MMC-HVDC system disconnection protection method based on transient current characteristic analysis

Abstract

The invention discloses an MMC-HVDC system disconnection protection method based on transient current characteristic analysis, and the method comprises the steps: collecting the transient current of a DC line, and carrying out the adaptive noise complete set empirical mode decomposition of the transient current to filter noise and obtain a reconstruction signal; when the reconstruction signal is lower than a preset broken line fault starting threshold value, broken line fault identification is started, the slope of a current characteristic component fitting curve is calculated according to a decomposition result, and if a plurality of curve slopes exist and are in positive and negative alternation or only one slope slope is zero, it is judged that a broken line fault occurs; calculating a Hausdorff distance between a positive current sequence and a negative current sequence according to the collected transient current so as to carry out fault pole selection discrimination, and calculating a direct current component of oscillation current according to an extreme point in a current characteristic component so as to carry out fault section discrimination; and executing cooperative MMC switching control according to a judgment result. According to the method, the technical problem that the fault section is difficult to quickly and accurately identify due to weak broken line fault characteristics in the flexible direct-current power transmission system is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of flexible direct current transmission system relay protection, in particular to a MMC-HVDC system line break protection method based on transient current characteristic analysis. BACKGROUND

[0002] Modular multilevel converter (MMC) based flexible direct current transmission (MMC-HVDC) technology has become the core direction of modern direct current grid development due to its significant advantages in control flexibility, scalability and large-scale new energy grid connection and consumption. However, the existing line protection is not perfect, which restricts the further development of flexible direct current transmission project, and it is urgent to research fast and reliable line protection.

[0003] At present, the protection research of direct current transmission line mainly focuses on short circuit fault, and the protection principle and scheme exploration of line break fault are relatively insufficient. In the direct current transmission line, the electrical characteristic change of line break fault is relatively weak, which is in sharp contrast to the obvious characteristics of traditional short circuit fault, which makes it difficult for traditional protection methods based on overcurrent, differential and other principles to effectively identify. At the same time, due to the lack of reliable fault feature extraction method and effective section identification principle, the existing protection scheme is difficult to quickly and accurately judge the specific section of fault, and also cannot reliably distinguish different types of faults such as single-pole line break and bipolar line break.

[0004] In addition, the noise interference, measurement error and line parameter change existing in the actual engineering operation environment further increase the difficulty of accurate identification of line break fault. These technical problems lead to the possibility of misoperation or refusal of existing protection system when dealing with line break fault, which affects the safe and stable operation of the system.

[0005] Therefore, how to quickly and accurately identify the line break section and provide a line protection method with good anti-interference ability has become a technical problem that needs to be solved by the technical personnel in the field. SUMMARY

[0006] In view of the deficiencies of the prior art, the present application provides a MMC-HVDC system line break protection method based on transient current characteristic analysis, which solves the technical problem that it is difficult to quickly and accurately identify the fault section in the flexible direct current transmission system due to the weak characteristics of line break fault by analyzing the slope characteristics and direct current components of the transient current fitting curve.

[0007] In order to solve the above technical problems, the present application adopts the following technical scheme:

[0008] A MMC-HVDC system line break protection method based on transient current characteristic analysis, comprising the following steps:

[0009] S1. Acquire the transient current of the DC line; perform adaptive noise complete set empirical mode decomposition on the transient current to filter out noise and obtain the reconstructed signal;

[0010] S2. When it is determined that the reconstructed signal is lower than the preset power failure fault initiation threshold, the disconnection fault identification in step S3 is executed.

[0011] S3. Based on the decomposition results of step S1, calculate the slope of the current characteristic component fitting curve to identify the open circuit fault: when the current characteristic component fitting curve has multiple slopes and alternates between positive and negative, or has only one slope and its value is zero, it is determined that the DC line has an open circuit fault, and step S4 is executed; otherwise, it is determined that the DC line has a non-open circuit fault, and step S1 is returned.

[0012] S4. Based on the acquired transient current, calculate the Hausdorff distance between the positive and negative current sequences of the DC line to identify the fault pole; at the same time, calculate the DC component of the oscillating current based on the extreme points in the current characteristic components to identify the fault section.

[0013] S5. Based on the judgment result of step S4, execute the corresponding collaborative MMC switching control.

[0014] As a preferred embodiment, in step S1, the reconstructed signal is obtained through the following steps:

[0015] S101. Perform adaptive noise complete set empirical mode decomposition on the transient current to obtain multiple intrinsic mode function components and residual terms;

[0016] S102. When the plurality of intrinsic mode function components satisfy the following formula, they are determined to be high-frequency components; otherwise, they are determined to be low-frequency components.

[0017] abs(∑IMF j (t))<IMF set ;

[0018] In the formula, abs(·) is the absolute value function; ∑IMF j (t) represents the summation of the j-th intrinsic mode function component over the entire data window length; IMF set Preset component threshold;

[0019] S103. After superimposing the low-frequency component with the residual term and filtering out noise, a reconstructed signal is obtained.

[0020] As a preferred embodiment, in step S2, the criterion for performing the disconnection fault identification is expressed as follows:

[0021] |X(t)|<X low

[0022] In the formula, X(t) is the reconstructed signal; X low The preset power outage fault start threshold is used.

[0023] As a preferred embodiment, in step S3, the current characteristic component fitting curve is obtained through the following steps:

[0024] S301. From the multiple intrinsic mode function components and residual terms obtained from step S1, select the component with the highest energy as the current characteristic component characterizing the current change trend.

[0025] S302. Extract all extreme points and the first and last points of the current characteristic component within a data window, and form a fitting curve for the current characteristic component by connecting the extreme points and the first and last points.

[0026] As a preferred embodiment, in step S301, the current characteristic component is represented as follows:

[0027] |Y(t)|=max(∑abs(IMF j (t)),r(t));

[0028] In the formula, Y(t) is the characteristic component of the current; max is the maximum value; and r(t) is the residual term.

[0029] As a preferred embodiment, in step S4, the Hausdorff distance between the positive and negative current sequences of the DC line is expressed as:

[0030]

[0031] In the formula, H(·) is the Hausdorff distance; A and B are the positive and negative current sequences of the DC line, respectively.

[0032] As a preferred embodiment, in step S4, the fault polarity selection judgment specifically includes:

[0033] If the Hausdorff distance satisfies the following formula, it is determined to be a bipolar open circuit fault; otherwise, it is determined to be a unipolar open circuit fault.

[0034] H < H SET ;

[0035] In the formula, H SET This is a preset distance threshold.

[0036] As a preferred embodiment, the DC component of the oscillating current is expressed as:

[0037]

[0038] In the formula, I CoDCE represents the DC component of the oscillating current; ave(·) represents the average value; E i Let be the current at the i-th extreme point on the fitted current characteristic component, i = 1, 2, ..., z.

[0039] As a preferred embodiment, in step S4, the fault segment identification specifically involves:

[0040] If the DC component of the oscillating current is approximately zero, it is determined to be an open circuit fault within the zone.

[0041] If the DC component of the oscillating current is approximately 1, it is determined to be an external open circuit fault.

[0042] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes a method for disconnection protection of an MMC-HVDC system based on transient current characteristic analysis as described above.

[0043] Compared with the prior art, the present invention has the following technical effects:

[0044] 1. This invention performs adaptive noise complete set empirical mode decomposition and reconstruction on the acquired DC line transient current, adaptively decomposing the noisy original signal into multiple intrinsic mode function components. By removing high-frequency components that represent noise and reconstructing the signal, it effectively suppresses high-frequency noise introduced by on-site electromagnetic interference, measurement errors, etc., and achieves high-precision signal purification and feature enhancement.

[0045] 2. This invention activates line protection by determining whether the reconstructed signal is below a preset power outage fault activation threshold. This criterion is based on the reconstructed signal after noise reduction processing, avoiding false activation caused by noise. At the same time, since a line breakage fault will cause the effective value of the current to drop significantly, the principle of the amplitude criterion based on the reconstructed signal is simple, and line protection can be quickly activated after the fault occurs, improving the speed of line protection.

[0046] 3. This invention identifies open circuit faults by analyzing the slope characteristics of the fitted curve of the current characteristic components. By utilizing the essential transient characteristics of open circuit faults, it achieves highly accurate identification. This method constructs a fitted curve by extracting extreme points, transforming the complex waveform identification problem into an intuitive slope pattern recognition problem. When there are multiple alternating positive and negative slopes (corresponding to oscillations), or only one zero slope (corresponding to a sudden drop and hold), these characteristics are significantly different from short circuit faults or other interference modes. Thus, this criterion has high discriminative power and can identify open circuit faults from other abnormal states with extreme accuracy.

[0047] 4. This invention comprehensively utilizes Hausdorff distance and DC component for fault pole selection and fault section identification. By calculating the Hausdorff distance between the transient current sequences of the positive and negative poles, the overall similarity of the current waveforms of the two poles can be effectively measured, thereby distinguishing between single-pole open circuits (large waveform differences) and double-pole open circuits (highly similar waveforms). Secondly, the DC component of the oscillating current obtained by calculating the extreme points of the current characteristic components accurately reflects the location attributes of the fault point: a DC component approaching zero indicates that the measurement point is electrically disconnected from the main grid (intra-zone fault), while a DC component maintained at its rated value indicates that the fault point is in another location (outtra-zone fault). Through the complementary use of the above two criteria, a comprehensive and accurate diagnosis of fault type and location is achieved. Attached Figure Description

[0048] To make the objectives, technical solutions, and advantages of the invention clearer, the invention will now be described in further detail with reference to the accompanying drawings, wherein:

[0049] Figure 1 This is a flowchart of the method proposed in the embodiments of the present invention;

[0050] Figure 2 The transient current of the disconnected circuit in this embodiment of the invention is obtained through adaptive noise complete set empirical mode decomposition diagram;

[0051] Figure 3 This is an example of an adaptive noise complete set empirical mode decomposition diagram in this invention.

[0052] Figure 4 This is a diagram showing the startup criterion results after a line breakage fault in an embodiment of the present invention;

[0053] Figure 5 This is a diagram showing the discrimination results of different wire breakage fault types in an embodiment of the present invention;

[0054] Figure 6 This is a slope diagram of the current fitting curve in an embodiment of the present invention;

[0055] Figure 7 This is a diagram of a flexible DC four-terminal ring network in an embodiment of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but only to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0057] The present invention will now be described in further detail with reference to the accompanying drawings.

[0058] Example 1:

[0059] Flexible DC transmission systems based on modular multilevel converters have demonstrated significant advantages in large-scale renewable energy integration and inter-regional interconnection. Their safe and stable operation is a prerequisite for realizing their technological value. However, existing relay protection research and practice mainly focus on short-circuit faults, with insufficient protection measures for open-circuit faults. Because the electrical characteristics change only slightly after an open-circuit fault, traditional protection methods based on overcurrent and differential principles are difficult to detect effectively. Furthermore, due to noise interference and transition resistance, existing methods lack rapid and accurate solutions for identifying open-circuit types and locating fault sections, severely restricting the reliability of flexible DC systems.

[0060] To address the aforementioned problems and shortcomings, this invention proposes a line break protection method for MMC-HVDC systems based on transient current characteristic analysis. This method achieves signal purification and feature enhancement through adaptive noise complete ensemble empirical mode decomposition (ERD), accurately captures the unique oscillation modes of line break faults using the slope characteristics of fitted curves, achieves fault pole selection by combining Hausdorff distance analysis, and completes section location through DC component detection. This method requires only local information to achieve rapid identification and accurate location of line break faults, and features clear principles, strong anti-interference capabilities, and good noise resistance, thus improving the operational reliability of flexible DC systems.

[0061] This invention provides a method for line-break protection of an MMC-HVDC system based on transient current characteristic analysis. The method includes the following steps: Figure 1 As shown:

[0062] S1. Acquire the transient current of the DC line; perform adaptive noise complete set empirical mode decomposition on the transient current to filter out noise and obtain the reconstructed signal;

[0063] In practice, since the location of the break point greatly affects the failure, the current oscillation frequency corresponding to different break points is also different, as shown in the following formula:

[0064]

[0065] In the formula, f is the frequency of current oscillation; v is the speed of light, which can theoretically be taken as 3 × 10⁻⁶. 8 m / s; l line This represents the distance between the measurement point and the break point. Under a break fault, a sinusoidal wave can be fitted. In this embodiment, a data point of one period is selected. When line l... line When the distance is 300km, the corresponding data window length is 4ms.

[0066] Next, the transient current from the disconnected circuit undergoes adaptive noise-complete ensemble empirical mode decomposition to obtain multiple intrinsic mode functions (IMFs) and residual terms r(t), thus yielding the reconstructed signal, such as... Figure 2 As shown; the reconstructed signal is obtained through the following steps:

[0067] S101. Perform adaptive noise complete set empirical mode decomposition on the transient current to obtain multiple intrinsic mode function components and residual terms;

[0068] S102. When the plurality of intrinsic mode function components satisfy the following formula, they are determined to be high-frequency components (IMFs). high Otherwise, it is determined to be a low-frequency component IMF. low ;

[0069] abs(∑IMF j (t))<IMF set ;

[0070] In the formula, abs(·) is the absolute value function; ∑IMF j (t) represents the summation of the j-th intrinsic mode function component over the entire data window length; IMF set Preset component threshold;

[0071] S103. The low-frequency component is superimposed with the residual term and then mixed with 20dB of white noise to obtain the reconstructed signal; the reconstructed signal is represented as:

[0072] X(t)=∑IMF low +r(t);

[0073] In the formula, X(t) is the reconstructed signal; IMF low It is a low-frequency component.

[0074] In this embodiment, by performing adaptive noise complete set empirical mode decomposition and reconstruction on the acquired DC line transient current, the noisy original signal is adaptively decomposed into multiple intrinsic mode function components. By removing high-frequency components that represent noise and reconstructing the signal, high-frequency noise introduced by on-site electromagnetic interference, measurement errors, etc. is effectively suppressed, and high-precision signal purification and feature enhancement are achieved.

[0075] S2. When it is determined that the reconstructed signal is lower than the preset power failure fault initiation threshold, the disconnection fault identification in step S3 is executed.

[0076] In practical applications, the current is normally at 1 p.u., but changes abruptly after a line breakage fault occurs. Therefore, the following criterion for line breakage fault identification is set: when the amplitude of the reconstructed signal is less than the preset power outage fault initiation threshold, it is determined that a line breakage fault has occurred.

[0077] |X(t)|<X low

[0078] In the formula, |X(t)| is the amplitude of the reconstructed signal; X low To preset the power outage fault trigger threshold, and to ensure that the maximum fluctuation of current and voltage under stable operation does not exceed 10%, X is selected. low =0.9.

[0079] When the fault initiation criterion is met, S1 = 1 is defined, and the data window length ΔT is taken forward and entered into other stages.

[0080] In this embodiment, line protection is activated by determining whether the reconstructed signal is lower than the preset power outage fault activation threshold. This criterion is based on the reconstructed signal after noise reduction, which avoids false activation caused by noise. At the same time, since a line failure will cause the effective value of the current to drop significantly, the principle of the amplitude criterion based on the reconstructed signal is simple and can quickly activate line protection after the fault occurs, thus improving the speed of line protection.

[0081] S3. Based on the decomposition results of step S1, calculate the slope of the current characteristic component fitting curve to identify the open circuit fault; when the current characteristic component fitting curve has multiple slopes that alternate between positive and negative, or has only one slope with a value of zero, then it is determined that the DC line has an open circuit fault, that is, |S n |>1 or|S n |=1 and L=0, where L is the slope of the calculated fitted curve. When either of these conditions is met, it is predicted that the line is broken, and S2=1 is defined; otherwise, it is determined that the DC line has a non-broken fault, and the process returns to step S1.

[0082] The fitting curve for the current characteristic component is obtained through the following steps:

[0083] S301. From the multiple intrinsic mode function components and residual terms obtained in step S1, the component with the highest energy is selected as the current characteristic component characterizing the current change trend; this current characteristic component is expressed as:

[0084] |Y(t)|=max(∑abs(IMF j (t)),r(t));

[0085] In the formula, Y(t) is the characteristic component of the current; max is the maximum value; and r(t) is the residual term.

[0086] S302. Extract all extreme points and the first and last points of the current characteristic component within a data window, and form a fitting curve for the current characteristic component by connecting the extreme points and the first and last points.

[0087] In this embodiment, open circuit faults are identified by analyzing the slope characteristics of the fitted curve of the current characteristic components. By utilizing the essential transient characteristics of open circuit faults, highly accurate identification is achieved. This method constructs a fitted curve by extracting extreme points, transforming the complex waveform recognition problem into an intuitive slope pattern recognition problem. When there are multiple alternating positive and negative slopes (corresponding to oscillations) or only one zero slope (corresponding to a sudden drop and hold), these features are significantly different from short circuit faults or other interference modes. Thus, the criterion has high discriminative power and can identify open circuit faults from other abnormal states with extreme accuracy.

[0088] S4. Based on the acquired transient current, calculate the Hausdorff distance between the positive and negative current sequences of the DC line to identify the fault pole; at the same time, calculate the DC component of the oscillating current based on the extreme points in the current characteristic components to identify the fault section.

[0089] In practical implementation, the Hausdorff distance between the positive and negative current sequences of the DC line is expressed as:

[0090]

[0091] In the formula, H(·) is the Hausdorff distance; A and B are the positive and negative current sequences of the DC line, respectively.

[0092] Under a bipolar open-circuit fault, the positive and negative currents are the same, and the calculated distance H is approximately zero. Considering the effects of measurement errors, calculation errors, and the unbalanced operation of the positive and negative poles, the distance H is slightly greater than zero. Therefore, H... SET The value of H should not be close to zero; under a single-pole open-circuit fault, the calculated H distance is greater than zero, and as the transition resistance increases, the calculated H distance decreases. To have a strong resistance to transition resistance, H... SETThe value of H should not be too high. Therefore, in this embodiment, H is selected. SET The value is 0.1.

[0093] When H < H SET If the condition is met, it is determined to be a bipolar open circuit fault, and S3 = 1 is defined; otherwise, it is determined to be a unipolar open circuit fault.

[0094] In practice, line breaks typically occur on overhead lines or cable lines, without considering internal line breaks within the converter station. After a line break fault occurs, the line current changes from the DC component I... CoDC Sine component I CoAC Together they form a whole. Under fault conditions within the zone, the DC component I... CoDC It drops sharply to zero, leaving only the sinusoidal component I. CoAC The current continues to decay; however, under an external open-circuit fault, the power flow is redistributed, affecting the DC component I of the oscillating current. CoDC The amplitude of the current varies depending on the amount of power transferred. Furthermore, the open circuit at the point of disconnection also affects the sinusoidal component I of the current in adjacent lines. CoAC The amplitude, considering that the current enters a new steady-state phase within hundreds of milliseconds after the line breakage in an external fault, can be considered as the DC component I of the oscillating current in the initial stage of the fault. CoDC The current value remains unchanged, and its amplitude equals the rated current before the fault. The DC component I of the oscillating current is calculated using the extreme points of the transient current. CoDC , is represented as:

[0095]

[0096] In the formula, I CoDC E represents the DC component of the oscillating current; ave(·) represents the average value; E i Let be the current at the i-th extreme point on the fitted current characteristic component, i = 1, 2, ..., z.

[0097] Therefore, the DC component I under the broken line in the area CoDC Approximately equal to zero, and define S4 = 1; DC component I under the disconnected line outside the zone. CoDC It is approximately equal to 1.

[0098] In this embodiment, Hausdorff distance and DC component are used together for fault pole selection and fault section identification. By calculating the Hausdorff distance between the positive and negative transient current sequences, the overall similarity of the current waveforms of the two poles can be effectively measured, thereby distinguishing between single-pole open circuit (large waveform difference) and double-pole open circuit (highly similar waveform). Secondly, the DC component of the oscillating current obtained by calculating the extreme points of the current characteristic components accurately reflects the location attribute of the fault point: the DC component approaching zero indicates that the measurement point is electrically disconnected from the main grid (intra-zone fault), while the DC component maintaining the rated value indicates that the fault point is in other locations (outtra-zone fault). The above two criteria complement each other and together achieve a comprehensive and accurate diagnosis of the fault type and location.

[0099] S5. Based on the judgment result of step S4, execute the corresponding collaborative MMC switching control.

[0100] In practice, based on the key decision information provided in step S4, such as "whether it is a single-pole or double-pole fault" and "whether it is an internal or external fault", the control system can take appropriate coping strategies (such as only disconnecting the faulty pole or disconnecting the entire line), thereby minimizing the power outage range, maintaining the power supply continuity of non-faulty areas, and guiding the system to quickly return to a safe and stable operating state, minimizing the losses caused by the fault.

[0101] Example 2:

[0102] To better understand the effectiveness of this embodiment, the simulation results of the embodiment of the present invention will be further described in detail below with reference to the accompanying drawings.

[0103] This embodiment uses a symmetrical bipolar system as the research object. The system parameters are shown in Table 1 to verify the line breakage protection performance proposed in this embodiment. The measuring device is as follows: Figure 3 As shown in the blue circle, the MMC submodule adopts an equivalent model, the MMC2 converter adopts constant active power control, and the MMC1 converter adopts constant DC voltage control.

[0104] Table 1 MMC-HVDC Converter Parameter Table

[0105]

[0106] Assume power is supplied from the MMC1 side of the converter to the MMC2 side, and define the current as flowing from the bus into the line in the positive direction. The sampling frequency for the line current is chosen to be 10kHz. Assume a line fault occurs at t=0ms, taking a break-through fault at the midpoint of the line as an example.

[0107] This embodiment incorporates a line break fault in the circuit to fully test the proposed protection scheme.

[0108] Startup Criterion Simulation Verification

[0109] A line open fault occurs, causing a sudden change in fault current. The reconstructed signal is obtained through adaptive noise complete set empirical mode decomposition, satisfying the initiation criterion |X(t)|<X. low , Figure 4 Simulation results of the fault start-up criteria for disconnection are presented.

[0110] Considering the long transmission line, there is a certain time delay in the propagation of fault information from the fault point to the measurement point. Tables 2 and 3 show the time when the proposed initiation criterion is met and the actual time when the fault occurs under different fault distances and noise levels. As can be seen from the tables, the initiation criterion can accurately pinpoint the time of fault occurrence, with a maximum error of no more than 2 data points, or 0.2 ms, which is within an acceptable range.

[0111] Table 2. Startup criterion satisfaction time under different fault distances - open circuit fault

[0112]

[0113] Note: TMMCi represents the time when the proposed start-up criterion is met on the MMCi side, and tMMCi represents the actual time when the fault occurs on the MMCi side.

[0114] Table 3. Startup criterion satisfaction time under different noise levels - open circuit fault

[0115]

[0116] Fault level selection simulation verification

[0117] Figure 5 The calculated H-distance is given under different fault types. Under a single-pole open-circuit fault, the H-distance calculated from the positive and negative pole currents is much greater than the setting threshold of 0.1. However, under a double-pole open-circuit fault, the calculated H-distance is close to zero and less than the setting threshold of 0.1.

[0118] Table 4 shows the calculated H distances for single-pole and double-pole open-circuit faults under different fault distances, which can correctly distinguish different types of faults.

[0119] Table 4 Results of Disconnection Fault Type Judgment - Disconnection Fault

[0120]

[0121] Note: H represents the maximum H-distance calculated on both sides of MMC1 and MMC2.

[0122] Disconnection simulation verification

[0123] Figure 6The current waveform of the open circuit fault is given. Under the open circuit fault, the sine wave is restored by using the IMF component selected by the current characteristic component. The fitted curve has three extreme points and four alternating positive and negative slopes, which can be judged as an open circuit fault.

[0124] Tables 5 and 6 present the calculated results of the slope characteristics of the fitted curves for open-circuit fault currents within the zone at different fault distances. Different locations of the open-circuit fault result in different fault current oscillation periods; the closer the open-circuit point, the shorter the oscillation period, and the more slopes of the corresponding curve. When the line has an open circuit at 10% of its points, the fitted curve yields 10 slopes, while at 100% of its points, the curve only has three slopes, both correctly identifying as open-circuit faults. When an open-circuit fault occurs at the beginning of the line, the line is not within the open-circuit loop, and the fault current drops sharply to zero and remains constant. The fitted curve has only one slope and is zero, consistent with theoretical analysis, and can also be considered an open-circuit fault.

[0125] Table 5. Discrimination results for different fault distances - Bipolar open circuit fault

[0126]

[0127] Table 6. Discrimination results for different fault distances - Positive electrode open circuit fault

[0128]

[0129]

[0130] As shown in Tables 5 and 6, under the condition of a line break fault within the zone, the DC component I of the fault current oscillation is calculated based on the extreme point. CoDC Both values ​​approach zero, satisfying the fault section criterion, and the fault is determined to be within the section. However, under an external line break fault, the DC component I of the current oscillation in the non-faulty line... CoDC This refers to the rated current of the non-faulty line before the fault. Figure 7 Middle Line L 34 Taking a broken line as an example, Table 7 shows the line L 13 and L 24 DC component I of the oscillating current CoDC The large calculated value is consistent with the characteristics of an out-of-area fault, and it can be determined to be an out-of-area line break fault.

[0131] Table 7. Results of fault identification outside the zone - open circuit fault

[0132]

[0133] Noise impact

[0134] Gaussian white noise was further added to the open circuit fault data, and the impact of signal-to-noise ratios of 20, 30, 40, and 50 dB on the proposed protection scheme was considered. The results are shown in Table 8. Because the current was decomposed in advance using an adaptive noise complete set empirical mode decomposition analysis reconstruction algorithm, the noise impact was reduced, and the proposed protection scheme exhibits strong noise immunity.

[0135] Table 8. Impact of Different Noise Levels - Line Breakage Faults in the Zone

[0136]

[0137] Example 3:

[0138] This embodiment discloses a computer device.

[0139] The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps described in the embodiments of the MMC-HVDC system disconnection protection method based on transient current characteristic analysis. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above system embodiments.

[0140] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the airport baggage system fault diagnosis agent.

[0141] The computer device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device may include, but is not limited to, a processor and memory.

[0142] In summary, this embodiment addresses the technical problem of weak fault characteristics and difficulty in rapid and accurate identification and location of open circuit faults in flexible DC transmission systems. It proposes an open circuit protection method for MMC-HVDC systems based on transient current characteristic analysis. This method first uses adaptive noise complete set empirical mode decomposition to denoise and reconstruct the transient current, extracting current characteristic components. Then, it uses the slope characteristics of the fitted curve to identify open circuit faults, distinguishes between unipolar and bipolar open circuits using Hausdorff distance, and determines whether the fault is within or outside the fault zone by the DC component of the oscillating current. Ultimately, it achieves fast, reliable open circuit protection that requires only local information.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described with reference to preferred embodiments, those skilled in the art should understand that various changes in form and detail can be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A method for line disconnection protection in an MMC-HVDC system based on transient current characteristic analysis, characterized in that, Includes the following steps: S1. Acquire the transient current of the DC line; perform adaptive noise complete set empirical mode decomposition on the transient current to filter out noise and obtain the reconstructed signal; S2. When it is determined that the reconstructed signal is lower than the preset power failure fault initiation threshold, the disconnection fault identification in step S3 is executed. S3. Based on the decomposition results of step S1, calculate the slope of the current characteristic component fitting curve to identify the open circuit fault: when the current characteristic component fitting curve has multiple slopes and alternates between positive and negative, or has only one slope and its value is zero, it is determined that the DC line has an open circuit fault, and step S4 is executed; otherwise, it is determined that the DC line has a non-open circuit fault, and step S1 is returned. S4. Based on the acquired transient current, calculate the Hausdorff distance between the positive and negative current sequences of the DC line to identify the fault pole; at the same time, calculate the DC component of the oscillating current based on the extreme points in the current characteristic components to identify the fault section. S5. Based on the judgment result of step S4, execute the corresponding collaborative MMC switching control.

2. The method for line disconnection protection of an MMC-HVDC system based on transient current characteristic analysis according to claim 1, characterized in that, In step S1, the reconstructed signal is obtained through the following steps: S101. Perform adaptive noise complete set empirical mode decomposition on the transient current to obtain multiple intrinsic mode function components and residual terms; S102. When the plurality of intrinsic mode function components satisfy the following formula, they are determined to be high-frequency components; otherwise, they are determined to be low-frequency components. abs(∑IMF j (t))<IMF set ; In the formula, abs(·) is the absolute value function; ∑IMF j (t) represents the summation of the j-th intrinsic mode function component over the entire data window length; IMF set Preset component threshold; S103. After superimposing the low-frequency component with the residual term and filtering out noise, a reconstructed signal is obtained.

3. The method for line disconnection protection of MMC-HVDC systems based on transient current characteristic analysis according to claim 2, characterized in that, In step S2, the criterion for performing the open circuit fault identification is expressed as follows: |X(t)|<X low In the formula, X(t) is the reconstructed signal; X low The preset power outage fault start threshold is used.

4. The method for line disconnection protection of an MMC-HVDC system based on transient current characteristic analysis according to claim 3, characterized in that, In step S3, the current characteristic component fitting curve is obtained through the following steps: S301. From the multiple intrinsic mode function components and residual terms obtained from step S1, select the component with the highest energy as the current characteristic component characterizing the current change trend. S302. Extract all extreme points and the first and last points of the current characteristic component within a data window, and form a fitting curve for the current characteristic component by connecting the extreme points and the first and last points.

5. The method for line disconnection protection of an MMC-HVDC system based on transient current characteristic analysis according to claim 4, characterized in that, In step S301, the current characteristic component is represented as: |Y(t)|=max(∑abs(IMF j (t)),r(t)); In the formula, Y(t) is the characteristic component of the current; max is the maximum value; and r(t) is the residual term.

6. The method for line disconnection protection of an MMC-HVDC system based on transient current characteristic analysis according to claim 5, characterized in that, In step S4, the Hausdorff distance between the positive and negative current sequences of the DC line is expressed as: In the formula, H(·) is the Hausdorff distance; A and B are the positive and negative current sequences of the DC line, respectively.

7. The method for line disconnection protection of an MMC-HVDC system based on transient current characteristic analysis according to claim 6, characterized in that, In step S4, the fault polarity selection judgment specifically includes: If the Hausdorff distance satisfies the following formula, it is determined to be a bipolar open circuit fault; otherwise, it is determined to be a unipolar open circuit fault. H<H SET ; In the formula, H SET This is a preset distance threshold.

8. The method for line disconnection protection of an MMC-HVDC system based on transient current characteristic analysis according to claim 7, characterized in that, The DC component of the oscillating current is expressed as: In the formula, I CoDC E represents the DC component of the oscillating current; ave(·) represents the average value; E i Let be the current at the i-th extreme point on the fitted current characteristic component, i = 1, 2, ..., z.

9. The method for line disconnection protection of an MMC-HVDC system based on transient current characteristic analysis according to claim 8, characterized in that, In step S4, the fault section identification specifically involves: If the DC component of the oscillating current is approximately zero, it is determined to be an open circuit fault within the zone. If the DC component of the oscillating current is approximately 1, it is determined to be an external open circuit fault.

10. A computer device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes a method for disconnection protection of an MMC-HVDC system based on transient current characteristic analysis as described in any one of claims 1-9.

Images