A train converter open-circuit fault positioning method and system

Abstract

The present application belongs to the technical field of circuit fault diagnosis, and particularly relates to a train converter open-circuit fault positioning method and system. An instant learning prediction model based on a variable near-neighbor sample set is constructed to predict output phase current, and a dynamic threshold fault detection function based on Z-score and an open-circuit fault sensitive factor based on current maximum value are adopted to realize rapid detection and step-by-step accurate positioning of faults. This method has the characteristics of intuitive modeling and no need for offline training, and shows good results and strong robustness to the online instant demand of dynamic industrial detection. The design of step-by-step fault diagnosis, which preliminarily locates first and then accurately locates, also effectively reduces the complexity of calculation, and improves the reliability and safety of the system. Through this method, the fault detection and diagnosis process of the train converter becomes more efficient and accurate.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of circuit fault diagnosis, and particularly relates to a train converter open-circuit fault positioning method and system. BACKGROUND

[0002] The train converter undertakes multiple responsibilities such as energy conversion, power output control and maintenance of system stability. With the development of electronic equipment towards higher integration and lightweight design, its application environment becomes increasingly complex and harsh. Under such a big environment, the converter gradually becomes one of the most easily damaged parts among many components. Once the converter fails, not only the control performance of the system will be damaged, but also the safety of the entire system or equipment may be threatened in extreme cases. The failure of the train converter is mainly divided into open-circuit and short-circuit types. Once the short-circuit fault occurs, the destruction speed is extremely fast and the destructive power is huge, and usually needs to rely on integrated hardware to prevent. The open-circuit fault is often difficult to detect in the early stage, and if the system runs for a long time in the presence of an open-circuit fault, it may cause a chain damage of other components. Therefore, how to diagnose the open-circuit fault of the train converter in time is crucial for ensuring the stable operation of the converter and preventing potential systemic risks. SUMMARY

[0003] The technical problem to be solved by the application is to overcome the deficiencies and defects mentioned in the background, and to provide a train converter open-circuit fault positioning method and system.

[0004] To solve the above technical problems, the technical solution provided by the application is:

[0005] In a first aspect, the application provides a train converter open-circuit fault positioning method, comprising:

[0006] S1: According to the current sampling samples and historical sample set of the three-phase output current and intermediate DC link capacitor voltage of the train converter, the distance between the samples is calculated to determine the total number of elements of the variable near-neighbor sample set;

[0007] S2: According to the total number of elements, an instantaneous learning prediction model of the three-phase output current of the train converter is constructed to obtain an output current residual value;

[0008] S3: A dynamic threshold fault detection function based on Z-score is constructed according to the output current residual value;

[0009] S4: Whether a fault occurs is judged according to the output current residual value and the dynamic threshold value;

[0010] S5: If a fault occurs, the output current and capacitor voltage after the fault are collected, and an open-circuit fault sensitive factor based on the maximum value of the current is established;

[0011] S6: constructing a preliminary fault positioning judgment function based on the sensitive factor according to the open-circuit fault sensitive factor, and determining a preliminary positioning result of the open-circuit fault of the train converter;

[0012] S7: constructing an accurate fault positioning judgment function according to the preliminary positioning result, and obtaining an accurate positioning result of the open-circuit fault of the train converter.

[0013] In a second aspect, the present application further provides a train converter open-circuit fault positioning system, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor implements the steps of the method of the first aspect when executing the computer program.

[0014] Compared with the prior art, the present application has the following beneficial effects:

[0015] The train converter open-circuit fault positioning method provided in the present application realizes rapid detection and step-by-step accurate positioning of faults by constructing an instant learning prediction model based on a variable near-neighbor sample set to predict output phase current, and simultaneously adopting a dynamic threshold fault detection function based on Z-score and an open-circuit fault sensitive factor based on current maximum value. This method has the characteristics of intuitive modeling and no need for offline training, and shows good effects and strong robustness to the online instant demand of dynamic industrial detection. The design of step-by-step fault diagnosis of preliminary positioning first and accurate positioning later also effectively reduces the complexity of calculation, and improves the reliability and safety of the system. Through this method, the fault detection and diagnosis process of the train converter becomes more efficient and accurate. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0017] Figure 1 is one of the flowcharts of a train converter open-circuit fault positioning method of a preferred embodiment of the present application;

[0018] Figure 2 is a three-level inverter main circuit topology of a preferred embodiment of the present application;

[0019] Figure 3 is a train converter open-circuit fault positioning method flowchart of a preferred embodiment of the present application;

[0020] Figure 4 is S a1An open-circuit fault detection result map;

[0021] Figure 5 S1 is the S of the preferred embodiment of the present application b2 An open-circuit fault preliminary positioning result map.

[0022] Figure 6 S1 is the S of the preferred embodiment of the present application b2 An open-circuit fault accurate positioning result map. DETAILED DESCRIPTION

[0023] In order to facilitate the understanding of the present application, the present application will be described in more detail and in a more complete, specific manner below in conjunction with the accompanying drawings and preferred embodiments, but the scope of protection of the present application is not limited to the following specific embodiments.

[0024] Unless otherwise defined, all the technical and scientific terms used herein have the same meanings as commonly understood by one of ordinary skill in the art. The technical terms used herein are only for the purpose of describing the specific embodiments and are not intended to limit the scope of protection of the present application.

[0025] Unless otherwise specified, the various raw materials, reagents, instruments and equipment used in the present application can be purchased from the market or can be prepared by existing methods.

[0026] Unless otherwise defined, the technical terms or scientific terms used in the present application should be understood as the usual meanings by those skilled in the art. The terms "first", "second" and similar words used in the present application do not represent any order, quantity or importance, but are only used to distinguish different components. Similarly, "one" or "a" and similar words do not represent a quantity limit, but represent the existence of at least one. The words "connected" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to represent relative positional relationships, and when the absolute position of the described object changes, the relative positional relationship also changes accordingly.

[0027] Please refer to Figure 1 The present application provides a train converter open-circuit fault positioning method, comprising:

[0028] S1: According to the current sampling samples and historical sample set of three-phase output current and intermediate DC link capacitor voltage of the train converter, the distance between the samples is calculated, and the total number of elements of the variable near-neighbor sample set is determined.

[0029] In this step, the current sampling sample refers to the sampling data of the three-phase output current of the train converter and the intermediate DC link capacitor voltage at the current moment. The historical sample set includes a normal state historical sample subset and a fault state historical sample subset. The normal state historical sample subset refers to a sampling data set of not less than one complete current period under normal state; the fault state historical sample subset refers to a sampling data set of not less than one complete current period under the single tube open circuit fault state of the train converter. The sampling data includes the three-phase output current of the train converter and the intermediate DC link capacitor voltage.

[0030] S2: According to the total number of elements, an instant learning prediction model of the three-phase output current of the train converter is constructed to obtain an output current residual value.

[0031] S3: A dynamic threshold fault detection function based on Z-score is constructed according to the output current residual value.

[0032] S4: Whether a fault occurs is judged according to the output current residual value and the dynamic threshold.

[0033] S5: If a fault occurs, the output current and the capacitor voltage after the fault are collected, and an open circuit fault sensitive factor based on the current maximum value is established.

[0034] S6: A preliminary fault positioning judgment function based on the sensitive factor is constructed according to the open circuit fault sensitive factor, and a preliminary positioning result of the open circuit fault of the train converter is determined.

[0035] S7: An accurate fault positioning judgment function is constructed according to the preliminary positioning result, and an accurate positioning result of the open circuit fault of the train converter is obtained.

[0036] The train converter open circuit fault positioning method realizes rapid detection and step-by-step accurate positioning of the fault by constructing an instant learning prediction model based on a variable near-neighbor sample set to predict the output phase current, and using a dynamic threshold fault detection function based on Z-score and an open circuit fault sensitive factor based on the current maximum value. This method has the characteristics of intuitive modeling and no need for offline training, and shows good effect and strong robustness to the online instant demand of dynamic industrial detection. The design of step-by-step fault diagnosis of preliminary positioning first and accurate positioning later also effectively reduces the complexity of calculation, and improves the reliability and safety of the system. Through this method, the fault detection and diagnosis process of the train converter becomes more efficient and accurate.

[0037] This embodiment takes a three-level inverter in a train converter as an example for illustration. The schematic diagram of the main circuit of the three-level inverter is shown in Figure 2 The DC side voltages U1 and U2 are 1800V, the DC side capacitors C1 and C2 are 16mF, and the motor parameters used are shown in Table 1.

[0038] Table 1 Motor parameters

[0039]

[0040] Optionally, the S1 comprises:

[0041] S11: calculating the distance between the sampling sample at the qth sampling moment and the samples in the normal state historical sample subset, and the formula is as follows:

[0042]

[0043] In the formula, d l is the distance between the sampling sample at the qth sampling moment and the lth sample in the normal state historical sample subset, l∈{1, 2,..., L}, L is the total number of samples in the normal state historical sample subset; f d is the calculation function of the distance between samples, i1[q] and i2[q] are any two-phase output phase currents in the sampling sample at the qth sampling moment, and are any two-phase output phase currents of the lth sample in the normal state historical sample subset, i1[q] and are the same-phase output phase current samples, i2[q] and are the same-phase output phase current samples; u n is the voltage of the nth capacitor in the sampling sample at the qth sampling moment, is the voltage of the nth capacitor in the lth sample in the normal state historical sample subset, where n=1, 2,..., N, and N is the total number of intermediate DC link capacitors.

[0044] S12: constructing a set of distances between the sampling sample at the qth sampling moment and each sample in the normal state historical sample subset, and the formula is as follows:

[0045]

[0046] D[q]={d1[q],...,d l [q],...,d L [q]};

[0047] In the formula, D[q] is the set of distances between the sampling sample at the qth sampling moment and the L samples in the normal state historical sample subset;

[0048] Arranging the elements in the set D[q] in ascending order to form an ascending distance set D′[q], which satisfies the following relationship:

[0049] D′[q]=sort(D[q])={d′1[q],...,d′ l ​[q],...,d′ L [q]};

[0050] wherein sort(·) is a function of arranging elements of a set in ascending order, d′ l [q] represents the value of the lth element in D[q] arranged in ascending order, and the elements in D′[q] satisfy the inequality: d′ l [q]≤d′ l+1 [q]≤...≤d′ L [q]};

[0051] For any two consecutive elements of the set D′[q], the difference between them is calculated to form a variable neighborhood sample set E[q], which satisfies the following relationship:

[0052] E[q]={e1,...,e l ,…,e L-1},e l =d′ l+1 [q]-d′ l [q];

[0053] wherein el represents the difference between adjacent elements d′ l+1 [q] and d′ l [q] in the ascending distance set D′[q].

[0054] Specifically, in the embodiment, i1[q] and i2[q] are i a [q] and i b [q], respectively, N=2, the intermediate DC link capacitor voltage includes u1[q] and u2[q], the sample similarity measurement method selects the Euclidean distance, and the specific calculation

[0055] is as follows:

[0056]

[0057] S13: calculating the total number of elements of the variable neighborhood sample set, and the formula is:

[0058] k[q]=argmax(E[q]);

[0059] wherein k[q] is the total number of elements of the variable neighborhood sample set of the sampling sample at the qth sampling moment, argmax(·) represents finding the element with the maximum value in the set and returning the index l∈{1,2,...,L-1} of the element.

[0060] Optionally, the S2 includes:

[0061] S21: establishing an instantaneous learning prediction model of the output phase current, which is represented as:

[0062]

[0063] In the formula, is the prediction value of the output phase current i3[q] at the qth sampling moment, i3[q] is the true value, the value of which is the output phase current sampling value at the qth sampling moment, i3[q] is the output phase current in the three-phase current except i1[q] and i2[q], f r (·) is a local prediction function of the random forest, the input of which includes the output phase currents i1[q], i2[q], the intermediate DC link capacitor voltages u1[q], u2[q], …, u n [q], …, u N [q], the total number of elements k[q] of the variable neighborhood sample set at the qth sampling moment, the normal state historical sample subset X normal .

[0064] S22: Calculate the prediction residual of the output phase current, the formula is:

[0065]

[0066] In the formula, r3[q] is the prediction residual of the output phase current i3[q] at the qth sampling moment.

[0067] Specifically, in the embodiment, the instantaneous learning prediction model is constructed for the C-phase output phase current, that is, r3[q] is r c [q].

[0068] Optionally, the S3 comprises:

[0069] S31: Construct a dynamic detection threshold upper limit based on Z-score, the formula is:

[0070] r 3_h [q] = μ3[q] + (z0*λ[q])σ3[q];

[0071] In the formula, r 3_h [q] is the dynamic detection threshold upper limit of r3[q], z0 is the initial threshold, λ[q] is the dynamic threshold adjustment factor at the qth sampling moment, μ3[q] and σ3[q] are the mean value and standard deviation of r3[q] respectively. Wherein, the calculation formula of μ3[q] is:

[0072] If t < 0, r3[t] = 0;

[0073] In the formula, t represents the sampling moment, T1 is the dynamic detection threshold sliding window size, which is a positive integer. The calculation formula of σ3[q] is:

[0074]

[0075] S32: Constructing the lower limit of the dynamic detection threshold based on Z-score, the formula is as follows:

[0076] r 3_lo [q] = μ3 [q] - (z0 * λ[q]) σ3 [q];

[0077] In the formula, r 3_lo [q] is the lower limit of the dynamic detection threshold of r3[q].

[0078] S33: Constructing the dynamic threshold adjustment factor, the formula is as follows:

[0079]

[0080] In the formula, λ[q] is the dynamic threshold adjustment factor at the qth sampling time, α is a constant greater than 0 and less than 1, w[q] is the cumulative number of samples of r3[t] continuously exceeding the dynamic detection threshold at the qth sampling time, w th is the number threshold of r3[t] continuously exceeding the dynamic detection threshold, and w th > 1. The calculation formula of w[q] is as follows:

[0081]

[0082] In the formula , w represents the number of times that r3[t] continuously exceeds the dynamic detection threshold minus 1.

[0083] S34: Constructing the dynamic threshold fault detection function, the formula is as follows:

[0084]

[0085] In the formula, R[q] is the fault detection function value at the qth sampling time. When R[q] = 1, it is determined that the system has failed. When R[q] = 0, it is determined that the system is in a normal state.

[0086] Optionally, the S5 comprises:

[0087] S51: Collecting the output phase current at the pth sampling time after the fault, calculating the maximum value of the output phase current, and the formula is as follows:

[0088]

[0089] In the formula, is the maximum output phase current of the x-phase bridge arm calculated in the sliding window T2 at the pth (p∈[q, q+O]) sampling time, O is the total number of output phase current samples collected after the fault, which is the integer value of the ratio of the output phase current period to the sampling interval, T2 is the size of the sliding window for calculating the maximum output phase current, which is not less than one output phase current period, i x [j] is the output phase current of the x-phase bridge arm at the jth sampling time.

[0090] Specifically, in this embodiment, the value of T2 is equal to one output phase current period:

[0091] S52: Calculate the minimum output phase current, the formula is:

[0092]

[0093] In the formula, is the minimum output phase current of the x-phase bridge arm calculated in the sliding window T2 at the pth (p∈[q, q+O]) sampling time.

[0094] S53: Construct the open-circuit fault sensitive factor based on the maximum output phase current, the formula is:

[0095]

[0096] In the formula, is the open-circuit fault sensitive factor of the maximum output phase current of the x-phase bridge arm at the pth (p∈[q, q+O]) sampling time, and ξ is a non-zero infinitesimal positive number.

[0097] Optionally, the S6 comprises:

[0098] S61: Construct a preliminary fault location judgment function, the formula is:

[0099]

[0100] In the formula, is the preliminary fault location judgment function at the pth sampling time, β th_x is the threshold value of the fault sensitive factor of the maximum output phase current of the x-phase bridge arm. When indicates that a fault occurs in the inner power switch tube of the x-phase bridge arm of the converter; when indicates that a fault occurs in the outer power switch tube of the three-phase bridge arm of the converter.

[0101] S62: Determine the preliminary fault location result, the formula is:

[0102]

[0103] wherein m[p] is the preliminary fault location result at the pth sampling time, the value of which is the number of the power switch tube in the converter, m[p] e {1, 2, …, M}, M is the total number of power switch devices.

[0104] Specifically, in the embodiment, m[p] e {1, 2, …, 12}. From the preliminary fault location result, it can be determined that the open-circuit fault is located at the inner side of the bridge arm or the outer side of the bridge arm, if it is at the inner side, it can be directly located to the fault phase, then m[p] has 2 different values; if it is at the outer side, m[p] has 6 different values.

[0105]

[0106] When the train three-level inverter occurs an open-circuit fault, the preliminary fault location result can determine whether the open-circuit fault is located at the inner side of the bridge arm or the outer side of the bridge arm, if it is at the inner side, it can be directly located to the fault phase, then m[p] has 2 different values; if it is at the outer side, m[p] has 6 different values.

[0107] Optionally, S7 comprises:

[0108] S71: combining the preliminary fault location result and the fault state historical sample subset, calculating the post-fault output phase current prediction residual, the formula is:

[0109]

[0110] wherein, is the post-fault output phase current prediction residual at the pth sampling time, is the prediction value of the post-fault output phase current at the pth sampling time, i3[p] is the post-fault output phase current value of any phase in the sampling sample at the pth sampling time. The calculation formula of is:

[0111]

[0112] wherein i1[p] and i2[p] are the post-fault output phase current values of any two phases (except the phase where i3[p] is located) in the sampling sample at the pth sampling time, k[p] is the total number of elements of the variable neighbor sample set of the sampling sample under fault at the pth sampling time, u n [p] is the voltage of the nth capacitor of the sampling sample under fault at the pth sampling time, X fault_m[p] is the fault state sampling data of the power switch tube numbered m[p] in the fault state historical sample subset X fault f. r f (·) is the local prediction function of the random forest under fault, the input of which includes the post-fault output phase current i1[p], i2[p] at the pth sampling time, the intermediate DC link capacitor voltage u1[p], u2[p], …, u n [p], …, u N ​[p], the total number of elements k[p] in the variable nearest neighbor set of the fault sample at the p-th sampling time, and the fault state history sample subset X. fault_m[p] .

[0113] Specifically, in this embodiment, the predicted residual of the output phase current after a fault is calculated based on the output phase current of phase C. That is

[0114] S72: Based on the residual prediction of the fault output phase current, construct an accurate fault location judgment function, as shown in the following formula:

[0115]

[0116] In the formula, H[p] is the value of the fault location and judgment function corresponding to the sampled sample at the p-th sampling time. It is based on the p-th sampling time. The fault detection function value, The calculation formula is:

[0117] In the formula, w m[p] [p] represents the statistics at the p-th sampling time. The cumulative number of samples, w, for continuous hyperdynamic detection threshold th_m[p] yes The threshold number of consecutive hyper-dynamic detections, w th_m[p] >1. w m[p] The formula for calculating [p] is:

[0118]

[0119] In the formula, express Subtract 1 from the number of consecutive times the dynamic detection threshold is exceeded. It is the p-th sampling time. Dynamic detection threshold upper limit, It is the p-th sampling time. Lower limit of dynamic detection threshold. The calculation formula is:

[0120]

[0121] In the formula, z 0_m[p] It is the initial threshold, λ m[p] [p] is the dynamic threshold adjustment factor at the p-th sampling time, μ 3_m[p] [p] and σ 3_m[p] [p] are respectively The mean and standard deviation. The calculation formula is:

[0122]

[0123] In the formula, μ 3_m[p] The calculation formula of [p] is:

[0124] If t < 0,

[0125] In the formula, T3 is a dynamic detection threshold sliding window size, which is a positive integer. σ 3_m[p] The calculation formula of [p] is:

[0126]

[0127] λ m[p] The calculation formula of [t] is:

[0128]

[0129] In the formula, α m[p] is a constant greater than 0 and less than 1.

[0130] The stable value of the accurate fault positioning judgment function value is calculated by judging the accurate fault positioning judgment function value, and the formula is:

[0131]

[0132] In the formula, H f is the stable value of the accurate positioning judgment function, H[O] is the accurate positioning judgment function value corresponding to the sampling sample at the Oth sampling time after the fault, and is the final value of the accurate positioning judgment function.

[0133] S73: Determine the accurate fault positioning result, and the formula is:

[0134] m o = f o (H f );

[0135] In the formula, m o is the accurate fault positioning result, the value of which is the number of the power switch tube in the converter, m o ∈{1,2,…,M}, f o (·) is a function of mapping H f to the number of the power switch tube in the converter.

[0136] It should be noted that in the present embodiment, the mapping function table of H f and the number of the power switch tube in the converter can be constructed as shown in Table 2. The fault diagnosis flow chart is as shown in Figure 3 .

[0137] Table 2 Relationship between Precise Fault Location Judgment Function Values ​​and Fault Location

[0138]

[0139] In this embodiment, S is used respectively a1 Open circuit fault detection process and S b2 The effectiveness of the method will be illustrated using the open-circuit fault diagnosis process as an example. First, the open-circuit fault detection method will be verified. a1 Open circuit fault detection results as follows Figure 4 As shown. By Figure 4 It can be seen that after the fault occurs at the 200th sample, the predicted residual of the output phase current increases rapidly after the 210th sample. The fault detection function value changes from 0 to 1 at the 214th sample, indicating that an open-circuit fault has been detected in the system.

[0140] Next, the open-circuit fault diagnosis method was verified. b2 Preliminary location results of open circuit fault are as follows Figure 5 As shown. By Figure 5 It can be seen that m[p] jumps to values ​​6 and 7 respectively at the 210th sample, indicating that an open-circuit fault has occurred in the power switch on the inner side of phase b; the results of the precise fault location judgment function are as follows: Figure 6 As shown, the precise location judgment function value H[p] stabilizes at the value 64 at the 214th sample. Referring to Table 2, it can be seen that it is S. b2 An open circuit fault has occurred.

[0141] In summary, this embodiment can detect and diagnose inverter open-circuit faults, improving the reliability and safety of the system.

[0142] This application also provides a train converter open-circuit fault location system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above-described method. This train converter open-circuit fault location system can implement various embodiments of the above-described train converter open-circuit fault location method and achieve the same beneficial effects; further details are omitted here.

[0143] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method of open circuit fault location for a train converter, characterized in that, The method comprises the following steps: S1: according to the current sampling samples and the historical sample set of the three-phase output current of the train converter and the capacitor voltage of the intermediate DC link, the distance between the samples is calculated, and the total number of elements of the variable neighbor sample set is determined; S2: according to the total number of elements, an instant learning prediction model of the three-phase output current of the train converter is constructed, and an output current residual value is obtained; S3: a dynamic threshold fault detection function based on Z-score is constructed according to the output current residual value; S4: whether a fault occurs is judged according to the output current residual value and the dynamic threshold; S5: if a fault occurs, the output current and the capacitor voltage after the fault are collected, and an open-circuit fault sensitive factor based on the maximum value of the current is established; S6: a preliminary fault positioning judgment function based on the sensitive factor is constructed according to the open-circuit fault sensitive factor, and a preliminary positioning result of the open-circuit fault of the train converter is determined; S7: an accurate fault positioning judgment function is constructed according to the preliminary positioning result, and an accurate positioning result of the open-circuit fault of the train converter is obtained; The S2 comprises: S21: an instant learning prediction model of the output phase current is established, which satisfies the following relationship: In the formula, is the predicted value of the output phase current i3[q] at the qth sampling moment, i3[q] is the true value, the value of which is the output phase current sampling value at the qth sampling moment, i3[q] is the output phase current in the three-phase current except i1[q] and i2[q], f r (i) is a random forest local prediction function, the input of which includes the output phase currents i1[q], i2[q], the intermediate DC link capacitor voltages u1[q], u2[q], …, u n [k-1] [q], …, u N [k-1] [q], k[q] represents the total number of elements of the variable neighborhood sample set of the sampling sample at the qth sampling moment, X normal represents a normal state historical sample subset; S22: the prediction residual of the output phase current is calculated, which satisfies the following relationship: In the formula, r3[q] is the prediction residual of the output phase current i3[q] at the qth sampling time; The S3 comprises: S31: a dynamic detection threshold upper limit based on Z-score is constructed, which satisfies the following relationship: r 3_h [q] = μ3[q] + (z0 * λ[q]) σ3[q]; In the formula, r 3_h [q] is a dynamic detection threshold upper limit of r3[q], z0 is an initial threshold, λ[q] is a dynamic threshold adjustment factor at the qth sampling time, and μ3[q] and σ3[q] are the mean and standard deviation of r3[q], respectively, wherein μ3[q] satisfies the following relationship: if t < 0, r3[t] = 0; In the formula, t represents a sampling time, T 1 is a dynamic detection threshold sliding window size, which is a positive integer, and σ3[q] satisfies the following relationship: S32: a dynamic detection threshold lower limit based on Z-score is constructed, which satisfies the following relationship: r 3_lo [q] = μ3[q] - (z0 * λ[q]) σ3[q]; wherein r 3_lo [q] is the lower dynamic detection threshold for r3[q] S33: a dynamic threshold adjustment factor is constructed, which satisfies the following relationship: In the formula, λ[q] is a dynamic threshold adjustment factor at the qth sampling time, α is a constant greater than 0 and less than 1, w[q] is the cumulative sample number of the r3[t] continuous over-dynamic detection threshold value counted at the qth sampling time, w th is the number threshold of the r3[t] continuous over-dynamic detection threshold value, w th >1, wherein the calculation formula of w[q] is as follows: In the formula denotes the number of times r3[t] continuously exceeds the dynamic detection threshold minus 1, r3[t] representing the prediction residual of the output phase current i3[q]; S34: a dynamic threshold fault detection function is constructed, which is as follows: In the formula, R[q] is the fault detection function value at the qth sampling time; The S5 comprises: S51: the output phase current at the pth sampling time after the fault is collected, the maximum value of the output phase current is calculated, which satisfies the following relationship: In the formula, is the maximum output phase current of the x (x∈{1,2,3}) phase bridge arm calculated in the sliding window T2 at the pth (p∈[q,q+O]) sampling time, O is the total number of output phase current samples collected after the fault, which is the integer value of the ratio of the output phase current period to the sampling interval, T 2 is the size of the sliding window for calculating the maximum output phase current, which is not less than one output phase current period, i x [j] is the output phase current of the x phase bridge arm at the jth sampling time. S52: the minimum value of the output phase current is calculated, which satisfies the following relationship: In the formula, is the minimum output phase current of the x (x e {1, 2, 3}) phase arm calculated in the sliding window T2 at the pth (p e [q, q+O]) sampling time. S53: an open-circuit fault sensitive factor based on the maximum value of the output phase current is constructed, which satisfies the following relationship: In the formula, is the open-circuit fault sensitive factor of the maximum value of the output phase current of the pth (p∈[q, q+O]) sampling time phase bridge arm of the converter, and ξ is a non-zero infinitesimal positive number.

2. The train converter open fault location method of claim 1, wherein, The S1 comprises: S11: the distance between the sampling sample at the qth sampling time and the samples in the normal state historical sample subset is calculated, which satisfies the following relationship: wherein d l [q] is the distance between the sampling sample at the qth sampling time and the lth sample in the normal state historical sample subset, l ∈ {1, 2, …, L}, L is the total number of samples in the normal state historical sample subset; f d (i) is a calculation function of the distance between samples, i1[q] and i2[q] are any two-phase output phase currents in the sampling sample at the qth sampling time, and are any two-phase output phase currents of the lth sample in the normal state historical sample subset, i1[q] and are any two-phase output phase currents of the lth sample in the normal state historical sample subset; u are any two-phase output phase currents of the lth sample in the normal state historical sample subset; u n [q] is the voltage of the nth capacitor in the sampling sample at the qth sampling time, is the voltage of the nth capacitor in the lth sample in the normal state historical sample subset, wherein n = 1, 2, …, N, and N is the total number of intermediate DC link capacitors; S12: the distance between the sampling sample at the qth sampling time and each sample in the normal state historical sample subset is constructed into a set, which satisfies the following relationship: D[q] = {d1[q],...,d l [q],...,d L [q]}; In the formula, D[q] is the set of distances between the sampling sample at the qth sampling time and the L samples in the normal state historical sample subset; The elements in the set D[q] are arranged in ascending order from small to large to form an ascending distance set D′[q], which satisfies the following relationship: D'[q] = sort(D[q]) = {d'1[q],...,d'k[q]} where k = |D[q]|. l D'[q] = sort(D[q]) = {d'1[q],...,d'k[q]} where k = |D[q]|. L D'[q] = sort(D[q]) = {d'1[q], where sort(i) is a function that sorts the set elements in ascending order, d' l [q] represents the value of the lth element in D[q] after sorting in ascending order, and the elements in D' [q] satisfy the inequality: d' l [q]≤…≤d′ l+1 [q]≤…≤d′ L [q] ; For any two consecutive elements of the set D′[q], the difference between them is calculated to form a variable neighbor sample set E[q], which satisfies the following relationship: E[q] = {e1,...,e l ,...,e L-1}, e l = d' l+1 [q] - d' l [q]; where e l denotes the difference between adjacent elements d' l+1 [q] in the ascending distance set D′[q] l [q] in the ascending distance set D′[q] S13: the total number of elements of the variable neighbor sample set is calculated, which satisfies the following relationship: k[q]=argmax(E[q]); Wherein, k[q] is the total number of variable neighbor sample set of sampling sample at the qth sampling moment, argmax(i) represents finding the element with the maximum value in the set and returning the index of the element, and l∈{1, 2,..., L-1}.

3. The train converter open fault location method of claim 1, wherein, The S4 comprises: When the fault detection function value at the qth sampling moment is equal to 1, it is determined that the system is in a fault state, and when the fault detection function value at the qth sampling moment is equal to 0, it is determined that the system is in a normal state.

4. The train converter open fault location method of claim 1, wherein, The S6 comprises: S61: constructing a preliminary fault location judgment function, satisfying the following relationship: In the formula, is the preliminary fault location judgment function at the pth sampling moment, β th_x is the threshold value of the fault sensitive factor of the maximum value of the x-phase bridge arm output phase current, and when indicates that a fault occurs in the inner power switch tube of the x-phase bridge arm of the converter; when indicates that a fault occurs in the outer power switch tube of the three-phase bridge arm of the converter; S62: determining a preliminary fault location result, satisfying the following relationship: Wherein, m[p] is the preliminary fault location result at the pth sampling moment, the value of which is the number of a power switch tube in the converter, m[p]∈{1, 2,..., M}, and M is the total number of power switch devices.

5. The train converter open fault location method of claim 1, wherein, The S7 comprises: S71: combining the preliminary fault location result and the historical sample subset of the fault state to calculate the fault output phase current prediction residual, satisfying the following relationship: In the formula, is the predicted residual of the output phase current after the fault at the pth sampling moment, is the predicted value of the output phase current after the fault at the pth sampling moment, and i3[p] is the value of any phase output phase current in the sampling sample after the fault at the pth sampling moment; The calculation formula of i3[p] is as follows: where i1[p] and i2[p] are the faulted output phase current values of any two phases except the phase where i3[p] is located in the sampling sample at the pth sampling moment, k[p] is the total number of elements of the variable neighborhood sample set of the sampling sample under fault at the pth sampling moment, u n [p] is the voltage of the nth capacitor of the sampling sample under fault at the pth sampling moment, X fault_m [ p ] is the fault state sampling data of the power switch tube numbered m[p] in the fault state historical sample subset X fault is the local prediction function of the random forest under fault, the input of which includes the output phase current i1[p], i2[p] under fault at the pth sampling moment, the intermediate DC link capacitor voltage u1[p], u2[p], …, u n [p], …, u N [p], the total number of elements k[p] of the variable neighborhood sample set of the sampling sample under fault at the pth sampling moment, the fault state historical sample subset X fault_m[p] ;​ S72: constructing an accurate fault location judgment function according to the fault output phase current prediction residual, satisfying the following relationship: In the formula, H[p] is a fault accurate positioning judgment function value corresponding to a sampling sample at a pth sampling moment, is a fault detection function value based on at a pth sampling moment, The calculation formula is as follows: In the formula, w m[p] [p] is the statistical value of the pth sampling time The cumulative sample number of consecutive over-dynamic detection threshold, w th_m[p] is The number threshold of consecutive over-dynamic detection threshold, w th_m[p] >1, w m[p] The calculation formula of [p] is as follows: In the formula, represents the number of times of continuously exceeding the dynamic detection threshold minus 1, is the pth sampling moment upper limit of the dynamic detection threshold, is the pth sampling moment lower limit of the dynamic detection threshold, The calculation formula of is as follows: In the formula, z 0_m[p] is an initial threshold value, λ m[p] [p] is a dynamic threshold adjustment factor at the pth sampling time, μ 3_m[p] [p] and σ 3_m[p] [p] are the mean value and the standard deviation of respectively, The calculation formula of is: In the formula, μ 3_m[p] The formula for calculating [p] is: If In the formula, T3 is a dynamic detection threshold sliding window size, which is a positive integer, is the prediction residual of the output phase current after the fault, and σ 3_m[p] The calculation formula is: λ m[p] The formula for calculating [t] is: wherein α is a constant greater than 0 and less than 1. m[p] is a constant greater than 0 and less than 1. Judging the value of the accurate fault location judgment function, calculating the stable value thereof, satisfying the following relationship: In the formula, H f is a stable value of the accurate fault positioning judgment function, H[O] is a value of the accurate fault positioning judgment function corresponding to the sampling sample at the Oth sampling time, and is a final value of the accurate fault positioning judgment function. S73: determining an accurate fault location result, satisfying the following relationship: m o = f o (H f ); where m o is the accurate fault location result, which is the number of the power switch tube in the converter, m o ∈{1,2,…,M}, f o (i) is a function that maps H f to the number of the power switch tube in the converter.

6. A train converter open fault location system comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, The processor implements the steps of the method of any one of claims 1 to 5 when executing the computer program.

Images