Converter fault diagnosis method, system and equipment based on mahalanobis distance

Abstract

The invention is suitable for the technical field of fault diagnosis, and provides a Mahalanobis distance-based converter fault diagnosis method, system and device, and the method comprises the steps: carrying out the segmented sampling of the obtained capacitor voltage data of a converter sub-module within a preset time; converting a one-dimensional array corresponding to the capacitor voltage data of the converter sub-module into a two-dimensional matrix array; based on the two-dimensional matrix array, calculating the mahalanobis distance from the output voltage of each sub-module to the total distribution at the current moment; judging whether the difference between the mahalanobis distance between the output voltage of each sub-module and the total distribution of the output voltage of each sub-module at the current moment and the mahalanobis distance between the output voltage of each sub-module and the total distribution of the output voltage of each sub-module in a set simulation model meets a preset condition or not; if yes, determining that the current sub-module has a fault; and if not, calculating the mahalanobis distance from the output voltage of the sub-module to the total distribution at the next moment. Based on the comparison of the mahalanobis distances, the accuracy of fault diagnosis is improved while the calculation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a converter fault diagnosis method, system and device based on Mahalanobis distance. Background Art

[0002] Modular multilevel converters have advantages such as modular structure, redundant configuration and high output waveform quality. These characteristics make multilevel photovoltaic grid-connected converters widely used in photovoltaic buildings and photovoltaic grid-connected. However, the current structure based on a large number of series submodules (SM) poses a challenge to the reliability of multilevel photovoltaic grid-connected converters. Multilevel photovoltaic grid-connected converters contain a large number of submodules. The open circuit failure of the insulated gate bipolar transistor (IGBT) in the SM will affect the normal operation of the multilevel photovoltaic grid-connected converter and may even cause damage to other components, eventually leading to the paralysis of the entire system. When the upper and lower tubes of the submodules of the multilevel photovoltaic grid-connected converter fail at the same time, the capacitor voltages of each module of the multilevel photovoltaic grid-connected converter are highly overlapped, and the capacitor voltage residual is very small. The fault diagnosis method based on the capacitor voltage change rate cannot quickly and accurately diagnose the fault problem.

[0003] In view of this, a converter fault diagnosis method, system and device based on Mahalanobis distance is needed. Summary of the invention

[0004] The embodiments of the present application provide a converter fault diagnosis method, system and device based on Mahalanobis distance, which are used to solve the problem that the fault diagnosis method based on capacitor voltage change rate cannot quickly and accurately diagnose the occurrence of faults.

[0005] A first aspect of an embodiment of the present application provides a converter fault diagnosis method based on Mahalanobis distance, comprising:

[0006] The capacitor voltage data of the converter submodule obtained by segmented sampling within a preset time;

[0007] Converting the one-dimensional array corresponding to the capacitor voltage data of the converter submodule into a two-dimensional matrix array;

[0008] Calculate the Mahalanobis distance between the output voltage of each submodule at the current moment and the overall distribution based on the two-dimensional matrix array;

[0009] Determine whether the difference between the Mahalanobis distance of the output voltage of each submodule at the current moment from the overall distribution and the Mahalanobis distance of the output voltage of each submodule in the set simulation model from the overall distribution meets a preset condition;

[0010] If yes, it is determined that the current submodule fails; if no, the Mahalanobis distance of the submodule output voltage from the overall distribution at the next moment is calculated.

[0011] Furthermore, the calculating of the Mahalanobis distance between the output voltage of each submodule at the current moment and the overall distribution based on the two-dimensional matrix array includes:

[0012] Transposing the two-dimensional matrix array, extracting eigenvectors after the matrix transposition to form a diagonal matrix;

[0013] The Mahalanobis distance between the output voltage of each submodule at the current moment and the overall distribution is calculated based on the formed diagonal matrix.

[0014] Furthermore, the expression for calculating the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment based on the formed diagonal matrix is ​​as follows:

[0015]

[0016] Where: S mn is the mnth element in the diagonal matrix, c is the mean vector of the population A, and k is the coefficient.

[0017] Furthermore, the set simulation model includes:

[0018] The voltage of the upper and lower bridge arms inside the converter is calculated by setting the circulating current control quantity, and a simulation model is generated based on the output voltage of each sub-module calculated in the steady-state link.

[0019] Furthermore, the expression of the output voltage of the submodule is as follows:

[0020]

[0021]

[0022] Where: U pa and U na are the submodule voltage reference values ​​after the reference voltage values ​​of the upper and lower bridge arms of the converter are evenly divided, m a is the output voltage modulation ratio, m con_a is the output voltage modulation ratio of the circulating current component, and cos{ωt+θ) is the voltage difference.

[0023] Furthermore, the determining whether the difference between the Mahalanobis distance of the overall distribution of the capacitance voltage distances of each submodule at the current moment and the Mahalanobis distance of the overall distribution of the capacitance voltage distances of each submodule in the set simulation model satisfies a preset condition includes:

[0024] It is determined whether the difference between the Mahalanobis distance of the overall distribution of the capacitance and voltage distances of each submodule at the current moment and the Mahalanobis distance of the overall distribution of the capacitance and voltage distances of each submodule in the set simulation model is greater than a set threshold.

[0025] A second aspect of an embodiment of the present application provides a converter fault diagnosis system based on Mahalanobis distance, comprising:

[0026] A capacitor voltage data acquisition unit, used for acquiring capacitor voltage data of a converter submodule by segmented sampling within a preset time;

[0027] An array conversion unit, used to convert the one-dimensional array corresponding to the capacitor voltage data of the converter submodule into a two-dimensional matrix array;

[0028] A Mahalanobis distance calculation unit, used for calculating the Mahalanobis distance of the output voltage of each submodule at the current moment from the overall distribution based on the two-dimensional matrix array;

[0029] A Mahalanobis distance comparison and judgment unit, used to judge whether the difference between the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment and the Mahalanobis distance of the output voltage of each submodule from the overall distribution in the set simulation model meets a preset condition;

[0030] A failure determination unit is used to determine that a current submodule has a failure when the difference between the Mahalanobis distance of the output voltage of each submodule at the current moment and the Mahalanobis distance of the output voltage of each submodule in the set model from the overall distribution meets a preset condition;

[0031] The next moment Mahalanobis distance calculation unit is used to calculate the Mahalanobis distance of the submodule output voltage from the overall distribution at the next moment when the difference between the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment and the Mahalanobis distance of the output voltage of each submodule from the overall distribution in the set model does not meet the preset condition.

[0032] A third aspect of an embodiment of the present application provides a computer device, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein when the processor executes the computer-readable instructions, the steps of the converter fault diagnosis method based on Mahalanobis distance as described in any one of the above are implemented.

[0033] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0034] The present application first converts the one-dimensional array corresponding to the capacitor voltage data of the converter submodule obtained by segmented sampling within a preset time into a two-dimensional array, and then calculates the Mahalanobis distance of the capacitor voltage of each submodule from the overall distribution at the current moment based on the two-dimensional matrix array, and then compares the calculated Mahalanobis distance of the submodule in the overall distribution with the corresponding calculated Mahalanobis distance in the set simulation model to determine whether the comparison result meets the set conditions. If so, the submodule is determined to be faulty. If not, the Mahalanobis distance of the submodule at the next moment is calculated, and so on, until the Mahalanobis distance of all submodules at all set moments is calculated. By calculating the value of the Mahalanobis distance for comparison to determine whether the corresponding submodule has a fault, the calculation process is simple and avoids the error caused by directly comparing the submodules, effectively improving the accuracy of diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic flow chart of an embodiment of a converter fault diagnosis method based on Mahalanobis distance in the present invention;

[0036] Figure 2 The hardware topology diagram of the multi-level photovoltaic grid-connected converter and its submodules in the present invention;

[0037] Figure 3 It is a structural block diagram of an embodiment of a converter fault diagnosis system based on Mahalanobis distance in the present invention;

[0038] Figure 4 The figure is a structural block diagram of a computer device in the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0040] In this embodiment, the converter fault diagnosis method based on Mahalanobis distance is used to improve the accuracy of submodule fault diagnosis. In this embodiment, the implementation method can be implemented in the system, in the server, or in the terminal, and is not specifically limited.

[0041] Embodiment 1

[0042] See also Figure 1 , an embodiment of a converter fault diagnosis method based on Mahalanobis distance in the present invention comprises the following steps:

[0043] S11. Segmented sampling of the converter submodule capacitor voltage data obtained within a preset time;

[0044] In this embodiment, the submodule fault analysis involves multiple types of faults. Each submodule with a single IGBT open circuit may fail, but the short-circuit fault is more harmful to the circuit. Here we only discuss the IGBT open circuit fault in the submodule, and the failure of a single IGBT may not implicate other IGBTs and cause their failure. In addition, through historical analysis of the charging and discharging state of the submodule after the converter failure, any IGBT failure will cause the capacitor voltage to produce an upward trend. Therefore, within a certain time period, the system samples the capacitor voltage data of each submodule in the converter multiple times in time periods to calculate the submodule voltage.

[0045] S12. Converting the one-dimensional array corresponding to the capacitor voltage data of the converter submodule into a two-dimensional matrix array;

[0046] After obtaining the capacitor voltage data of the converter submodule, the collected one-dimensional array is converted into a two-dimensional matrix array. The conversion into a two-dimensional matrix array can improve the data accuracy, so that the subsequent Mahalanobis distance calculation can adapt to the algorithm calculation logic and data preprocessing.

[0047] For example, the capacitor voltage data of each submodule in multiple time periods obtained can represent the voltage of the submodule, and the expression can be expressed as follows:

[0048]

[0049] Where: i∈(1,N),

[0050] S13. Calculate the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment based on the two-dimensional matrix array;

[0051] Calculating the Mahalanobis distance also includes the following steps:

[0052] 1. Transpose the two-dimensional matrix array and extract the eigenvectors after the matrix transposition to form a diagonal matrix;

[0053] 2. Based on the formed diagonal matrix, the Mahalanobis distance of the capacitance and voltage of each submodule at the current moment from the overall distribution is calculated.

[0054] Based on the above two-dimensional matrix expression, the data matrix is ​​an m×n matrix. The matrix is ​​transposed to obtain the transposed matrix. For example, if the above matrix is ​​represented by A, the transposed matrix is ​​A. T , if there is the following:

[0055] AA T =P1P T ; A T A=QΛ2Q T ; then A=P∑Q T; Where P = (p1, p2, p2, ..., p m ), and p1,p2,p2,…,p m AA T The characteristic vector of m ) is A T The eigenvector of A, ∑ is an m×n matrix, and the eigenvalues ​​of the elements on the diagonal arranged from large to small are taken as singular values. Based on the above, the improved expression of Mahalanobis distance is as follows:

[0056]

[0057] Where: S mn is the mnth element in the diagonal matrix, c is the mean vector of the population A, and k is the coefficient.

[0058] The improved Mahalanobis distance calculation formula is used to calculate the Mahalanobis distance of the voltage of each submodule at the current moment from the overall voltage.

[0059] S14. Determine whether the difference between the Mahalanobis distance of the output voltage of each submodule at the current moment from the overall distribution and the Mahalanobis distance of the output voltage of each submodule of the set simulation model from the overall distribution meets the preset conditions;

[0060] In this embodiment, the simulation model is set to calculate the voltage of the upper and lower bridge arms inside the converter by setting the circulating current control amount, and to generate the simulation model based on the output voltage of each submodule in the steady state link. Figure 2 The submodule of the multi-level photovoltaic grid-connected converter is composed of capacitors, IGBTs and diodes, and the entire multi-level photovoltaic grid-connected converter is composed of several such submodules. Such a structure results in a very large number of IGBTs in the multi-level photovoltaic grid-connected converter, and the multi-level photovoltaic grid-connected converter is thus very susceptible to IGBT open-circuit faults. Therefore, the simulation model is set to represent the output voltage when the multi-level photovoltaic grid-connected converter is operating normally as a fault, where the expression of the output voltage is as follows:

[0061]

[0062] Where: U pa and U na are the submodule voltage reference values ​​after the reference voltage values ​​of the upper and lower bridge arms of the converter are evenly divided, m a is the output voltage modulation ratio, m con_a is the output voltage modulation ratio of the circulating current component, and cos(ωt+θ) is the voltage difference.

[0063] After calculating the reference voltage values ​​of the upper and lower bridge arms of the converter, the voltage reference value of the normal working state of the submodule is obtained by dividing the voltage equally, thereby changing its conduction or shutdown by controlling the submodule voltage.

[0064] S15. When the difference between the Mahalanobis distance of the output voltage of each submodule at the current moment and the Mahalanobis distance of the output voltage of each submodule in the set simulation model meets the preset conditions, it is determined that the current submodule fails;

[0065] S16. When the difference between the Mahalanobis distance of each submodule output voltage from the overall distribution at the current moment and the Mahalanobis distance of each submodule output voltage from the overall distribution in the set simulation model does not meet the preset condition, calculate the Mahalanobis distance of the submodule output voltage from the overall distribution at the next moment.

[0066] In step S15-S16, judging whether the difference between the Mahalanobis distance of the output voltage of each submodule at the current moment and the Mahalanobis distance of the output voltage of each submodule in the set simulation model and the overall distribution satisfies the preset condition is to judge whether the difference between the two is greater than the set threshold value, wherein the setting of the threshold value is set according to the type of single IGBT open circuit, when the calculated difference is greater than the set threshold value, it can be determined that the current submodule has a fault; when the calculated difference is less than or equal to the set threshold value, the Mahalanobis distance at the next moment is continued to be calculated, and so on, until the Mahalanobis distance corresponding to the output voltage of all submodules is calculated. It should be noted that before comparing the Mahalanobis distance, calculating the Mahalanobis distance of the output voltage of the submodule in the set simulation model in the overall distribution is to obtain the output voltage of the submodule. It can be seen from the above expression of the output voltage that the Mahalanobis distance is calculated using the submodule voltage reference value.

[0067] The above embodiment calculates the Mahalanobis distance of the overall distribution of each sub-module distance and compares it with the Mahalanobis distance calculated by the sub-module output voltage in the set simulation model. The calculation process of the set simulation model is simple and does not require the use of complex intelligent algorithms, which effectively improves the calculation efficiency. The judgment is based on the improved Mahalanobis distance calculation method, which improves the diagnostic accuracy and significantly shortens the diagnostic time.

[0068] Embodiment 2

[0069] See also Figure 3 , an embodiment of a converter fault diagnosis system based on Mahalanobis distance in the present invention comprises the following steps:

[0070] The capacitor voltage data acquisition unit 101 is used to acquire the capacitor voltage data of the converter submodule by segmented sampling within a preset time;

[0071] An array conversion unit 102, used to convert the one-dimensional array corresponding to the capacitor voltage data of the converter submodule into a two-dimensional matrix array;

[0072] A Mahalanobis distance calculation unit 103, used for calculating the Mahalanobis distance between the output voltage of each submodule at the current moment and the overall distribution based on the two-dimensional matrix array;

[0073] A Mahalanobis distance comparison and judgment unit 104 is used to judge whether the difference between the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment and the Mahalanobis distance of the output voltage of each submodule from the overall distribution in the set simulation model meets a preset condition;

[0074] A failure determination unit 105 is used to determine that a current submodule has a failure when the difference between the Mahalanobis distance of the output voltage of each submodule at the current moment and the Mahalanobis distance of the output voltage of each submodule in the set model from the overall distribution meets a preset condition;

[0075] The next moment Mahalanobis distance calculation unit 106 is used to calculate the Mahalanobis distance of the submodule output voltage from the overall distribution at the next moment when the difference between the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment and the Mahalanobis distance of the output voltage of each submodule from the overall distribution in the set model does not meet the preset condition.

[0076] For the specific definition of the fault diagnosis system, please refer to the definition of the fault diagnosis method above, which will not be repeated here. Each module in the above fault diagnosis system can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0077] Embodiment 3

[0078] See also Figure 4 The present invention provides a computer device, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the above method are implemented.

[0079] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0080] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored. In addition, each functional unit in each embodiment of the present invention can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units.

[0081] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0082] It can be understood that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A converter fault diagnosis method based on Mahalanobis distance, characterized in that: include: The capacitor voltage data of the converter submodule obtained by segmented sampling within a preset time; Converting the one-dimensional array corresponding to the capacitor voltage data of the converter submodule into a two-dimensional matrix array; Calculate the Mahalanobis distance between the output voltage of each submodule at the current moment and the overall distribution based on the two-dimensional matrix array; Determine whether the difference between the Mahalanobis distance of the output voltage of each submodule at the current moment from the overall distribution and the Mahalanobis distance of the output voltage of each submodule in the set simulation model from the overall distribution meets a preset condition; If yes, it is determined that the current submodule fails; if no, the Mahalanobis distance of the submodule output voltage from the overall distribution at the next moment is calculated.

2. The converter fault diagnosis method based on Mahalanobis distance according to claim 1, characterized in that: The calculating the Mahalanobis distance of the output voltage of each submodule at the current moment from the overall distribution based on the two-dimensional matrix array includes: Transposing the two-dimensional matrix array, extracting eigenvectors after the matrix transposition to form a diagonal matrix; The Mahalanobis distance between the output voltage of each submodule at the current moment and the overall distribution is calculated based on the formed diagonal matrix.

3. The converter fault diagnosis method based on Mahalanobis distance according to claim 2, characterized in that: The expression for calculating the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment based on the formed diagonal matrix is ​​as follows: Where: S mn is the mnth element in the diagonal matrix, c is the mean vector of the population A, and k is the coefficient.

4. The converter fault diagnosis method based on Mahalanobis distance according to claim 1, characterized in that: The simulation model set includes: The voltage of the upper and lower bridge arms inside the converter is calculated by setting the circulating current control quantity, and a simulation model is generated based on the output voltage of each sub-module calculated in the steady-state link.

5. The converter fault diagnosis method based on Mahalanobis distance according to claim 4, characterized in that: The expression of the output voltage of the submodule is as follows: Where: U pa and U na are the submodule voltage reference values ​​after the reference voltage values ​​of the upper and lower bridge arms of the converter are evenly divided, m a is the output voltage modulation ratio, m con_a is the output voltage modulation ratio of the circulating current component, and cos(ωt+θ) is the voltage difference.

6. The converter fault diagnosis method based on Mahalanobis distance according to claim 5, characterized in that: The determining whether the difference between the Mahalanobis distance of the overall distribution of the capacitance and voltage distances of each submodule at the current moment and the Mahalanobis distance of the overall distribution of the capacitance and voltage distances of each submodule in the set simulation model meets a preset condition includes: It is determined whether the difference between the Mahalanobis distance of the overall distribution of the capacitance and voltage distances of each submodule at the current moment and the Mahalanobis distance of the overall distribution of the capacitance and voltage distances of each submodule in the set simulation model is greater than a set threshold.

7. A converter fault diagnosis system based on Mahalanobis distance, characterized in that: include: A capacitor voltage data acquisition unit, used for acquiring capacitor voltage data of a converter submodule by segmented sampling within a preset time; An array conversion unit, used to convert the one-dimensional array corresponding to the capacitor voltage data of the converter submodule into a two-dimensional matrix array; A Mahalanobis distance calculation unit, used for calculating the Mahalanobis distance of the output voltage of each submodule at the current moment from the overall distribution based on the two-dimensional matrix array; A Mahalanobis distance comparison and judgment unit, used to judge whether the difference between the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment and the Mahalanobis distance of the output voltage of each submodule from the overall distribution in the set simulation model meets a preset condition; A failure determination unit is used to determine that a current submodule has a failure when the difference between the Mahalanobis distance of the output voltage of each submodule at the current moment and the Mahalanobis distance of the output voltage of each submodule in the set model from the overall distribution meets a preset condition; The next moment Mahalanobis distance calculation unit is used to calculate the Mahalanobis distance of the submodule output voltage from the overall distribution at the next moment when the difference between the Mahalanobis distance of the output voltage of each submodule from the overall distribution at the current moment and the Mahalanobis distance of the output voltage of each submodule from the overall distribution in the set model does not meet the preset condition.

8. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the steps of the converter fault diagnosis method based on Mahalanobis distance as described in any one of claims 1 to 6 are implemented.

Images