the extreme line and the primary adjustment model library, and generating a corresponding adjustment instruction based on the primary adjustment model; Run the adjustment instruction to repair the fault, and based on the obtained operating state data of the collector line after the fault is repaired, test the current fan blade collector line to judge the effectiveness of the adjustment instruction.
[0029] Embodiment 2: When establishing the first-level adjustment model library, it also includes: Classify the fault types based on historical fault data to obtain the first-level fault type set A, A = {a1, a2…ai…an}; Among them, ai represents the first type of fault of the i-th type, and n represents the total number of first-level fault types; Obtain the operating state data of each monitoring point corresponding to the first type of fault ai of the i-th type; Obtain the operating state evaluation value of each monitoring point based on the operating state data of the monitoring point; Construct a first-level fault identification model by combining the operating state evaluation values of all monitoring points and the environmental data of the current fan blade collector line; Determine the first-level fault type of the current fan blade collector line by combining the obtained operating state evaluation values of each monitoring point at the current monitoring time node and the first-level fault identification model; Set the adjustment instruction for each first-level fault type based on historical fault data to generate the corresponding first-level adjustment model; Generate a first-level adjustment model library by combining all the first-level adjustment models.
[0030] In this embodiment, all fault records related to the fan blade collector line are extracted from the historical fault database. These records should contain detailed information such as the time and location of the fault, the description of the fault phenomenon, and the relevant parameters detected for the fault.
[0031] Clean the extracted fault data to remove the incorrect data, duplicate data, and incomplete data to ensure the accuracy and availability of the data.
[0032] Fault feature analysis: For each cleaned fault record, analyze its fault features. For example, for a fault caused by a short circuit in the line, analyze the location of the short circuit point, the current change during the short circuit, the voltage fluctuation situation, and the data features of the monitoring points related to the short circuit.
[0033] Classify the faults according to the similarity of the fault features. For example, classify all electrical equipment damages caused by lightning strikes into one category, and classify the faults caused by the insulation performance degradation due to line aging into another category, etc., to obtain the first-level fault type set A.
[0034] According to the normal operation requirements and technical specifications of the collector line of the fan blades, an operation status evaluation index system is established for each monitoring point. For example, for a voltage monitoring point, the evaluation indexes may include whether the voltage is within ±5% of the rated voltage, whether the voltage fluctuation frequency is less than a certain threshold, etc.; for a temperature monitoring point, the evaluation index may be whether the temperature is within the normal operation temperature range of the equipment, etc.
[0035] Corresponding weights are set for each evaluation index, and the determination of the weights can be based on methods such as expert experience and fault impact analysis. For example, for the voltage monitoring point of a key device (such as a core transformer), the weight of its voltage stability index may be relatively high.
[0036] According to the established evaluation index system and weights, the operation status data of each monitoring point are calculated to obtain the operation status evaluation value of each monitoring point. For example, for a certain voltage monitoring point, if its voltage is within ±3% of the rated voltage and the fluctuation frequency is small, a relatively high evaluation value is calculated according to the set indexes and weights; if the voltage deviates greatly from the rated value or fluctuates frequently, a relatively low evaluation value is calculated.
[0037] For each type of primary fault, search the historical fault data for the adjustment instructions adopted when successfully repairing this type of fault in the past. These adjustment instructions may include the adjusted values of voltage and current, control commands for certain devices (such as the opening and closing of switches, the tap adjustment of transformers, etc.).
[0038] Analyze the found adjustment instructions and summarize the effective adjustment strategies for each type of fault. For example, for the fault type caused by too high voltage, the adjustment instruction may be to reduce the input voltage to 90% of the rated value and continuously observe the voltage change.
[0039] According to the summarized adjustment strategies, construct a primary adjustment model corresponding to each type of primary fault. If the adjustment strategy involves mathematical calculations (such as calculating the tap position of the transformer to be adjusted according to the voltage deviation), then these calculation relationships are constructed into a mathematical model. For example, for a certain type of fault, its adjustment model can be a linear or non-linear function based on parameters such as voltage deviation and current deviation.
[0040] Example 3: When constructing the primary fault identification model, it also includes: Generate a primary fault mapping table by combining historical fault data and historical environmental data corresponding to the fault occurrence time; Obtain the types of primary faults that occurred under the same environmental data based on the primary fault mapping table; Among them, the types of primary faults that occurred under the same environmental data are one or more; Obtain the operation status evaluation values of each monitoring point corresponding to each first-level fault type under the current environmental data; Generate the distribution characteristics of the evaluation values based on the operation status evaluation values of all monitoring points; Construct a first-level fault identification model by combining historical environmental data and the distribution characteristics of the evaluation values.
[0041] In this embodiment, a first-level fault mapping table is generated: When generating the first-level fault mapping table by combining historical fault data and historical environmental data corresponding to the fault occurrence time, the environmental data is classified more meticulously. In addition to common factors such as temperature and humidity, factors such as the electromagnetic interference intensity and dust content in the environment are also considered for their impact on faults.
[0042] To more accurately reflect the relationship between faults and environmental data, the frequency information of fault occurrence is added to the mapping table. For example, under a certain specific environmental condition, the occurrence frequency of a certain fault type is relatively high, and this information helps to raise the vigilance for this fault type in actual fault diagnosis.
[0043] Obtain the fault type and evaluation value under the same environment: When obtaining the first-level fault types occurring under the same environmental data based on the first-level fault mapping table, data mining techniques such as association rule mining are used to find the hidden association patterns between environmental data and fault types.
[0044] When obtaining the operation status evaluation values of each monitoring point corresponding to each first-level fault type under the current environmental data, consider the corrective effect of environmental data on the evaluation values. For example, in a high-temperature environment, the normal operating temperature range of the equipment may change, and correspondingly, the calculation of the operation status evaluation value of the monitoring point also needs to be adjusted.
[0045] Embodiment 4: When generating the distribution characteristics of the evaluation values, it also includes: Based on historical fault data, divide the operation status of the fan blade collector line into normal operation status and abnormal operation status; Obtain the set of operation status evaluation values of each monitoring point when the fan blade collector line is in the normal operation status corresponding to the current environmental data; And generate a normal operation interval [a, b] based on the set of operation status evaluation values of each monitoring point in the normal operation status; Wherein, a is the left endpoint of the interval of the operation status evaluation value of the current monitoring point in the normal operation status, and b is the right endpoint of the interval of the operation status evaluation value of the current monitoring point in the normal operation status; Based on the normal operation interval [a, b], screen the operation status evaluation values of each obtained monitoring point, and obtain the operation status evaluation values of the monitoring points not within the normal operation interval [a, b] to generate abnormal monitoring points; Divide the area of abnormal monitoring points, and generate the geometric distribution characteristics of abnormal monitoring points by judging the density of abnormal monitoring points within the area; Obtain the variance, mean, maximum value, and minimum value of the evaluation value of abnormal monitoring points to generate the numerical distribution characteristics of abnormal monitoring points; Integrate the geometric distribution characteristics and numerical distribution characteristics of abnormal monitoring points to generate the distribution characteristics of the evaluation value.
[0046] In this embodiment, when generating the distribution characteristics of the evaluation value: when dividing the operating state of the fan blade collector line based on historical fault data, the clustering analysis method is used. In addition to simply dividing it into normal operating state and abnormal operating state, the level of the abnormal operating state can be further subdivided (such as mild abnormality, moderate abnormality, severe abnormality).
[0047] When generating the normal operating interval [a, b], dynamically adjust the range of the interval according to the operating characteristics of different fans and the statistical analysis of historical data. For example, for new fans, since the equipment performance is better, the normal operating interval can be relatively narrow; for old fans, the normal operating interval may need to be appropriately widened.
[0048] When analyzing the geometric distribution characteristics of abnormal monitoring points, in addition to using the method of rectangular area division and screening, different shapes of area division such as circular and triangular can also be tried, and the method that can best reflect the distribution law of abnormal monitoring points can be selected according to the actual situation.
[0049] When obtaining the variance, mean, maximum value, and minimum value of the evaluation value of abnormal monitoring points, consider the time series characteristics of the data. For example, calculate the variance within different time periods and observe the change trend of the variance to more comprehensively understand the numerical distribution characteristics of abnormal monitoring points.
[0050] Embodiment 5: When generating the geometric distribution characteristics of abnormal monitoring points, it further includes: Construct a rectangular area to contain all abnormal monitoring points; Perform division and screening on the rectangular area to obtain the first geometric distribution; Perform division and screening on the first geometric distribution again to obtain the geometric distribution characteristics of abnormal monitoring points; The division and screening include: Divide the current rectangular area into 9 sub-rectangular areas of the same size of 3×3; Obtain the number of abnormal monitoring points in each sub-rectangular area; Eliminate the m sub-rectangular areas with the smallest number of abnormal monitoring points; Among them, the value of m is selected based on actual accuracy requirements; Obtain the positions based on the remaining sub-rectangular areas.
[0051] Example 6: When judging the validity of the adjustment instruction, it further includes: Determine the fault area based on the adjustment instruction, and determine the early warning area based on the fault area; Separate the fault area and the early warning area through the adjustment instruction; After repairing the fault area, conduct the first test on the current fault area through a simulation signal to generate a first test result; If the first test result is passed, connect the fault area and the early warning area, and conduct the second test on the connected fault area and early warning area through a simulation signal to generate a second test result; If the second test result is passed, make the collector line of the fan blades operate normally, and obtain the change curve of the operation state data of each monitoring point on the collector line of the fan blades; Generate a third test result based on the change curve of the operation state data; Judge the validity of the adjustment instruction based on the third test result.
[0052] Example 7: When determining the early warning area based on the fault area, it includes: Obtain the monitoring points connected to the fault area based on the topological structure and electrical characteristics of the collector line; Predict the fault propagation path based on the current fault type; Correct the propagation path based on the environmental data at the current monitoring time node to generate a second propagation path; Construct an early warning area by combining the second propagation path and the monitoring points connected to the fault area; The starting boundary of the early warning area is the monitoring point connected to the fault area; The ending boundary of the early warning area is delimited by the second propagation path.
[0053] In this embodiment, determining the fault area and the early warning area: When determining the fault area based on the adjustment instruction, according to the electrical connection relationship of the collector line and the fault propagation principle, accurately define the boundary of the fault area. For example, if it is a line short - circuit fault, determine the affected line segment as the fault area according to the position of the short - circuit point and the direction of the short - circuit current.
[0054] When determining the early warning area based on the fault area, consider the difference in the fault diffusion speed under different fault types. For fault types with a faster diffusion speed (such as some electrical breakdown faults), expand the scope of the early warning area; for fault types with a slower diffusion speed (such as faults caused by slow equipment aging), the early warning area can be relatively smaller.
[0055] When correcting the diffusion path according to the environmental data at the current monitoring time node, consider the obstructive or facilitative effect of environmental factors on fault diffusion. For example, in a humid environment, faults may be more likely to spread along the surface of the line, while in a dry environment, fault diffusion may be more confined to the interior of the line.
[0056] Example 8: When generating the second test result, it further includes: Set the parameters of the simulation signal according to the fault type; Start testing from one end of the fault area to the warning area in sequence, use the simulation signal to test each node or device, and record the signal response data of each test point; Based on the comparison between the change situation of the simulation signal when passing through the fault area and the warning area and the preset transformation situation, judge the connection status between the repaired fault area and the warning area; Among them, the change situation includes: the attenuation degree and distortion condition of the signal; Generate the second test result based on the connection status between the repaired fault area and the warning area.
[0057] Example 9: When judging the effectiveness of the adjustment instruction based on the third test result, it further includes: Calculate the volatility of the change curve based on the operation status data change curves of each monitoring point. By comparing the volatility with the first preset value, judge the line stability and generate the stability evaluation value c1; Conduct a comparative analysis on the same parameter change curves of different monitoring points. By checking whether there are local differences exceeding the second preset value, judge the executability of the adjustment instruction and generate the executability evaluation value c2; Generate the effectiveness evaluation value c3 of the adjustment instruction based on the stability evaluation value and the executability evaluation value of the line; C3 = w1*c1 + w2*c2; Among them, w1 is the weight of the stability evaluation value c1, and w2 is the weight of the executability evaluation value c2.
[0058] In this embodiment, when conducting the first test on the current fault area through the simulation signal after repairing the fault area, parameters such as the frequency and amplitude of the simulation signal should simulate the working conditions in actual operation as much as possible. For example, if the voltage frequency of the collector line is 50Hz when the fan is running normally, then the frequency of the simulation signal is also set to 50Hz.
[0059] When conducting the second test, in addition to paying attention to the connection status between the repaired fault area and the warning area, it is also necessary to check whether the electrical performance after connection has returned to the normal level. For example, measure whether parameters such as the line resistance and inductance after connection are within the normal range.
[0060] When judging the effectiveness of the adjustment instruction based on the results of the third test, the weights w1 and w2 of the stability evaluation value c1 and the executability evaluation value c2 are dynamically adjusted according to factors such as the importance of the fan and the severity of the fault. For example, for key fan equipment, the weight w1 of stability can be set higher; for less serious faults, the weight w2 of executability can be appropriately increased.
[0061] Finally, it should be noted that: Obviously, those skilled in the art can make various changes and deformations 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 claims of the present invention and its equivalent technologies, the present invention also intends to include these changes and deformations.
[0062] The above 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 method for recovering from a tripping fault of a collector circuit of a fan blade, characterized in that, Including: Setting the positions of monitoring points based on historical fault data, and obtaining the operation status data of the collector line based on the monitoring points; Judging the operation status of the collector line based on the obtained operation status data of each monitoring point; Constructing a first-level adjustment model library based on historical fault data; Combining the current operation status of the extreme line and the first-level adjustment model library to select the corresponding fault handling model, and generating the corresponding adjustment instruction based on the first-level adjustment model; Running the adjustment instruction to repair the fault, and judging the effectiveness of the adjustment instruction by testing the operation status data of the current fan blade collector line after the fault is repaired.
2. The method for recovering from the tripping fault of the collector circuit of the fan blade according to claim 1, characterized in that When constructing the first-level adjustment model library, it also includes: Classifying the fault types based on historical fault data to obtain the first-level fault type set A, A = {a1, a2... ai... an}; Among them, ai represents the first type of fault of the i-th type, and n represents the total number of first-level fault types; Obtaining the operation status data of each monitoring point corresponding to the first type of fault ai of the i-th type; Obtaining the operation status evaluation value of each monitoring point based on the operation status data of the monitoring point; Constructing a first-level fault identification model by combining the operation status evaluation values of all monitoring points and the environmental data of the current fan blade collector line; Determining the first-level fault type of the current fan blade collector line by combining the operation status evaluation values of each monitoring point at the current monitoring time node and the first-level fault identification model; Setting the adjustment instruction for each first-level fault type based on historical fault data to generate the corresponding first-level adjustment model; Combining all the first-level adjustment models to generate a first-level adjustment model library.
3. The method for recovering from the tripping fault of the collector circuit of the fan blade according to claim 2, characterized in that, When constructing the first-level fault identification model, it also includes: Generating a first-level fault mapping table by combining historical fault data and historical environmental data corresponding to the fault occurrence time; Obtaining the first-level fault types occurring under the same environmental data based on the first-level fault mapping table; Among them, the first-level fault types occurring under the same environmental data are one or more; Obtaining the operation status evaluation value of each monitoring point corresponding to each first-level fault type under the current environmental data; Generating the distribution characteristics of the evaluation values based on the operation status evaluation values of all monitoring points; Constructing a first-level fault identification model by combining historical environmental data and the distribution characteristics of the evaluation values.
4. The method for recovering from the tripping fault of the collector circuit of the fan blade according to claim 3, characterized in that, When generating the distribution characteristics of the evaluation values, it also includes: Dividing the operation status of the fan blade collector line based on historical fault data to generate a normal operation status and an abnormal operation status; Obtaining the set of operation status evaluation values of each monitoring point when the fan blade collector line is in the normal operation status corresponding to the current environmental data; And generating a normal operation interval [a, b] based on the set of operation status evaluation values of each monitoring point in the normal operation status; Among them, a is the left endpoint of the interval of the operation status evaluation value of the current monitoring point in the normal operation status, and b is the right endpoint of the interval of the operation status evaluation value of the current monitoring point in the normal operation status; Screening the operation status evaluation values of the obtained monitoring points based on the normal operation interval [a, b], and obtaining the operation status evaluation values of the monitoring points not within the normal operation interval [a, b] to generate abnormal monitoring points; Divide the area of abnormal monitoring points, and generate the geometric distribution characteristics of abnormal monitoring points by judging the density of abnormal monitoring points in the area; Obtain the variance, mean, maximum value and minimum value of the evaluation value of abnormal monitoring points to generate the numerical distribution characteristics of abnormal monitoring points; Integrate the geometric distribution characteristics and numerical distribution characteristics of abnormal monitoring points to generate the distribution characteristics of evaluation values.
5. The method for recovering from the tripping fault of the collector circuit of the fan blade according to claim 4, characterized in that, When generating the geometric distribution characteristics of abnormal monitoring points, it also includes: Construct a rectangular area to contain all abnormal monitoring points; Divide and screen the rectangular area to obtain the first geometric distribution; Perform division and screening on the first geometric distribution again to obtain the geometric distribution characteristics of abnormal monitoring points; The division and screening include: Divide the current rectangular area into 9 sub-rectangular areas of the same size of 3×3; Obtain the number of abnormal monitoring points in each sub-rectangular area; Eliminate the m sub-rectangular areas with the smallest number of abnormal monitoring points; Among them, the value of m is selected based on actual accuracy requirements; Obtain the positions based on the remaining sub-rectangular areas.
6. The method for restoring the tripping fault of the collector line of the fan blade according to claim 5, characterized in that, When judging the effectiveness of the adjustment instruction, it also includes: Determine the fault area based on the adjustment instruction, and determine the warning area based on the fault area; Separate the fault area and the warning area through the adjustment instruction; After repairing the fault area, conduct the first test on the current fault area through a simulated signal to generate the first test result; If the first test result is passed, connect the fault area and the warning area, and conduct the second test on the connected fault area and warning area through a simulated signal to generate the second test result; If the second test result is passed, make the collector line of the wind turbine blades operate normally, and obtain the change curve of the operating state data of each monitoring point of the collector line of the wind turbine blades; Generate the third test result based on the change curve of the operating state data; Judge the effectiveness of the adjustment instruction based on the third test result.
7. The method for restoring the tripping fault of the collector circuit of the fan blade according to claim 6, wherein, When determining the warning area based on the fault area, it includes: Obtain the monitoring points connected to the fault area based on the topological structure and electrical characteristics of the collector line; Predict the diffusion path of the fault based on the current fault type; Correct the diffusion path based on the environmental data at the current monitoring time node to generate the second diffusion path; Construct a warning area by combining the second diffusion path and the monitoring points connected to the fault area; The starting boundary of the warning area is the monitoring point connected to the fault area; The ending boundary of the warning area is delimited by the second diffusion path.
8. The method for recovering from the tripping fault of the collector circuit of the fan blade according to claim 7, characterized in that, When generating the second test result, it also includes: Set the parameters of the simulated signal according to the fault type; Start testing from one end of the fault area to the warning area in sequence, use the simulated signal to test each node or device, and record the signal response data of each test point; Based on the comparison between the change situation of the simulated signal when passing through the fault area and the warning area and the preset transformation situation, judge the connection status of the repaired fault area and the warning area; Among them, the change situation includes: the attenuation degree and distortion condition of the signal; The connection status of the repaired fault area and the warning area generates the second test result.
9. The method for recovering from the tripping fault of the collector circuit of the fan blade according to claim 8, wherein, When judging the effectiveness of the adjustment instruction based on the third test result, it also includes: Calculate the volatility of the change curve based on the change curve of the operating status data at each monitoring point, and judge the line stability by comparing the volatility with the first preset value to generate the stability evaluation value c1; Conduct a comparative analysis on the change curves of the same parameter at different monitoring points, and judge the executability of the adjustment instruction by checking whether there are local differences exceeding the second preset value to generate the executability evaluation value c2; Generate the effectiveness evaluation value c3 of the adjustment instruction based on the stability evaluation value and the executability evaluation value of the line; C3 = w1 * c1 + w2 * c2; Among them, w1 is the weight of the stability evaluation value c1, and w2 is the weight of the executability evaluation value c2.