MMC converter fault diagnosis method and system based on horse field system

Abstract

The invention discloses an MMC current converter fault diagnosis method based on a horse field system. According to the method, through collected MMC converter submodule voltage characteristic data, submodule voltage data is extracted, a fault diagnosis characteristic system is constructed, and the mahalanobis distance of the data is calculated. And constructing a reference space of the horse field system by using the calculated mahalanobis distance, and then carrying out validity confirmation on the constructed reference space to judge whether the conditions can be met or not. The construction of an orthogonal table, the signal-to-noise ratio and the signal-to-noise ratio gain are introduced, and the reference space is optimized. And finally, calculating unknown data by introducing calculation of a threshold value, and obtaining whether the MMC converter sub-module has an open-circuit fault or not. The fault of the MMC can be diagnosed through the horse field system, the problem that the operation process is too long due to too many parameters is solved, and due to the fact that the Mahalanobis distance is introduced, data can be converted into dimensionless data, and the final fault diagnosis can be more accurate.

Description

Technical Field

[0001] This invention relates to the technical field of open-circuit fault diagnosis of MMC converter submodules, and particularly to a fault diagnosis method and system for MMC converters based on the Martin system. Background Technology

[0002] With in-depth research on MMC converter systems, various control technologies have been proposed, but improving system safety remains the primary goal of controller design. MMC converters possess a large number of submodules and power switching devices, each of which is a potential point of failure. While the series connection of numerous identical submodules increases the difficulty of system fault detection, it is also its advantage. The multi-modal structure allows a faulty MMC converter to operate in an internal fault state for a short period, using redundant submodules to replace the faulty point and avoid permanent system failure. However, to fully leverage the advantages of MMC converters, the fault diagnosis system design must accomplish the following: monitoring during safe operation, eliminating the need for shutdown during fault conditions, quickly determining the system fault state, locating the faulty submodule, bypassing it, and then activating redundant modules. Summary of the Invention

[0003] In view of the aforementioned existing problems, the present invention is proposed.

[0004] Therefore, this invention provides a fault diagnosis method for MMC converters based on the Martin system. By utilizing selected key variables to diagnose open-circuit faults in sub-modules of the MMC converter, the method reduces the slow diagnosis caused by too many feature variables and improves the accuracy of diagnosis.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: a fault diagnosis method for MMC converters based on the Martin system, comprising: extracting submodule voltage data from collected MMC converter submodule voltage characteristic data, constructing a fault diagnosis feature system, and calculating the Mahalanobis distance of the data; constructing a Martin system reference space using the calculated Mahalanobis distance, and verifying the validity of the constructed reference space; optimizing the reference space by introducing the construction of orthogonal arrays, signal-to-noise ratio, and signal-to-noise ratio gain; and calculating unknown data by introducing threshold calculation to obtain fault diagnosis of the MMC converter submodule.

[0006] As a preferred embodiment of the fault diagnosis method for MMC converters based on the Martin system described in this invention, the voltage characteristic data of the MMC converter submodules includes dimensional data of the maximum, minimum, mean, peak ripple, and effective values ​​of the MMC submodule voltages, as well as dimensionless data of variance, crest factor, margin factor, waveform factor, impulse factor, skewness factor, and kurtosis factor.

[0007] As a preferred embodiment of the fault diagnosis method for MMC converters based on the Martin system described in this invention, wherein: the Martin distance of the calculated data includes, let the unknown MMC voltage data sample be Z'=(z 10 ,z 20 ,…,z k0 ), ' represents the transpose of the matrix. Before determining the affiliation of a sample, the samples are standardized, that is, Z' is transformed into T' = (t 10 , t 20 , ..., t k0 The calculation formula is:

[0008]

[0009] The Mahalanobis distance from the voltage characteristic samples of the MMC submodule to the reference space is calculated as follows:

[0010]

[0011] Where k represents the number of samples.

[0012] As a preferred embodiment of the fault diagnosis method for MMC converters based on the Martin system described in this invention, the construction of the Martin system reference space includes: calculating the sample mean and root mean square error of the voltage characteristic indicators of each MMC submodule, standardizing the data, and dividing the feature X... i Convert to Y i Calculate the correlation coefficient matrix between standardized features; calculate the inverse of the correlation coefficient matrix.

[0013] The sample mean and standard deviation of the voltage characteristic indicators of each MMC submodule:

[0014]

[0015]

[0016] in, x represents the sample mean. ij σ represents the voltage value of the j-th sample point in the i-th submodule, n represents the total number of sample points, and σ represents the voltage value of the j-th sample point in the i-th submodule. i Indicates the sample mean squared error. This represents the sample mean of the voltage of the i-th submodule;

[0017] The data standardization refers to standardizing the data under each voltage characteristic in the reference space, that is, X... i :N(μ i ,σ i ) Convert to Y i :N(0,1), the calculation formula is:

[0018]

[0019] As a preferred embodiment of the MMC converter fault diagnosis method based on the Martin system described in this invention, the construction of the Martin system reference space further includes a correlation coefficient matrix between the standardized features:

[0020]

[0021]

[0022] The calculation of the standardized feature Y i Correlation matrix between (i = 1, ..., k):

[0023]

[0024]

[0025] Among them, R ij This represents the correlation coefficient between the i-th voltage feature and the i'-th voltage feature;

[0026] For samples belonging to the baseline space, the Mahalanobis distance is less than or equal to 1, and at most does not exceed a certain threshold; for samples not belonging to the baseline space, the Mahalanobis distance has a larger value, thereby determining whether the baseline space can effectively distinguish abnormal samples in the MMC module outside the baseline space.

[0027] As a preferred embodiment of the fault diagnosis method for MMC converters based on the Martin system described in this invention, the optimization of the reference space includes: selecting an appropriate two-level orthogonal table based on the number of features, using the feature as the first level to indicate "using the feature", and using the feature as the second level to indicate "not using the feature"; arranging the feature as a controllable factor inside the table, and arranging d samples that do not belong to the reference space outside the table;

[0028] Calculate the Mahalanobis distance from the reference point in the reference space under different voltage characteristic combinations under abnormal conditions, and calculate the S / N ratio for each voltage characteristic combination. The S / N ratio formula is as follows:

[0029]

[0030] Where l represents the number of observations;

[0031] Range analysis was used to screen variables and finally determine the effective combination of features to form a new reference space. The validity of the newly constructed reference space was confirmed again, and abnormal voltage data of MMC submodules outside the reference space were collected again to monitor the validity of the newly built reference space.

[0032] As a preferred embodiment of the fault diagnosis method for MMC converters based on the Martin system described in this invention, the fault diagnosis includes determining the relationship between unknown samples and the reference space.

[0033] The category of unknown MMC submodule data samples can be determined based on the preliminary threshold calculation formula, which is as follows:

[0034]

[0035] Among them, MD n(max) MD represents the maximum Mahalanobis distance in normal samples of the MMC submodule. a(min) This represents the minimum Mahalanobis distance among abnormal samples in the MMC submodule. Different thresholds will be given for different problems. If the Mahalanobis distance is greater than the threshold, it is determined that the sample does not belong to the population or has a low probability of belonging to the population. If it exceeds the threshold, it means that the MMC submodule has an open circuit fault. At this time, an early warning or shutdown of the converter will be issued.

[0036] Another objective of this invention is to provide a fault diagnosis system for MMC converters based on the Martin system, which can solve the problem that when there are too many characteristic variables in the collected historical operating data of MMC, it will affect the overall diagnostic operation speed, and when the data changes too much, it will cause faults. This system can diagnose MMC faults more accurately.

[0037] As a preferred embodiment of the MMC converter fault diagnosis system based on the Martin system described in this invention, it includes: an operating data extraction module, a reference space construction module, an effectiveness verification module, a reference space optimization module, a threshold determination module, and a fault diagnosis module.

[0038] The data extraction module is responsible for extracting voltage characteristic data of submodules from the MMC converter;

[0039] The reference space construction module uses the extracted voltage feature data and the calculated Mahalanobis distance to construct the reference space of the Martin system. The reference space is used as a standard framework for comparing and judging normal and fault states.

[0040] The validity verification module verifies the validity of the constructed benchmark space to ensure it meets the requirements of the diagnostic system.

[0041] The reference space optimization module introduces concepts such as orthogonal arrays, signal-to-noise ratio (SNR), and SNR gain to filter selected features and optimize the reference space.

[0042] The threshold determination module calculates and determines the threshold, which is the standard for judging whether unknown data indicates that the submodule has an open circuit fault.

[0043] The fault diagnosis module uses the aforementioned constructed and optimized baseline space and thresholds to perform fault diagnosis on the sub-modules of the MMC converter.

[0044] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement the steps of a fault diagnosis method for an MMC converter based on the Martin system.

[0045] A computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of a fault diagnosis method for an MMC converter based on the Martin system.

[0046] The beneficial effects of this invention are as follows: the Martin system can diagnose faults in MMC, solving the problem of excessive parameters causing excessively long operation time. By introducing the Martin distance, the data can be converted into dimensionless data, making the final fault diagnosis more accurate. Compared with other fault diagnosis methods, the Martin system method based on this invention has the advantages of avoiding the influence of irrelevant features, resulting in high diagnostic accuracy and faster operation with fewer features. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0048] Figure 1 This is a schematic flowchart of a fault diagnosis method for an MMC converter based on the Martin system, provided as an embodiment of the present invention.

[0049] Figure 2 This is a flowchart illustrating the operation of an MMC converter fault diagnosis method based on the Martin system, provided as an embodiment of the present invention, during fault diagnosis.

[0050] Figure 3 This invention provides an embodiment of a fault diagnosis method for an MMC converter based on the Martin system, showing the operation of an open-circuit fault in an MMC submodule.

[0051] Figure 4 This is a schematic diagram of the working module of an MMC converter fault diagnosis system based on the Martin system, provided as an embodiment of the present invention. Detailed Implementation

[0052] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0053] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0054] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0055] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0056] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0057] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0058] Example 1

[0059] Reference Figures 1-3 This is the first embodiment of the present invention, which provides a fault diagnosis method for MMC converters based on the Martin system, including:

[0060] S1: By collecting the voltage characteristic data of the MMC converter submodules, extract the submodule voltage data, construct a fault diagnosis feature system, and calculate the Mahalanobis distance of the data.

[0061] Furthermore, the voltage characteristic data of the MMC converter submodule includes dimensional data of the maximum, minimum, average, peak ripple, and effective values ​​of the MMC submodule voltage, as well as dimensionless data of variance, crest factor, margin factor, waveform factor, impulse factor, skewness factor, and kurtosis factor.

[0062] It should be noted that the Mahalanobis distance of the calculated data includes, let the unknown MMC voltage data sample be Z' = (z 10 ,z 20 ,…,z k0 Z' represents the transpose of the matrix. Before determining the affiliation of a sample, the samples are standardized, i.e., Z' is transformed into T' = (t...). 10 , t 20 , ..., t k0 The calculation formula is:

[0063]

[0064] The Mahalanobis distance from the voltage characteristic samples of the MMC submodule to the reference space is calculated as follows:

[0065]

[0066] Where k represents the number of samples.

[0067] S2: Construct the Martin system reference space using the calculated Mahalanobis distance, and verify the validity of the constructed reference space.

[0068] Furthermore, constructing the baseline space begins with data collection, specifically data from the submodules of MMC operating normally. The data structure table for the baseline space is shown in Table 1.

[0069] Table 1 Reference Space Data Table

[0070]

[0071] X in Table 1 j (i = 1, ..., k) refers to the measurement items. ij (i = 1, ..., k; j = 1, ..., n) represents the measurement data of the sample under each measurement item. The reference space is composed of healthy samples from these characteristic data. Generally, if the number of variables k < 3, calculating the Mahalanobis distance will be meaningless. However, if k is too large, it will lead to excessively long calculation time. The data characteristics of the reference space are calculated, and the collected data information is processed uniformly. Table 2 shows the voltage characteristics of the MMC submodules that need to be collected.

[0072] Table 2 Voltage Characteristic Expressions of MMC Submodules

[0073]

[0074]

[0075] Furthermore, the construction of the Martin system reference space includes calculating the sample mean and root mean square error of the voltage characteristic indicators of each MMC submodule, standardizing the data, and converting the feature X... i Convert to Y i Calculate the correlation coefficient matrix between standardized features; calculate the inverse of the correlation coefficient matrix.

[0076] The sample mean and standard deviation of the voltage characteristic indicators of each MMC submodule:

[0077]

[0078]

[0079] in, x represents the sample mean. ij σ represents the voltage value of the j-th sample point in the i-th submodule, n represents the total number of sample points, and σ represents the voltage value of the j-th sample point in the i-th submodule. i Indicates the sample mean squared error. This represents the sample mean of the voltage of the i-th submodule;

[0080] The data standardization refers to standardizing the data under each voltage characteristic in the reference space, that is, X... i :N(μ i ,σ i ) Convert to Y i :N(0,1), the calculation formula is:

[0081]

[0082] Furthermore, the correlation coefficient matrix between the standardized features:

[0083]

[0084]

[0085] The calculation of the standardized feature Y i Correlation matrix between (i = 1, ..., k):

[0086]

[0087]

[0088] Among them, R ij This represents the correlation coefficient between the i-th voltage feature and the i'-th voltage feature;

[0089] For samples belonging to the baseline space, the Mahalanobis distance is less than or equal to 1, and at most does not exceed a certain threshold; for samples not belonging to the baseline space, the Mahalanobis distance has a larger value, thereby determining whether the baseline space can effectively distinguish abnormal samples in the MMC module outside the baseline space.

[0090] S3: The reference space is optimized by introducing the construction of orthogonal arrays, signal-to-noise ratio and signal-to-noise ratio gain.

[0091] Furthermore, the optimization of the reference space includes selecting an appropriate two-level orthogonal table based on the number of features, using the feature as the first level to indicate "using the feature" and using the feature as the second level to indicate "not using the feature"; arranging the feature as a controllable factor inside the table and arranging d samples that do not belong to the reference space outside the table; Table 3 shows a schematic diagram of the inner table design.

[0092] Table 3. Inner table design for baseline space optimization

[0093]

[0094] Each experiment in the table defines a set of feature combinations. For example, experiment number 3 in the table means that the original five features in the baseline space are reduced to three features ABE. If this combination is verified to be optimal, then only the combination of features ABE will be used when making predictions and evaluations based on the baseline space. The effectiveness of the feature combinations will be tested below.

[0095] Calculate the Mahalanobis distance from the reference point in the reference space under different voltage characteristic combinations under abnormal conditions, and calculate the S / N ratio for each voltage characteristic combination. The S / N ratio formula is as follows:

[0096]

[0097] Where l represents the number of observations;

[0098] Range analysis was used to screen variables and finally determine the effective combination of features to form a new reference space. The validity of the newly constructed reference space was confirmed again, and abnormal voltage data of MMC submodules outside the reference space were collected again to monitor the validity of the newly built reference space.

[0099] S4: The threshold calculation is introduced to calculate unknown data and obtain fault diagnosis of MMC converter submodule.

[0100] Furthermore, the fault diagnosis includes determining the relationship between unknown samples and the baseline space; and classifying unknown MMC submodule data samples according to a preliminary threshold calculation formula, the formula of which is:

[0101]

[0102] Among them, MD n(max) MD represents the maximum Mahalanobis distance in normal samples of the MMC submodule. a(min) This represents the minimum Mahalanobis distance among abnormal samples in the MMC submodule. Different thresholds will be given for different problems. If the Mahalanobis distance is greater than the threshold, it is determined that the sample does not belong to the population or has a low probability of belonging to the population. If it exceeds the threshold, it means that the MMC submodule has an open circuit fault. At this time, an early warning or shutdown of the converter will be issued.

[0103] Example 2

[0104] One embodiment of the present invention provides a fault diagnosis method for MMC converters based on the Martin system. To verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0105] Submodule voltage data of a certain MMC converter during normal operation and fault operation were collected. This data includes 100 sets of data during normal operation and 50 sets of data during fault operation, with the fault type being submodule open circuit fault. The data were processed and analyzed according to the method of this invention.

[0106] This invention addresses the problem of excessively long operation times due to too many parameters. By introducing Mahalanobis distance, it transforms data into dimensionless data, making the fault diagnosis process simpler and more efficient. It also improves the accuracy of fault diagnosis. By constructing a Mahalanobis system reference space and filtering and optimizing selected features, this invention can more accurately diagnose faults in MMC converters. Furthermore, the introduction of threshold calculation makes fault diagnosis results more explicit and reliable. By setting reasonable thresholds, this invention can effectively distinguish between normal operation and fault operation data.

[0107] Table 4

[0108]

[0109]

[0110] By comparing normal and fault data, it can be found that the method of the present invention can effectively diagnose open-circuit faults in MMC converter submodules. In the normal data, the Mahalanobis distance of all data is less than the threshold of 0.8, and therefore it is correctly identified as normal operation. In the fault data, the Mahalanobis distance of all data is greater than the threshold of 0.8, and therefore it is correctly identified as fault operation. This indicates that the method of the present invention has high accuracy and reliability.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0112] Example 3

[0113] The third embodiment of the present invention differs from the first two embodiments in that:

[0114] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0115] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0116] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0117] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0118] Example 4

[0119] Reference Figure 4 As an embodiment of the present invention, a fault diagnosis system for MMC converters based on the Martin system is provided, characterized in that it includes an operating data extraction module, a reference space construction module, an effectiveness verification module, a reference space optimization module, a threshold determination module, and a fault diagnosis module;

[0120] The data extraction module is responsible for extracting voltage characteristic data of submodules from the MMC converter.

[0121] The reference space construction module uses extracted voltage feature data and calculated Mahalanobis distance to construct the reference space of the Martin system. The reference space is used as a standard framework for comparing and judging normal and fault states.

[0122] The validity verification module verifies the validity of the constructed baseline space to ensure it meets the requirements of the diagnostic system.

[0123] The reference space optimization module introduces concepts such as orthogonal arrays, signal-to-noise ratio (SNR), and SNR gain to filter selected features and optimize the reference space.

[0124] The threshold determination module calculates and determines the threshold, which is the standard for judging whether unknown data indicates that the submodule has an open circuit fault.

[0125] The fault diagnosis module uses the aforementioned constructed and optimized baseline space and thresholds to perform fault diagnosis on the sub-modules of the MMC converter.

[0126] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A fault diagnosis method for MMC converters based on the Martin system, characterized in that: include, By collecting the voltage characteristic data of the MMC converter submodules, the submodule voltage data is extracted, a fault diagnosis feature system is constructed, and the Mahalanobis distance of the data is calculated. The Martin system reference space is constructed using the calculated Mahalanobis distance, and the validity of the constructed reference space is verified. The reference space is optimized by introducing the construction of orthogonal arrays, signal-to-noise ratio, and signal-to-noise ratio gain; The introduction of threshold calculations enables the calculation of unknown data and the generation of fault diagnoses for MMC converter submodules.

2. The fault diagnosis method for MMC converters based on the Martin system as described in claim 1, characterized in that: The voltage characteristic data of the MMC converter submodule includes dimensional data of the maximum, minimum, average, peak ripple, and effective values ​​of the MMC submodule voltage, as well as dimensionless data of variance, crest factor, margin factor, waveform factor, impulse factor, skewness factor, and kurtosis factor.

3. The fault diagnosis method for MMC converters based on the Martin system as described in claim 2, characterized in that: The Mahalanobis distance of the calculated data includes, let the unknown MMC voltage data sample be Z'=(z 10 ,z 20 ,…,z k0 Z' represents the transpose of the matrix. Before determining the affiliation of a sample, the samples are standardized, i.e., Z' is transformed into T' = (t...). 10 ,t 20 ,…,t k0 The calculation formula is: The Mahalanobis distance from the voltage characteristic samples of the MMC submodule to the reference space is calculated as follows: Where k represents the number of samples.

4. The fault diagnosis method for MMC converters based on the Martin system as described in claim 3, characterized in that: The construction of the Martin system reference space includes calculating the sample mean and root mean square error of the voltage characteristic indicators of each MMC submodule, standardizing the data, and transforming the characteristic X... i Convert to Y i Calculate the correlation coefficient matrix between standardized features; calculate the inverse of the correlation coefficient matrix. The sample mean and standard deviation of the voltage characteristic indicators of each MMC submodule: in, x represents the sample mean. ij σ represents the voltage value of the j-th sample point in the i-th submodule, n represents the total number of sample points, and σ represents the voltage value of the j-th sample point in the i-th submodule. i Indicates the sample mean squared error. This represents the sample mean of the voltage of the i-th submodule; The data standardization refers to standardizing the data under each voltage characteristic in the reference space, that is, X... i :N(μ i ,σ i ) Convert to Y i :N(0,1), the calculation formula is:

5. The fault diagnosis method for MMC converters based on the Martin system as described in claim 4, characterized in that: The construction of the Martin system baseline space also includes the correlation coefficient matrix between the standardized features: The calculation of the standardized feature Y i Correlation matrix between (i = 1, ..., k): Among them, R ij This represents the correlation coefficient between the i-th voltage feature and the i'-th voltage feature; For samples belonging to the baseline space, the Mahalanobis distance is less than or equal to 1, and at most does not exceed a certain threshold; for samples not belonging to the baseline space, the Mahalanobis distance has a larger value, thereby determining whether the baseline space can effectively distinguish abnormal samples in the MMC module outside the baseline space.

6. The fault diagnosis method for MMC converters based on the Martin system as described in claim 5, characterized in that: The optimization of the baseline space includes selecting an appropriate two-level orthogonal table based on the number of features, using the features as the first level to represent "using features" and using the features as the second level to represent "not using features"; arranging the features as controllable factors inside the table, and arranging d samples that do not belong to the baseline space outside the table; Calculate the Mahalanobis distance from the reference point in the reference space under different voltage characteristic combinations under abnormal conditions, and calculate the S / N ratio for each voltage characteristic combination. The S / N ratio formula is as follows: Where l represents the number of observations; Range analysis was used to screen variables and finally determine the effective combination of features to form a new reference space. The validity of the newly constructed reference space was confirmed again, and abnormal voltage data of MMC submodules outside the reference space were collected again to monitor the validity of the newly built reference space.

7. The fault diagnosis method for MMC converters based on the Martin system as described in claim 6, characterized in that: The fault diagnosis includes determining the relationship between unknown samples and the reference space. The category of unknown MMC submodule data samples can be determined based on the preliminary threshold calculation formula, which is as follows: Among them, MD n(max) MD represents the maximum Mahalanobis distance in normal samples of the MMC submodule. a(min) This represents the minimum Mahalanobis distance among abnormal samples in the MMC submodule. Different thresholds will be given for different problems. If the Mahalanobis distance is greater than the threshold, it is determined that the sample does not belong to the population or has a low probability of belonging to the population. If it exceeds the threshold, it means that the MMC submodule has an open circuit fault. At this time, an early warning or shutdown of the converter will be issued.

8. A system employing the MMC converter fault diagnosis method based on the Martin system as described in any one of claims 1 to 7, characterized in that: It includes a runtime data extraction module, a baseline space construction module, a validity verification module, a baseline space optimization module, a threshold determination module, and a fault diagnosis module; The data extraction module is responsible for extracting voltage characteristic data of submodules from the MMC converter; The reference space construction module uses the extracted voltage feature data and the calculated Mahalanobis distance to construct the reference space of the Martin system. The reference space is used as a standard framework for comparing and judging normal and fault states. The validity verification module verifies the validity of the constructed benchmark space to ensure it meets the requirements of the diagnostic system. The reference space optimization module introduces the concepts of orthogonal arrays, signal-to-noise ratio (SNR), and SNR gain to filter selected features and optimize the reference space. The threshold determination module calculates and determines the threshold, which is the standard for judging whether unknown data indicates that the submodule has an open circuit fault. The fault diagnosis module uses the aforementioned constructed and optimized baseline space and thresholds to perform fault diagnosis on the sub-modules of the MMC converter.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images