Method and device for setting current protection of distribution network based on fault current curve

Abstract

This invention provides a method and apparatus for setting current protection in distribution networks based on fault current curves. The method includes: calculating fault current curves for the protection device under multiple scenarios based on a sample matrix of operating states of the distribution network, generating a corresponding short-circuit current matrix; clustering the fault current curves of each sample in the short-circuit current matrix to generate multiple sets of fault current influence states, a comprehensive fault current curve corresponding to each set of fault current influence states, and a protection setting matrix; and matching the current operating state information based on the sets of fault current influence states and the protection setting matrix to generate a target protection setting combination. This enables adaptive matching of protection settings when there are fluctuations in the output of distributed power sources and changes in topology, effectively improving the accuracy of current protection under complex operating conditions, ensuring the coordination and unity of protection selectivity and sensitivity, and reducing the risk of over-tripping or failure to operate.

Description

Technical Field

[0001] This invention relates to the field of distribution network protection technology, and in particular to a method and apparatus for setting distribution network current protection based on fault current curves. Background Technology

[0002] As the global energy structure transitions towards cleaner and lower-carbon energy sources, the penetration rate of distributed generation (DG) such as photovoltaic (PV) and wind power in distribution networks continues to increase, forming active distribution networks with multiple DG connections. The application of active distribution networks improves energy efficiency and enhances power supply flexibility, becoming an important direction for smart grid development. However, the high proportion of DG connections also poses serious challenges to the fault protection of distribution networks. On the one hand, DG output exhibits significant randomness and volatility, influenced by natural factors such as climate; on the other hand, the distribution network topology dynamically changes due to switching operations. These factors lead to complex and varied fault scenarios in active distribution networks, with fault current amplitude and distribution characteristics fundamentally different from those in traditional passive distribution networks. Traditional current protection schemes based on fixed settings are no longer fully adaptable to these changes.

[0003] For fault protection in active distribution networks, existing technologies mainly include: traditional fixed-set protection schemes, which calculate and maintain the setpoints offline based on the initial topology and rated parameters, but ignore the changes in fault current distribution after the connection of distributed generation (DG), resulting in deviations between the protection range and the preset value when DG output fluctuates; setpoint correction schemes based on DG output experience values, which statically correct the setpoints using the maximum or average output obtained from historical statistics, but static experience values ​​are difficult to track the dynamic changes in DG output, and deviations still exist in scenarios with continuous output fluctuations; fault current calculation schemes based on single topology scenarios, which perform offline setpoint adjustments for a few preset topology scenarios, but have limited coverage of actual topology changes such as multi-switch combination switching, and the setpoints do not completely match the actual fault current characteristics when the preset scenario is exceeded; and DG output scenario partitioning schemes based on simple clustering, which result in insufficient accuracy and stability in distinguishing fault current contribution characteristics by the clustering results.

[0004] In summary, existing technologies for fault protection in active distribution networks primarily focus on fault current characteristic analysis and setting, but their adaptability to fluctuations in distributed generation (DG) output and dynamic topology changes needs improvement. When DG output or network topology changes, existing solutions have room for improvement in terms of the accuracy of protection range, the comprehensiveness of setting adjustments, and the effectiveness of classifying different operating states. Therefore, it is necessary to research fault assessment and protection setting optimization technologies that can better adapt to fluctuations in DG output and changes in topology.

[0005] This section is intended to provide background or context for the embodiments of this application set forth in the claims. The description herein is not an admission that it is prior art simply because it is included in this section. Summary of the Invention

[0006] One objective of this invention is to provide a distribution network current protection setting method based on fault current curves. This method can cluster operating states based on the similarity of the impact of fault current curves, generating protection settings adapted to different state sets. This enables adaptive matching of protection settings when distributed power output fluctuates and topology changes occur, effectively improving the accuracy of current protection operation under complex operating conditions, ensuring the coordination and unity of protection selectivity and sensitivity, and reducing the risk of cascading tripping or failure to operate. Another objective of this invention is to provide a distribution network current protection setting device based on fault current curves. A further objective of this invention is to provide a computer-readable medium. A final objective of this invention is to provide a computer device.

[0007] To achieve the above objectives, this invention discloses a method for setting distribution network current protection based on fault current curves, comprising: Based on the operating status sample matrix of the distribution network, the protection device is used to calculate the fault current curves for multiple scenarios and generate the corresponding short-circuit current matrix. Cluster the fault current curves of each sample in the short-circuit current matrix to generate multiple fault current influence state sets and the comprehensive fault current curve corresponding to each fault current influence state set. Generate a protection setting matrix based on the comprehensive fault current curve corresponding to each fault current influence state set; Based on the fault current influence state set and the protection setting matrix, the collected current operating status information is matched to generate the target protection setting combination.

[0008] Preferably, based on the operating status sample matrix of the distribution network, multi-scenario fault current curves are calculated for the protection device to generate a corresponding short-circuit current matrix, including: Multiple fault location points are set along the topology line within the protection influence area of ​​the protection device; Based on different topology states and distributed power supply operating states, the fault current of the protection device is calculated according to the preset information matrix when a fault occurs at each fault location point. Based on the fault current of the protection device, multiple sample fault current curves corresponding to the protection device are plotted to form a short-circuit current matrix.

[0009] Preferably, the fault current curves of each sample in the short-circuit current matrix are clustered to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set, including: Feature extraction is performed on the fault current curves of each sample in the short-circuit current matrix to generate an offset feature vector; Based on the offset feature vector, the fault current curves of each sample are clustered to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set.

[0010] Preferably, feature extraction is performed on the fault current curves of each sample in the short-circuit current matrix to generate an offset feature vector, including: The fault current curves without distributed power supply access were selected from the fault current curves of each sample as the benchmark fault current curves. Based on the baseline fault current curve and the sample fault current curve, the maximum relative offset, minimum relative offset, and average relative offset are generated. Based on the preset reference segmented protection range, the reference fault current curve is matched to determine the reference segmented protection setting group. Based on the benchmark segmented protection setting group, the fault current curves of each sample are matched to determine the segmented protection range of the sample. Based on the sample segment protection range and the benchmark segment protection range, the segment relative offset is generated; An offset feature vector is generated based on the segmented relative offset, the maximum relative offset, the minimum relative offset, and the average relative offset.

[0011] Preferably, a protection setting matrix is ​​generated based on the comprehensive fault current curve corresponding to each fault current influence state set, including: Based on the preset segmented protection setting coefficients, the fault current of the comprehensive fault current curve is calculated to generate the protection setting group corresponding to the protection device. A protection setting matrix is ​​generated based on the protection setting group corresponding to each protection device.

[0012] Preferably, based on the fault current influence state set and the protection setting matrix, the collected current operating state information is matched to generate a target protection setting combination, including: Generate the current fault current curve based on the current operating status information; Similarity calculations are performed based on the current fault current curve and multiple fault current influence state sets to determine the target influence state set; The target influence state set is matched with the protection setting matrix to generate the target protection setting combination.

[0013] This invention also discloses a distribution network current protection setting device based on fault current curves, comprising: The multi-scenario fault current curve calculation unit is used to calculate the multi-scenario fault current curve of the protection device based on the operating status sample matrix of the distribution network, and generate the corresponding short-circuit current matrix. The curve clustering unit is used to cluster the fault current curves of each sample in the short-circuit current matrix to generate multiple fault current influence state sets and the comprehensive fault current curve corresponding to each fault current influence state set. The protection setting matrix calculation unit is used to generate the protection setting matrix based on the comprehensive fault current curve corresponding to each fault current influence state set. The target protection setting combination matching unit is used to match the collected current operating status information based on the fault current influence state set and the protection setting matrix to generate the target protection setting combination.

[0014] The present invention also discloses a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method described above.

[0015] The present invention also discloses a computer device, including a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, wherein the processor executes the program to implement the method described above.

[0016] The present invention also discloses a computer program product, including a computer program / instruction, which, when executed by a processor, implements the method described above.

[0017] This invention calculates fault current curves for protection devices based on a sample matrix of operating states in the distribution network, generating a corresponding short-circuit current matrix. It then clusters the fault current curves in each sample of the short-circuit current matrix to generate multiple sets of fault current influence states and a comprehensive fault current curve corresponding to each set. Based on the comprehensive fault current curve for each set, a protection setting matrix is ​​generated. Using the fault current influence state sets and the protection setting matrix, the collected current operating state information is matched to generate a target protection setting combination. This invention can cluster operating states based on the similarity of fault current curve influences, generating protection settings adapted to different state sets. This enables adaptive matching of protection settings when distributed power output fluctuates and topology changes, effectively improving the accuracy of current protection under complex operating conditions, ensuring the coordination and unity of protection selectivity and sensitivity, and reducing the risk of cascading tripping or failure to operate. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart of a distribution network current protection setting method based on fault current curves provided for embodiments of the present invention; Figure 2 A flowchart of another distribution network current protection setting method based on fault current curve provided in an embodiment of the present invention; Figure 3 A topology diagram of a 10kV distribution network connected to a distributed power source is provided in an embodiment of the present invention; Figure 4 A trend graph showing the sum of squared errors as a function of the number of clusters, provided for an embodiment of the present invention; Figure 5 This is a schematic diagram of clustering results for a high-proportion distributed power supply access processing scenario provided by an embodiment of the present invention; Figure 6 A graph showing the change of short-circuit current over time measured at the protection device under different distributed power output states and fault types, provided as an embodiment of the present invention. Figure 7 Another curve showing the change of short-circuit current over time measured at the protection device under different distributed power output states and fault types, provided as an embodiment of the present invention; Figure 8 A schematic diagram illustrating the sensitivity of a protection device 5 under different output levels of a distributed power source, provided in an embodiment of the present invention. Figure 9 A schematic diagram of the structure of a distribution network current protection setting device based on a fault current curve provided in an embodiment of the present invention; Figure 10 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0020] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] To facilitate understanding of the technical solution provided in this application, the relevant content of the technical solution is explained below. In distribution networks, staged current protection (such as Stage I, Stage II, and Stage III) is typically configured. Its protection settings are based on the curve of fault current changing with fault distance, i.e., the fault current curve. In traditional passive distribution networks, this curve is determined solely by the system power source; the fault current monotonically decreases with increasing line length, resulting in a relatively fixed protection range. After the integration of distributed power sources, changes in their output and network topology switching states alter the flow path and amplitude distribution of the fault current, causing the fault current curve corresponding to the same protection installation location to shift and distort. This leads to an expansion or contraction of the protection range set according to the fixed curve. To adapt to this change, cluster analysis can be performed on the fault current curves under different operating conditions. States with similar impacts on protection action characteristics are grouped into one category, and a representative curve for each category is generated. The corresponding protection settings are then calculated based on this representative curve. During online operation, the current actual fault current curve is matched with various representative curves, and the matched settings are quickly retrieved, thereby achieving adaptive adjustment of the protection parameters.

[0022] After distributed generation (DG) is connected, its output fluctuations and changes in the distribution network topology reshape the flow path and amplitude distribution of fault currents, causing changes in the shape of the measured fault current curve at the protection installation point. This, in turn, extends or reduces the operating range of staged current protection, potentially leading to problems such as cascading trips, insufficient sensitivity, or even protection failure to operate. Existing adaptive protection methods typically use the output level of DG, load size, or source-load combination as the basis for scenario division. This approach struggles to directly characterize the actual impact of different operating states on the fault current curve and protection range, easily resulting in deviations between the state classification results and the actual requirements of the protection settings. To address these issues, this invention uses the similarity of the impact of fault current curves as the criterion for operating state division. By extracting the offset of the fault current curve and the offset of the protection range, operating states with similar impacts on protection operating characteristics are grouped into the same state set, and typical fault current curves that can be used for setting calculations are generated accordingly. During system online operation, based on the degree of matching between the current measured fault current curve and each typical curve, the corresponding protection settings are invoked, thereby improving the adaptability of current protection to changes in DG output and topology adjustments.

[0023] This invention proposes an adaptive fault assessment and protection setting method for distribution networks based on the similarity of fault current curve influence. This method does not use the output of distributed generation sources, load levels, or source-load combinations as direct clustering inputs. Instead, it uses the overall influence characteristics of the fault current curves at the protection installation location under different operating states as the classification basis. For the same protection device, if the fault current curves formed under various topology states and distributed generation operating states exhibit similar patterns in amplitude range, curve change rate, protection range offset direction, and offset magnitude, these operating states are merged into the same fault current influence state set. In this way, operating states within the same influence state set correspond to similar protection action response characteristics and setting adjustment requirements, thereby significantly improving the pertinence and adaptability of protection setting.

[0024] The following example uses a distribution network current protection setting device based on a fault current curve as the execution subject to illustrate the implementation process of the distribution network current protection setting method based on a fault current curve provided in this embodiment of the invention. It is understood that the execution subject of the distribution network current protection setting method based on a fault current curve provided in this embodiment of the invention includes, but is not limited to, a distribution network current protection setting device based on a fault current curve.

[0025] Figure 1 A flowchart of a distribution network current protection setting method based on fault current curves is provided for an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes: Step 101: Based on the operating status sample matrix of the distribution network, calculate the fault current curves of the protection device for multiple scenarios and generate the corresponding short-circuit current matrix.

[0026] In this embodiment of the invention, operating status samples under different distributed power outputs and different switching state combinations are collected in advance to form an operating status sample matrix. For each protection device, multiple fault location points are set along the line within its protection influence domain. The fault current flowing through the protection device when a short circuit occurs at each fault location under each sample state is calculated one by one, thereby obtaining multiple fault current curves, which are then summarized to form a short circuit current matrix.

[0027] This step involves pre-calculating short-circuit current data covering various operating states offline, providing a complete input basis for subsequent clustering and setting, and avoiding redundant online calculations.

[0028] Step 102: Cluster the fault current curves of each sample in the short-circuit current matrix to generate multiple fault current influence state sets and the comprehensive fault current curve corresponding to each fault current influence state set.

[0029] In this embodiment of the invention, the fault current curve corresponding to each sample state in the short-circuit current matrix is ​​regarded as a data object. Feature vectors that can reflect the changes in curve shape and protection range are extracted. Clustering algorithms (such as k-means) are used to divide each sample fault current curve into several categories. The sample fault current curves in each category have similar offset characteristics and are called a fault current influence state set. All fault current curves in the same state set are combined (e.g., the average value or median value is taken) to generate a representative comprehensive fault current curve.

[0030] This step compresses a large number of operating states into a few sets of influencing states, significantly reducing the computational complexity of offline tuning and online matching, while ensuring the consistency of protection behavior under similar states.

[0031] Step 103: Generate the protection setting matrix based on the comprehensive fault current curve corresponding to each fault current influence state set.

[0032] In this embodiment of the invention, for each fault current-affected state set, based on its comprehensive fault current curve, the corresponding protection settings for stage I, stage II, and stage III are calculated according to the setting rules of staged current protection (e.g., stage I is set according to the fault at the end of the line, stage II is coordinated with adjacent lines, etc.), forming a protection setting group. The above process is repeated for all protection devices and all their state sets, and all protection setting groups are summarized into a protection setting matrix.

[0033] This step directly maps the offline clustering results to specific protection settings, ensuring that the settings under each state set match the fault current characteristics of that state, thus providing a data foundation for online adaptive switching.

[0034] Step 104: Based on the fault current influence state set and the protection setting matrix, match the collected current operating status information to generate the target protection setting combination.

[0035] In this embodiment of the invention, when the system is running online, it collects real-time operating information such as distributed power output and switch status, and calculates the fault current curve under the current state based on this operating information. The fault current curve is then compared with the comprehensive fault current curve of each fault current-affected state set obtained during the offline phase (e.g., calculating the Euclidean distance), and the most similar state set is selected as the matching result. The setting group of the corresponding protection device under this state set is read from the protection setting matrix and used as the target protection setting combination for the current application.

[0036] This step enables online real-time matching and adaptive switching of protection settings, allowing the protection settings to automatically adjust in response to changes in distributed power output and topology, effectively improving the accuracy and flexibility of protection.

[0037] In the technical solution provided by this invention, based on the operating status sample matrix of the distribution network, multi-scenario fault current curves are calculated for the protection device to generate a corresponding short-circuit current matrix; the fault current curves of each sample in the short-circuit current matrix are clustered to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set; a protection setting matrix is ​​generated based on the comprehensive fault current curve corresponding to each fault current influence state set; based on the fault current influence state set and the protection setting matrix, the collected current operating status information is matched to generate a target protection setting combination. This allows for clustering of operating states based on the similarity of fault current curve influences, generating protection settings adapted to different state sets. This enables adaptive matching of protection settings when distributed power output fluctuates and topology changes, effectively improving the accuracy of current protection operation under complex operating conditions, ensuring the coordination and unity of protection selectivity and sensitivity, and reducing the risk of cascading tripping or failure to operate.

[0038] Figure 2 A flowchart of another distribution network current protection setting method based on fault current curves provided in this embodiment of the invention is shown below. Figure 2 As shown, the method includes: Step 201: Collect sample information on the operating status of the power distribution network.

[0039] In this embodiment of the invention, each step is performed by a distribution network current protection setting device based on the fault current curve.

[0040] In this embodiment of the invention, the grid structure information of the distribution network changes during operation. The change in topology is characterized by the closing status of switches. Sample information of the distribution network under different operating states is collected to form a data set representing a possible operating scenario of the distribution network.

[0041] In this embodiment of the invention, the operational status sample information includes, but is not limited to, line switch status, power switch status, distributed generation output, network topology, and line impedance parameters. The collection range covers the typical fluctuation range of distributed generation output (e.g., from 0 to rated current) and various possible switch combination states (including single switch switching and multi-switch combination switching). Each sample corresponds to a set of defined switch statuses and distributed generation output values, used to characterize an operational scenario.

[0042] This step collects sample information covering the output fluctuation range and topology change range of distributed power sources, providing complete raw data for the subsequent construction of the operating status sample matrix, ensuring that offline calculations can cover various operating conditions that may occur in actual operation.

[0043] Step 202: Perform topological state encoding on the running status sample information to construct the running status sample matrix.

[0044] In this embodiment of the invention, topology state coding refers to the process of converting the open / closed state of a switch into a numerical code.

[0045] Specifically, the circuit switch status includes circuit switch status containing a single switch and circuit switch status containing two switches. For circuit switch status containing a single switch, closing is recorded as 1, and opening is recorded as 0. The status value of each circuit switch is recorded sequentially from the power supply side until the end of the circuit, forming a circuit switch matrix A. For circuit switch status containing two switches, it is recorded as 1 only when both switches are closed, and 0 when either switch is open or both are open. The status value of each circuit switch is recorded sequentially from the power supply side until the end of the circuit, and then the switch status is recorded in the opposite direction and arranged to form a circuit switch matrix A. This is denoted as: , where s is the line switch number.

[0046] For the power switch status, if the power switch is closed, its status is recorded as 1; if the power switch is open, its status is recorded as 0. The power switches are encoded to form a power switch matrix B. This matrix is ​​denoted as... Where 'o' represents the power switch number.

[0047] Based on the topological connections in the distribution network topology, an correlation matrix M is defined. pq Correlation matrix M pq The elements in the matrix describe the relationship between the current flowing into bus p and the current flowing to line q at the fault location. The correlation matrix M pq The dimension is determined by the number of buses and lines, and each element takes a value of 0 or 1, with the specific value defined according to the current correlation:

[0048] Connect the line switch matrix A and the correlation matrix M pq Multiplying them together yields the topological description matrix T. RE.p , denoted as: .

[0049] Construct the line impedance matrix Z based on the line impedance parameters. L.p Line impedance matrix Z L.p The impedance matrix Z is constructed based on the line connection relationships. For bus p, the impedance values ​​of all lines within its power supply range are obtained; the impedance value of each line is calculated by multiplying the impedance per unit length of the line by the line length; all line impedance values ​​are arranged in order of line number to construct the line impedance matrix Z. L.p .

[0050] The topological description matrix T RE.p The line impedance matrix Z corresponding to bus p L.p Multiplying these yields the associated impedance matrix Z of the bus voltage p.RE.p , denoted as: .

[0051] Based on the output of the distributed power source, construct the power supply output current matrix I. S The dimension of this matrix is ​​equal to the number of distributed power sources. For each distributed power source, if its output current is 0, the corresponding element has a value of 0; if its output current is not 0, the corresponding element has the output current value. The output current values ​​of all distributed power sources are arranged in order of their power source numbers to construct a power source output current matrix I. S .

[0052] Connect the power switch matrix B and the power output current matrix I. S Multiplying them together yields the power supply current matrix I, which takes into account the topology. RE , denoted as: .

[0053] The topological description matrix T RE.p The line impedance matrix Z corresponding to bus p L.p Multiplying these yields the associated impedance matrix Z of the bus voltage p. RE.p Line impedance matrix Z L.p It can be a diagonal matrix or constructed based on the line connection relationship, with the diagonal elements being the impedance values ​​of the corresponding lines.

[0054] Connect the power switch matrix B and the power output current matrix I. S Multiplying them together yields the power supply current matrix I, which takes into account the topology. RE Power supply output current matrix I S It is a diagonal matrix, and the diagonal elements are the output current values ​​of each distributed power source under the current sample (which can be calculated from the output power and voltage).

[0055] Repeat the above encoding process for all operating state sample information, and calculate the correlation impedance matrix Z. RE.p Power supply current matrix I RE and the topological description matrix T RE.p The samples are stored in a unified running state sample matrix E, denoted as: Each column or row of the operating status sample matrix E corresponds to a sample's complete topology code and electrical parameter information.

[0056] This step uses system topology state encoding and matrix transformation to quantize discrete switch states and continuous electrical parameters into a structured operating state sample matrix. This provides standardized and complete input data for subsequent multi-scenario fault current curve calculations, ensuring consistency and reproducibility of calculations between different samples.

[0057] Step 203: Set multiple fault location points along the topology line within the protection influence area of ​​the protection device.

[0058] In this embodiment of the invention, the protection influence domain of each protection device is determined. The protection influence domain refers to the line range within which the protection device can sense fault current and may operate, typically including the line where the protection device is located and its downstream adjacent lines.

[0059] Specifically, within the protection influence zone, multiple fault location points are set along the topology line. These fault location points are set either by dividing the line according to its length ratio or by continuously selecting points along the line according to a set step size. The setup should cover all line segments within the protection influence zone, including main lines and branch lines. Each fault location point corresponds to a fault distance *l*, representing the actual distance from the installation point of the protection device along the line to the fault point.

[0060] As an alternative, if the fault location points are divided according to the proportion of the line length, they can be set at 10%, 20%, ..., 90%, 100% of the line length.

[0061] As an alternative, if the fault location is selected continuously along the line according to the set step size, a fault location point can be set every 0.5km or 1km.

[0062] Step 204: Calculate the fault current of the protection device at each fault location point according to different topology states and distributed power supply operating states, based on the preset information matrix.

[0063] In this embodiment of the invention, for each fault location, the short-circuit fault current is calculated under different topology states and distributed power supply operating states. The combination of topology states and distributed power supply operating states is derived from operating state sample information. Each sample corresponds to a set of defined topology states (i.e., line switch states and power switch states) and distributed power supply operating states (i.e., output of each distributed power supply). The preset information matrix is ​​derived from the operating state sample matrix, which stores the associated impedance matrix Z of each sample. RE.p Power supply current matrix I RE and the topological description matrix T RE.p .

[0064] Under different topology states and distributed power supply operating states, short-circuit fault calculations are performed for the current fault location. Short-circuit fault types can include three-phase short circuits, two-phase short circuits, etc. This invention uses a three-phase short circuit as an example for illustration. The calculation process includes the following sub-steps: (1) Modify the network topology parameters according to the location of the fault. Introduce a fault branch at the fault point and treat the fault point as a new node.

[0065] (2) Determine the network structure after the fault based on the topology association information matrix and the current sample's switching state; determine the network structure after the fault based on the association impedance matrix Z. RE.p This allows us to obtain the actual connection relationship between nodes and branches in the current sample.

[0066] (3) The iterative method is used to solve for the node voltage and distributed generation output after the fault. Initially, it is assumed that the output of all distributed generation is its rated value or the value given in the current sample; calculate the voltage of each node; recalculate the output current of the distributed generation based on the node voltage (for inverter-type DG, its output current is limited by the control strategy, usually a constant current source or a current source affected by voltage). Repeat the above process until the convergence condition is met. The convergence condition can be that the maximum change in the voltage of all nodes in two adjacent iterations is less than a set threshold (e.g., 10). -4 (pu), or the maximum change in the output of all distributed power sources is less than a set threshold.

[0067] (4) After convergence, calculate the fault current flowing through each protection device based on the voltage difference between adjacent nodes and the line impedance. Specifically, for the line segment where protection device k is located, the voltages at its two end nodes are U... left and U right The line impedance is Z line The fault current flowing through the protection device is I. k = (U left - U right ) / Z line .

[0068] (5) Record the current fault location point I under the current sample state (the m-th distributed power supply operating state and the n-th topology state). j The fault current value flowing through the k-th protection device when a fault occurs is denoted as: .

[0069] For each protection device k, for all sample states (all combinations of m and n) and all fault location points I j (j=1,2,…,r), repeat the above calculation process.

[0070] Step 205: Based on the fault current of the protection device, draw multiple sample fault current curves corresponding to the protection device to form a short-circuit current matrix.

[0071] In this embodiment of the invention, for the k-th protection device, a set of fault current sequences corresponding to the fault location points is obtained under different topology states and distributed power supply operating states; this sequence is regarded as a curve, with the fault distance l as the abscissa and the fault current I as the ordinate. kPlot the sample fault current curves corresponding to this protection device on the ordinate. Each sample fault current curve reflects the variation of fault current with fault distance under specific distributed power source operating conditions and topology.

[0072] Arrange the fault current sequence corresponding to all sample states of the k-th protection device according to the fault point, different topology states, and distributed power supply operating states to form a short-circuit current matrix:

[0073] Among them, F k This represents the short-circuit current matrix of the k-th protection device; Indicates the protection device number; Indicates the operating status of distributed power sources; Represents the topological state; represents the first Each fault location point , This represents the total number of fault locations. It represents the fault current flowing through the k-th protection device when a fault occurs at the j-th fault location under the m-th distributed power source operating state and the n-th topology state.

[0074] This step normalizes a large amount of discrete fault current data into a structured short-circuit current matrix. Each sample state corresponds to a complete fault current curve, providing a standardized input data format for subsequent cluster analysis based on curve similarity, which facilitates feature extraction and state classification.

[0075] Step 206: Extract features from the fault current curves of each sample in the short-circuit current matrix to generate offset feature vectors.

[0076] In this embodiment of the invention, numerical features that characterize the impact on protection operation characteristics are extracted from each sample fault current curve, and these features are combined into an offset feature vector. The offset feature vector includes: maximum relative offset, minimum relative offset, average relative offset, relative offset of protection range I, relative offset of protection range II, and relative offset of protection range III.

[0077] In this embodiment of the invention, step 206 specifically includes: Step 2061: Select the sample fault current curve without distributed power supply access from each sample fault current curve as the benchmark fault current curve.

[0078] In this embodiment of the invention, for the k-th protection device, under the m-th distributed power supply operating state and the n-th topology state, the fault current curves flowing through the protection device at different fault locations are denoted as follows: , .in,l Indicates the distance to the fault. This represents the fault current flowing through the k-th protection device under the corresponding operating condition.

[0079] In this embodiment of the invention, "no distributed power source access" means that the output current of all distributed power sources is zero, and the grid connection switches of all distributed power sources are in the open state. Specifically, sample fault current curves without distributed power source access are selected from each sample fault current curve, and the selected sample fault current curves are determined as the reference fault current curves, denoted as: , The subscript 0 indicates the baseline state. The baseline fault current curve reflects the intrinsic relationship between fault current and fault distance when there is no distributed generation influence.

[0080] Step 2062: Based on the reference fault current curve and the sample fault current curve, generate the maximum relative offset, minimum relative offset and average relative offset.

[0081] In this embodiment of the invention, for each sample fault current curve (excluding the reference curve itself) in the short-circuit current matrix, its absolute offset relative to the reference fault current curve at each fault location point is calculated:

[0082] in, It represents the absolute offset of the fault current of the k-th protection device when a fault occurs at the j-th fault location under the m-th distributed power source operating state and the n-th topology state. Indicates the first Each fault location point , This represents the total number of fault locations. This indicates that when a fault occurs at the j-th fault location in the m-th distributed power source operating state and the n-th topology state, the current flowing through the j-th fault location is... Fault current of each protection device; Indicates the flow through the first when no distributed power source is connected. The fault current of the protection device.

[0083] like This indicates that the current operating state has increased the fault current at the protection device; if This indicates that the current operating state reduces the fault current at the protection device; if This indicates that the current operating state causes no change in the fault current at the protection device.

[0084] To eliminate the inconsistency in characteristic dimensions caused by differences in rated current between different protection devices, the relative fault current offset is further calculated and used to represent the degree of change in the fault current curve.

[0085] in, It represents the relative offset of the fault current of the k-th protection device when a fault occurs at the j-th fault location under the m-th distributed power source operating state and the n-th topology state; It represents the absolute offset of the fault current of the k-th protection device when a fault occurs at the j-th fault location under the m-th distributed power source operating state and the n-th topology state. Indicates the flow through the first when no distributed power source is connected. The fault current of the protection device.

[0086] It is worth noting that when If the value is zero or extremely small, a very small positive number can be set as a denominator protection, or the point can be excluded.

[0087] After obtaining the relative offsets of all r fault location points, calculate the maximum relative offset, minimum relative offset, and average relative offset:

[0088] in, The maximum relative offset of the fault current of the kth protection device when a fault occurs at each fault location in the mth distributed power source operating state and the nth topology state represents the maximum upward offset of the fault current curve relative to the reference curve in the current operating state. The relative offset of the fault current of the k-th protection device when a fault occurs at the j-th fault location under the m-th distributed power source operating state and the n-th topology state. The minimum relative offset of the fault current of the k-th protection device when a fault occurs at each fault location under the m-th distributed power source operating state and the n-th topology state is represented by the maximum downward offset of the fault current curve relative to the reference curve under the current operating state. The average relative offset of the fault current of the k-th protection device when a fault occurs at each fault location under the m-th distributed power supply operating state and the n-th topology state represents the overall offset of the fault current curve. For the first Each fault location point , This represents the total number of fault locations.

[0089] Step 2063: Based on the preset reference segmented protection range, match the reference fault current curve to determine the reference segmented protection setting group.

[0090] In this embodiment of the invention, the segmented protection range under a preset baseline state refers to the expected line length range protected by each segment (Segment I, Segment II, and Segment III) in the staged current protection. For example, the protection range of Segment I can be preset to 80% of the total line length, the protection range of Segment II to 120% of the total line length and extending to adjacent lines, and the protection range of Segment III to the total line length as a backup. These preset values ​​are determined based on engineering experience or distribution network protection configuration specifications.

[0091] Specifically, the preset I-segment reference protection range boundary Substitute a distance value into the reference fault current curve Solve the equation The corresponding fault current value is obtained. ; Set the preset II-stage reference protection range boundary Substitute into the reference fault current curve Solve the equation The corresponding fault current value is obtained. Similarly, by substituting the preset III-stage reference protection range boundary into the reference fault current curve, the corresponding fault current value is obtained. Therefore, the baseline segmented protection setting group is obtained, denoted as: .

[0092] Step 2064: Match the fault current curves of each sample according to the benchmark segmented protection setting group to determine the sample segmented protection range.

[0093] In this embodiment of the invention, for each sample fault current curve, under the m-th distributed power source operating state and the n-th topology state, the reference segmented protection setting group remains unchanged. Let the current protection range boundaries of segment I, segment II, and segment III corresponding to the current fault current curve be respectively... , and It satisfies:

[0094] Find the current protection range boundary that satisfies the above formula on the corresponding fault current curve. , and This constitutes the protected area for sample segmentation.

[0095] Step 2065: Generate the relative offset of segments based on the sample segment protection range and the benchmark segment protection range.

[0096] In this embodiment of the invention, the segmented relative offset includes the relative offsets of segment I, segment II, and segment III. Specifically, for each segment (I, II, III), the relative offset of the sample segment protection range relative to the reference segment protection range is calculated:

[0097] in, , , These are the relative offsets of segments I, II, and III, respectively, representing the degree of offset of the current protection range boundary of segments I, II, and III relative to the reference state. , and These are the current protection boundaries for sections I, II, and III, respectively. , , These are the reference protection range boundaries for sections I, II, and III, respectively.

[0098] In this embodiment of the invention, when the relative offset of the segments is greater than 0, it indicates that the protection range has been expanded, which may cause over-level operation; when the relative offset of the segments is less than 0, it indicates that the protection range has been shortened, which may lead to decreased sensitivity or failure to operate.

[0099] Step 2066: Generate an offset feature vector based on the segmented relative offset, maximum relative offset, minimum relative offset, and average relative offset.

[0100] In this embodiment of the invention, for the k-th protection device in the m-th distributed power supply operating state and the n-th topology state, the segmented relative offset, maximum relative offset, minimum relative offset, and average relative offset are combined to construct an offset feature vector of the fault current curve influence. for:

[0101] Each component of the offset feature vector is a dimensionless real number, which can be directly used in subsequent clustering algorithms. Steps 2061 to 2066 are performed on each sample fault current curve (except for the baseline curve) in the short-circuit current matrix to obtain a set of offset feature vectors, each vector corresponding to a distributed power source operating state and a topology state.

[0102] This step transforms the original fault current curve data into a quantified offset feature vector. This offset feature vector reflects both the impact of the current operating state on the fault current curve itself and on the protection action range, providing a standardized feature input for subsequent clustering analysis based on impact similarity.

[0103] Step 207: Based on the offset feature vector, cluster the fault current curves of each sample to generate multiple fault current influence state sets and the comprehensive fault current curve corresponding to each fault current influence state set.

[0104] In this embodiment of the invention, step 207 specifically includes: Step 2071: Using the offset feature vector as the clustering object, classify the fault current curves of each sample under different topological states and distributed power supply operating states, and generate multiple fault current influence state sets.

[0105] In this embodiment of the invention, the offset feature vector is used as the input of the clustering algorithm. The clustering algorithm automatically divides the samples into several clusters based on the similarity between vectors (usually using Euclidean distance). The offset feature vectors within each cluster are close to each other, while the offset feature vectors between clusters are significantly different.

[0106] In this embodiment of the invention, the clustering algorithms that can be used include, but are not limited to, k-means, k-means++, hierarchical clustering, spectral clustering, and density clustering (such as DBSCAN). If k-means or k-means++ is used, the number of clusters must be specified in advance. The number of clusters can be determined using objective criteria such as the elbow method (inflection point of the SSE descent curve) or the silhouette coefficient method.

[0107] After clustering, each cluster corresponds to a set of fault current influence states. Let Ω be the set of fault current influence states for the k-th protection device. k,c This set contains the sample fault current curves corresponding to all sample states assigned to this cluster:

[0108] in, This represents the state set affected by the c-th fault current corresponding to the k-th protection device; q represents the number of fault current curves in this state set. This represents the a-th fault current curve within the state set. .

[0109] It is worth noting that although the sample states within the same state set may differ in original parameters such as distributed power output and topology switch status, they are similar in offset feature vectors. This means that these states have similar characteristics in their impact on the fault current curve and protection range at the protection device, and therefore can share the same set of protection settings.

[0110] Step 2072: Generate the corresponding comprehensive fault current curve based on the fault current influence state set.

[0111] In this embodiment of the invention, the set of states affecting the c-th fault current of the k-th protection device is Ω. k,c The set contains q sample fault current curves. A comprehensive calculation method is used to calculate the comprehensive current value of the q sample fault current curves, which is then used as the value of the comprehensive fault current curve at that location. By processing all r fault location points sequentially, a comprehensive fault current curve representing the state set can be obtained. The comprehensive fault current curve is a representative curve formed by comprehensively calculating the current values ​​of all fault current curves in a set of fault current influence states at the same fault location point.

[0112] In this embodiment of the invention, the comprehensive calculation methods include, but are not limited to, the median method, the average method, the cluster center method, the weighted average method, and the envelope value method. Specifically, the median method involves taking the median of the fault current curves of q samples at that position, which can reduce the influence of abnormal curves. The average method involves taking the average of the fault current curves of q samples at that position. The cluster center method involves restoring the cluster vectors generated by the clustering algorithm itself to curves. The weighted average method involves assigning different weights to different sample states (e.g., weighting based on occurrence probability). The envelope value method involves taking the upper envelope (maximum value) or the lower envelope (minimum value) to improve adaptability to extreme operating conditions. This embodiment of the invention does not limit the specific selection of the comprehensive calculation method.

[0113] Taking the average method as an example, for the same fault location The fault currents corresponding to each sample fault current curve within the fault current influence state set are combined to generate the corresponding comprehensive fault current curve:

[0114] in, Indicates the first The first protection device in the Under the state set affected by the fault current, at the... The combined fault current at each fault location point; Indicates the first Each fault location point; q is the number of sample fault current curves; This indicates the state set containing the first... The fault current curve Fault current at each fault location point .

[0115] This step, through clustering based on offset feature vectors, groups a large number of original operating state samples into a limited set of fault current-affected states, significantly reducing the computational complexity of offline setting and online matching. Since each state within the same set has a similar impact on protection action characteristics, it becomes possible to share a common set of protection settings, avoiding redundant calculations caused by setting settings individually for each sample state. Furthermore, the comprehensive fault current curve further condenses multiple curves within the state set into a single representative curve, providing a unified and simplified input for subsequent protection setting calculations.

[0116] Step 208: Calculate the fault current of the comprehensive fault current curve according to the preset segmented protection setting coefficients, and generate the protection setting group corresponding to the protection device.

[0117] In this embodiment of the invention, the segmented protection setting coefficients include the I-stage protection setting coefficient, the II-stage protection setting coefficient, and the III-stage protection setting coefficient.

[0118] Specifically, for segment I, the expected protection range boundary of segment I for the k-th protection device is preset. (The unit is distance, such as line length or kilometers from the installation point of the protection device), then the protection setting value of segment I under the c-th state set can be expressed as:

[0119] in, Indicates the first The first protection device in the The protection setting of section I under the state set affected by each fault current; This represents the setting coefficient for the first stage of protection; This represents the fault current corresponding to the boundary of the expected protection range in section I of the comprehensive fault current curve.

[0120] For stage II, the expected protection range boundary of stage II for the kth protection device is pre-set. (The unit is distance, such as line length or kilometers from the installation point of the protection device), then the II-stage protection setting under the c-th state set can be expressed as:

[0121] in, Indicates the first The first protection device in the The II-stage protection settings under the state set affected by each fault current; Indicates the setting coefficient for stage II protection; This indicates the fault current corresponding to the boundary of the expected protection range of Section II or the coordination point of adjacent lines on the comprehensive fault current curve.

[0122] For Section III, the impact of the maximum line load current and the normal operating output of the distributed power source on the load current at the protection installation location under the fault current influence state set is determined as follows:

[0123] in, Indicates the first The first protection device in the III-stage protection settings under a set of fault current-affected states; This represents the setting coefficient for the third-stage protection; Indicates the first The flow passes through the state set of the nth state. The maximum load current of each protection device.

[0124] The first The first protection device in the The protection setting values ​​for stages I, II, and III, calculated under each fault current influence state set, form the corresponding protection setting set. :

[0125] In a power distribution network, multiple protection devices are usually configured. The above calculation is repeated for all protection devices and their corresponding fault current influence state sets to generate protection setting groups for each protection device.

[0126] This step calculates the protection setting set based on the comprehensive fault current curve, allowing all operating states within the same fault current-affected state set to share the same set of settings. This avoids individual setting for each sample and significantly reduces the amount of offline calculation. Furthermore, since the comprehensive curve represents the common characteristics of all curves within the state set, the settings calculated based on it have good adaptability and operational accuracy within that state set.

[0127] Step 209: Generate a protection setting matrix based on the protection setting group corresponding to each protection device.

[0128] In this embodiment of the invention, the protection setting groups corresponding to all protection devices are summarized to form a protection setting matrix. .

[0129] Furthermore, after the protection setting matrix is ​​calculated, it is stored in the host computer of the protection device or the power distribution automation system for online operation.

[0130] This step centrally stores the protection setting groups corresponding to each protection device as a protection setting matrix, achieving standardization and retrieval of the setting results. During online operation, the target setting group can be found in a constant time simply by using the current protection device number and the matched state set number, ensuring real-time setting switching.

[0131] Step 210: Generate the current fault current curve based on the current operating status information.

[0132] In this embodiment of the invention, during the online operation of the distribution network, current operating status information is collected in real time. This current operating status information includes: line switch status, power switch status, distributed generation output (active power, reactive power, or output current), network topology, and line impedance parameters.

[0133] Specifically, the current operating status information is topologically encoded to construct a current operating status matrix; the fault current of the protection device at each fault location is calculated based on the current operating status matrix; and multiple sample fault current curves corresponding to the protection device are plotted based on the fault current of the protection device. For example: the first... The fault current curve of each protection device under the current operating condition is denoted as: .

[0134] It is worth noting that the coding method is the same as in step 202, the calculation method of the fault current is the same as in step 204, and the curve plotting method is the same as in step 205, so they will not be repeated here.

[0135] Step 211: Based on the current fault current curve and multiple fault current influence state sets, perform similarity calculations to determine the target influence state set.

[0136] In this embodiment of the invention, for the k-th protection device, the offline stage has been obtained Each fault current affects a set of states, and each fault current affects a set of states c (c=1,2,……). (This corresponds to a comprehensive fault current curve) .

[0137] Specifically, the current fault current curve The similarity of each comprehensive fault current curve is compared with that of the other curves, and the set of fault current influence states corresponding to the most similar representative curve is selected as the target influence state set to which the current operating state belongs.

[0138] In this embodiment of the invention, Euclidean distance is used as the similarity measure. Since the fault current amplitudes at different fault locations may differ significantly, a relative distance is used to avoid high-current points dominating the distance calculation. For discrete fault locations... (r is the total number of fault locations), the distance between the current fault current curve and the c-th comprehensive fault current curve. The calculation formula is as follows:

[0139] in, This represents the distance between the current fault current curve and the c-th comprehensive fault current curve; r is the total number of fault location points; Indicates the first One fault location point; Indicates the current fault current curve of the first... Fault current at each fault location point; This indicates the c-th comprehensive fault current curve. The fault current at each fault location.

[0140] It is worth noting that the denominator uses the current value of a representative curve as a normalization reference to eliminate the influence of differences in current magnitude at different locations. When the value is zero or extremely small, a very small positive number can be set as the denominator.

[0141] The distances between the current fault current region and each comprehensive fault current curve are compared, and the fault current influence state set corresponding to the comprehensive fault current curve with the smallest distance is selected as the target influence state set to which the current operating state belongs.

[0142] in, This indicates that the target affects the set of states; This indicates the distance between the current fault current curve and the c-th comprehensive fault current curve.

[0143] This step uses relative Euclidean distance for curve matching, which can effectively eliminate the influence of the difference in current amplitude at different fault locations on similarity judgment, accurately identify the state set to which the current operating state belongs, and provide a reliable basis for subsequently calling the correct protection settings.

[0144] Step 212: Match the target influence state set with the protection setting matrix to generate the target protection setting combination.

[0145] In this embodiment of the invention, after determining the target influence state set, the current protection device k and the target influence state set are used as the basis for determining the target influence state set. Protection setting matrix generated and stored from the offline stage Read the corresponding protection setting group The protection setting group will be read. The target protection setting combination that the protection device k is actually used in under the current operating state is denoted as: Protection constant matrix The system stores the protection settings for each protection device under each fault current influence state set, specifically for stages I, II, and III.

[0146] Furthermore, the target protection setting combination is sent to the corresponding protection device for subsequent fault judgment.

[0147] Furthermore, during operation, changes in the operating status are continuously monitored. When the calculated target impact state set changes, the target protection setting combination corresponding to the new target impact state set is retrieved from the protection setting matrix to achieve adaptive switching of the protection settings; if the target impact state set does not change, that is, the current operating status still belongs to the original target impact state set, the original target protection setting combination remains unchanged to avoid unnecessary setting switching.

[0148] This step enables online adaptive matching and recall of protection settings. When changes in operating status lead to changes in the matching state set, the system automatically switches to the corresponding setting group, ensuring that the protection settings always adapt to the current fault current characteristics. When the state set remains unchanged, the settings are kept constant, avoiding instability caused by frequent switching. This method effectively improves the dynamic adaptability and operational reliability of active distribution network current protection.

[0149] In a specific embodiment of the present invention: Figure 3 A topology diagram of a 10kV distribution network connected to distributed power sources is provided for an embodiment of the present invention, as shown below. Figure 3 As shown, A is the system power access point, supplying power downstream through lines AB, AE, and AH; B, C, D, E, F, G, H, and I are key nodes in the network, connecting different loads and distributed power sources respectively. Loads 1 to 4 are located at nodes D, G, I, and H respectively, representing the user's power load. f 1 and f 2 represents simulated fault points, located near points H and E respectively, used to test the protection operation logic. Feeder AD section is connected to distributed power source 1, and feeder AG section is connected to distributed power sources 2 and 3, representing the supporting role of the distributed power sources in the system. Circuit breaker ( Figure 3Dark squares (in the middle) are placed on each line segment to achieve fault isolation and topology switching. Protection devices are configured on one side near the system power source. The distribution network voltage level is 10 kV, the system impedance is (0.5 + j1.2) Ω / km, and overhead cable JKLGYJ model parameters are used. The impedance per unit length of the line is (0.26 + j0.25) Ω / km. The line lengths of AB and AE sections are 3 km, AH section is 2 km, BC, EF, and HI sections are 4 km, and CD and FG sections are 3 km. A simulation model is constructed that includes distributed generation (DG) operating condition switching and network topology adjustment to simulate the system with changing operating modes. A time-varying simulation scenario is set to simulate the following key operations, including changes in DG output and the closing of line switches. The operating status of all switches and DG output information are recorded.

[0150] Figure 4 A trend graph of the sum of squared errors as a function of the number of clusters is provided as an embodiment of the present invention, such as... Figure 4 As shown, this reflects the change in model fit goodness under different numbers of clusters. The horizontal axis represents the number of clusters L, ranging from 2 to 9 with an interval of 1; the vertical axis represents the sum of squared errors, ranging from 0.1 to 1 with an interval of 0.1. The curve characteristics show that as the number of clusters L increases, the sum of squared errors generally decreases, indicating a gradual improvement in clustering performance; however, when L increases from 4 to 5, a significant inflection point appears in the sum of squared errors (…). Figure 4 (As shown in the underlined circle), the rate of decline slows significantly, indicating that adding new clusters no longer significantly improves model performance, exhibiting a diminishing marginal return effect. While further increasing the number of clusters can continue to reduce the sum of squared errors, the optimization gains gradually weaken, while increasing computational complexity and the risk of overfitting. Therefore, choosing L=5 as the optimal number of clusters achieves the best balance between computational efficiency and model generalization ability, while ensuring good clustering results and data structure expressiveness.

[0151] Clustering uses a nonlinear dimensionality reduction algorithm to map complex data in the original multidimensional feature space to a two-dimensional visualization plane by preserving the local similarity structure between sample points in high-dimensional data. This reveals the inherent nonlinear distribution patterns and potential laws of the data, thus achieving dimensionality reduction visualization of the data. Figure 5 This is a schematic diagram illustrating the clustering results in a high-proportion distributed power supply access processing scenario provided by an embodiment of the present invention, as shown below. Figure 5As shown, the horizontal and vertical axes represent Main Feature 1 and Main Feature 2, respectively, both ranging from -100 to 80, with intervals of 20. Both axes are low-dimensional projected coordinates optimized by a nonlinear dimensionality reduction algorithm, reflecting the relative positional relationship of each data point in the reduced space. The density and separation of the clusters reflect the typicality and distinguishability of various power output states. Different colored points represent the five distributed power source power output state sets divided after clustering: blue represents distributed power source power output state set 1, brown represents distributed power source power output state set 2, orange represents distributed power source power output state set 3, purple represents distributed power source power output state set 4, and green represents distributed power source power output state set 5. Each category corresponds to a group of power output scenarios with similar operating characteristics.

[0152] This invention effectively overcomes the problems of difficulty in modeling and parameter sensitivity in traditional mechanism modeling when dealing with high-dimensional, nonlinear, and strongly coupled systems. It can automatically identify typical output modes under complex operating scenarios, providing data-driven decision support for subsequent distribution network operation optimization, protection setting, and fault response strategy formulation, and significantly improving the intelligence level of operation analysis of systems with a high proportion of distributed power sources.

[0153] By recording the switch information and the output states of different distributed power sources after clustering, iterative calculations are performed offline using programming to obtain the fault current flowing through each protection point. Then, the setting value of each protection is obtained by combining the setting formula. Since the calculation is performed offline, there is no high requirement for speed, and the data processing capability of a general microcomputer is sufficient.

[0154] (1) Protection operation status when the output of distributed power source changes: To test the performance of the proposed method under varying distributed power source output, the calculation and simulation results of protection device 5 are used as an example for verification. Table 1 shows the protection setting values ​​under different distributed power source output state sets: Table 1

[0155] Figure 6 This invention provides a graph showing the change of short-circuit current over time measured at the protection device under different distributed power supply output states and fault types. Figure 6 The data shown corresponds to distributed power source output state set 1. For example... Figure 6 As shown, the horizontal axis represents time t, in seconds (s), ranging from 0.35 to 0.55, with intervals of 0.1; the vertical axis represents fault current, in kiloamperes (kA), ranging from 1.6 to 3.4, with intervals of 0.6. set (I) and I set(II) These represent the protection setting values ​​for Stage I and Stage II, respectively, which are 3.106kA and 1.681kA. The fault scenarios included are three-phase short-circuit faults occurring at 50%, 70%, and 90% of the line's length, marked by the green dashed line I. f50% =3.159kA, red dashed line I f70% =2.864kA and the yellow dashed line I f 90% =2.671kA and a two-phase short-circuit fault occurring at the end of the line, marked as blue dashed line I. f(2) =2.280kA. As can be seen from the curve, all fault currents significantly exceed I. set (I) and increases as the fault point gets closer to the power supply side, verifying that the protection can operate reliably and quickly under the distributed power supply output state set 1.

[0156] Figure 7 This is another graph showing the change of short-circuit current over time measured at the protection device under different distributed power output states and fault types, provided as an embodiment of the present invention. Figure 7 The data shown corresponds to distributed power generation output status set 5. For example... Figure 7 As shown, the horizontal axis represents time t, in seconds (s), ranging from 0.35 to 0.55, with intervals of 0.1; the vertical axis represents fault current, in kiloamperes (kA), ranging from 0.5 to 4.0, with intervals of 0.5. set (I) and I set (II) These represent the protection setting values ​​for Stage I and Stage II, respectively, which are 3.427kA and 1.447kA. The fault scenarios included are three-phase short-circuit faults occurring at 50%, 70%, and 90% of the line's length, marked by the green dashed line I. f50% =3.837kA, red dashed line I f70% =3.378kA and the yellow dashed line I f 90% =3.012kA and a two-phase short-circuit fault occurring at the end of the line, marked as blue dashed line I. f(2) =1.902kA. Due to the different distributed power output levels and distribution, the system's equivalent impedance and power flow direction change under this condition, resulting in changes in the amplitude of each fault current. Figure 6 Differences exist, especially in cases of remote faults and two-phase short circuits. Nevertheless, all fault currents are still higher than Iset(I), indicating that the proposed method can still guarantee the sensitivity and selectivity of protection under distributed power supply output state set 5, thus verifying the robustness and adaptability of the strategy under various typical operating conditions.

[0157] Depend on Figure 7It can be seen that under both typical output conditions, protection device 5 can operate correctly according to the fault location and type. When the fault occurs at 50% of the line, regardless of whether the system is in distributed power supply output state set 1 or distributed power supply output state set 5, the measured three-phase short-circuit fault current exceeds the corresponding I-stage setting values ​​of 3.106 kA and 3.427 kA, respectively, satisfying the operating conditions of instantaneous overcurrent protection. Therefore, the I-stage of protection device 5 can operate reliably and quickly. When the fault location moves to 70% or 90% of the line, although the fault current decreases, it is still higher than the corresponding II-stage setting values ​​of 1.681 kA (distributed power supply output state set 1) and 1.447 kA (distributed power supply output state set 5). At the same time, when a two-phase short-circuit fault occurs at the end of the line, the fault current is also greater than the II-stage setting value. This indicates that under the aforementioned remote fault and asymmetrical fault conditions, the second stage of protection device 5 can be correctly activated and complete fault isolation, verifying the effectiveness and adaptability of the proposed protection setting strategy under different DG output scenarios.

[0158] (2) Protection operation status when the topology changes: To test the performance of the proposed improved method, considering the distributed power supply (DP) connection for the three feeder sets, protection device 8 of feeder 3 was selected to verify the effectiveness of the proposed method. A three-phase short-circuit fault occurred at the outlet of the downstream line where protection device 8 is located. The DP output was in output state set 3 and processing state set 5. The protection operation status is shown in Table 2. Table 2

[0159] In this context, √ indicates that the correct action is protected; × indicates that the incorrect action is protected.

[0160] As shown in Table 2, under various network topology changes (simulated by tripping different combinations of switches, such as tripping switch 2, tripping switch 3, and tripping switches 2 and 5), the traditional fixed setting method repeatedly resulted in erroneous actions (marked as "×") because it did not dynamically adjust with the system operation mode. Especially when tripping switches 3, 6, 7 or their combinations, although the fault current exceeded the traditional I-stage setting of 3.579 kA, the actual short-circuit current was complexly distributed due to the influence of DG output and topology reconfiguration. The traditional method misjudged it as an intra-zone fault, leading to erroneous operation. In contrast, the present invention dynamically adjusts the I-stage setting value based on the output state set identified by clustering (e.g., the setting value is 3.986–4.461 kA under output state set 3 and 3.669–4.636 kA under output state set 5), achieving correct operation in all operating conditions (marked as "√").

[0161] The results show that the present invention can not only effectively avoid the risk of maloperation of traditional protection under topology switching and distributed power output fluctuations, but also maintain high sensitivity and selectivity under different operating conditions, fully demonstrating its strong adaptability to complex operating conditions in distribution networks with a high proportion of distributed power sources.

[0162] Figure 8 This is a schematic diagram illustrating the sensitivity of a protection device 5 under different output levels of a distributed power source, as provided in an embodiment of the present invention. Figure 8 As shown, the horizontal axis represents the distributed generation output state sets, including distributed generation output state sets 1 to 5; the vertical axis represents sensitivity, ranging from 1.2 to 1.5, with intervals of 0.1. The light blue dotted line represents sensitivity level II; the dark blue dotted line represents sensitivity level III. Figure 8 It can be seen that under different distributed power source output state sets, the sensitivity of protection device 5 in stages II and III remains at a high level and shows a stable trend with changes in output state. Specifically, the sensitivity of stage II is generally maintained above 1.38 between output state sets 1 and 4, reaching a maximum of 1.40, and even in output state set 5, it remains above 1.30. The sensitivity of stage III gradually increases from 1.20 in output state set 1 to approximately 1.27 in output state set 4, and slightly decreases in output state set 5 but remains above 1.20. This indicates that the proposed method effectively improves the sensitivity of protection in inverter-type distributed power source access scenarios through a dynamic tuning strategy, especially ensuring sufficient sensitivity margin when the output of distributed power sources fluctuates significantly. Compared to the problem of insufficient sensitivity caused by the influence of distributed power sources in traditional fixed-value methods, this invention can adapt to various operating conditions, significantly enhance the upstream overcurrent protection's ability to detect remote faults, thereby improving the overall reliability and adaptability of the protection system in distribution networks with a high proportion of distributed power sources.

[0163] It is worth noting that the fault assessment and setting optimization scheme proposed in this invention, which combines network topology and distributed power output adaptation, aims to solve the problem of maloperation / failure to operate of traditional protection systems caused by the nonlinearity of fault currents and the variability of topology in distribution networks with a high proportion of distributed power output. As an alternative, the same core objective can also be achieved through communication-based longitudinal protection schemes, transient quantity-based fault protection schemes, multi-agent system (MAS)-based collaborative protection schemes, and model predictive control (MPC)-based protection schemes.

[0164] Among them, the communication-based longitudinal protection scheme shares measurement data through high-speed communication between distribution terminals and uses current differential or directional longitudinal protection principles to determine the fault range. Because it does not rely on steady-state current amplitude settings, its protection performance is largely unaffected by distributed power source output fluctuations and topology changes, and it can reliably identify faults within and outside the protection zone, thus effectively avoiding cascading tripping or failure to operate. Its implementation relies on a highly reliable, low-latency communication network and is suitable for urban distribution networks with good communication infrastructure.

[0165] The fault protection scheme based on transient quantities uses high-frequency components in the transient process after a fault occurs as criteria, extracting transient features through techniques such as wavelet transform to replace the traditional steady-state current setting. Since the transient process is mainly determined by system impedance and line parameters, the output of distributed generation sources and topology changes have a relatively small impact on the transient features. Therefore, this scheme can also address the challenges brought by a high proportion of distributed generation sources, achieving rapid and accurate fault diagnosis. However, this scheme requires high-sampling-rate measurement devices and has high requirements for communication reliability and time delay, making it suitable for industrial power distribution networks with high speed requirements.

[0166] The cooperative protection scheme based on a multi-agent system (MAS) abstracts protection devices in the distribution network into agents. Each agent collects local information and performs distributed collaborative decision-making with neighboring nodes to jointly complete fault location and action determination. Through information interaction and coordination, this scheme can adapt to changes in distributed power output and topology reconfiguration, avoiding protection maloperation or failure to operate, thus achieving the objectives of this invention. This scheme needs to address the communication coordination and decision-making conflicts among agents.

[0167] Model predictive control (MPC) utilizes historical data and meteorological information to construct predictive models such as Long Short-Term Memory (LSTM) networks. This allows for the early prediction of short-term power output trends from distributed generation sources, and, combined with switching operation plans, anticipates topology changes, thereby updating protection settings in advance. This approach proactively adapts to distributed generation fluctuations and topology changes, achieving the invention's objectives. Its performance depends on the accuracy of the predictive model, and it exhibits good applicability in areas with clear distributed generation output patterns and high prediction accuracy.

[0168] It is worth noting that the acquisition, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. The user information in the embodiments of this application was obtained through legal and compliant means, and the acquisition, storage, use, and processing of user information have been authorized and agreed upon by the client.

[0169] It is worth noting that the information collected in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant countries and regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation portals are provided for users to choose to authorize or refuse.

[0170] It is worth noting that the technical solution provided in this application provides users with a corresponding operation entry point, allowing users to choose to agree to or reject the automated decision-making result; if the user chooses to reject, the process will proceed to the expert decision-making process.

[0171] The technical solution of the distribution network current protection setting method based on fault current curves provided in this invention involves calculating fault current curves for the protection device in multiple scenarios based on the distribution network's operating state sample matrix, generating a corresponding short-circuit current matrix; clustering the fault current curves of each sample in the short-circuit current matrix to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set; generating a protection setting matrix based on the comprehensive fault current curve corresponding to each fault current influence state set; and matching the collected current operating state information with the fault current influence state set and the protection setting matrix to generate a target protection setting combination. This method can cluster operating states based on the similarity of fault current curve influences, generating protection settings adapted to different state sets. This enables adaptive matching of protection settings when distributed power output fluctuates and topology changes, effectively improving the accuracy of current protection operation under complex operating conditions, ensuring the coordination and unity of protection selectivity and sensitivity, and reducing the risk of cascading tripping or failure to operate.

[0172] Figure 9 This is a schematic diagram of a distribution network current protection setting device based on a fault current curve, provided in an embodiment of the present invention. This device is used to execute the aforementioned distribution network current protection setting method based on a fault current curve, such as... Figure 9 As shown, the device includes: a multi-scenario fault current curve calculation unit 11, a curve clustering unit 12, a protection setting matrix calculation unit 13, and a target protection setting combination matching unit 14.

[0173] The multi-scenario fault current curve calculation unit 11 is used to calculate the multi-scenario fault current curve of the protection device based on the operating status sample matrix of the distribution network, and generate the corresponding short-circuit current matrix.

[0174] The curve clustering unit 12 is used to cluster the fault current curves of each sample in the short-circuit current matrix to generate multiple fault current influence state sets and the comprehensive fault current curve corresponding to each fault current influence state set.

[0175] The protection setting matrix calculation unit 13 is used to generate the protection setting matrix based on the comprehensive fault current curve corresponding to each fault current influence state set.

[0176] The target protection setting combination matching unit 14 is used to match the collected current operating status information based on the fault current influence state set and the protection setting matrix to generate the target protection setting combination.

[0177] In this embodiment of the invention, the multi-scenario fault current curve calculation unit 11 is specifically used to set multiple fault location points along the topology line within the protection influence domain of the protection device; calculate the fault current of the protection device when each fault location point is faulty according to different topology states and distributed power supply operating states, based on a preset information matrix; and draw multiple sample fault current curves corresponding to the protection device based on the fault current of the protection device to form a short-circuit current matrix.

[0178] In this embodiment of the invention, the curve clustering unit 12 is specifically used to extract features from each sample fault current curve in the short-circuit current matrix and generate an offset feature vector; based on the offset feature vector, the fault current curves of each sample are clustered to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set.

[0179] In this embodiment of the invention, the curve clustering unit 12 is specifically used to select sample fault current curves without distributed power supply access from each sample fault current curve as reference fault current curves; generate maximum relative offset, minimum relative offset, and average relative offset based on the reference fault current curve and the sample fault current curves; match the reference fault current curves based on a preset reference segmented protection range to determine the reference segmented protection setting group; match each sample fault current curve based on the reference segmented protection setting group to determine the sample segmented protection range; generate segmented relative offsets based on the sample segmented protection range and the reference segmented protection range; and generate an offset feature vector based on the segmented relative offset, maximum relative offset, minimum relative offset, and average relative offset.

[0180] In this embodiment of the invention, the protection setting matrix calculation unit 13 is specifically used to calculate the fault current of the comprehensive fault current curve according to the preset segmented protection setting coefficient, generate the protection setting group corresponding to the protection device, and generate the protection setting matrix according to the protection setting group corresponding to each protection device.

[0181] In this embodiment of the invention, the target protection setting combination matching unit 14 is specifically used to generate the current fault current curve based on the current operating status information; perform similarity calculation based on the current fault current curve and multiple fault current influence state sets to determine the target influence state set; and match the target influence state set with the protection setting matrix to generate the target protection setting combination.

[0182] In this embodiment of the invention, based on the operating status sample matrix of the distribution network, the protection device performs multi-scenario fault current curve calculations to generate a corresponding short-circuit current matrix; the fault current curves of each sample in the short-circuit current matrix are clustered to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set; a protection setting matrix is ​​generated based on the comprehensive fault current curve corresponding to each fault current influence state set; based on the fault current influence state set and the protection setting matrix, the collected current operating status information is matched to generate a target protection setting combination. This allows for clustering of operating states based on the similarity of fault current curve influences, generating protection settings adapted to different state sets. This enables adaptive matching of protection settings when distributed power output fluctuates and topology changes, effectively improving the accuracy of current protection operation under complex operating conditions, ensuring the coordination and unity of protection selectivity and sensitivity, and reducing the risk of cascading tripping or failure to operate.

[0183] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer device, specifically, a computer device can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0184] This invention provides a computer device, including a memory and a processor. The memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions. When the program instructions are loaded and executed by the processor, they implement the steps of the above-described embodiment of the distribution network current protection setting method based on fault current curves. For a detailed description, please refer to the above-described embodiment of the distribution network current protection setting method based on fault current curves.

[0185] The following is for reference. Figure 10 It shows a schematic diagram of the structure of a computer device 600 suitable for implementing the embodiments of this application.

[0186] like Figure 10As shown, the computer device 600 includes a central processing unit (CPU) 601, which can perform various appropriate tasks and processes based on programs stored in read-only memory (ROM) 602 or programs loaded from storage section 608 into random access memory (RAM) 603. The RAM 603 also stores various programs and data required for the operation of the computer device 600. The CPU 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0187] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal feedback (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed in storage section 608 as needed.

[0188] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program including program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611.

[0189] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0190] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0191] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0192] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0193] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0194] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0195] The acquisition, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.

[0196] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0197] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0198] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0199] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0200] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for setting distribution network current protection based on fault current curves, characterized in that, The method includes: Based on the operating status sample matrix of the distribution network, the protection device is used to calculate the fault current curves for multiple scenarios and generate the corresponding short-circuit current matrix. Cluster the fault current curves of each sample in the short-circuit current matrix to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set. Generate a protection setting matrix based on the comprehensive fault current curve corresponding to each fault current influence state set; Based on the fault current influence state set and the protection setting matrix, the collected current operating state information is matched to generate a target protection setting combination.

2. The method for setting distribution network current protection based on fault current curves according to claim 1, characterized in that, The step of calculating multi-scenario fault current curves for the protection device based on the operating status sample matrix of the distribution network to generate a corresponding short-circuit current matrix includes: Multiple fault location points are set along the topology line within the protection influence area of ​​the protection device; According to different topology states and distributed power supply operating states, the fault current of the protection device is calculated based on the preset information matrix when a fault occurs at each fault location point. Based on the fault current of the protection device, multiple sample fault current curves corresponding to the protection device are plotted to form a short-circuit current matrix.

3. The method for setting distribution network current protection based on fault current curves according to claim 1, characterized in that, The step of clustering the fault current curves of each sample in the short-circuit current matrix to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set includes: Feature extraction is performed on the fault current curves of each sample in the short-circuit current matrix to generate an offset feature vector; Based on the offset feature vector, the fault current curves of each sample are clustered to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set.

4. The method for setting distribution network current protection based on fault current curves according to claim 3, characterized in that, The step of extracting features from the fault current curves of each sample in the short-circuit current matrix to generate an offset feature vector includes: The fault current curves without distributed power supply access were selected from the fault current curves of each sample as the benchmark fault current curves. Based on the reference fault current curve and the sample fault current curve, the maximum relative offset, minimum relative offset and average relative offset are generated. Based on the preset reference segmented protection range, the reference fault current curve is matched to determine the reference segmented protection setting group. Based on the aforementioned benchmark segmented protection setting group, the fault current curves of each sample are matched to determine the segmented protection range of the sample. Based on the sample segment protection range and the benchmark segment protection range, the segment relative offset is generated; An offset feature vector is generated based on the segmented relative offset, the maximum relative offset, the minimum relative offset, and the average relative offset.

5. The method for setting distribution network current protection based on fault current curves according to claim 1, characterized in that, The step of generating a protection setting matrix based on the comprehensive fault current curve corresponding to each fault current influence state set includes: Based on the preset segmented protection setting coefficients, the fault current of the comprehensive fault current curve is calculated to generate the protection setting group corresponding to the protection device. A protection setting matrix is ​​generated based on the protection setting group corresponding to each protection device.

6. The method for setting distribution network current protection based on fault current curves according to claim 1, characterized in that, The step of matching the collected current operating status information with the fault current influence state set and the protection setting matrix to generate a target protection setting combination includes: Generate the current fault current curve based on the current operating status information; Based on the current fault current curve and multiple fault current influence state sets, a similarity calculation is performed to determine the target influence state set; The target influence state set is matched with the protection setting matrix to generate the target protection setting combination.

7. A distribution network current protection setting device based on fault current curves, characterized in that, The device includes: The multi-scenario fault current curve calculation unit is used to calculate the multi-scenario fault current curve of the protection device based on the operating status sample matrix of the distribution network, and generate the corresponding short-circuit current matrix. The curve clustering unit is used to cluster the fault current curves of each sample in the short-circuit current matrix to generate multiple fault current influence state sets and a comprehensive fault current curve corresponding to each fault current influence state set. The protection setting matrix calculation unit is used to generate the protection setting matrix based on the comprehensive fault current curve corresponding to each fault current influence state set. The target protection setting combination matching unit is used to match the collected current operating status information based on the fault current influence state set and the protection setting matrix to generate a target protection setting combination.

8. A computer-readable medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the distribution network current protection setting method based on fault current curves as described in any one of claims 1 to 6.

9. A computer device comprising a memory and a processor, the memory for storing information including program instructions, and the processor for controlling the execution of the program instructions, characterized in that, When the program instructions are loaded and executed by the processor, the distribution network current protection setting method based on the fault current curve as described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the distribution network current protection setting method based on the fault current curve as described in any one of claims 1 to 6.

Images