Commutation failure precursor detection method, system, medium and device based on feature trajectory comparison

Abstract

The embodiment of the application relates to the technical field of high-voltage direct current transmission, and discloses a commutation failure precursor detection method and system based on feature trajectory comparison, a medium and equipment. The method comprises the following steps: sampling real-time current of a commutation valve from a commutation starting moment, constructing a real-time waveform representing the current based on the real-time current of the commutation valve; sampling grid operation parameters to generate an ideal waveform corresponding to each sampling moment in the real-time waveform; weighting the distance between the real-time waveform and the ideal waveform by a preset weight coefficient to determine a trajectory deviation index; comparing the trajectory deviation index with a preset discrimination threshold, and pre-judging whether a commutation failure exists according to the comparison result, which can more comprehensively and accurately quantify the deviation of the actual commutation process from the ideal state, so that the judgment basis is more sufficient and the accuracy is higher.

Description

Technical Field

[0001] This invention relates to the field of high voltage direct current transmission technology, and in particular to a method, system, medium, and equipment for detecting precursors of commutation failure based on characteristic trajectory comparison. Background Technology

[0002] In high-voltage direct current (HVDC) transmission systems, grid-commutated converters are a mainstream technology for power conversion. However, commutation failure is a serious fault that threatens the safe and stable operation of the system, potentially leading to severe consequences such as interruption of DC power transmission and overcurrent in the converter valves. Therefore, rapid and accurate detection of commutation failure is crucial.

[0003] Existing commutation failure detection technologies primarily rely on analyzing electrical quantities during the commutation process. One approach involves measuring key indicators such as the arc extinction angle at the end of the commutation process and comparing them to fixed safety thresholds. This type of method is essentially a "post-hoc" detection, only making a judgment when the commutation process is about to end or even when an anomaly has already occurred. It inherently suffers from detection delays and cannot provide early warning time for the system to implement preventative control measures.

[0004] Another approach attempts to predict faults in advance by constructing characteristic vectors of electrical quantities during commutation and comparing them with an ideal reference. However, existing references are typically static models pre-calculated and stored offline under typical operating conditions. When the actual operating conditions of the power grid, such as load levels and AC system voltage, change, this static reference cannot accurately reflect a "healthy" commutation process under the current conditions. This results in poor adaptability of the detection method and makes it prone to misjudgment or missed detection when the operating conditions deviate from typical conditions. Therefore, existing technologies struggle to provide rapid, accurate, and reliable early warnings of commutation failure precursors. Summary of the Invention

[0005] Therefore, it is necessary to propose a method, system, medium, and device for detecting commutation failure precursors based on feature trajectory comparison to address the above problems.

[0006] A method for detecting commutation failure precursors based on feature trajectory comparison, the method comprising:

[0007] Starting from the moment of commutation, the real-time current of the converter valve is sampled, and a real-time waveform representing the current is constructed based on the real-time current of the converter valve.

[0008] The power grid operating parameters are sampled to generate an ideal waveform corresponding to each sampling moment in the real-time waveform.

[0009] The trajectory deviation index is determined by weighting the distance between the real-time waveform and the ideal waveform using preset weighting coefficients.

[0010] The trajectory deviation index is compared with a preset discrimination threshold, and it is predicted whether the commutation failure exists according to a comparison result.

[0011] The real-time current of the converter valve is sampled from the commutation start time, and a real-time waveform representing the current is constructed based on the real-time current of the converter valve, and specifically includes:

[0012] The real-time current of the converter valve is sampled from the commutation start time.

[0013] At each sampling time, a time sequence feature vector is determined according to the real-time current of the converter valve, and the time sequence feature vector includes a normalized instantaneous amplitude, a normalized instantaneous change rate and a normalized instantaneous curvature.

[0014] The real-time waveform representing the current is obtained based on the time sequence feature vector of each sampling time.

[0015] At each sampling time, a time sequence feature vector is determined according to the real-time current of the converter valve, and the time sequence feature vector includes a normalized instantaneous amplitude, a normalized instantaneous change rate and a normalized instantaneous curvature, and specifically includes:

[0016] The DC current before commutation of the converter valve starts is collected.

[0017] The normalized instantaneous amplitude is determined according to a ratio of the real-time current of the converter valve to the DC current before commutation.

[0018] The normalized instantaneous change rate is determined according to a ratio of a first derivative of the real-time current of the converter valve to the DC current before commutation.

[0019] The normalized instantaneous curvature is determined according to a ratio of a second derivative of the real-time current of the converter valve to the DC current before commutation.

[0020] The grid operating parameters are sampled to generate an ideal waveform corresponding to each sampling time in the real-time waveform, and specifically includes:

[0021] The grid operating parameters are sampled, and the grid operating parameters include an effective value of an inverter-side line voltage, a fundamental angular frequency, a commutation voltage offset angle, a transformer ratio, a commutation inductance, and a DC current before commutation.

[0022] The grid operating parameters are substituted into a theoretical model of the converter valve current to determine the real-time current of the converter valve in the normal commutation process.

[0023] At each sampling moment, an ideal timing feature vector is determined according to the real-time current of the commutation valve in the normal commutation process, and the ideal timing feature vector includes an ideal normalized instantaneous amplitude, an ideal normalized instantaneous rate of change, and an ideal normalized instantaneous curvature.

[0024] An ideal waveform corresponding to each sampling moment in the real-time waveform is obtained based on the ideal timing feature vector of each sampling moment.

[0025] The ideal timing feature vector is determined by substituting the grid operation parameter into a theoretical model of the commutation valve current, and specifically includes:

[0026] The theoretical model of the commutation valve current is determined, and the theoretical model of the commutation valve current is:

[0027] ;

[0028] Wherein, is the real-time current of the commutation valve in the normal commutation process, is the real-time current of the conduction valve in the normal commutation process, is the effective value of the inverter side line voltage, is the fundamental angular frequency, is the commutation voltage offset angle, β is the advance trigger angle, is the transformer ratio, π is the reference point, representing 180° electrical angle, is the commutation inductance, , is the commutation start time, , is the theoretical commutation end time, t is the current sampling moment, , is the DC current before commutation, is the trigger angle, is the turn-off angle.

[0029] The real-time current of the commutation valve in the normal commutation process is determined by substituting the grid operation parameter into the theoretical model of the commutation valve current.

[0030] The trajectory deviation index is determined by weighting the distance between the real-time waveform and the ideal waveform by a preset weight coefficient, and specifically includes:

[0031] According to

[0032] For any sampling moment, the distance between the two corresponding timing feature vectors in the real-time waveform and the ideal waveform is weighted to obtain the weighted distance between any two sampling moments, wherein For weighted distance, , , To set the preset weighting coefficients, , For the first j The temporal feature vector of the ideal waveform at each sampling time. x ∈ (1, 2, 3), For the first i The time-series feature vector in the real-time waveform at each sampling time. x ∈ (1, 2, 3), ti For the i-th sampling time, tj For the first j Each sampling time.

[0033] A distance matrix is ​​constructed using the weighted distance between any two sampling times.

[0034] In the distance matrix, the path with the smallest cumulative sum of weighted distances from the sampling time at the start of commutation to the current sampling time is determined, and the cumulative sum of weighted distances on the path is used as the trajectory deviation index.

[0035] The step of comparing the trajectory deviation index with a preset discrimination threshold and predicting whether commutation failure exists based on the comparison result specifically includes:

[0036] If the trajectory deviation index is greater than the preset discrimination threshold, then commutation failure exists.

[0037] If the trajectory deviation index is less than or equal to the preset discrimination threshold, then there is no commutation failure.

[0038] A commutation failure precursor detection system based on feature trajectory comparison, the system comprising:

[0039] The real-time waveform construction module is used to sample the real-time current of the converter valve from the start of commutation and construct a real-time waveform representing the current based on the real-time current of the converter valve.

[0040] The ideal waveform generation module is used to sample the power grid operating parameters and generate an ideal waveform corresponding to each sampling moment in the real-time waveform.

[0041] The trajectory deviation index acquisition module is used to determine the trajectory deviation index by weighting the distance between the real-time waveform and the ideal waveform using preset weighting coefficients.

[0042] The commutation failure precursor detection module is used to compare the trajectory deviation index with a preset detection threshold and predict whether commutation failure exists based on the comparison result.

[0043] A computer readable storage medium stores a computer program, the computer program is executed by a processor, so that the processor executes the steps of the method.

[0044] A computer device comprises a memory and a processor, the memory stores a computer program, the computer program is executed by the processor, so that the processor executes the steps of the method.

[0045] By adopting the embodiment of the present application, the following beneficial effects are achieved:

[0046] The present application can capture subtle waveform distortion in the early stage of fault occurrence by constructing a real-time waveform representing current, significantly improving the sensitivity and predictability of detection, and improving fault detection to fault warning. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0048] Among them:

[0049] Figure 1 A flowchart of an embodiment of a commutation failure precursor detection method based on feature trajectory comparison provided by the present application;

[0050] Figure 2 A flowchart of another embodiment of a commutation failure precursor detection method based on feature trajectory comparison provided by the present application;

[0051] Figure 3 A structural schematic diagram of an embodiment of a commutation failure precursor detection system based on feature trajectory comparison provided by the present application;

[0052] Figure 4 A structural schematic diagram of an embodiment of a device provided by the present application;

[0053] Figure 5 Structure diagram of an embodiment of the medium provided by the present application. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0055] As shown in Figure 1 , Figure 1 Flowchart of an embodiment of a commutation failure precursor detection method based on feature trajectory comparison provided by the present application. The commutation failure precursor detection method based on feature trajectory comparison comprises the following steps:

[0056] S101: Sampling the real-time current of the commutation valve from the commutation start time, and constructing a real-time waveform representing the current based on the real-time current of the commutation valve.

[0057] Exemplarily, the real-time current of the commutation valve is sampled from the commutation start time; at each sampling time, a time sequence feature vector is determined according to the real-time current of the commutation valve, and the time sequence feature vector comprises a normalized instantaneous amplitude, a normalized instantaneous rate of change and a normalized instantaneous curvature. Specifically, the DC current before the commutation valve commutation starts is collected; the normalized instantaneous amplitude is determined according to the ratio of the real-time current of the commutation valve to the DC current before the commutation starts; the normalized instantaneous rate of change is determined according to the ratio of the first derivative of the real-time current of the commutation valve to the DC current before the commutation starts; and the normalized instantaneous curvature is determined according to the ratio of the second derivative of the real-time current of the commutation valve to the DC current before the commutation starts.

[0058] Further, the real-time waveform representing the current is obtained based on the time sequence feature vector at each sampling time , as shown in the following formula:

[0059] ;

[0060] Among them, is the normalized instantaneous amplitude, is the normalized instantaneous rate of change (slope), is the normalized instantaneous curvature.

[0061] S102: Sampling the power grid operation parameters to generate an ideal waveform corresponding to each sampling time in the real-time waveform.

[0062] Exemplarily, the grid operating parameters are sampled, including: the effective value of the inverter side line voltage, the fundamental angular frequency, the commutation voltage offset angle, the transformer ratio, the commutation inductance, and the DC current before commutation starts;

[0063] The grid operating parameters are substituted into the theoretical model of the converter valve current to determine the real-time current of the converter valve in the normal commutation process;

[0064] At each sampling time, the ideal timing feature vector is determined according to the real-time current of the converter valve in the normal commutation process, including: ideal normalized instantaneous amplitude, ideal normalized instantaneous rate of change, and ideal normalized instantaneous curvature;

[0065] Based on the ideal timing feature vector at each sampling time, the ideal waveform corresponding to each sampling point in the real-time waveform is obtained , as shown in the following formula:

[0066] ;

[0067] wherein, is the ideal normalized instantaneous amplitude, is the ideal normalized instantaneous rate of change (slope), is the ideal normalized instantaneous curvature.

[0068] S103: The distance between the real-time waveform and the ideal waveform is weighted by a preset weight coefficient to determine the trajectory deviation index.

[0069] Exemplarily, to measure the deviation degree of the real-time waveform at any sampling time from the ideal waveform , the weighted distance between any two feature vectors in the feature trajectory is defined as :

[0070] ;

[0071] wherein, is the weighted distance, , , is a preset weight coefficient, and is set to , is the timing feature vector in the ideal waveform at the i-th sampling time, j ∈(1, 2, 3), x is the timing feature vector in the real-time waveform at the i-th sampling time, ∈(1, 2, 3), i is the i-th sampling time, x is the i-th sampling time, ti is the i-th sampling time, tj is the i-th sampling time, j ​a sampling time.

[0072] Further, a distance matrix is constructed by the weighted distance between any two sampling time points; in the distance matrix, the path with the minimum cumulative sum of weighted distance from the sampling time point when commutation starts to the current sampling time point is determined, and the cumulative sum of weighted distance on the path is taken as the trajectory deviation index.

[0073] It should be noted that when ti and tj are the same ( i = j ). This case represents the comparison of two points at the same sampling time. For example, the feature difference between the 5th sampling time of the real-time waveform and the 5th sampling time of the ideal waveform is compared. If the actual commutation process is perfectly synchronized in time with the theoretical model without any delay or advance, the best matching path is a series of points ti = tj . The total deviation calculated will mainly consist of these points.

[0074] When ti and tj are not the same ( i ≠ j ). This case represents the comparison of "cross-time-point". For example, the 5th sampling time of the real-time waveform ( ti ), whose waveform features (amplitude, slope, curvature) are actually more similar to the 4th sampling time of the ideal waveform ( tj , here i > j ). The real-time waveform lags behind the ideal waveform on the time axis.

[0075] S104: Compare the trajectory deviation index with the preset discrimination threshold, and pre-judge whether there is a commutation failure according to the comparison result.

[0076] Exemplarily, the trajectory deviation index is compared with the preset discrimination threshold, if the trajectory deviation index is greater than the preset discrimination threshold, there is a commutation failure; if the trajectory deviation index is less than or equal to the preset discrimination threshold, there is no commutation failure.

[0077] From the above description, firstly, the application can capture the subtle waveform distortion in the early stage of fault occurrence by constructing the real-time waveform representing the current, significantly improving the sensitivity and predictability of detection, and upgrading the fault detection to fault warning. Secondly, by dynamically generating the ideal waveform of the current according to the grid operation parameters as the judgment reference, the detection method can automatically adapt to the changes of the system operation state, fundamentally solving the misjudgment and omission problem caused by the traditional static reference, and improving the adaptability and reliability of the detection. Finally, by presetting the weight coefficient to weight the distance between the real-time waveform and the ideal waveform, the trajectory deviation index is determined; the trajectory deviation is compared with the preset discrimination threshold, and whether there is a commutation failure is predicted according to the comparison result, which can more comprehensively and accurately quantify the deviation of the actual commutation process from the ideal state, so that the judgment basis is more sufficient and the accuracy is higher.

[0078] As shown in Figure 2 , Figure 2 a flowchart of another embodiment of a commutation failure precursor detection method based on feature trajectory comparison provided by the application. A commutation failure precursor detection method based on feature trajectory comparison, the method comprises:

[0079] S201: sampling the real-time current of the commutation valve from the commutation starting time.

[0080] Exemplarily, at the starting time of the commutation process (for example, the trigger pulse of a certain valve is detected, and the last group of valves in series are conducting), the system starts to sample the real-time current of the commutation valve to be turned off at a high frequency.

[0081] S202: At each sampling time, determine the time sequence feature vector according to the real-time current of the commutation valve, and the time sequence feature vector includes: normalized instantaneous amplitude, normalized instantaneous change rate and normalized instantaneous curvature.

[0082] S203: Obtain the real-time waveform representing the current based on the time sequence feature vector at each sampling time.

[0083] Exemplarily, at each sampling time , a real-time waveform representing the current is constructed in real time as follows:

[0084] ;

[0085] Among them, is the normalized instantaneous amplitude, is the normalized instantaneous change rate (slope), is the normalized instantaneous curvature.

[0086] Specifically, the normalized instantaneous amplitude of the current is as follows:

[0087] ;

[0088] wherein, is the DC current before commutation starts, is the real-time current of the converter valve.

[0089] This component reflects the relative size of the current, and through normalization processing, the influence of system load level changes is eliminated, focusing on the waveform shape itself.

[0090] The normalized instantaneous rate of change of current (slope) is as follows:

[0091] ;

[0092] wherein, is the DC current before commutation starts, is the real-time current of the converter valve,

[0093] This component is numerically equal to the tangent slope of the valve current waveform at the moment of , reflecting the "instantaneous speed" of the current drop. In normal commutation, the slope should be negative and have a large absolute value; when commutation failure is about to occur, the current will be reversed and the drop will become slow, and the absolute value of the slope will decrease significantly.

[0094] The normalized instantaneous curvature of the current is as follows:

[0095] ;

[0096] wherein, is the DC current before commutation starts, is the real-time current of the converter valve.

[0097] This component is numerically equal to the second derivative of the valve current waveform, reflecting the speed of change of the slope, i.e., the "bending degree" or "acceleration" of the waveform. This is the most sensitive indicator for capturing failure precursors. The concave degree of the normal waveform is significant (curvature is positive and large), and when a failure precursor appears, the waveform will be "forcibly flattened by external force", the concave trend will slow down, and the curvature will change significantly first.

[0098] S204: Sample the grid operating parameters, including: inverter-side line voltage effective value, fundamental angular frequency, commutation voltage offset angle, transformer ratio, commutation inductance, and DC current before commutation starts.

[0099] S205: Substitute the grid operating parameters into the theoretical model of the converter valve current to determine the real-time current of the converter valve in the normal commutation process.

[0100] Exemplarily, the grid operating parameters are sampled, including: the line voltage effective value on the inverter side, the fundamental angular frequency, the commutation voltage offset angle, the transformer ratio, the commutation inductance, and the DC current before commutation.

[0101] wherein the theoretical model of the current of the converter valve in the normal commutation process is:

[0102] ;

[0103] wherein, is the real-time current of the converter valve in the normal commutation process, is the real-time current of the conducting valve in the normal commutation process, is the line voltage effective value on the inverter side, β is the leading trigger angle, π is the reference point, representing 180° electrical angle, i.e. “natural commutation point”, is the fundamental angular frequency, is the commutation voltage offset angle, is the transformer ratio, is the commutation inductance, t is the current sampling time, , is the commutation start time, , is the theoretical commutation end time, , is the DC current before commutation, is the trigger angle, is the turn-off angle.

[0104] Further, the above grid operating parameters are substituted into the theoretical model of the current of the converter valve to determine the real-time current of the converter valve in the normal commutation process.

[0105] S206: At each sampling time, the ideal timing feature vector is determined according to the real-time current of the converter valve in the normal commutation process, and the ideal timing feature vector includes: ideal normalized instantaneous amplitude, ideal normalized instantaneous rate of change, and ideal normalized instantaneous curvature.

[0106] Exemplarily, the ideal normalized instantaneous amplitude is as follows:

[0107] ;

[0108] wherein, is the DC current before commutation, is the real-time current of the converter valve in the normal commutation process.

[0109] The ideal normalized instantaneous rate of change (slope) is as follows:

[0110] ;

[0111] in, This is the DC current before commutation begins. This represents the real-time current of the commutator valve during normal commutation.

[0112] The ideal normalized instantaneous curvature is shown in the following equation:

[0113] ;

[0114] in, This is the DC current before commutation begins. This represents the real-time current of the commutator valve during normal commutation.

[0115] S207: Obtain the ideal waveform corresponding to each sampling point in the real-time waveform based on the ideal time-series feature vector at each sampling time.

[0116] For example, the ideal waveform corresponding to each sampling point in the real-time waveform is obtained based on the ideal time-series feature vector at each sampling time. :

[0117] ;

[0118] in, For ideal normalized instantaneous amplitude, The ideal normalized instantaneous rate of change (slope). For ideal normalized instantaneous curvature.

[0119] S208: The trajectory deviation index is determined by weighting the distance between the real-time waveform and the ideal waveform using preset weighting coefficients.

[0120] For example, to measure the real-time waveform at any sampling time For ideal waveform The degree of deviation is defined as the weighted distance between any two feature vectors in the feature trajectory. for:

[0121] ;

[0122] in, For weighted distance, , , To set the preset weighting coefficients, , For the first j The temporal feature vector of the ideal waveform at each sampling time. x ∈ (1, 2, 3), For the firsti The time-series feature vector in the real-time waveform at each sampling time. x ∈ (1, 2, 3), ti For the i-th sampling time, tj For the first j Each sampling time.

[0123] Furthermore, a distance matrix is ​​constructed using the weighted distance between any two sampling times. In the distance matrix, the path with the smallest cumulative sum of weighted distances from the sampling time at the start of commutation to the current sampling time is determined, and the cumulative sum of weighted distances along the path is used as the trajectory deviation index.

[0124] S209: If the trajectory deviation index is greater than the preset discrimination threshold, there is a commutation failure.

[0125] S210: If the trajectory deviation index is less than or equal to the preset discrimination threshold, then there is no commutation failure.

[0126] For example, the trajectory deviation is compared with a preset discrimination threshold. If the trajectory deviation index is greater than the preset discrimination threshold, there is a commutation failure; if the trajectory deviation index is less than or equal to the preset discrimination threshold, there is no commutation failure.

[0127] like Figure 3 As shown, Figure 3 This is a schematic diagram of an embodiment of a commutation failure precursor detection system based on feature trajectory comparison provided by the present invention. A commutation failure precursor detection system 10 based on feature trajectory comparison includes:

[0128] The real-time waveform construction module 11 is used to sample the real-time current of the converter valve from the start of commutation and construct a real-time waveform characterizing the current based on the real-time current of the converter valve.

[0129] The ideal waveform generation module 12 is used to sample the power grid operating parameters and generate an ideal waveform corresponding to each sampling moment in the real-time waveform.

[0130] The trajectory deviation index acquisition module 13 is used to determine the trajectory deviation index by weighting the distance between the real-time waveform and the ideal waveform using preset weighting coefficients.

[0131] The commutation failure precursor detection module 14 is used to compare the trajectory deviation index with a preset detection threshold and predict whether commutation failure exists based on the comparison result.

[0132] For example, in the real-time waveform construction module 11, the real-time current of the converter valve is sampled from the start of commutation. At each sampling moment, a timing feature vector is determined based on the real-time current of the converter valve. The timing feature vector includes: normalized instantaneous amplitude, normalized instantaneous rate of change, and normalized instantaneous curvature. The real-time waveform representing the current is obtained based on the timing feature vector at each sampling moment.

[0133] In the ideal waveform generation module 12, the power grid operating parameters are sampled, including: the effective value of the inverter side line voltage, the fundamental angular frequency, the commutation voltage offset angle, the transformer turns ratio, the commutation inductance, and the DC current before commutation begins. The power grid operating parameters are substituted into the theoretical model of the converter valve current to determine the real-time current of the converter valve during normal commutation. At each sampling moment, the ideal time-series feature vector is determined based on the real-time current of the converter valve during normal commutation. The ideal time-series feature vector includes: the ideal normalized instantaneous amplitude, the ideal normalized instantaneous rate of change, and the ideal normalized instantaneous curvature. Based on the ideal time-series feature vector at each sampling moment, the ideal waveform corresponding to each sampling point in the real-time waveform is obtained.

[0134] In the trajectory deviation index acquisition module 13, the distance between the real-time waveform and the ideal waveform is weighted by a preset weighting coefficient to determine the trajectory deviation index.

[0135] In the commutation failure precursor detection module 14, if the trajectory deviation index is greater than the preset detection threshold, then there is a commutation failure; if the trajectory deviation index is less than or equal to the preset detection threshold, then there is no commutation failure.

[0136] As described above, the system provided by this invention collects current data in real time during the commutation process, constructs a time-series feature vector, and compares it with a dynamically generated ideal trajectory. By calculating the deviation between the two, the trajectory deviation is finally compared with a preset discrimination threshold. The system determines whether there is a commutation failure through a preset discrimination standard. This not only improves the speed and accuracy of fault detection, but also enables the prediction of early signs of faults, thus providing a strong guarantee for the stable operation of the power system.

[0137] like Figure 4 As shown, Figure 4 This is a schematic diagram of an embodiment of the device provided by the present invention. The device 20 includes a memory 21 and a processor 22. The memory 21 stores a computer program, and the processor 22 executes the computer program during operation to achieve, for example... Figure 1 and Figure 2 The method shown.

[0138] The specific technical details of the phase failure precursor detection method based on feature trajectory comparison implemented by the device 20 executing the computer program have been described in detail in the above method steps, and thus will not be described again.

[0139] As shown in Figure 5 , Figure 5 is a structural schematic diagram of an embodiment of the medium provided by the present application. The medium 30 stores at least a computer program 31, and the computer program 31 is executed by the processor 22 to implement the method as shown in Figure 1 and Figure 2 , and the detailed method can be referred to the above, which will not be described again. In an embodiment, the medium 30 can be a storage chip, a hard disk or a mobile hard disk or an optical disc, and other readable and writable storage tools, and can also be a server and the like.

[0140] In addition, the processes depicted in the figures do not necessarily have to be executed in the specific order shown or in consecutive order to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0141] Each of the embodiments in the specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, for the device, equipment and non-volatile computer readable storage medium embodiments, since they are basically similar to the method embodiments, they are described more simply, and the related parts can be referred to the part of the method embodiment.

[0142] The device, equipment, non-volatile computer readable storage medium and method provided by the embodiments of the present application are corresponding, so the device, equipment and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding device, equipment and non-volatile computer storage medium will not be described again.

[0143] The system, device, module or unit described in the above embodiments can be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer. Specifically, the computer may, for example, be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0144] For the sake of description, the above-described apparatus is described with various units in function for convenience. Of course, the functions of the units can be implemented in one or more software and / or hardware in implementing the specification. It is to be understood that the embodiments of the specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of the specification can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0145] The specification is described with reference to flowcharts and / or block diagrams of methods, apparatus (system) and computer program products according to embodiments of the specification. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts 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, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the flow Figure 1 one or more flows and / or blocks Figure 1 an apparatus that implements the functions specified in one or more flows or blocks.

[0146] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatus, which implements the flow Figure 1 one or more flows and / or blocks Figure 1 an apparatus that implements the functions specified in one or more flows or blocks.

[0147] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a process for implementing the flow Figure 1 one or more flows and / or blocks Figure 1 an apparatus that implements the functions specified in one or more flows or blocks.

[0148] In a typical configuration, the computing device includes one or more processors (CPU), input / output interface, network interface, and memory.

[0149] Memory can include, without being limited to, non- transitory storage in computer-readable media, random access memory (RAM), and / or read-only memory (ROM), such as flash memory, among others. Memory is an example of computer-readable media.

[0150] Computer-readable media includes permanent and non- permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, without being limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disks (DVDs) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. According to the definitions provided herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0151] It is also important to note that the terms "comprises", "comprising", or other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0152] The description can be described in the general context of computer- executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The manner in which the computer-executable instructions are executed includes the use of a

[0153] The various embodiments in the specification are described in progressive manner, and the same or similar parts among the various embodiments can be mutually referred to, and each embodiment focuses on the difference from other embodiments. In particular, the system embodiments are described more simply because they are basically similar to the method embodiments, and the relevant parts can be referred to the part of the method embodiments.

[0154] The above only describes the preferred embodiments of the present application, and of course cannot limit the scope of the present application, and the equivalent changes made according to the claims of the present application are still within the scope of the present application.

Claims

1. A method for detecting commutation failure precursors based on feature trajectory comparison, characterized in that, The method includes: Starting from the moment of commutation, the real-time current of the converter valve is sampled, and a real-time waveform representing the current is constructed based on the real-time current of the converter valve. Sampling of grid operating parameters generates an ideal waveform corresponding to each sampling moment in the real-time waveform. This sampling specifically includes: sampling grid operating parameters, including: the effective value of the inverter side line voltage, the fundamental angular frequency, the commutation voltage offset angle, the transformer turns ratio, the commutation inductance, and the DC current before commutation begins; substituting these grid operating parameters into the theoretical model of the converter valve current to determine the real-time current of the converter valve during normal commutation; and at each... At each sampling time, an ideal time-series feature vector is determined based on the real-time current of the converter valve during normal commutation. This ideal time-series feature vector includes: ideal normalized instantaneous amplitude, ideal normalized instantaneous rate of change, and ideal normalized instantaneous curvature. Based on the ideal time-series feature vector at each sampling time, an ideal waveform corresponding to each sampling time in the real-time waveform is obtained. Substituting the grid operating parameters into the theoretical model of the converter valve current to determine the real-time current of the converter valve during normal commutation specifically includes: determining the theoretical model of the converter valve current, where the theoretical model of the converter valve current is: ； in, This represents the real-time current of the commutator valve during normal commutation. This represents the real-time current of the valve during normal commutation. This is the effective value of the inverter side line voltage. The fundamental angular frequency, This refers to the commutation voltage offset angle. Here, π represents the transformer turns ratio, π is the reference point representing 180° electrical angle, and β is the lead firing angle. Here, t is the commutation inductor, and t is the current sampling time. , The moment of commutation start. , This is the theoretical end of the commutation. , This is the DC current before commutation begins. For trigger angle, The shut-off angle is determined by substituting the power grid operating parameters into the theoretical model of the converter valve current, thus determining the real-time current of the converter valve during normal commutation. The trajectory deviation index is determined by weighting the distance between the real-time waveform and the ideal waveform using preset weighting coefficients. The trajectory deviation index is compared with a preset discrimination threshold, and the commutation failure is predicted based on the comparison result.

2. The method for detecting commutation failure precursors based on feature trajectory comparison according to claim 1, characterized in that, The process of sampling the real-time current of the converter valve from the start of commutation and constructing a real-time waveform representing the current based on the real-time current of the converter valve specifically includes: The real-time current of the converter valve is sampled from the moment of commutation. At each sampling moment, a time-series feature vector is determined based on the real-time current of the converter valve. The time-series feature vector includes: normalized instantaneous amplitude, normalized instantaneous rate of change, and normalized instantaneous curvature. The real-time waveform representing the current is obtained based on the time-series feature vector at each sampling time.

3. The method for detecting commutation failure precursors based on feature trajectory comparison according to claim 2, characterized in that, At each sampling moment, a time-series feature vector is determined based on the real-time current of the converter valve. This time-series feature vector includes: normalized instantaneous amplitude, normalized instantaneous rate of change, and normalized instantaneous curvature, specifically including: Collect the DC current before the commutation of the converter valve begins; The normalized instantaneous amplitude is determined based on the ratio of the real-time current of the converter valve to the DC current before commutation begins. The normalized instantaneous rate of change is determined based on the ratio of the first derivative of the real-time current of the converter valve to the DC current before commutation begins. The normalized instantaneous curvature is determined based on the ratio of the second derivative of the real-time current of the converter valve to the DC current before commutation begins.

4. A method for detecting commutation failure precursors based on feature trajectory comparison according to claim 1 or 2, characterized in that, The process of determining the trajectory deviation index by weighting the distance between the real-time waveform and the ideal waveform using preset weighting coefficients specifically includes: according to For any sampling time, the weighted distances between the two corresponding time-series feature vectors of the real-time waveform and the ideal waveform are calculated to obtain the weighted distance between any two sampling times. For weighted distance, , , To set the preset weighting coefficients, , Let be the temporal feature vector in the ideal waveform at the j-th sampling time. x ∈ (1, 2, 3), For the first i The time-series feature vector in the real-time waveform at each sampling time. x ∈ (1, 2, 3), ti For the first i Each sampling time, tj For the first j Each sampling time; Construct a distance matrix using the weighted distance between any two sampling times; In the distance matrix, the path with the smallest cumulative sum of weighted distances from the sampling time at the start of commutation to the current sampling time is determined, and the cumulative sum of weighted distances on the path is used as the trajectory deviation index.

5. The method for detecting commutation failure precursors based on feature trajectory comparison according to claim 4, characterized in that, The step of comparing the trajectory deviation index with a preset discrimination threshold and predicting whether commutation failure exists based on the comparison result specifically includes: If the trajectory deviation index is greater than the preset discrimination threshold, then a commutation failure exists; If the trajectory deviation index is less than or equal to the preset discrimination threshold, then there is no commutation failure.

6. A commutation failure precursor detection system based on feature trajectory comparison, characterized in that, The system includes: The real-time waveform construction module is used to sample the real-time current of the converter valve from the start of commutation, construct a real-time waveform representing the current based on the real-time current of the converter valve, and sample grid operating parameters to generate an ideal waveform corresponding to each sampling moment in the real-time waveform. Specifically, this includes: sampling grid operating parameters, including: RMS value of inverter side line voltage, fundamental angular frequency, commutation voltage offset angle, transformer turns ratio, commutation inductance, and DC current before commutation begins; substituting the grid operating parameters into the theoretical model of the converter valve current to determine the commutation during normal commutation. The real-time current of the valve; at each sampling moment, an ideal time-series feature vector is determined based on the real-time current of the converter valve during normal commutation. The ideal time-series feature vector includes: ideal normalized instantaneous amplitude, ideal normalized instantaneous rate of change, and ideal normalized instantaneous curvature; based on the ideal time-series feature vector at each sampling moment, an ideal waveform corresponding to each sampling moment in the real-time waveform is obtained; the step of substituting the power grid operating parameters into the theoretical model of the converter valve current to determine the real-time current of the converter valve during normal commutation specifically includes: determining the theoretical model of the converter valve current, wherein the theoretical model of the converter valve current is: ； in, This represents the real-time current of the commutator valve during normal commutation. This represents the real-time current of the valve during normal commutation. This is the effective value of the inverter side line voltage. The fundamental angular frequency, This refers to the commutation voltage offset angle. Here, π represents the transformer turns ratio, π is the reference point representing 180° electrical angle, and β is the lead firing angle. Here, t is the commutation inductor, and t is the current sampling time. , The moment of commutation start. , This is the theoretical end of the commutation. , This is the DC current before commutation begins. For trigger angle, The shut-off angle is determined by substituting the power grid operating parameters into the theoretical model of the converter valve current, thus determining the real-time current of the converter valve during normal commutation. An ideal waveform generation module is used to sample power grid operating parameters and generate an ideal waveform corresponding to each sampling moment in the real-time waveform. The trajectory deviation index acquisition module is used to determine the trajectory deviation index by weighting the distance between the real-time waveform and the ideal waveform using preset weighting coefficients. The commutation failure precursor detection module is used to compare the trajectory deviation index with a preset detection threshold and predict whether commutation failure exists based on the comparison result.

7. A computer-readable storage medium, characterized in that, The device stores a computer program that, when executed by a processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 5.

8. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 5.

Images