CLCC parameter optimization method based on AC side fault data set

Abstract

The invention discloses a CLCC parameter optimization method based on an alternating current side fault data set. The method comprises the following steps: implementing alternating current side fault parallel simulation on a target direct current project in simulation software; analyzing the multi-dimensional data, and identifying a commutation failure and a non-commutation failure; identifying the boundary between the commutation failure and the non-commutation failure by using an SVM boundary algorithm; for a real target direct current project, a fault oscillograph is used for collecting waveforms; and according to the confidence interval of each fault component, judging whether to continue to adopt the LCC mode for the receiving end or to switch from the LCC mode to the CLCC mode. The invention provides a more accurate and intelligent commutation failure prediction and processing method, the stability of a power grid is remarkably improved, the service life of equipment is remarkably prolonged, and the response capability of a system is remarkably improved.

Description

Technical Field

[0001] The present disclosure relates to the field of direct current (DC) transmission, and in particular to a method for optimizing CLCC parameters based on an AC side fault data set. Background Art

[0002] The Controllable Commutation Converter (CLCC) is an evolution of the traditional six-pulse converter (LCC). It uses parallel hybrid valves to replace HVDC commutation arms and is compatible with the LCC mode. In the CLCC mode, the timing of commutation handover between the main and auxiliary branches is crucial. Current switching methods rely primarily on empirical grid settings, but this approach has its drawbacks. Premature switching, before a commutation failure has occurred, can burden the auxiliary branches and threaten the CLCC's service life. Switching too late can fail to effectively prevent commutation failures.

[0003] In addition, in the prior art, the mode switching from LCC to CLCC is based on the number of times the lightning arrester operates and the triggering angle v12 valve shuts off the voltage within 2 seconds. The problem is that this method has the defect of inaccurate switching.

[0004] How to conveniently identify potential commutation failures and switch CLCC at the appropriate time is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] To solve the above technical problems, the present disclosure discloses a CLCC parameter optimization method based on an AC side fault dataset, comprising the following steps:

[0006] S100, performing AC side fault parallel simulation on the target DC project in simulation software to obtain multi-dimensional data;

[0007] S200, analyzing the multi-dimensional data and identifying commutation failure and non-commutation failure;

[0008] S300, using an SVM boundary algorithm to identify a boundary between a commutation failure and a non-commutation failure, and further obtaining confidence intervals for each fault component corresponding to the high-risk boundary; wherein the boundary is used as a high-risk boundary for switching the receiving end from the LCC mode to the CLCC mode; the high-risk boundary is represented by a portion of the multidimensional data; wherein the number of fault components is equal to the dimension of the multidimensional data;

[0009] S400, using a fault recorder to collect waveforms of the currents of three phases a, b, and c on the receiving-end converter valve side of the actual target DC project;

[0010] S500. Identify data with the same number of dimensions as the multidimensional data from the collected waveform, and determine whether to continue using the LCC mode for the receiving end or switch from the LCC mode to the CLCC mode based on the data with the same number of dimensions and the confidence interval of each fault component.

[0011] Preferably, step S100 includes the following steps:

[0012] S101. Establish a target DC engineering model in PSCAD, wherein the target DC engineering model includes a sending end and a receiving end, wherein the receiving end includes the following three branches connected in parallel: a branch corresponding to an overvoltage component, a branch corresponding to a harmonic component, and a branch corresponding to a DC component;

[0013] S102. For a ground fault on the receiving AC side, reasonably set amplitudes and scales for each of the five variables of the three branches, namely, the DC component, harmonic component, overvoltage component, phase difference between the overvoltage and DC component from the time of occurrence to the zero-crossing point, and phase difference between the harmonic and fundamental wave, to ensure the effectiveness and accuracy of the analysis; wherein the five variables serve as independent variables for the simulation;

[0014] S103, after the amplitude setting and scale setting are completed, when the sending end adopts the LCC mode and the receiving end adopts the LCC mode, PSCAD is used to perform parallel simulation of the AC side fault;

[0015] Preferably, in step S102:

[0016] The DC component is divided into 7 scales from 0 to 24% of the DC component amplitude.

[0017] Preferably, in step S102:

[0018] The harmonic components are divided into 7 scales from 0 to 25% of the harmonic component amplitude.

[0019] Preferably, in step S102:

[0020] The overvoltage component is divided into 7 scales from 100% to 130% of the overvoltage component amplitude.

[0021] Preferably, in step S102:

[0022] The phase difference from zero to the moment when the overvoltage and DC component occur to the zero-crossing point is divided into 24 scales every 15 degrees within the range of 0 to 360 degrees.

[0023] Preferably, in step S102:

[0024] The phase difference between zero and the fundamental wave is divided into 24 scales every 15 degrees within the range of 0 to 360 degrees.

[0025] Preferably, in step S100,

[0026] During the simulation, the currents I of the three corresponding channels a, b, and c on the receiving end converter valve side are recorded. a , I b , I c ; Wherein, the current I on the receiving end converter valve side a , I b , I c as the dependent variable of the simulation.

[0027] Preferably,

[0028] Each channel includes 5 working conditions: overvoltage component V ov , harmonic component V h , DC component V DC , the phase difference between the overvoltage and DC component occurrence time and the zero crossing point , and the phase difference between the harmonics and the fundamental .

[0029] Preferably, in step S200,

[0030] If I a , I b , I c If the phase difference is 120 degrees, it is considered as non-commutation failure;

[0031] If I a , I b , I c If the phases do not differ by 120 degrees, it is considered a commutation failure.

[0032] The present disclosure has the following beneficial effects:

[0033] This paper uses innovative technical means, such as multi-dimensional data analysis based on AC-side fault data sets, combined with SVM algorithms and Radviz visualization technology, to provide a more accurate and intelligent commutation failure prediction and processing method, significantly improving the stability of the power grid, the service life of equipment, and the responsiveness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 1 is a flow chart of a CLCC parameter optimization method based on an AC side fault data set in one embodiment of the present disclosure;

[0035] Figure 2 This is a schematic diagram of commutation failure in another embodiment of the present disclosure;

[0036] Figure 3Schematic diagram of a multi-infeed DC transmission system of the East China Power Grid in another embodiment of the present disclosure;

[0037] Figure 4 FIG1 is a Radviz diagram of commutation failure and non-commutation failure in another embodiment of the present disclosure;

[0038] Figure 5 FIG1 is a Radviz diagram of a commutation failure with SVM boundaries marked in another embodiment of the present disclosure;

[0039] Figure 6 This is a schematic diagram of a mode when a fault occurs in another embodiment of the present disclosure;

[0040] Figure 7 、 Figure 8 This is a schematic diagram comparing another embodiment of the present disclosure, after optimization and before optimization using the optimization method disclosed in the present disclosure. DETAILED DESCRIPTION

[0041] In order to make those skilled in the art understand the technical solutions disclosed in this disclosure, the following will be combined with the embodiments and related appendixes. Figures 1 to 8 , the technical solutions of various embodiments are described, and the embodiments described are part of the embodiments of the present disclosure, rather than all of the embodiments. Mention of "embodiment" in this article means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present disclosure. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It will be understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0042] The use of cross hatching and / or shading in the accompanying drawings is generally used to make the boundaries between adjacent components clear. As such, unless otherwise indicated, the presence or absence of cross hatching or shading does not convey or indicate any preference or requirement for the specific materials, material properties, dimensions, proportions, commonalities between the components shown, and / or any other characteristics, attributes, properties, etc. of the components. In addition, in the accompanying drawings, the sizes and relative sizes of the components may be exaggerated for clarity and / or descriptive purposes. When the exemplary embodiments can be implemented differently, the specific process sequence can be performed in a different order than described. For example, two successively described processes can be performed substantially simultaneously or in an order opposite to the order described. In addition, the same figure numbers represent the same components.

[0043] When a component is referred to as being “on,” “over,” “connected to,” or “coupled to” another component, the component may be directly on, directly connected to, or directly coupled to the other component, or intervening components may be present. However, when a component is referred to as being “directly on,” “directly connected to,” or “directly coupled to” another component, there are no intervening components present. For this purpose, the term “connected” may refer to a physical connection, an electrical connection, etc., with or without intervening components.

[0044] For descriptive purposes, the present disclosure may use spatially relative terms such as "below," "beneath," "under," "down," "above," "upper," "above," "higher," and "side (e.g., as in "sidewall")," to describe the relationship of one component to another (other) component as shown in the accompanying drawings. The spatially relative terms are intended to encompass different orientations of the device in use, operation, and / or manufacture in addition to the orientation depicted in the accompanying drawings. For example, if the device in the drawings is turned over, a component described as "below" or "beneath" another component or feature would then be positioned "above" the other component or feature. Thus, the exemplary term "below" can encompass both the "above" and "below" orientations. Furthermore, the device may be otherwise oriented (e.g., rotated 90 degrees or at other orientations), and as such, the spatially relative descriptors used herein should be interpreted accordingly.

[0045] The terms used herein are for the purpose of describing specific embodiments and are not intended to be restrictive. As used herein, unless the context clearly indicates otherwise, the singular forms "one (kind, person)" and "said (the)" are also intended to include plural forms. In addition, when the terms "comprise" and / or "include" and their variations are used in this specification, the features, integral bodies, steps, operations, parts, assemblies and / or their groups stated are explained, but the presence or addition of one or more other features, integral bodies, steps, operations, parts, assemblies and / or their groups is not excluded. It should also be noted that, as used herein, the terms "substantially", "approximately" and other similar terms are used as approximate terms and not as degree terms, so that they are used to explain the inherent deviations of the measured values, calculated values ​​and / or the values ​​provided that will be recognized by those of ordinary skill in the art.

[0046] In the description of this specification, the description with reference to the terms "one embodiment / method", "some embodiments / methods", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment / method or example are included in at least one embodiment / method or example of the present disclosure. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment / method or example. Moreover, the specific features, structures, materials or characteristics described may be combined in an appropriate manner in any one or more embodiments / methods or examples. In addition, those skilled in the art may combine and combine different embodiments / methods or examples described in this specification and the features of different embodiments / methods or examples, unless they are contradictory.

[0047] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the present disclosure, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0048] In one embodiment, combined Figure 1 As shown, the present disclosure discloses a CLCC parameter optimization method based on an AC side fault data set, comprising the following steps:

[0049] S100, performing AC side fault parallel simulation on the target DC project in simulation software to obtain multi-dimensional data;

[0050] S200, analyzing the multi-dimensional data and identifying commutation failure and non-commutation failure;

[0051] S300, using an SVM boundary algorithm to identify a boundary between a commutation failure and a non-commutation failure, and further obtaining confidence intervals for each fault component corresponding to the high-risk boundary; wherein the boundary is used as a high-risk boundary for switching the receiving end from the LCC mode to the CLCC mode; the high-risk boundary is represented by a portion of the multidimensional data; wherein the number of fault components is equal to the dimension of the multidimensional data;

[0052] S400, using a fault recorder to collect waveforms of the currents of three phases a, b, and c on the receiving-end converter valve side of the actual target DC project;

[0053] S500. Identify data with the same number of dimensions as the multidimensional data from the collected waveform, and determine whether to continue using the LCC mode for the receiving end or switch from the LCC mode to the CLCC mode based on the data with the same number of dimensions and the confidence interval of each fault component.

[0054] Typically, multi-dimensional data includes: DC component, harmonic component, overvoltage, phase difference between the overvoltage and DC component occurrence time and the zero crossing point, and phase difference between harmonic and fundamental wave, totaling 5 dimensions of data.

[0055] It can be understood that the inventive concept of the present disclosure is that the steps S100 to S300 are used to obtain multi-dimensional data of the target DC engineering model in a parallel simulation manner in the simulation software, and to identify commutation failure and non-commutation failure based on the features therein. Figure 4 , non-commutation failure is shown as the point in the red area, and commutation failure is shown as the point in the blue area. After being able to identify commutation failure and non-commutation failure, the SVM boundary algorithm can be further used to identify the boundary between commutation failure and non-commutation failure, see Figure 5 The dashed closed line shown is drawn, and the confidence interval of the fault component corresponding to the boundary is obtained. Then, through steps S400 to S500, the relevant data extracted from the waveform of the actual target DC project and the confidence interval of each fault component are determined to determine whether to continue using the LCC mode for the receiving end or switch from the LCC mode to the CLCC mode.

[0056] In another embodiment, step S100 includes the following steps:

[0057] S101. Establish a target DC engineering model in PSCAD, wherein the target DC engineering model includes a sending end and a receiving end, wherein the receiving end includes the following three branches connected in parallel: a branch corresponding to an overvoltage component, a branch corresponding to a harmonic component, and a branch corresponding to a DC component;

[0058] S102. For a ground fault on the receiving AC side, reasonably set amplitudes and scales for each of the five variables of the three branches, namely, the DC component, harmonic component, overvoltage component, phase difference between the overvoltage and DC component from the time of occurrence to the zero-crossing point, and phase difference between the harmonic and fundamental wave, to ensure the effectiveness and accuracy of the analysis; wherein the five variables serve as independent variables for the simulation;

[0059] S103, after the amplitude setting and scale setting are completed, when the sending end adopts the LCC mode and the receiving end adopts the LCC mode, PSCAD is used to perform parallel simulation of the AC side fault;

[0060] In another embodiment, in step S102:

[0061] The DC component is divided into 7 scales from 0 to 24% of the DC component amplitude.

[0062] In another embodiment, in step S102:

[0063] The harmonic components are divided into 7 scales from 0 to 25% of the harmonic component amplitude.

[0064] In another embodiment, in step S102:

[0065] The overvoltage component is divided into 7 scales from 100% to 130% of the overvoltage component amplitude.

[0066] In another embodiment, in step S102:

[0067] The phase difference from zero to the moment when the overvoltage and DC component occur to the zero-crossing point is divided into 24 scales every 15 degrees within the range of 0 to 360 degrees.

[0068] In another embodiment, in step S102:

[0069] The phase difference between zero and the fundamental wave is divided into 24 scales every 15 degrees within the range of 0 to 360 degrees.

[0070] In another embodiment, in step S100,

[0071] During the simulation, the currents I of the three corresponding channels a, b, and c on the receiving end converter valve side are recorded. a , I b , I c ; Wherein, the current I on the receiving end converter valve side a , I b , I c as the dependent variable of the simulation.

[0072] In summary, it can be understood that in the parallel simulation, 7*7*7*24*24, or 197568 times of simulation data will be obtained, and the values ​​of the following three dependent variables can be obtained in each simulation: the current I on the receiving end converter valve side a , I b , I c , and knowing the values ​​of the corresponding independent variables: DC component, harmonic component, overvoltage, the phase difference between the overvoltage and DC component occurrence time and the zero crossing point, and the phase difference between the harmonic and the fundamental wave.

[0073] In another embodiment,

[0074] Each channel includes 5 working conditions: overvoltage component V ov , harmonic component V h , DC component V DC , the phase difference between the overvoltage and DC component occurrence time and the zero crossing point , and the phase difference between the harmonics and the fundamental .

[0075] In another embodiment, in step S200,

[0076] If I a , I b , I c If the phase difference is 120 degrees, it is considered as non-commutation failure;

[0077] If I a , I b , I c If the phases do not differ by 120 degrees, it is considered a commutation failure.

[0078] Further, exemplary,

[0079] Harmonics include the third harmonic.

[0080] This is because, through actual simulation and verification, this disclosure found that the third harmonic can account for up to 25% of AC side faults, which has a significant impact on commutation failure. Therefore, this disclosure focuses on the third harmonic, and other harmonics are not the main analysis objects due to their low proportion and sensitivity.

[0081] In another embodiment,

[0082] The multidimensional data in the multidimensional fault dataset is mapped into a two-dimensional space to reflect the relationship between key variables.

[0083] In another embodiment,

[0084] Presented as a Radviz diagram to identify potential commutation failure risk areas.

[0085] In another embodiment, the method further comprises the steps of:

[0086] S400 , calculating the confidence interval of the key variable by using the SVM algorithm to provide a criterion for switching the CLCC mode.

[0087] It can be understood that since the criterion is based on the confidence interval, the criterion has the property of being precise.

[0088] The criterion determines whether the CLCC mode needs to be enabled. Exemplarily, when the CLCC mode (CLCC_STATE) does not need to be enabled, the LCC mode (LCC_STATE) is enabled.

[0089] Furthermore, the criterion needs to satisfy one or more of the following constraints: it is beneficial to avoid unnecessary switching, it is beneficial to reduce the burden on the auxiliary branch, and it is beneficial to extend the service life of the CLCC.

[0090] It should be noted that the present disclosure determines the critical conditions of commutation failure based on the theoretical model of commutation failure. Figure 2 .

[0091] Commutation voltage u zero-crossing offset When the trigger angle is constant, the condition for non-commutation failure is to achieve the minimum area S required for commutation. cr , which is expressed as follows:

[0092] .

[0093] According to the existing technology, in order to achieve a sufficient commutation area S cr , the commutation overlap angle can be increased, but this will cause the arc extinction angle γ to decrease, thereby reducing the system's time and ability to self-recover before and after the current zero crossing, and increasing the risk of commutation failure.

[0094] The present disclosure can analyze the impact of five key variables, including DC component, harmonic component, and overvoltage component, on commutation failure through simulation tools and / or mathematical modeling. Specifically:

[0095] First, the DC component of the AC side fault is extracted, and the voltage waveforms at different amplitudes are simulated to quantify its impact on the commutation area.

[0096] Secondly, extract the harmonic components, especially the low-order harmonics (such as the 2nd and 3rd harmonics), and analyze their impact on the voltage waveform distortion;

[0097] Thirdly, the overvoltage component is extracted to study its influence on the current zero-crossing offset;

[0098] Finally, considering that the DC component, harmonic component, and overvoltage component also involve the corresponding two phases, that is; therefore, by setting the amplitude and corresponding phase range of the DC component, harmonic component, and overvoltage component, combined with simulation results and mathematical modeling, the contribution of each variable to the commutation failure can be quantified, revealing its role mechanism in the commutation failure.

[0099] It can be understood that in order to more accurately reflect power grid faults, the study expands the key variables of the three types of faults into five specific key variables, and analyzes their impact on commutation failure through these variables. In order to accurately control and analyze these key factors, each variable can be reasonably divided and scaled to ensure the accuracy and effectiveness of the analysis.

[0100] Furthermore, the present disclosure can implement Radviz visualization analysis in the following ways:

[0101] Using Radviz visualization analysis technology, the five fault variables are mapped into a two-dimensional graph to visually demonstrate their impact on the system. In the Radviz graph, each variable is connected to anchor points fixed on the circumference via a spring model. The position of the data points reflects their relative distance from each anchor point, revealing the impact of each dimension on the system state and obtaining characteristic classification points for commutation failure and non-commutation failure under the multidimensional fault array. The degree of influence of each dimension is reflected by the change in the distance between the data point and the anchor point, which helps analyze the relationship between the fault variables. Figure 5 , the closer to the overvoltage component V ov , harmonic component V h , DC component V DC , the phase difference between the overvoltage and DC component occurrence time and the zero crossing point , and the phase difference between the harmonics and the fundamental The corresponding anchor points indicate that the stronger the spring tension is, the more significant the impact on the system is, and the system is more likely to fail in commutation.

[0102] Furthermore, the present disclosure can incorporate an SVM boundary algorithm through the following exemplary approach. For example, by incorporating a support vector machine (SVM) boundary algorithm, 197,568 multidimensional data sets were analyzed to identify high-risk fault zones in power grid operation. By identifying the boundary between commutation failure and non-commutation failure, CLCC mode was activated within the confidence interval.

[0103] In a more specific example,

[0104] The PSCAD / EMTDC simulation tool was used to establish a model of a multi-infeed DC transmission system in a certain power grid. Figure 3 As shown, the short-circuit fault point is set on the 220kV outgoing line side of Nanqiao Station.

[0105] The specific technical routes are as follows:

[0106] The three components of the AC side fault are divided into five specific key variables, namely the DC component amplitude V DC , harmonic component amplitude V h , overvoltage amplitude V ov , the phase difference between the overvoltage and DC component occurrence time and the zero crossing point , and the phase difference between the harmonics and the fundamental The third harmonic accounts for up to 25% of the AC side faults and has a significant impact on commutation failure. Therefore, this paper focuses on the third harmonic. Other harmonics are not the main analysis objects due to their low proportion and sensitivity.

[0107] Based on multiple experiments and simulations, if the harm of AC-side faults can be avoided as much as possible by switching the receiving end from LCC mode to CLCC mode, the amplitude range of the DC component is set to 0-24%, divided into 7 scales; the amplitude range of the harmonic component is set to 0-25%, divided into 6 scales; the amplitude range of the overvoltage component is set to 1 to 1.3 pu (this is based on the fact that under normal circumstances, the maximum overvoltage amplitude allowed by the system is 1.3 times the normal value), divided into 7 scales; the range of the two phase differences is 0 to 360°, and each is divided into 24 scales.

[0108] Based on these classifications, when setting AC-side faults in PSCAD, the Multiple Run module and multi-threaded parallel simulation capabilities can efficiently simulate the grid's response under different fault conditions. This simulation system can run multiple fault scenarios in parallel, record key parameters of the DC control and protection system, and obtain and record the characteristic classification points of commutation failure and non-commutation failure under the multi-dimensional fault array.

[0109] Then, the data are normalized and visualized using Radial Visualization technology. Figure 4 As shown in Figure 2, the characteristic classification points of commutation failure and non-commutation failure under the multi-dimensional fault array are obtained.

[0110] Using the SVM algorithm, a visualization of the commutation failure is drawn and the boundaries are marked, such as Figure 5 As shown, the confidence intervals for the five key variables of commutation failure are calculated. DC does not need to be changed because the patent law recognizes the numerical range expressed by the ± sign.

[0111] For example, the confidence intervals of various variables are shown in the following table:

[0112]

[0113] It should be noted that when the following logical conditions are met, the receiving end needs to switch from LCC mode to CLCC mode:

[0114] i) When the DC component V DC falls within its corresponding confidence interval, and the phase difference between the overvoltage and the DC component from the time of occurrence to the zero crossing point When it falls within the corresponding confidence interval, the receiving end needs to switch from LCC mode to CLCC mode;

[0115] ii) When the harmonic component V h falls within its corresponding confidence interval, and the phase difference between the harmonic and the fundamental wave When it falls within the corresponding confidence interval, the receiving end needs to switch from LCC mode to CLCC mode;

[0116] iii) When the overvoltage component Vov falls within its corresponding confidence interval, and the phase difference between the overvoltage and DC component from the time of occurrence to the zero crossing point is When it falls within the corresponding confidence interval, the receiving end needs to switch from LCC mode to CLCC mode. When a commutation failure occurs, real-time monitoring and recording can be carried out by introducing a fault recording device. The fault recording device decomposes the AC side voltage waveform into three key variables: DC component, harmonic component, and overvoltage component. The confidence intervals calculated by the SVM algorithm for these variables are compared with the high-risk intervals for commutation failure. If the value of any variable is close to these high-risk intervals, the system will automatically initiate the corresponding commutation failure measures, see Figure 6 .

[0117] When a power grid fault occurs, the fault recording and playback device automatically collects grid waveform data and calculates key variables such as DC components, harmonic components, overvoltage components, and phase difference. The confidence intervals calculated using the SVM algorithm serve as the criterion. If these variables fall into the high-risk range, the risk of commutation failure is determined. At this point, the system activates CLCC mode.

[0118] Compared with the existing technology, the disclosed proposal has the following technical advantages:

[0119] (1) Accurate phase switching timing judgment to extend the service life of CLCC equipment

[0120] This disclosure uses confidence intervals calculated using the SVM algorithm to accurately determine whether to enable CLCC mode, avoiding premature or delayed switching, reducing the burden on auxiliary branches, and extending the service life of CLCC equipment. This addresses the inaccuracies of traditional empirical methods in determining switching timing, ensuring greater grid stability.

[0121] This paper uses a multidimensional data analysis method based on the AC side fault data set and calculates the confidence intervals of key variables (such as DC components, harmonic components, overvoltage components, etc.) through the SVM algorithm, providing a more accurate criterion for determining whether to switch to CLCC mode, thereby avoiding unnecessary switching and extending equipment life.

[0122] (2) Optimize CLCC mode switching to reduce unnecessary auxiliary branch burden

[0123] The present disclosure avoids premature switching tasks when no commutation failure occurs through scientific optimization of commutation switching timing, reduces the burden on auxiliary branches, thereby improving system stability and extending equipment service life.

[0124] Using Radviz visualization technology, multidimensional fault data is mapped into a two-dimensional space, intuitively revealing the relationships between different fault components. This supports accurate judgment of whether the system has entered a high-risk area for commutation failure and optimizes the timing of switching to CLCC mode.

[0125] (3) Improve the prediction accuracy of commutation failure

[0126] The present invention can accurately predict the risk of commutation failure and take real-time protective measures based on the calculated confidence interval. Compared with traditional methods, this data-driven prediction method provides higher accuracy and real-time performance.

[0127] Combining the SVM algorithm with the AC-side fault dataset, this paper calculates the confidence intervals of key variables to accurately determine whether the power grid has entered a high-risk range for commutation failure, thereby identifying fault risks in advance and responding.

[0128] See also Figure 7 and Figure 8 In another embodiment, the present disclosure implements a simulation of a three-phase ground fault of 0.1s to compare the optimization effects of the CLCC mode. Specifically,

[0129] During a 0.1-second three-phase ground fault simulation, we optimized the CLCC mode switching for a single-phase example before and after the optimization, and compared the waveforms before and after the optimization. This comparison shows that after optimizing the CLCC parameters, the system can accurately determine potential risks on the AC side when an AC fault occurs and effectively identify the possibility of commutation failure. This optimization significantly improves the CLCC's response accuracy during faults, ensuring grid stability and the timely activation of fault protection mechanisms.

[0130] Before optimization, CLCC mode switching relied heavily on specific grid control parameters, such as the trigger angle threshold and the number of arrester trips. Because these parameters couldn't accurately reflect commutation failures, CLCC mode switching was inaccurate, making it impossible to accurately identify and address commutation failure risks, thus reducing the grid's emergency response capabilities.

[0131] The optimization method disclosed in this disclosure can more accurately judge and respond to power grid faults, avoid commutation failures and other potential problems, and improve the reliability and fault recovery capabilities of the system.

[0132] After CLCC optimization: valve current commutation is successful

[0133] exist Figure 7In the example, the optimized CLCC mode was able to mitigate the risk of commutation failure. Despite slight current waveform instability, the commutation process was still successfully completed. This demonstrates that the optimized CLCC mode has greater robustness and fault tolerance in the face of AC faults, ensuring stable grid operation.

[0134] Before CLCC optimization: commutation failure

[0135] In contrast, Figure 8 Commutation failure occurred during the optimization process, and the valve current waveform could not commutate normally. This was because the pre-optimized CLCC mode could not accurately judge the state of the power grid, resulting in the failure to effectively identify and control the risk of commutation failure, which led to the expansion of the power grid fault and seriously affected the safe and stable operation of the power grid.

[0136] Further,

[0137] This disclosure can be combined with the following technical means to further optimize system performance:

[0138] Reactive Power Compensation and Trigger Angle Optimization: Based on this disclosure, the system's responsiveness to AC-side faults can be enhanced by optimizing the reactive power compensation strategy. Adjusting the reactive power compensation device's response time and compensation amount can further improve voltage stability and reduce commutation failures caused by voltage fluctuations. Furthermore, optimizing the trigger angle adjustment strategy and dynamically adjusting the commutation trigger angle based on grid status and real-time feedback signals allows for more precise control of commutation timing and improves system stability.

[0139] Dynamic control threshold adjustment: A dynamic control threshold adjustment mechanism can be introduced to automatically adjust the switching threshold based on the grid operating status and real-time measurement data. For example, the activation conditions of CLCC mode can be dynamically adjusted based on AC voltage fluctuations, enabling the system to flexibly respond to different fault conditions and improving the system's fault tolerance and adaptability.

[0140] Integration of deep learning algorithms: Deep learning algorithms are used to predict and analyze grid fault patterns. Through large-scale data training, deep learning models can automatically learn the grid's operating characteristics and fault characteristics, thereby optimizing the regulation strategies for CLCC mode switching and reactive power compensation. This approach can further improve the prediction accuracy of commutation failures and enhance the intelligence level of the grid system.

[0141] It should be noted that existing technologies may continue to employ switching strategies based on fixed empirical values. This approach relies on static rules of grid operation and cannot dynamically adjust to actual fault conditions. This can easily lead to commutation failures or premature switching, resulting in system instability. Existing technologies may also rely solely on traditional reactive power compensation methods, neglecting the prediction and prevention of commutation failures. While this can provide a certain degree of regulation for voltage fluctuations, it may fail to provide an effective emergency response in the event of a fault, thereby reducing grid stability. Existing technologies may also rely on traditional static fault response mechanisms, making switching decisions based on set rules and lacking intelligent analysis and real-time adjustment capabilities. This makes it difficult for the grid to adapt to complex operating conditions, reducing the system's ability to respond to dynamic faults and its recovery speed. This further highlights the importance of the intelligent optimization approach based on multidimensional data analysis, SVM algorithms, and Radviz visualization technology disclosed in this disclosure, providing a precise and efficient CLCC optimization solution that significantly improves grid stability and fault protection. While other alternatives may attempt to replace these approaches with static values ​​or traditional methods, they struggle to achieve the optimization effects of the disclosed solution in terms of dynamic response, prediction accuracy, and fault protection.

[0142] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present disclosure.

Claims

1. A CLCC parameter optimization method based on AC side fault data set, characterized in that: The method comprises the following steps: S100, performing AC side fault parallel simulation on the target DC project in simulation software to obtain multi-dimensional data; S200, analyzing the multi-dimensional data and identifying commutation failure and non-commutation failure; S300, using an SVM boundary algorithm to identify a boundary between a commutation failure and a non-commutation failure, and further obtaining confidence intervals for each fault component corresponding to the high-risk boundary; wherein the boundary is used as a high-risk boundary for switching the receiving end from the LCC mode to the CLCC mode; the high-risk boundary is represented by a portion of the multidimensional data; wherein the number of fault components is equal to the dimension of the multidimensional data; S400, using a fault recorder to collect waveforms of the currents of three phases a, b, and c on the receiving-end converter valve side of the actual target DC project; S500. Identify data with the same number of dimensions as the multidimensional data from the collected waveform, and determine whether to continue using the LCC mode for the receiving end or switch from the LCC mode to the CLCC mode based on the data with the same number of dimensions and the confidence interval of each fault component.

2. The method according to claim 1, characterized in that Preferably, step S100 includes the following steps: S101. Establish a target DC engineering model in PSCAD, wherein the target DC engineering model includes a sending end and a receiving end, wherein the receiving end includes the following three branches connected in parallel: a branch corresponding to an overvoltage component, a branch corresponding to a harmonic component, and a branch corresponding to a DC component; S102. For a ground fault on the receiving AC side, reasonably set amplitudes and scales for each of the five variables of the three branches, namely, the DC component, harmonic component, overvoltage component, phase difference between the overvoltage and DC component from the time of occurrence to the zero-crossing point, and phase difference between the harmonic and fundamental wave, to ensure the effectiveness and accuracy of the analysis; wherein the five variables serve as independent variables for the simulation; S103 , after the amplitude setting and the scale setting are completed, PSCAD is used to perform parallel simulation of AC side faults when the sending end adopts the LCC mode and the receiving end adopts the LCC mode.

3. The method according to claim 2, characterized in that In step S102: The DC component is divided into 7 scales from 0 to 24% of the DC component amplitude.

4. The method according to claim 2, characterized in that In step S102: The harmonic components are divided into 7 scales from 0 to 25% of the harmonic component amplitude.

5. The method according to claim 2, characterized in that In step S102: The overvoltage component is divided into 7 scales from 100% to 130% of the overvoltage component amplitude.

6. The method according to claim 2, characterized in that In step S102: The phase difference from zero to the moment when the overvoltage and DC component occur to the zero-crossing point is divided into 24 scales every 15 degrees within the range of 0 to 360 degrees.

7. The method according to claim 2, characterized in that In step S102: The phase difference between zero and the fundamental wave is divided into 24 scales every 15 degrees within the range of 0 to 360 degrees.

8. The method according to claim 1, characterized in that In step S100, During the simulation, the currents I of the three corresponding channels a, b, and c on the receiving end converter valve side are recorded. a , I b , I c ; Wherein, the current I on the receiving end converter valve side a , I b , I c as the dependent variable of the simulation.

9. The method according to claim 8, characterized in that Each channel includes 5 working conditions: overvoltage component V ov , harmonic component V h , DC component V DC , the phase difference between the overvoltage and DC component occurrence time and the zero crossing point , and the phase difference between the harmonics and the fundamental .

10. The method according to claim 8, characterized in that In step S200, If I a , I b , I c If the phase difference is 120 degrees, it is considered as non-commutation failure; If I a , I b , I c If the phases do not differ by 120 degrees, it is considered a commutation failure.

Images