Wind power digital twin co-evolution detection method based on signal distribution characteristics

Abstract

This invention discloses a wind power digital twin co-evolution detection method based on signal distribution characteristics, comprising: acquiring wind turbine signal data generated by the digital twin and wind turbine signal data collected by sensors; preprocessing the two types of wind turbine signal data respectively to ensure that the characteristic frequency amplitude error of the preprocessed wind turbine signal data is less than a preset threshold; interpolating the two types of preprocessed wind turbine signal data to the original data volume respectively; performing time-series sequence distribution transformation on the two interpolated results respectively to obtain the time-series sequence signal distribution probability; and calculating the overall difference value based on the two time-series sequence signal distribution probabilities as the output of the twin co-evolution detection result. This invention aims to verify the degree of co-evolution of the digital twin system by comparing the signal distribution of the measured values ​​and the processed digital twin values, so that the digital twin system can comprehensively and accurately reflect the operating status of the wind turbine unit.

Description

Technical Field

[0001] This invention belongs to the field of digital twin technology for wind turbine cluster service quality, specifically relating to a wind power digital twin collaborative evolution detection method based on signal distribution characteristics. Background Technology

[0002] With the development of the wind power industry, the number and scale of wind turbine units are constantly expanding. How to effectively assess and improve the service quality of wind turbine clusters has become a problem that needs to be solved. Currently, digital twin technology is being applied in aerospace, manufacturing, healthcare, urban development, energy, and other fields. These applications and explorations have demonstrated the significant advantages of digital twin technology. The core idea of ​​digital twins is to fully utilize physical data and mechanisms to construct ultra-realistic virtual copies, monitoring, controlling, and optimizing the physical world through interaction and iteration between the virtual and physical worlds. During the long-term operation of wind turbine clusters, various faults may occur, affecting power generation efficiency and reducing grid stability. The detection and verification technology of the digital twin system for wind turbine cluster service quality is crucial for improving the reliability and performance of wind turbine clusters. It can predict faults, optimize operation, diagnose problems, and provide intelligent management to improve the reliability and performance of wind turbine clusters, reduce downtime and maintenance costs, and ultimately achieve more efficient wind energy utilization. One of the key aspects of digital twin technology lies in its "co-evolution" capability, that is, how the virtual model updates synchronously with the real-time changes of the physical system to accurately reflect its current state. The operating characteristics of wind turbines change continuously throughout their lifecycle, requiring digital twin systems to possess strong co-evolution capabilities. Co-evolution ensures that digital twin systems can not only accurately simulate the initial state but also continuously adjust during long-term operation, enabling the virtual and physical systems to evolve synchronously and reflect subtle changes in actual operation.

[0003] To ensure this co-evolution, digital twin systems rely on high-quality sensor data input and a high-precision model update mechanism. During data transmission, the data collected by sensors is often massive and diverse. How to effectively process and assess the quality of this real-time data to ensure the digital twin model can adjust in real time is a key focus of current research. Accurate verification of co-evolution can improve the overall performance of the digital twin system, enabling it to more accurately reflect the operating status of wind turbine units. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a wind power digital twin co-evolution detection method based on signal distribution characteristics, which addresses the above-mentioned problems in the prior art. The present invention aims to verify the degree of co-evolution of the digital twin system by comparing the signal distribution of the measured value and the processed digital twin value, so as to enable the digital twin system to comprehensively and accurately reflect the operating status of the wind turbine.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A wind power digital twin co-evolution detection method based on signal distribution characteristics includes:

[0007] S1, acquire the fan signal data generated by the digital twin and the fan signal data collected by the sensor, wherein the fan signal data is one of the fan's temperature, radial vibration, axial vibration, voltage, current and power;

[0008] S2, preprocess the two types of wind turbine signal data respectively so that the characteristic frequency amplitude error of the preprocessed wind turbine signal data is less than a preset threshold.

[0009] S3, interpolates the two preprocessed fan signal data to the original data volume respectively;

[0010] S4, perform time series distribution transformation on the two interpolation results respectively to obtain the time series signal distribution probability;

[0011] S5 calculates the overall difference value based on the probability distribution of the two time series signals and outputs it as the twin co-evolution detection result.

[0012] Optionally, step S4 includes: dividing the wind turbine signal data into M regions between the maximum and minimum values, and calculating the probability distribution of the time-series signal in any m-th region according to the following formula:

[0013] ,

[0014] In the above formula, Let be the probability distribution of the time-series signal in the m-th region. Let m be the total number of data points in the m-th region. This represents the total number of data points in the entire wind turbine signal data.

[0015] Optionally, the functional expression for calculating the population difference value in step S5 is:

[0016] ,

[0017] In the above formula, The total difference value is M, where M is the number of regions between the maximum and minimum values ​​of the wind turbine signal data. The probability distribution of the time-series signal data of the wind turbine generated by the digital twin is in the 1st... The percentage of data for each region The probability distribution data of the time series signal corresponding to the wind turbine signal data collected by the sensor in the 1st... The percentage of data for each district.

[0018] Optionally, the calculation function expression for the characteristic frequency amplitude error in step S2 is:

[0019]

[0020] In the above formula, For characteristic frequency amplitude error, The characteristic frequency amplitude of the preprocessed wind turbine signal data. The characteristic frequency amplitude of the original wind turbine signal data.

[0021] Optionally, step S2 includes:

[0022] S2.1, Set the initial door width for the revolving door algorithm SDA;

[0023] S2.2, The rotating door algorithm SDA is used to process the two types of fan signal data respectively;

[0024] S2.3: Convert the two types of wind turbine signal data into frequency domain signals using Fast Fourier Transform (FFT) and extract the characteristic frequency amplitude of the original wind turbine signal data. Convert the processed data from step S2.2 into frequency domain signals using FFT and extract the characteristic frequency amplitude of the preprocessed wind turbine signal data. Calculate the characteristic frequency amplitude error of the two types of preprocessed wind turbine signal data based on the characteristic frequency amplitude of the preprocessed wind turbine signal data and the characteristic frequency amplitude of the original wind turbine signal data. If both characteristic frequency amplitude errors are less than a preset threshold, proceed to step S2.4; otherwise, update the door width of the revolving door algorithm SDA and proceed to step S2.2.

[0025] Optionally, after processing the two types of wind turbine vibration signal data using the revolving door algorithm SDA in step S2.2, the method further includes processing the two types of wind turbine vibration signal data using an anomaly algorithm, including, for each current data point, processing the data based on given anomaly parameters. The length formed above and below the current data point is... The system determines whether the next data point is abnormal based on the abnormal range. If the next data point exceeds the abnormal range, it is deleted, and the data point is reconstructed using linear interpolation based on the two preceding data points.

[0026] Optionally, after processing the two types of wind turbine vibration signal data using the revolving door algorithm SDA in step S2.2, the process further includes processing the two types of wind turbine vibration signal data using data segment merging, including detecting the increased uphill sections of data points based on changes in data points, and performing uphill section merging according to the following formula:

[0027] ,

[0028] In the above formula, The current moment; Total time; for The amplitude of the vibration signal at any given moment; for The amplitude of the vibration signal at any given moment; The data segments representing opposing trends where the amplitude between two identical trends is less than a preset threshold, and the data segments with small fluctuations are detected and merged to process the two types of wind turbine vibration signal data. This includes identifying data segments in the wind turbine vibration signal data that continuously show upward and downward changes over a time period, with the amplitude of both the upward and downward trends less than a preset threshold, as small fluctuation data segments. Its function expression is:

[0029] ;

[0030] In the above formula, ~ For small jitter data segments The first to c monotonic data segments, and any d-th monotonic data segment The function expression is:

[0031] ,

[0032] In the above formula, and The first A monotonic data segment The starting and ending points; for each small jitter data segment The system iterates through two adjacent monotonic data segments in turn, and calculates the equivalent revolving door start point and door width parameters for the small-amplitude jitter data segments according to the following formula:

[0033] ,

[0034] ,

[0035] ,

[0036] In the above formula, for The signal amplitude at that moment; For a specific point The amplitude; for The signal amplitude at that moment; The rotating door point of the rotating door algorithm SDA The signal amplitude; The current moment; For the gate point time; For a specific point The moment; The rotating door point of the rotating door algorithm SDA The moment; The lower support point of the rotating door in the rotating door algorithm SDA The amplitude; Support points on the rotating door of the rotating door algorithm SDA The moment; The lower support point of the rotating door in the rotating door algorithm SDA The moment; Support points on the rotating door of the rotating door algorithm SDA The amplitude; Gate width parameter for small jitter data segments The starting point signal amplitude; a specific point Set the door point of the rotating door algorithm SDA. ;

[0037] The signal amplitudes of the current two monotonic data segments are obtained based on the equivalent revolving door starting point and door width parameters detected by the small jitter data segments. upper revolving door signal amplitude and the amplitude of the lower revolving door signal Determine the detection conditions of the revolving door algorithm. If the condition is true, then merge the two monotonic data segments obtained from the traversal.

[0038] Furthermore, the present invention also provides a wind power digital twin co-evolution detection system based on signal distribution characteristics, comprising a microprocessor and a memory interconnected thereto, wherein the microprocessor is programmed or configured to execute the wind power digital twin co-evolution detection method based on signal distribution characteristics.

[0039] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program or instructions that are programmed or configured to execute the wind power digital twin co-evolution detection method based on signal distribution characteristics by a processor.

[0040] In addition, the present invention also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the wind power digital twin co-evolution detection method based on signal distribution characteristics by a processor.

[0041] Compared with existing technologies, the present invention has the following advantages: The present invention includes acquiring wind turbine signal data generated by digital twins and wind turbine signal data collected by sensors; preprocessing the two types of wind turbine signal data respectively so that the characteristic frequency amplitude error of the preprocessed wind turbine signal data is less than a preset threshold; interpolating the two types of preprocessed wind turbine signal data to the original data volume respectively; performing time-series sequence distribution transformation on the two interpolated results respectively to obtain the time-series sequence signal distribution probability; calculating the overall difference value based on the two time-series sequence signal distribution probabilities as the output of the twin co-evolution detection result. The present invention verifies the degree of co-evolution of the digital twin system by comparing the signal distribution after processing the measured value and the digital twin value, so that the digital twin system can comprehensively and accurately reflect the operating status of the wind turbine unit. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the basic process of the method in an embodiment of the present invention.

[0043] Figure 2 This is a schematic diagram of the rotating door algorithm SDA in an embodiment of the present invention.

[0044] Figure 3 This is a schematic diagram of the exception algorithm in an embodiment of the present invention.

[0045] Figure 4 This is a schematic diagram of data segment merging and wavelet dithering data segment processing in an embodiment of the present invention.

[0046] Figure 5 This is a schematic diagram of the original vibration signal data collected in an embodiment of the present invention.

[0047] Figure 6 This is a schematic diagram comparing the amount of data in an embodiment of the present invention.

[0048] Figure 7 The above are comparisons of time-domain signals after amplitude error preprocessing at different characteristic frequencies in embodiments of the present invention. (a) is the time-domain signal after preprocessing the original signal, (b) is the time-domain signal after preprocessing the signal with an amplitude error of 0.3%, (c) is the time-domain signal after preprocessing the signal with an amplitude error of 2.2%, and (d) is the time-domain signal after preprocessing the signal with an amplitude error of 4.0%.

[0049] Figure 8 This is a schematic diagram illustrating the process of interpolating wind turbine signal data to the original data volume in an embodiment of the present invention.

[0050] Figure 9 The data shown is the fan signal data before conversion in this embodiment of the invention.

[0051] Figure 10 The probability distribution of the time sequence signal obtained by conversion in the embodiments of the present invention is shown.

[0052] Figure 11 This is a comparison of the probability distribution of time-series signals obtained by converting two types of wind turbine signal data in an embodiment of the present invention.

[0053] Figure 12 This is a comparison of the probability distribution of the first set of time sequence signals for the axial vibration of the tower in this embodiment of the invention.

[0054] Figure 13 This is a comparison of the probability distribution of the second set of time sequence signals for the axial vibration of the tower in this embodiment of the invention. Detailed Implementation

[0055] like Figure 1 As shown, the wind power digital twin collaborative evolution detection method based on signal distribution characteristics in this embodiment includes:

[0056] S1, acquire the wind turbine signal data (digital twin data) generated by the digital twin and the wind turbine signal data (acquired data) collected by the sensor. The wind turbine signal data is one of the following: wind turbine temperature, radial vibration, axial vibration, voltage, current and power.

[0057] S2, preprocess the two types of wind turbine signal data respectively so that the characteristic frequency amplitude error of the preprocessed wind turbine signal data is less than a preset threshold.

[0058] S3, interpolates the two preprocessed fan signal data to the original data volume respectively;

[0059] S4, perform time series distribution transformation on the two interpolation results respectively to obtain the time series signal distribution probability;

[0060] S5 calculates the overall difference value based on the probability distribution of the two time series signals and outputs it as the twin co-evolution detection result.

[0061] In this embodiment, the calculation function expression for the characteristic frequency amplitude error in step S2 is as follows:

[0062]

[0063] In the above formula, For characteristic frequency amplitude error, The characteristic frequency amplitude of the preprocessed wind turbine signal data. The characteristic frequency amplitude of the original wind turbine signal data.

[0064] The Rotating Door Algorithm (SDA) is a well-known algorithm. Originally proposed for signal trend fitting, the SDA principle is as follows: Figure 2As shown, the measurement data consists of various measurement data points. During the measurement data point retrieval process, the rotating door algorithm (SDA) selects a gate width at the beginning of each iteration, and subsequent data points are connected to this gate width as the upper and lower boundaries. One boundary is fixed when it reaches its maximum value, while the other boundary continues to update as the data points change. When the two boundary lines are parallel, the current measurement data point is selected as the end point of the iteration, and a new iteration begins. The detection method of the rotating door algorithm for the current measurement data point is as follows:

[0065]

[0066] In the above formula, The signal amplitude at the current detection point; This refers to the signal amplitude of the upper revolving door. The amplitude of the lower rotating door signal. Step S2 in this embodiment includes:

[0067] S2.1, Set the initial door width for the revolving door algorithm SDA;

[0068] S2.2, The rotating door algorithm SDA is used to process the two types of fan signal data respectively;

[0069] S2.3: Convert the two types of wind turbine signal data into frequency domain signals using Fast Fourier Transform (FFT) and extract the characteristic frequency amplitude of the original wind turbine signal data. Convert the processed data from step S2.2 into frequency domain signals using FFT and extract the characteristic frequency amplitude of the preprocessed wind turbine signal data. Calculate the characteristic frequency amplitude error of the two types of preprocessed wind turbine signal data based on the characteristic frequency amplitude of the preprocessed wind turbine signal data and the characteristic frequency amplitude of the original wind turbine signal data. If both characteristic frequency amplitude errors are less than a preset threshold, proceed to step S2.4; otherwise, update the door width of the revolving door algorithm SDA and proceed to step S2.2.

[0070] As an optional implementation, in step S2.2 of this embodiment, after processing the two types of wind turbine vibration signal data using the revolving door algorithm SDA, the method further includes processing the two types of wind turbine vibration signal data using an anomaly algorithm, including, for each current data point, processing the data based on given anomaly parameters. The length formed above and below the current data point is... The algorithm determines whether the next data point is abnormal by considering the anomaly range. If the next data point exceeds the anomaly range, it is deleted, and the deleted data point is reconstructed using linear interpolation based on the two preceding data points. Anomaly algorithms can be used to identify erroneous data points after signal preprocessing by this type of method. Figure 3 As shown, abnormal parameters This determines the tolerance level of each data point for anomalies, based on the length formed above and below the current data point. The abnormal range is used to determine whether the next data point is abnormal. For example, in the figure, at the 3-second mark, the fourth data point is outside the abnormal range of the previous point, so it is considered an abnormal data point and is deleted. Then, based on the two data points preceding this point, linear interpolation is used to reconstruct the data point according to the linear relationship. The reconstructed data point is the reconstructed point.

[0071] After obtaining the preprocessed data segments using the Rotating Door Algorithm (SDA) and anomaly algorithms, some data segments still exhibit the same trend or continuous fluctuation. For these data segments, the parameter and resolution adaptive algorithm primarily handles data segment merging and small-amplitude jitter data segments. In step S2.2 of this embodiment, after processing the two types of wind turbine vibration signal data using the Rotating Door Algorithm (SDA), the algorithm further includes processing the two types of wind turbine vibration signal data using data segment merging. This includes detecting the increase in uphill sections based on changes in data points and performing uphill section merging according to the following formula:

[0072] ,

[0073] In the above formula, The current moment; Total time; for The amplitude of the vibration signal at any given moment; for The amplitude of the vibration signal at any given moment; A collision (BUMP) event refers to a data segment where the amplitude between two identical trends is less than a preset threshold. The data segment merging stage mainly handles redundant points that still exist after processing by the rotating door algorithm, and merges these points. Figure 4 This is a schematic diagram illustrating data segment merging and wavelet dithering data segment processing in this embodiment. Figure 4 Data points with an abscissa between 0s and 3.5s represent a continuously ascending slope. During the continuous forward detection, a downhill slope with an abscissa between 3.5s and 7s was found. Therefore, the data point at 3.5s is set as the end point of the previous uphill slope and possibly the start point of the next downhill slope. Similarly, the data point at 7s is also considered as the end point of the previous segment and possibly the start point of the next segment.

[0074] During the aforementioned merging process, some insignificant slope events are still retained, namely, continuous upward and downward changes over a period of time, with both upward and downward trend segments having small amplitudes. These data segments are defined as small-amplitude jitter data segments. Therefore, step S2.2 of this embodiment also includes using small-amplitude jitter data segment detection merging to process the two types of wind turbine detection signals separately, including: identifying data segments in the wind turbine vibration signal data that continuously exhibit upward and downward changes over a period of time, with both upward and downward trend segments having amplitudes less than a preset threshold, as small-amplitude jitter data segments. Its function expression is:

[0075] ;

[0076] In the above formula, ~ For small jitter data segments The first to c monotonic data segments, and any d-th monotonic data segment The function expression is:

[0077] ,

[0078] In the above formula, and The first A monotonic data segment The starting and ending points; for each small jitter data segment The system iterates through two adjacent monotonic data segments in turn, and calculates the equivalent revolving door start point and door width parameters for the small-amplitude jitter data segments according to the following formula:

[0079] ,

[0080] ,

[0081] ,

[0082] In the above formula, for The signal amplitude at that moment; For a specific point The amplitude; for The signal amplitude at that moment; The rotating door point of the rotating door algorithm SDA The signal amplitude; The current moment; For the gate point time; For a specific point The moment; The rotating door point of the rotating door algorithm SDA The moment; The lower support point of the rotating door in the rotating door algorithm SDA The amplitude; Support points on the rotating door of the rotating door algorithm SDA The moment; The lower support point of the rotating door in the rotating door algorithm SDA The moment; Support points on the rotating door of the rotating door algorithm SDA The amplitude; Gate width parameter for small jitter data segments The starting point signal amplitude; a specific point Set the door point of the rotating door algorithm SDA. ;

[0083] The signal amplitudes of the current two monotonic data segments are obtained based on the equivalent revolving door starting point and door width parameters detected by the small jitter data segments. upper revolving door signal amplitude and the amplitude of the lower revolving door signal Determine the detection conditions of the revolving door algorithm. If the condition is true, then merge the two monotonic data segments obtained from the traversal.

[0084] In step S2.3 of this embodiment, updating the gate width of the revolving door algorithm SDA means adding a preset gate width precision to the gate width of the revolving door algorithm SDA. For example, in this embodiment, the gate width precision is 0.0001, so the expression for updating the gate width of the revolving door algorithm SDA is: In the above formula, The door width is set for the updated revolving door algorithm SDA. Given the door width of the original revolving door algorithm SDA, then the updated door width of the revolving door algorithm SDA can be set. The door width as a new revolving door algorithm SDA Jump to step S2.2 and continue iteratively executing the rotating door algorithm SDA.

[0085] Figure 5 This is a schematic diagram of the raw vibration signal data collected in this embodiment. Figure 5 As shown, in this embodiment, a vibration signal with 10,000 data points is selected, and the compression of the data volume before and after processing is investigated by using the amplitude error of the characteristic frequency of different signal frequencies. Figure 6 This is a diagram illustrating the comparison of data volume in this embodiment, such as... Figure 6As shown, when the characteristic frequency amplitude error begins to increase, the data volume begins to decrease sharply, but when the error reaches 0.3%, the change in data volume is not significant. Since the signal frequency domain consists of multiple characteristic frequencies with different amplitudes, when the gate width is greater than a certain value, characteristic frequencies with corresponding amplitudes will be processed. Figure 6 As the error increases, a "plateau phenomenon" occurs in the amount of data.

[0086] Figure 7 This document compares the time-domain signals after preprocessing amplitude errors at different characteristic frequencies in this embodiment of the invention. (a) is the time-domain signal after preprocessing the original signal; (b) is the time-domain signal after preprocessing the signal with an amplitude error of 0.3%; (c) is the time-domain signal after preprocessing the signal with an amplitude error of 2.2%; and (d) is the time-domain signal after preprocessing the signal with an amplitude error of 4.0%. See also... Figure 7 It is known that as the characteristic frequency amplitude error increases, the vibration signal gradually becomes sparse. While reducing noise, fault information is also continuously lost. Therefore, considering both the reduction in data volume and the preservation of fault features, this paper selects a frequency domain characteristic frequency amplitude error of 0.3% as the threshold for adaptive gate width selection. Step S2 uses the rotating door algorithm SDA to preprocess the two types of fan detection signals, aiming to reduce the characteristic frequency amplitude error of the preprocessed fan detection signal. Less than 0.3%.

[0087] like Figure 8 As shown, in step S3 of this embodiment, interpolating the preprocessed wind turbine signal data to the original data includes:

[0088] S3.1 Input the pre-processed signal, i.e., the pre-processed fan signal data;

[0089] S3.2, Locate the missing point;

[0090] S3.3, Create a complete sequence range corresponding to the original data volume;

[0091] S3.4, Fill the known data into the interpolation array;

[0092] S3.5 combines the interpolated data with the complete sequence number.

[0093] In step S3.4 of this embodiment, when filling the known data into the interpolation array, the interpolation array uses cubic spline interpolation, and the following conditions must be met:

[0094] Condition 1: At each node, the spline function is equal to the value of the data point.

[0095] and ;

[0096] In the above formula, For the first Segment spline function at nodes The value at that location, For data points at nodes The value at that location, For the first Segment spline function at nodes The value at that location, For data points at nodes The values ​​at the points are: nodes, which are data points. Before interpolation, these values ​​are preprocessed by algorithms such as rotating gates and are discontinuous. Spline interpolation is used to fill in the missing data. A spline function is composed of polynomials, each determined by two adjacent data points. Any two adjacent polynomials and their derivatives are continuous at the connection point.

[0097] Condition 2: At each internal node, the first and second derivatives of the spline are continuous:

[0098] and ;

[0099] In the above formula, and They are respectively The first and second derivatives, and for The first and second derivatives, For the first Segment spline function at nodes The value at;

[0100] Condition 3: The second derivative is zero at the boundary:

[0101] and ;

[0102] In the above formula, For the function at the node The second derivative, For the (n-1)th segment of the spline function at the node The second derivative, This is the starting node of the spline. This is the termination node of the spline.

[0103] The polynomial for cubic spline interpolation can be written as:

[0104] ,

[0105] In the above formula, For the first Segment spline function at nodes The value, For the first The constant term coefficient of the segment spline function, For the first The coefficients of the linear term of the segment spline function, For the first The coefficients of the quadratic term in a segment spline function. For the first The coefficients of the cubic term in a segment spline function, For the first The starting point of the segment spline.

[0106] In this embodiment, step S4 includes: dividing the wind turbine signal data into M regions between the maximum and minimum values, and calculating the probability distribution of the time-series signal in any m-th region according to the following formula:

[0107] ,

[0108] In the above formula, Let be the probability distribution of the time-series signal in the m-th region. Let m be the total number of data points in the m-th region. This represents the total number of data points in the entire wind turbine signal data. In this embodiment, the wind turbine signal data before conversion in step S4 is as follows: Figure 9 As shown, the probability distribution of the converted time-series signal is as follows: Figure 10 As shown. Figure 11 This invention presents a comparison of the probability distributions of time-series signals obtained from the conversion of two types of wind turbine signal data in an embodiment of the invention. Figures 9-11 As can be seen, the method in this embodiment can greatly reduce the amount of data processing, and is especially suitable for wind turbine signal data of digital twin systems with high sampling rates.

[0109] In this embodiment, the functional expression for calculating the overall difference value in step S5 is:

[0110] ,

[0111] In the above formula, The total difference value is M, where M is the number of regions between the maximum and minimum values ​​of the wind turbine signal data. The probability distribution of the time-series signal data of the wind turbine generated by the digital twin is in the 1st... The percentage of data for each region The probability distribution data of the time series signal corresponding to the wind turbine signal data collected by the sensor in the 1st... The data proportion of each region. The overall difference value TD can effectively quantify the performance data of wind power digital twin co-evolution. For example, in this embodiment, tower axial vibration is used as wind turbine detection data. The probability distribution of time series signals obtained from the wind turbine signal data generated by the two sets of digital twins (digital twin data) and the wind turbine signal data collected by the sensors (collected data) is compared. Figure 12 and Figure 13 As shown, for Figure 12 For the first set of data, the calculated population variance value (TD) is 0.0000. Regarding... Figure 13 The second set of data yielded an overall difference value (TD) of 0.0750.

[0112] In summary, the wind power digital twin co-evolution detection method based on signal distribution characteristics in this embodiment includes acquiring wind turbine signal data generated by the digital twin and wind turbine signal data collected by sensors; preprocessing the two types of wind turbine signal data to ensure that the characteristic frequency amplitude error of the preprocessed wind turbine signal data is less than a preset threshold; interpolating the two types of preprocessed wind turbine signal data to the original data volume; performing time-series sequence distribution transformation on the two interpolated results to obtain the time-series sequence signal distribution probability; and calculating the overall difference value based on the two time-series sequence signal distribution probabilities as the output of the twin co-evolution detection result. This embodiment compares the time-series sequence signal distribution probabilities obtained by transforming the two types of wind turbine signal data. This method can greatly reduce the amount of data processing, and is especially suitable for wind turbine signal data of high sampling rate digital twin systems. By comparing the signal distribution of the measured values ​​and the processed digital twin values, the degree of co-evolution of the digital twin system is verified, so that the digital twin system can comprehensively and accurately reflect the operating status of the wind turbine unit.

[0113] Furthermore, this embodiment also provides a wind power digital twin co-evolution detection system based on signal distribution characteristics, including a microprocessor and a memory interconnected, wherein the microprocessor is programmed or configured to execute the wind power digital twin co-evolution detection method based on signal distribution characteristics.

[0114] In addition, this embodiment also provides a computer-readable storage medium storing a computer program or instructions that are programmed or configured to execute the wind power digital twin co-evolution detection method based on signal distribution characteristics by a processor.

[0115] In addition, this embodiment also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the wind power digital twin co-evolution detection method based on signal distribution characteristics through a processor.

[0116] Those skilled in the art will understand that the technical solutions provided by the embodiments of this application may be in the form of a method, system, or computer program product. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product embodied on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create an implementation for the process. Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0117] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A wind power digital twin collaborative evolution detection method based on signal distribution characteristics, characterized in that, include: S1, acquire the fan signal data generated by the digital twin and the fan signal data collected by the sensor, wherein the fan signal data is one of the fan's temperature, radial vibration, axial vibration, voltage, current and power; S2, preprocess the two types of wind turbine signal data respectively so that the characteristic frequency amplitude error of the preprocessed wind turbine signal data is less than a preset threshold. S3, interpolates the two preprocessed fan signal data to the original data volume respectively; S4, perform time series distribution transformation on the two interpolation results respectively to obtain the time series signal distribution probability; S5, calculate the overall difference value based on the probability distribution of the two time series signals and output it as the twin co-evolution detection result; Step S2 includes: S2.1, Set the initial door width for the revolving door algorithm SDA; S2.2, The rotating door algorithm SDA is used to process the two types of fan signal data respectively; S2.3: Convert the two types of wind turbine signal data into frequency domain signals using Fast Fourier Transform (FFT) and extract the characteristic frequency amplitude of the original wind turbine signal data. Convert the processed data from step S2.2 into frequency domain signals using FFT and extract the characteristic frequency amplitude of the preprocessed wind turbine signal data. Calculate the characteristic frequency amplitude error of the two types of preprocessed wind turbine signal data based on the characteristic frequency amplitude of the preprocessed wind turbine signal data and the characteristic frequency amplitude of the original wind turbine signal data. If both characteristic frequency amplitude errors are less than a preset threshold, proceed to step S3; otherwise, update the door width of the revolving door algorithm SDA and proceed to step S2.

2.

2. The wind power digital twin collaborative evolution detection method based on signal distribution characteristics according to claim 1, characterized in that, Step S4 includes: dividing the wind turbine signal data into M regions between the maximum and minimum values, and calculating the probability distribution of the time series signal in any m-th region according to the following formula: ， In the above formula, Let be the probability distribution of the time-series signal in the m-th region. Let m be the total number of data points in the m-th region. This represents the total number of data points in the entire wind turbine signal data.

3. The wind power digital twin collaborative evolution detection method based on signal distribution characteristics according to claim 2, characterized in that, The functional expression for calculating the population difference value in step S5 is: ， In the above formula, The total difference value is M, where M is the number of regions between the maximum and minimum values ​​of the wind turbine signal data. The probability distribution of the time-series signal data of the wind turbine generated by the digital twin is in the 1st... The percentage of data for each region The probability distribution data of the time series signal corresponding to the wind turbine signal data collected by the sensor in the 1st... The percentage of data for each district.

4. The wind power digital twin collaborative evolution detection method based on signal distribution characteristics according to claim 1, characterized in that, The expression for the calculation function of the characteristic frequency amplitude error in step S2 is: In the above formula, For characteristic frequency amplitude error, The characteristic frequency amplitude of the preprocessed wind turbine signal data. The characteristic frequency amplitude of the original wind turbine signal data.

5. The wind power digital twin collaborative evolution detection method based on signal distribution characteristics according to claim 1, characterized in that, After processing the two types of wind turbine vibration signal data using the Revolving Door Algorithm (SDA) in step S2.2, the process also includes processing the two types of wind turbine vibration signal data using an anomaly algorithm, including, for each current data point, processing the data based on given anomaly parameters. The length formed above and below the current data point is... The system determines whether the next data point is abnormal based on the abnormal range. If the next data point exceeds the abnormal range, it is deleted, and the data point is reconstructed using linear interpolation based on the two preceding data points.

6. The wind power digital twin collaborative evolution detection method based on signal distribution characteristics according to claim 1, characterized in that, Step S2.2, after processing the two types of wind turbine vibration signal data using the Revolving Door Algorithm (SDA), also includes processing the two types of wind turbine vibration signal data using data segment merging. This includes detecting the increased uphill sections of data points based on changes in data points, and performing uphill section merging according to the following formula: ， In the above formula, The current moment; Total time; for The amplitude of the vibration signal at any given moment; for The amplitude of the vibration signal at any given moment; The data segments representing opposing trends where the amplitude between two identical trends is less than a preset threshold, and the data segments with small fluctuations are detected and merged to process the two types of wind turbine vibration signal data. This includes identifying data segments in the wind turbine vibration signal data that continuously show upward and downward changes over a time period, with the amplitude of both the upward and downward trends less than a preset threshold, as small fluctuation data segments. Its function expression is: ； In the above formula, ~ For small jitter data segments The first to c monotonic data segments, and any d-th monotonic data segment The function expression is: ， In the above formula, and The first A monotonic data segment The starting and ending points; for each small jitter data segment The system iterates through two adjacent monotonic data segments in turn, and calculates the equivalent revolving door start point and door width parameters for the small-amplitude jitter data segments according to the following formula: ， ， ， In the above formula, for The signal amplitude at that moment; For a specific point The amplitude; for The signal amplitude at that moment; The rotating door point of the rotating door algorithm SDA The signal amplitude; The current moment; For the gate point time; For a specific point The moment; The rotating door point of the rotating door algorithm SDA The moment; The lower support point of the rotating door in the rotating door algorithm SDA The amplitude; Support points on the rotating door of the rotating door algorithm SDA The moment; The lower support point of the rotating door in the rotating door algorithm SDA The moment; Support points on the rotating door of the rotating door algorithm SDA The amplitude; Gate width parameter for small jitter data segments The starting point signal amplitude; a specific point Set the door point of the rotating door algorithm SDA. ; The signal amplitudes of the current two monotonic data segments are obtained based on the equivalent revolving door starting point and door width parameters detected by the small jitter data segments. upper revolving door signal amplitude and the amplitude of the lower revolving door signal Determine the detection conditions of the revolving door algorithm. If the condition is true, then merge the two monotonic data segments obtained from the traversal.

7. A wind power digital twin collaborative evolution detection system based on signal distribution characteristics, comprising a microprocessor and a memory interconnected, characterized in that, The microprocessor is programmed or configured to execute the wind power digital twin co-evolution detection method based on signal distribution characteristics as described in any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program or instructions, characterized in that, The computer program or instructions are programmed or configured to execute, via a processor, the wind power digital twin co-evolution detection method based on signal distribution characteristics as described in any one of claims 1 to 6.

9. A computer program product, comprising a computer program or instructions, characterized in that, The computer program or instructions are programmed or configured to execute, via a processor, the wind power digital twin co-evolution detection method based on signal distribution characteristics as described in any one of claims 1 to 6.

Images