Electric leakage test method for mining combined low-voltage feed system

Abstract

The invention relates to the technical field of general measurement of electrical variables, in particular to an electric leakage test method for a mining combined low-voltage feed system, and the method comprises the steps: carrying out the slicing operation of a data sequence collected in a preset time period, and obtaining a plurality of subsequences corresponding to a slicing window; performing feature extraction operation on the subsequences corresponding to the target loops in the running state to obtain system features of the mining combined low-voltage feed system in the slice window; determining a target scene according to a matching result of the system feature of the slice window and a plurality of preset scene features; based on the target scene, performing theoretical capacitance current separation operation on zero-sequence current contained in the sub-sequence of the loop to obtain residual current of each sampling point in the sub-sequence of the loop; and according to the change trend of the residual current in the sub-sequence of the loop, determining an electric leakage test result of the mining combined low-voltage feed system in the slice window corresponding to the sub-sequence.

Description

Technical Field

[0001] This application relates to the field of electrical variable measurement technology, and in particular to a leakage current testing method for a combined low-voltage power supply system for mining applications. Background Technology

[0002] The combined low-voltage power supply system for mines is the core equipment for underground power supply in coal mines. This system typically employs a multi-circuit parallel power supply method to simultaneously power various loads such as tunneling machines, coal mining machines, conveyors, and pumping stations. The unique characteristics of the underground working environment determine the highly dynamic nature of this system's operation: the continuous advancement of the working face can extend the power supply distance from hundreds of meters to thousands of meters, involving frequent cable extensions and retrievals; the rapid switching of load types at different operational stages involves alternating between heavy-load tunneling and light-load maintenance; and the load distribution of multiple circuits changes in real-time with production scheduling, exhibiting significant time-varying and uncertainties in the system.

[0003] Existing leakage current detection technologies are primarily designed for power grids with fixed topologies. These technologies identify leakage faults by establishing a steady-state capacitive current benchmark. However, in mine-use combined power supply systems, normal cable length variations themselves cause significant changes in the ground capacitive current. The amplitude of this capacitive current fluctuation caused by topology changes (treating cable length changes as topology changes) is often on the same order of magnitude as the leakage current, making it difficult for traditional detection methods based on fixed benchmark comparisons to distinguish between the two. Therefore, a better leakage current testing scheme for mine-use combined low-voltage power supply systems is needed to overcome these difficulties. Summary of the Invention

[0004] This application provides a leakage current testing method for a mine-use combined low-voltage power supply system, which improves the accuracy of leakage current testing for the mine-use combined low-voltage power supply system.

[0005] In a first aspect, this application provides a method for testing leakage current in a mine-use combined low-voltage power supply system, the method comprising: The data sequence collected within a preset time period is sliced ​​to obtain multiple sub-sequences corresponding to the slicing window. The data sequence includes the data sequence of the loop in the mining combined low-voltage power supply system. Feature extraction is performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window; Based on the matching results between the system characteristics of the slice window and multiple preset scene characteristics, the target scene to which the slice window belongs is determined; Based on the target scenario, a theoretical capacitance current separation operation is performed on the zero-sequence current contained in the subsequence of the circuit to obtain the residual current at each sampling point in the subsequence of the circuit. Based on the variation trend of residual current in the subsequence of the circuit, the leakage current test result of the mining combined low-voltage power supply system in the slice window corresponding to the subsequence is determined.

[0006] Furthermore, the system features include the correlation between topology and zero-sequence current. The feature extraction operation performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: Differential processing operations are performed on the cable length and zero-sequence current contained in the sub-sequence of the target loop to obtain the first differential sequence corresponding to the cable length and the second differential sequence corresponding to the zero-sequence current. A correlation measurement operation is performed on the first difference sequence and the second difference sequence of the same slice window to obtain the correlation degree of the target loop in the slice window; The correlation degree of the mine-use combined low-voltage power supply system in the slice window is determined based on the correlation degree of each target circuit in the same slice window.

[0007] Furthermore, the system features include the coordination coefficient between the correlation degrees of the various loops. The feature extraction operation performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: The correlation degree of each target circuit in the mining combined low-voltage power supply system within the same slice window is statistically analyzed by probability distribution to obtain the distribution results; Based on the distribution results, a synergy measurement operation is performed on the correlation degree of different target circuits to obtain the synergy coefficient of the mine-use combined low-voltage power supply system in the slice window.

[0008] Furthermore, the system features include the residual rate characterizing the contribution of the remaining component in the zero-sequence current. The feature extraction operation performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: A linear regression fitting operation is performed on the zero-sequence current contained in the subsequence of the target loop to obtain a fitting function, and the current corresponding to the sampling point in the fitting function is determined as the correlation component related to the topological change. The associated component is removed from the zero-sequence current contained in the subsequence of the target loop to obtain the residual component that is independent of the topology change, and the proportion of the residual component in the zero-sequence current is used as the residual rate. The residual rate of the mine-use combined low-voltage power supply system in the slice window is determined based on the residual rate of each target circuit at each sampling point in the slice window.

[0009] Furthermore, the process of determining the scene features includes: Slicing multiple sample data sequences yields multiple sub-sample sequences corresponding to the slice window, wherein the sample data sequences include the sample data sequences of the loop; Feature extraction is performed on the sub-sample sequences corresponding to each target circuit to obtain the sample system features of the mine-use combined low-voltage power supply system in the slice window; Based on the characteristics of the sample system, the sub-sample sequences are clustered to obtain multiple clusters; Based on the scene to which the labeled sub-sample sequences in the cluster belong, the scene to which the cluster belongs is determined, and based on the sample system characteristics of each sub-sample sequence in the cluster, the scene characteristics corresponding to the cluster are determined.

[0010] Furthermore, when the target scenario is a single-loop operation scenario, the step of performing theoretical capacitance current separation on the zero-sequence current contained in the sub-sequence of the loop based on the target scenario to obtain the residual current at each sampling point in the sub-sequence of the loop includes: Based on the cable length of each sampling point in the subsequence corresponding to the target circuit, determine the theoretical capacitance current of each sampling point in the subsequence corresponding to the target circuit; Using the theoretical capacitance current, a theoretical capacitance current separation operation is performed on the zero-sequence current of each sampling point in the subsequence corresponding to the target circuit to obtain the residual current of each sampling point in the subsequence corresponding to the target circuit.

[0011] Furthermore, determining the leakage current test result of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence based on the variation trend of the residual current in the sub-sequence of the circuit includes: Based on the cable length of each sampling point in the subsequence corresponding to the target loop, determine the cable length variation data corresponding to the target loop; Based on the data related to cable stress changes at each sampling point in the subsequence corresponding to the target loop, determine the stress change data of the cable corresponding to the target loop; The leakage current test results of the target circuit are determined based on the length change data and / or stress change data of the target circuit.

[0012] Furthermore, when the target scenario is a multi-circuit alternating operation scenario, determining the leakage current test result of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence based on the changing trend of the residual current in the sub-sequence of the circuit includes: The theoretical capacitance current of each sampling point in the subsequence corresponding to each target circuit is superimposed to obtain the total theoretical capacitance current of the mine combined low-voltage power supply system. The residual current of each sampling point in the subsequence corresponding to each target circuit is superimposed to obtain the total residual current of the mine-use combined low-voltage power supply system. The total theoretical capacitance current and the total residual current are superimposed to obtain the reconfigured total zero-sequence current of the mine-use combined low-voltage power supply system. Based on the deviation between the total zero-sequence current and the reconstructed total zero-sequence current, the leakage current test result of the mine-use combined low-voltage power supply system is determined, where the total zero-sequence current is the zero-sequence current at each sampling point in the data sequence of the mine-use combined low-voltage power supply system.

[0013] Furthermore, after determining that the leakage current test result of the mine-use combined low-voltage power supply system indicates the presence of a leakage current fault, the method further includes: In the case where only one of the target circuits in the mining combined low-voltage power supply system exhibits an increasing trend in residual current, it is determined whether the residual current of the target circuit is greater than a first preset value. If the residual current of the target circuit is greater than the first preset value, determine whether the growth trend of the residual current of the target circuit is synchronized with the growth trend of the total residual current. If the growth trend is synchronized, determine that the leakage test result of the target circuit is a leakage fault. In the case where the residual current of multiple target circuits in the mining combined low-voltage power supply system shows an increasing trend, the average value of the residual current of each target circuit is determined. Determine the ratio of the average residual current of the target circuit to the total residual current, and identify the target circuit corresponding to the largest ratio as the primary leakage circuit, and the target circuits corresponding to other ratios as secondary leakage circuits.

[0014] Furthermore, when the target scenario is a shutdown scenario, the step of performing theoretical capacitance current separation on the zero-sequence current contained in the sub-sequence of the circuit based on the target scenario to obtain the residual current at each sampling point in the sub-sequence of the circuit includes: The zero-sequence current contained in the subsequence of the circuit is decomposed in the frequency domain to obtain the first current containing the low-frequency component; The first current is correlated with environmental parameters to obtain a correlation coefficient. If the correlation coefficient is greater than a second preset value, the first current is taken as the theoretical capacitance current, and the part of the zero-sequence current other than the first current is taken as the residual current. The step of determining the leakage current test result of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence based on the variation trend of the residual current in the sub-sequence of the circuit includes: If the absolute value of the residual current in the subsequence of the circuit is greater than a third preset value and the fluctuation amplitude is less than a fourth preset value, the fluctuation amplitude of the environmental parameter in the time interval corresponding to the subsequence is determined. If the fluctuation range of the environmental parameters is less than the fifth preset value, the leakage test result of the circuit is determined to be that leakage exists.

[0015] Secondly, this application provides a leakage current testing device for a mine-use combined low-voltage power supply system, the device comprising: The slicing module is used to slice the data sequence collected within a preset time period to obtain multiple sub-sequences corresponding to the slicing window. The data sequence includes the data sequence of the loop in the mine combined low-voltage power supply system. The feature module is used to perform feature extraction operations on the sub-sequences corresponding to each target loop in the running state to obtain the system features of the mine-use combined low-voltage power supply system in the slice window; The matching module is used to determine the target scene to which the slice window belongs based on the matching results between the system characteristics of the slice window and multiple preset scene characteristics; The separation module is used to perform theoretical capacitance current separation operation on the zero-sequence current contained in the subsequence of the circuit based on the target scenario, so as to obtain the residual current of each sampling point in the subsequence of the circuit. The testing module is used to determine the leakage current test results of the mining combined low-voltage power supply system in the slice window corresponding to the subsequence based on the changing trend of the residual current in the subsequence of the circuit.

[0016] Thirdly, embodiments of this application also provide a leakage current testing system for a mine-use combined low-voltage power supply system. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the methods described above.

[0017] Fourthly, this application provides a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the above-described method. The computer-readable storage medium may be volatile or non-volatile.

[0018] Fifthly, this application provides an electronic device, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above-described method when executing instructions stored in the memory.

[0019] Sixthly, this application provides a computer program product including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above-described method.

[0020] In the embodiments of this specification, after feature extraction of the sub-sequences obtained from slicing, the target scene to which the slice window corresponding to the sub-feature belongs is determined based on the matching of the obtained system features and scene features. Then, the theoretical capacitance current in the zero-sequence current of the sub-sequence is separated based on the target scene. Finally, the leakage current test result is determined based on the changing trend of the separated residual current. Based on the coupling characteristics of topology dynamics and leakage current signals in mine combined power supply systems, this invention improves the accuracy of leakage current test result judgment under topology dynamic changes through scene matching operation and theoretical capacitance current separation operation matched with the target scene. This effectively solves the problem of misjudgment caused by the similarity between capacitance current fluctuation and leakage current fluctuation in traditional methods, and improves the practicality and reliability of the leakage current protection system using this leakage current test method in mine combined power supply scenarios. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0022] Figure 1 The diagram above illustrates a flow chart of a leakage current testing method for a mine-use combined low-voltage power supply system. Figure 2 The diagram above illustrates a flow chart of a leakage current testing device for a combined low-voltage power supply system used in mining.

[0023] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0024] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0025] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0026] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0027] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0028] Figure 1 A flowchart illustrating a leakage current testing method for a mine-use combined low-voltage power supply system according to an embodiment of this disclosure is provided. This method can be applied to a leakage current testing device for a mine-use combined low-voltage power supply system. The leakage current testing device for the mine-use combined low-voltage power supply system can be a terminal device, a server, or other processing equipment. The terminal device can be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, vehicle-mounted device, wearable device, etc.

[0029] In some possible implementations, the leakage current testing method for the mine-use combined low-voltage power supply system can be implemented by the processor calling computer-readable instructions stored in memory.

[0030] like Figure 1 As shown, the leakage current testing method for the mine-use combined low-voltage power supply system may include: Step S11: Slice the data sequence collected within a preset time period to obtain multiple sub-sequences corresponding to the slicing window.

[0031] The data sequence may include the data sequences of each circuit in the mine-use combined low-voltage power supply system, and may also include the data sequence of the mine-use combined low-voltage power supply system. At each sampling point of the data sequence of each circuit, data such as zero-sequence current, cable length, switch status, and load current in the circuit can be collected; at each sampling point of the data sequence of the mine-use combined low-voltage power supply system, the total zero-sequence current of the entire mine-use combined low-voltage power supply system can be collected.

[0032] At the outgoing end of each feeder circuit, a zero-sequence current transformer can be installed to collect the zero-sequence current signal of each circuit in real time. Specifically, the zero-sequence current transformer can adopt a through-core structure, with all three phase conductors passing through the transformer core simultaneously. When the system is operating normally, the vector sum of the three-phase currents is zero, and the zero-sequence current transformer has no output; when a single-phase ground fault or insulation degradation occurs, the zero-sequence current transformer outputs the zero-sequence current component. The acquisition module of the mine-use combined low-voltage feeder system can synchronously sample the zero-sequence current of each circuit according to a fixed sampling period, forming multi-channel time-series data.

[0033] In a mining combined low-voltage power supply system, a zero-sequence current transformer can be installed on the neutral grounding wire of the power source or generator to collect the total zero-sequence current of the mining combined low-voltage power supply system in real time.

[0034] An encoder can be used to install a cable length sensor on the cable reel to obtain the real-time cable length of each loop. Specifically, the cable length sensor can record the cumulative rotation angle of the reel. By multiplying the reel radius by the rotation angle, the actual length of cable unwound from the cable reel can be obtained.

[0035] In each power supply circuit, a power supply switch and a current transformer can be installed. The status of the auxiliary contacts of the power supply switch can be used to determine the on / off status of each circuit (i.e., the switch status); the current transformer can be used to collect the effective value of the three-phase operating current (i.e., the load current) of each circuit, thereby determining the load operating status of the circuit.

[0036] By aligning and storing the zero-sequence current, cable length, switch status, load current, etc. in the above data according to a unified timestamp, a data sequence of the loop can be established; by aligning and storing the total zero-sequence current, etc. in the above data according to a unified timestamp, a data sequence of the mining combined low-voltage power supply system can be established.

[0037] For a data sequence, it is possible to slice the data into a series of overlapping (e.g., overlapping for 9 minutes) fixed-length slice windows W, using a fixed duration T (e.g., 10 minutes) as the length of a slice window.k Where k=1,2,3…, each slice window W k Both the data sequence and the subsequence contain a subsequence of length T. Clearly, the same sampling points in the data sequence and the subsequence contain the same data.

[0038] All the above data must be standardized when performing time-series curve fitting and projection in this specification to eliminate dimensional differences.

[0039] Step S12: Perform feature extraction on the sub-sequences corresponding to each target circuit in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window.

[0040] Among them, the target circuit is the circuit in the mine-use combined low-voltage power supply system that is in operation. The state of the circuit can be determined from the switching state data of the circuit in the sub-sequence. The state includes non-operational state (shutdown or standby) and operational state (neither shut down nor standby).

[0041] As mentioned above, the data sequences of each circuit contain a wider variety of data types. Therefore, feature extraction operations can be performed on the subsequences corresponding to each target circuit. The extracted circuit features can be processed (such as integration) to obtain the system features of the mining combined low-voltage power supply system in the slice window corresponding to the subsequence.

[0042] This specification does not specifically limit the feature types and categories extracted from the subsequences; please refer to the subsequent embodiments, which will not be elaborated here. The system features of a mine-use combined low-voltage power supply system can have multiple categories. In the presence of multiple system features, coupling operations such as splicing can be used to obtain coupled system features, which can then be used for subsequent matching operations with scene features.

[0043] Step S13: Determine the target scene to which the slice window belongs based on the matching results between the system features of the slice window and multiple preset scene features.

[0044] Specifically, multiple scenarios can be set, each with corresponding scenario features. The method for extracting features from subsequences can be the same as the method for extracting scenario features. In one possible implementation, the process of determining the scenario features includes: Slicing multiple sample data sequences yields multiple sub-sample sequences corresponding to the slice window, wherein the sample data sequences include the sample data sequences of the loop; Feature extraction is performed on the sub-sample sequences corresponding to each target circuit to obtain the sample system features of the mine-use combined low-voltage power supply system in the slice window; Based on the characteristics of the sample system, the sub-sample sequences are clustered to obtain multiple clusters; Based on the scene to which the labeled sub-sample sequences in the cluster belong, the scene to which the cluster belongs is determined, and based on the sample system characteristics of each sub-sample sequence in the cluster, the scene characteristics corresponding to the cluster are determined.

[0045] The time length of the sample data sequence can be a preset time period, and multiple sample event sequences constitute a historical dataset. Furthermore, similar to the aforementioned data sequences, the sample data sequence can also include sample data sequences of each circuit in the mine-use combined low-voltage power supply system, and can also include sample data sequences of the entire mine-use combined low-voltage power supply system. Each sampling point in the circuit's sample data sequence can include data such as timestamps, cable lengths of each circuit, load currents of each circuit, and switch statuses of each circuit; the sample data sequence of the mine-use combined low-voltage power supply system can include data such as timestamps and the system's total zero-sequence current.

[0046] Operations such as slicing the sample data sequence and extracting features from sub-sample sequences can be performed in the same way as those for the data sequence, and will not be repeated here. Unlike the data sequence, maintenance records can be used to annotate portions of some sample data sequences. These annotations can include leakage events or operational scenarios. These operational scenarios can include single-circuit operations (such as only the tunneling machine advancing slowly), multi-circuit alternating operations (such as alternating operations of the coal mining machine and the conveyor), and shutdowns (such as pump stations shut down while energized).

[0047] Based on the characteristics of the sample system, clustering operations can be performed on each sub-sample sequence. Specifically, this clustering operation can employ the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to perform unsupervised clustering of the sub-sample sequences across all slice windows. In the clustering algorithm, algorithm parameters (neighborhood radius, minimum number of samples, etc.) can be determined through grid search combined with silhouette coefficients to obtain a stable result with the highest silhouette score (at least 4 clusters). The clustering distance between sub-sample sequences can be the Euclidean distance between the feature vectors of the sample system.

[0048] Furthermore, since the sample data sequences are labeled, cluster information can be determined based on the labeled sub-sample sequences within the obtained clusters. Specifically, the scene of the labeled sub-sample sequences within a cluster can be used as the scene to which the cluster belongs. Further, the sample system features of each sub-sample sequence within the cluster can be averaged (this can be a weighted average, with weights related to distance from the cluster center, smaller weights for farther clusters and larger weights for closer clusters) to obtain the scene features of the cluster. The scene features corresponding to this cluster can serve as the "fingerprint center" of its respective scene.

[0049] After clustering, it can be observed that the clusters for the three scenarios exhibit significant differences: (1) The clustering of the slow-progressing scenario is characterized by: ① high coupling correlation (≥0.8); ② low cooperation coefficient entropy (≤0.3); ③ low residual rate (≤5%); ④ strong temporal stability of the feature vectors, that is, the coefficient of variation (ratio of standard deviation to mean) of the three dimensions (coupling correlation, cooperation coefficient entropy, residual rate) of all feature vectors in M ​​consecutive slice windows (e.g., M=10) is <0.15. This strong temporal stability can reflect the strong correlation and low interference characteristics between the slow change of the single-loop topology and the current response.

[0050] (2) The clustering of multi-loop alternating operation scenarios is characterized by: ① discrete coupling correlation (0.2-0.7); ② high synergy coefficient entropy (≥0.6); ③ medium to low residual rate (5%-12%); ④ frequent temporal fluctuations of feature vectors, that is, the coefficient of variation (ratio of standard deviation to mean) of all three dimensions (coupling correlation, synergy coefficient entropy, residual rate) of all feature vectors in M ​​consecutive windows (e.g., M=10) is >0.5. This frequent temporal fluctuation reflects the disorder of correlation patterns caused by asynchronous changes in multi-loop topology.

[0051] (3) The clustering of residual current scenarios after shutdown is characterized by: ① extremely low coupling correlation (≤0.2); ② low cooperation coefficient entropy (≤0.4); ③ medium to high residual rate (12%-30%); ④ and the topology change rate is close to zero (the sum of the cable length changes of all loops in this window is <10 meters). This topology change rate is close to zero, reflecting the characteristic that non-topological factors (environmental drift, leakage current) dominate the current fluctuation when there is no topology change.

[0052] In the above process, clustering operations can be used to obtain multiple clusters with significant differences, which can effectively distinguish the boundary conditions of multiple scenarios and improve the accuracy of matching the scene features and system features corresponding to the clusters.

[0053] Traditional detection methods based on fixed benchmark values ​​often fail in leakage current testing of combined low-voltage power supply systems in mines because they confuse capacitive current fluctuations caused by topological changes with actual leakage current fluctuations, and both have amplitudes of the same order of magnitude. For example, a topological change such as cable extension can lead to an increase in capacitance to ground, which in turn increases the capacitive zero-sequence current. Similarly, an increase in resistive leakage current caused by insulation degradation also manifests as an increase in zero-sequence current.

[0054] The scenarios corresponding to the three clusters mentioned above are typical examples of situations where the fluctuations in capacitive current caused by topological changes due to changes in cable length are confused with the fluctuations in actual leakage current.

[0055] Specifically, in slow-advance scenarios, the cable extends at an extremely low speed as the tunneling operation progresses, corresponding to a gradual change in capacitive current. The rate of change of this current is comparable to the rate of increase in leakage current caused by the slow deterioration of insulation. At the same time, their time-domain characteristics highly overlap, making it difficult to distinguish between the two.

[0056] In multi-circuit alternating operation scenarios, the cable lengths of each circuit change asynchronously (extension, stabilization, and recycling coexist), and the total zero-sequence current of the system exhibits complex fluctuations. At this time, if a leakage occurs in a stable circuit, its signal is easily overwhelmed by the background current generated by the topology changes of other circuits.

[0057] In the residual current scenario after shutdown, when the load is off but the cable is energized, the topology is stable. However, at this time, the capacitive current drift caused by changes in ambient temperature and humidity is difficult to distinguish numerically from the small leakage current, and there is a lack of effective dynamic reference for identification.

[0058] Therefore, in the process of matching system features and scene features, the matching operation can be performed in the three scenarios mentioned above. This matching operation can be to match the real-time system feature vector with each scene fingerprint (i.e., scene features), calculate the Mahalanobis distance between the real-time system features and each fingerprint center, and select the scene fingerprint with the smallest Mahalanobis distance. Its label is the preliminary recognition result of the current system features.

[0059] Specifically, when the matching result indicates that the system features of the slice window do not fully meet all the judgment conditions of the above three scenarios, the scenario that is closest among the three scenarios can be classified as the target scenario; when the matching result indicates that the system features of the slice window do not fully meet all the judgment conditions of the above three scenarios, and there are more than two instances where the system feature does not match the cluster cluster performance of the most matching scenario (refer to the four cluster cluster performances corresponding to each of the above three scenarios), the subsequence corresponding to the system feature is classified as an unconfused sequence.

[0060] This unambiguous sequence is used to indicate that the scene corresponding to the slice window does not belong to the scene where the known capacitance current fluctuation and the actual leakage current fluctuation are confused (i.e., the three scenarios mentioned above). Therefore, after the subsequence is divided into an unambiguous sequence, the subsequent theoretical capacitance current separation operation does not need to be performed. That is, the traditional leakage fault detection based on alarm threshold can be directly used, or other types of leakage fault detection strategies can be determined according to the actual situation. This specification does not limit this.

[0061] Step S14: Based on the target scenario, perform theoretical capacitance current separation operation on the zero-sequence current contained in the subsequence of the circuit to obtain the residual current of each sampling point in the subsequence of the circuit.

[0062] Step S15: Based on the variation trend of the residual current in the subsequence of the circuit, determine the leakage current test result of the mining combined low-voltage power supply system in the slice window corresponding to the subsequence.

[0063] The theoretical capacitive current is calculated based on the objective properties of the circuit (such as cable length). The residual current is the current remaining after subtracting the theoretical capacitive current from the zero-sequence current in the circuit. The leakage current test results can indicate whether there are leakage circuits in the mine-use combined low-voltage power supply system, and also whether there are leakage issues in each circuit of the mine-use combined low-voltage power supply system.

[0064] The fundamental difference between the capacitive current fluctuation caused by the topological change of cable length and the actual leakage current fluctuation is that the change of capacitive current has a deterministic linear correspondence with the change of cable length, and this change is reversible. When the cable is recycled, the capacitive current will decrease synchronously. However, the change of leakage current is not directly related to the cable length. It is caused by the irreversible deterioration of the insulation material. Even if the cable length remains unchanged or is shortened, the leakage current will continue to exist or even continue to increase.

[0065] The confusion between capacitive current and actual leakage current varies across different scenarios. For example, when the work area advances, causing the cable to lengthen (i.e., a topological change occurs), the capacitance to ground of a single-loop system increases, and the capacitive current naturally rises; when the cable is shortened during retraction, the capacitive current decreases accordingly. This change in capacitive current caused by topological change exhibits a gradual characteristic, and its time-domain evolution is highly similar to the leakage current growth process caused by slow insulation degradation. The two are difficult to distinguish in terms of macroscopic characteristics such as the rate of change of amplitude and duration. In multi-loop parallel systems, the changes in cable length in different loops often occur asynchronously, causing the total zero-sequence current of the system to exhibit complex non-stationary fluctuation characteristics.

[0066] Therefore, different theoretical capacitance current separation operations can be used for different target scenarios to obtain residual currents. The leakage current test results can then be determined based on the residual currents, enabling the implementation of targeted hierarchical separation strategies. Specifically, after calculating the theoretical capacitance current, it can be separated from the zero-sequence current of each circuit to obtain the residual current, which indirectly reflects the leakage current state. By analyzing the relevant data of the residual current (such as its changing trend), the leakage current test results of the corresponding slice window of the sub-sequence can be determined. For example, if the residual current continues to increase even when the cable length corresponding to a circuit decreases, it can be determined that leakage exists in that circuit.

[0067] In traditional leakage current testing methods, setting a high alarm threshold to avoid false alarms due to topology changes reduces the sensitivity to actual leakage current; conversely, lowering the threshold increases sensitivity, leading to numerous false alarms due to frequent topology changes. This detection dilemma caused by the dynamic nature of the power supply topology severely restricts the practicality and reliability of leakage current protection systems in mining combined power supply scenarios.

[0068] In the embodiments of this specification, after feature extraction of the sub-sequences obtained from slicing, the target scene to which the slice window corresponding to the sub-feature belongs is determined based on the matching of the obtained system features and scene features. Then, the theoretical capacitance current in the zero-sequence current of the sub-sequence is separated based on the target scene. Finally, the leakage current test result is determined based on the changing trend of the separated residual current. Based on the coupling characteristics of topology dynamics and leakage current signals in mine combined power supply systems, this invention improves the accuracy of leakage current test result judgment under topology dynamic changes through scene matching operation and theoretical capacitance current separation operation matched with the target scene. This effectively solves the problem of misjudgment caused by the similarity between capacitance current fluctuation and leakage current fluctuation in traditional methods, and improves the practicality and reliability of the leakage current protection system using this leakage current test method in mine combined power supply scenarios.

[0069] As mentioned earlier, there can be multiple categories of system features, and the change in leakage current is not directly related to the change in cable length (i.e., topology change). Clearly, the greater the component of leakage current in the zero-sequence current, the weaker the correlation between the zero-sequence current and the topology. Therefore, the correlation between the topology (mainly cable length in this specification) and the zero-sequence current can be considered a system feature. In one possible implementation, the system feature includes the correlation between the topology and the zero-sequence current. The feature extraction operation performed on the sub-sequences corresponding to each target circuit in the operating state to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: Differential processing operations are performed on the cable length and zero-sequence current contained in the sub-sequence of the target loop to obtain the first differential sequence corresponding to the cable length and the second differential sequence corresponding to the zero-sequence current. A correlation measurement operation is performed on the first difference sequence and the second difference sequence of the same slice window to obtain the correlation degree of the target loop in the slice window; in one embodiment of the present invention, the correlation degree of the slice window is calculated by taking the Pearson correlation coefficient of the first difference sequence and the second difference sequence as the correlation degree of the target loop in the slice window.

[0070] The correlation degree of the mine-use combined low-voltage power supply system within the same slice window is determined based on the correlation degree of each target circuit in the same slice window. In this embodiment of the invention, one method for calculating the correlation degree of the mine-use combined low-voltage power supply system within a slice window is to use the average correlation degree of each target circuit within the slice window as the correlation degree of the mine-use combined low-voltage power supply system within the slice window.

[0071] The differential processing operation can be a first-order or second-order differential processing operation. The first differential sequence is the sequence formed by the differential result of the cable length data, and the second differential sequence is the sequence formed by the differential result of the zero-sequence current. The window size of the first and second differential sequences is the same as that of the aforementioned subsequences.

[0072] In one example, for the i-th circuit in a mine-use combined low-voltage power supply system, we can first calculate its corresponding subsequence W in the slice window. k Cable length sequence L within i The first-order difference sequence ΔL of (t) i (i.e., the first difference sequence), and calculate its corresponding subsequence W in the slice window. k The zero-sequence current signal I inside i The first-order difference sequence ΔI of (t) i (i.e., the second difference sequence).

[0073] After obtaining ΔL i and ΔI i Then, ΔL corresponding to the same slice window can be calculated. i and ΔI i The Pearson correlation coefficient p(k,i) represents the degree of linear correlation between the change in cable length and the change in zero-sequence current within the slice window.

[0074] Furthermore, for N loops in the "running" state (neither stopped nor standby), the subsequence W corresponding to the slice window can be calculated. k The average correlation within the slice window is used to obtain the subsequence W corresponding to the slice window. k The corresponding topology-current binary coupling correlation degree can be used as a system feature vector.

[0075] Since residual current is a strong indicator of leakage current, its existence affects the correlation between the topology and the zero-sequence current. In the above process, incorporating the correlation between the topology and the zero-sequence current into the determination of system characteristics can improve the quality of system characteristics and thus improve the accuracy of leakage current test results determined based on system characteristics.

[0076] During the operation of a mine-use combined low-voltage power supply system, fluctuations in system state and inconsistencies in the state of each circuit can affect the correlation of subsequences of different circuits at different time periods. To improve the richness of system features, derived features related to correlation can be added to the system features. In one possible implementation, the system features include the coordination coefficient between the correlations of the circuits. The feature extraction operation performed on the subsequences corresponding to each target circuit in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: The correlation degree of each target circuit in the mining combined low-voltage power supply system within the same slice window is statistically analyzed by probability distribution to obtain the distribution results; Based on the distribution results, a synergy measurement operation is performed on the correlation degree of different target circuits to obtain the synergy coefficient of the mine-use combined low-voltage power supply system in the slice window.

[0077] Specifically, subsequence W can be collected. k The correlation values ​​of all running loops within the corresponding slice window are {p(k,1), p(k,2)...p(k,N)}. These correlation values ​​are divided into several equally wide numerical intervals (the number of intervals matches the number of running loops, which can be set to N), and the probability distribution of the p-values ​​in each numerical interval is statistically analyzed as the distribution result. By calculating the Shannon entropy corresponding to this distribution result, the cooperative consistency of the multi-loop topology-current response is quantified. This Shannon value can be called the cooperative coefficient of multi-loop coupling.

[0078] Clearly, the lower the Shannon entropy value, the stronger the synergy, and the more uniform the pattern of topological influence on each loop; the higher the Shannon entropy value, the weaker the synergy, and the greater the difference in the correlation between topological changes and current responses of each loop.

[0079] The above process quantifies the quality of correlation by incorporating the synergy coefficient into the calculation of system features, thereby improving both the richness and quality of system features.

[0080] In one possible implementation, the system features include a residual rate characterizing the contribution of the remaining component in the zero-sequence current. The feature extraction operation performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: A linear regression fitting operation is performed on the zero-sequence current contained in the subsequence of the target loop to obtain a fitting function, and the current corresponding to the sampling point in the fitting function is determined as the correlation component related to the topological change. The associated component is removed from the zero-sequence current contained in the subsequence of the target loop to obtain the residual component that is independent of the topology change, and the proportion of the residual component in the zero-sequence current is used as the residual rate. The residual rate of the mine-use combined low-voltage power supply system in the slice window is determined based on the residual rate of each target circuit at each sampling point in the slice window.

[0081] Among them, the correlated component is the component of the zero-sequence current that is related to topology changes, and the remaining component is the component of the zero-sequence current data that is not related to the correlated component.

[0082] Specifically, the zero-sequence currents in the same subsequence can be fitted by linear regression. The fitting result can be a linear fitting function (a straight line function), which represents the part of the zero-sequence current that can be directly explained by topological dynamic changes. Therefore, the current corresponding to the sampling point in the fitting function can be used as the correlation component.

[0083] Furthermore, the residual component can be obtained by subtracting the associated component corresponding to the zero-sequence current from the zero-sequence current at each sampling point. After obtaining the residual component, the proportion of the residual component in the zero-sequence current can be directly used as the residual rate, which can characterize the contribution of non-topological factors to the fluctuation of the zero-sequence current.

[0084] The above process quantifies the contribution of non-topological factors to zero-sequence current fluctuations by incorporating the residual rate into the system characteristics, thereby improving both the richness and quality of the system characteristics.

[0085] After simultaneously determining the correlation degree, synergy coefficient, and residual rate, a three-dimensional coupled feature vector of "correlation degree - synergy coefficient - residual rate" can be formed for each slice window, serving as the core input for scene recognition. After identifying the target scene, a hierarchical separation strategy can be specifically constructed based on different scenes to achieve the separation of theoretical capacitive current in zero-sequence current. In one possible implementation, when the target scene is a single-loop operation scenario, the theoretical capacitive current separation operation is performed on the zero-sequence current contained in the sub-sequence of the loop based on the target scene, obtaining the residual current at each sampling point in the sub-sequence of the loop, including: Based on the cable length of each sampling point in the subsequence corresponding to the target circuit, determine the theoretical capacitance current of each sampling point in the subsequence corresponding to the target circuit; Using the theoretical capacitance current, a theoretical capacitance current separation operation is performed on the zero-sequence current of each sampling point in the subsequence corresponding to the target circuit to obtain the residual current of each sampling point in the subsequence corresponding to the target circuit.

[0086] The single-circuit operation scenario indicates that only one circuit in the mining combined low-voltage power supply system is in operation, and the equipment in that circuit (such as a tunneling machine) is in a slow-moving state.

[0087] In this scenario, the cable length of the circuit gradually changes, resulting in a superposition of capacitive current and resistive leakage current caused by the slow degradation of insulation. Furthermore, the growth rates and temporal trends of the capacitive and leakage currents are highly coupled. Therefore, for this scenario, the core of feature extraction is "zero-sequence current-cable length correlation residual time series analysis," and the core of leakage current detection lies in separating the interpretable capacitive current from the unexplainable residual current.

[0088] Under ideal, fault-free conditions, the zero-sequence current change caused by the slow change in cable length (corresponding to the slow advancement or retraction of equipment) should be purely capacitive. Therefore, the theoretical capacitive current can be determined based on the synchronously acquired cable length time-series data.

[0089] Specifically, the capacitance to ground per unit cable length C is pre-defined according to the cable type. l (pF / km) and cable phase voltage U φ (V) can be used to calculate the time-series data of the theoretical capacitive current caused by changes in cable length, i.e., the real-time cable length L. i (t) and capacitance to ground per unit length C l The product of these two values ​​yields the real-time total capacitance to ground C(t) of the cable. Substituting this into the standard formula, the real-time theoretical capacitance current I can be obtained. C (t)=ω×C(t)×U φ , where ω is the angular frequency (the power system frequency is fixed at f=50Hz, and the angular frequency = 2πf=314rad / s).

[0090] By subtracting the theoretical capacitance current time series data from the actual zero-sequence current time series data at each sampling point, the residual current sequence can be obtained. If there is no leakage fault in the system, the residual current should be close to zero and show no obvious increasing trend, containing only a small amount of measurement noise; conversely, if there is insulation deterioration leakage, the average slope of the residual current sequence within the slice window is positive and shows an overall increasing trend.

[0091] Since there are various reasons for changes in theoretical capacitive current in single-circuit operation scenarios, such as changes in cable length and cable stress, after obtaining the residual current, the leakage current test results can be refined based on the specific residual current change trend, thereby improving the accuracy of the leakage current test results. In one possible implementation, determining the leakage current test results of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence of the circuit based on the change trend of the residual current in the sub-sequence includes: Based on the cable length of each sampling point in the subsequence corresponding to the target loop, determine the cable length variation data corresponding to the target loop; Based on the data related to cable stress changes at each sampling point in the subsequence corresponding to the target loop, determine the stress change data of the cable corresponding to the target loop; The leakage current test results of the target circuit are determined based on the length change data and / or stress change data of the target circuit.

[0092] Specifically, the length variation data and / or stress variation data can be determined first, and then the leakage current test results can be determined based on the obtained data: (1) If the cable length stops extending or begins to be retracted, the theoretical capacitive current will stop increasing or decrease synchronously. If the residual current continues to increase and does not adjust in the opposite direction with the change of cable length, the leakage test result is determined to be leakage.

[0093] Therefore, the turning point t, where the cable length stops increasing or begins to shorten, can be identified within the subsequence. r Calculate the average slope k of the residual current in the time periods before and after the inflection point, respectively. tr1 k tr2 If the cable length stops increasing, k tr2 If the residual trend is still significantly greater than zero (greater than 0.3 is considered significant), then the residual trend is determined to be irreversible, a strong indication of leakage, and the leakage test result can be determined to indicate the presence of leakage. It should be understood that the slope in this specification uses a normalized slope value.

[0094] (2) Combined with the cable mechanical stress change period (i.e. the period of frequent dragging and bending, which can be identified by the sudden change of the drum rotation angular velocity, the sudden change threshold can be set to the historical average angular acceleration, and the period exceeding the threshold is the stress change period), if the residual current suddenly increases during the period (the growth slope is greater than the average slope of the rising segment of the residual current in the window, then it is judged as a sudden increase), and the growth trend continues after the mechanical stress disappears, it indicates that the mechanical stress accelerates the insulation deterioration and causes leakage, and the leakage test result can be judged to be that there is leakage.

[0095] (3) If the average slope of the residual current sequence of zero-sequence current in the window is positive (representing an increase) and satisfies any of the above conditions (the increase is irreversible or strongly correlated with mechanical stress events), then the leakage test result can be determined to be a leakage, and the circuit can be marked as a suspected fault circuit.

[0096] In the above process, the accuracy of judging the leakage test results when the topology changes dynamically is improved by separating the theoretical capacitive current in the zero-sequence current and determining the leakage test results in the single-loop operation scenario.

[0097] In one possible implementation, when the target scenario is a multi-circuit alternating operation scenario, determining the leakage current test result of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence based on the variation trend of the residual current in the sub-sequence of the circuit includes: The theoretical capacitance current of each sampling point in the subsequence corresponding to each target circuit is superimposed to obtain the total theoretical capacitance current of the mine combined low-voltage power supply system. The residual current of each sampling point in the subsequence corresponding to each target circuit is superimposed to obtain the total residual current of the mine-use combined low-voltage power supply system. The total theoretical capacitance current and the total residual current are superimposed to obtain the reconfigured total zero-sequence current of the mine-use combined low-voltage power supply system. Based on the deviation between the total zero-sequence current and the reconstructed total zero-sequence current, the leakage current test result of the mine-use combined low-voltage power supply system is determined, where the total zero-sequence current is the zero-sequence current at each sampling point in the data sequence of the mine-use combined low-voltage power supply system.

[0098] The scenario of multiple circuits operating alternately is used to indicate that multiple circuits in a mining combined low-voltage power supply system are in operation.

[0099] At this point, referring to the calculation method for residual current in a single-loop operation scenario described above, the residual current of each loop in the mine-use combined low-voltage power supply system can be obtained. Then, the total zero-sequence current measured in real-time by the mine-use combined low-voltage power supply system can be compared with the sum of the residual currents of multiple loops. Based on the deviation, it can be determined whether leakage current exists in the mine-use combined low-voltage power supply system. This deviation can be the difference between the total zero-sequence current and the reconstructed total zero-sequence current.

[0100] Specifically, for each circuit, its actual zero-sequence current can be subtracted from its theoretical capacitive current (i.e., theoretical capacitor current) to obtain the residual current of each circuit; by vector superimposing the residual currents of all circuits in the mine-use combined low-voltage power supply system, the total residual current of the mine-use combined low-voltage power supply system can be obtained; at the same time, the theoretical capacitive currents of all circuits can be vector superimposed to obtain the total theoretical capacitive current of the system.

[0101] Furthermore, the total theoretical capacitive current and the total residual current can be superimposed to obtain the reconstructed total zero-sequence current. The reconstructed total zero-sequence current is then compared with the actual collected total zero-sequence current. If the reconstruction deviation is significant (calculate the average reconstruction deviation within the entire subsequence; if it is greater than or equal to 15 mA, it is considered a significant deviation) and shows a trend of increasing (perform linear fitting on the deviation time-series curve to obtain the slope and goodness of fit; if the slope is greater than or equal to 0.5 mA per minute and the goodness of fit is greater than or equal to 0.6, it is considered a significant trend), it indicates that there is leakage current contributing to the mine-use combined low-voltage power supply system. Conversely, if the reconstruction deviation is extremely small and shows no trend change, it indicates that the residual current in each circuit is merely measurement noise, and there is no leakage fault in the mine-use combined low-voltage power supply system.

[0102] In this scenario, asynchronous changes in the topology of multiple loops lead to disordered superposition of the total zero-sequence current vector. After determining that a leakage fault exists in the system, the leakage loop can be further identified to achieve accurate location of the leakage and improve the intelligence of the leakage test method for mine-use combined low-voltage power supply systems. However, in this scenario, the leakage signal of a single loop can be masked by capacitive current fluctuations in other loops. Therefore, in this scenario, the core of accurate leakage loop identification lies in locating the leakage loop by decomposing the contributions of each loop. In one possible implementation, after determining that the leakage test result of the mine-use combined low-voltage power supply system indicates the presence of a leakage fault, the method further includes; In the case where only one of the target circuits in the mine-use combined low-voltage power supply system exhibits an increasing trend in residual current, it is determined whether the residual current of the target circuit is greater than a first preset value. If the residual current of the target circuit is greater than the first preset value, determine whether the growth trend of the residual current of the target circuit is synchronized with the growth trend of the total residual current. If the growth trend is synchronized, determine that the leakage test result of the target circuit is a leakage fault. In the case where the residual current of multiple target circuits in the mining combined low-voltage power supply system shows an increasing trend, the average value of the residual current of each target circuit is determined. Determine the ratio of the average residual current of the target circuit to the total residual current, and identify the target circuit corresponding to the largest ratio as the primary leakage circuit, and the target circuits corresponding to other ratios as secondary leakage circuits.

[0103] Specifically, when performing time-series analysis on the residual current of each circuit, if the residual current of a certain circuit (i.e., the target circuit) is continuously non-zero (i.e., greater than the first preset value, which is 0) and shows an increasing trend, and this trend is completely synchronized with the growth trend of the total residual current, and the residual current of other circuits does not show an obvious trend change, then it is determined that there is a leakage fault in that circuit.

[0104] Whether the growth trends of the residual current of the target circuit and the total residual current are synchronized can be measured by the Pearson correlation coefficient between the residual current of the target circuit and the total residual current. If the correlation coefficient is greater than 0.85, it is determined to be highly synchronized.

[0105] If multiple target circuits simultaneously exhibit an increase in residual current, the primary and secondary leakage circuits can be identified by comparing the contribution ratio of each target circuit's residual current to the total residual current (the ratio of its average residual current to the average total residual current). Specifically, the ratio of the average residual current of a target circuit to the total residual current can be used as the contribution ratio of that target circuit to the total residual current. When the ratio of a particular target circuit is the largest, that target circuit can be identified as the primary leakage circuit, and the remaining target circuits as secondary leakage circuits.

[0106] In the above process, the leakage current test results under the multi-circuit alternating operation scenario are determined by the deviation between the total zero-sequence current and the reconstructed total zero-sequence current, which improves the accuracy of the leakage current test results judgment when the topology changes dynamically in this scenario.

[0107] In one possible implementation, when the target scenario is a shutdown scenario, the step of performing theoretical capacitance current separation on the zero-sequence current contained in the sub-sequence of the circuit based on the target scenario to obtain the residual current at each sampling point in the sub-sequence of the circuit includes: The zero-sequence current contained in the subsequence of the circuit is decomposed in the frequency domain to obtain the first current containing the low-frequency component; The first current is correlated with environmental parameters to obtain a correlation coefficient. If the correlation coefficient is greater than a second preset value, the first current is taken as the theoretical capacitance current, and the part of the zero-sequence current other than the first current is taken as the residual current. The step of determining the leakage current test result of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence based on the variation trend of the residual current in the sub-sequence of the circuit includes: If the absolute value of the residual current in the subsequence of the circuit is greater than a third preset value and the fluctuation amplitude is less than a fourth preset value, the fluctuation amplitude of the environmental parameter in the time interval corresponding to the subsequence is determined. If the fluctuation range of the environmental parameters is less than the fifth preset value, the leakage test result of the circuit is determined to be that leakage exists.

[0108] Among them, the shutdown scenario is used to indicate that the mine-use combined low-voltage power supply system is in a no-circuit operating state, but the collected data sequence shows that the cables in the mine-use combined low-voltage power supply system are energized.

[0109] In this scenario, the topology is stable (cable length remains unchanged), but changes in ambient temperature and humidity can cause capacitive current drift. Both this capacitive current and the small leakage current exhibit small fluctuations in zero-sequence current under no-load conditions, making them numerically indistinguishable and lacking an effective dynamic reference for leakage current identification.

[0110] In one example, leakage current can be detected by separating periodic drift (i.e., current drift caused by environmental changes, which is periodic due to the stable changes in the downhole environment) from non-periodic leakage current signals. Specifically, Fourier transform can be used to perform frequency domain decomposition on the zero-sequence current signal to separate the high-frequency and low-frequency components of the fluctuating signal, and the current signal corresponding to the low-frequency component is called the first current.

[0111] Since the low-frequency components of the zero-sequence current signal are mostly related to environmental changes, while the high-frequency components contain potential noise and leakage signals, the correlation analysis can be performed between the separated first current and the time-series data of downhole environmental parameters (such as humidity and temperature). By averaging multiple correlation values ​​corresponding to the same subsequence, the correlation coefficient of the subsequence can be obtained. If the correlation coefficient is greater than 0.7 (i.e., the second preset value), it is confirmed that the first current is a capacitive current drift caused by the environment, and it is removed from the zero-sequence current of the loop to obtain the residual fluctuation component, which is the residual current.

[0112] Next, the leakage test results can be determined based on the stability and trend of the residual fluctuation component (i.e., residual current): if the residual current approaches zero (e.g., ±0.01mA, 0.01 is the third preset value) and there is no obvious upward, downward or sudden trend, it is judged as measurement noise and there is no leakage fault; if the residual component is a stable non-zero value (the value fluctuation range within 15 minutes is ≤5%, 5% is the fourth preset value) and continues to exist after the environmental parameters stabilize (fluctuation range ≤5%, 5% is the fifth preset value), it indicates that there is a stable resistive leakage current, and it is judged that there is a leakage fault in the mine combined low-voltage power supply system.

[0113] In the above process, after separating the first current through frequency domain decomposition, the stability and trend line of the residual current other than the first current are used to determine the leakage current test results, which improves the accuracy of the leakage current test results when the topology is stable in the shutdown scenario.

[0114] After confirming a leakage fault, the system using the leakage test method can immediately output a fault signal. This can alert staff to promptly investigate potential insulation problems or enable precise tripping of the faulty circuit via the power supply switch control module, preventing the fault from spreading. Simultaneously, it can record synchronous datasets (zero-sequence current, cable length, load status, environmental parameters, etc.) at the moment the fault occurs, providing data support for subsequent insulation degradation analysis and fault tracing.

[0115] The present invention also provides a leakage current testing device for a combined low-voltage power supply system for mining applications. Figure 2 This diagram illustrates a leakage current testing device for a mine-use combined low-voltage power supply system according to an embodiment of this specification. The leakage current testing device for this mine-use combined low-voltage power supply system can be a terminal device, a server, or other processing equipment. The terminal device can be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, vehicle-mounted device, wearable device, etc.

[0116] In some possible implementations, the leakage current testing device for the mine-use combined low-voltage power supply system can be implemented by a processor calling computer-readable instructions stored in memory.

[0117] like Figure 2 As shown, the leakage current testing device 20 for the mine-use combined low-voltage power supply system may include: Slicing module 21 is used to slice the data sequence collected within a preset time period to obtain multiple sub-sequences corresponding to the slicing window. The data sequence includes the data sequence of the loop in the mining combined low-voltage power supply system. Feature module 22 is used to perform feature extraction operations on the sub-sequences corresponding to each target loop in the running state to obtain the system features of the mine combined low-voltage power supply system in the slice window; The matching module 23 is used to determine the target scene to which the slice window belongs based on the matching results between the system features of the slice window and multiple preset scene features; The separation module 24 is used to perform theoretical capacitance current separation operation on the zero-sequence current contained in the sub-sequence of the circuit based on the target scenario, so as to obtain the residual current of each sampling point in the sub-sequence of the circuit. Test module 25 is used to determine the leakage current test result of the mining combined low-voltage power supply system in the slice window corresponding to the subsequence based on the changing trend of the residual current in the subsequence of the circuit.

[0118] Thirdly, embodiments of this application also provide a leakage current testing system for a mine-use combined low-voltage power supply system. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the methods described above.

[0119] This invention is now complete.

[0120] In summary, in the embodiments of this specification, after feature extraction of the sub-sequences obtained from slicing, the target scene to which the slice window corresponding to the sub-feature belongs is determined based on the matching of the obtained system features and scene features. Then, the theoretical capacitance current in the zero-sequence current of the sub-sequence is separated based on the target scene. Finally, the leakage current test result is determined based on the changing trend of the separated residual current. Based on the coupling characteristics of topology dynamics and leakage current signals in the mine combined power supply system, this invention improves the accuracy of leakage current test result judgment under topology dynamic changes through scene matching operation and theoretical capacitance current separation operation matched with the target scene. This effectively solves the problem of misjudgment caused by the similarity between capacitance current fluctuation and leakage current fluctuation in traditional methods, and improves the practicality and reliability of the leakage current protection system using this leakage current test method in the mine combined power supply scenario.

[0121] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for testing leakage current in a mine-use combined low-voltage power supply system, characterized in that, The method includes: The data sequence collected within a preset time period is sliced ​​to obtain multiple sub-sequences corresponding to the slicing window. The data sequence includes the data sequence of the loop in the mining combined low-voltage power supply system. Feature extraction is performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window; Based on the matching results between the system characteristics of the slice window and multiple preset scene characteristics, the target scene to which the slice window belongs is determined; Based on the target scenario, a theoretical capacitance current separation operation is performed on the zero-sequence current contained in the subsequence of the circuit to obtain the residual current at each sampling point in the subsequence of the circuit. Based on the variation trend of residual current in the subsequence of the circuit, the leakage current test result of the mining combined low-voltage power supply system in the slice window corresponding to the subsequence is determined.

2. The leakage current testing method for a mine-use combined low-voltage power supply system according to claim 1, characterized in that, The system features include the correlation between topology and zero-sequence current. The feature extraction operation performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: Differential processing operations are performed on the cable length and zero-sequence current contained in the sub-sequence of the target loop to obtain the first differential sequence corresponding to the cable length and the second differential sequence corresponding to the zero-sequence current. A correlation measurement operation is performed on the first difference sequence and the second difference sequence of the same slice window to obtain the correlation degree of the target loop in the slice window; The correlation degree of the mine-use combined low-voltage power supply system in the slice window is determined based on the correlation degree of each target circuit in the same slice window.

3. The leakage current testing method for a mine-use combined low-voltage power supply system according to claim 2, characterized in that, The system features include the coordination coefficient between the correlation degrees of the various loops. The feature extraction operation performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: The correlation degree of each target circuit in the mining combined low-voltage power supply system within the same slice window is statistically analyzed by probability distribution to obtain the distribution results; Based on the distribution results, a synergy measurement operation is performed on the correlation degree of different target circuits to obtain the synergy coefficient of the mine-use combined low-voltage power supply system in the slice window.

4. The leakage current testing method for a combined low-voltage power supply system for mining applications according to claim 2, characterized in that, The system features include the residual rate, which characterizes the contribution of the remaining component in the zero-sequence current. The feature extraction operation performed on the sub-sequences corresponding to each target loop in operation to obtain the system features of the mine-use combined low-voltage power supply system in the slice window includes: A linear regression fitting operation is performed on the zero-sequence current contained in the subsequence of the target loop to obtain a fitting function, and the current corresponding to the sampling point in the fitting function is determined as the correlation component related to the topological change. The associated component is removed from the zero-sequence current contained in the subsequence of the target loop to obtain the residual component that is independent of the topology change, and the proportion of the residual component in the zero-sequence current is used as the residual rate. The residual rate of the mine-use combined low-voltage power supply system in the slice window is determined based on the residual rate of each target circuit at each sampling point in the slice window.

5. The leakage current testing method for a mine-use combined low-voltage power supply system according to claim 1, characterized in that, The process of determining the scene features includes: Slicing multiple sample data sequences yields multiple sub-sample sequences corresponding to the sliced ​​window, wherein the sample data sequences include the sample data sequences of the loop; Feature extraction is performed on the sub-sample sequences corresponding to each target circuit to obtain the sample system features of the mine-use combined low-voltage power supply system in the slice window; Based on the characteristics of the sample system, the sub-sample sequences are clustered to obtain multiple clusters; Based on the scene to which the labeled sub-sample sequences in the cluster belong, the scene to which the cluster belongs is determined, and based on the sample system characteristics of each sub-sample sequence in the cluster, the scene characteristics corresponding to the cluster are determined.

6. The leakage current testing method for a mine-use combined low-voltage power supply system according to claim 1, characterized in that, When the target scenario is a single-loop operation scenario, the step of performing theoretical capacitance current separation on the zero-sequence current contained in the sub-sequence of the loop based on the target scenario to obtain the residual current at each sampling point in the sub-sequence of the loop includes: Based on the cable length of each sampling point in the subsequence corresponding to the target circuit, determine the theoretical capacitance current of each sampling point in the subsequence corresponding to the target circuit; Using the theoretical capacitance current, a theoretical capacitance current separation operation is performed on the zero-sequence current of each sampling point in the subsequence corresponding to the target circuit to obtain the residual current of each sampling point in the subsequence corresponding to the target circuit.

7. The leakage current testing method for a mine-use combined low-voltage power supply system according to claim 6, characterized in that, The step of determining the leakage current test result of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence based on the variation trend of the residual current in the sub-sequence of the circuit includes: Based on the cable length of each sampling point in the subsequence corresponding to the target loop, determine the cable length variation data corresponding to the target loop; Based on the data related to cable stress changes at each sampling point in the subsequence corresponding to the target loop, determine the stress change data of the cable corresponding to the target loop; The leakage current test results of the target circuit are determined based on the length change data and / or stress change data of the target circuit.

8. The leakage current testing method for a mine-use combined low-voltage power supply system according to claim 1, characterized in that, In the case where the target scenario is a multi-circuit alternating operation scenario, determining the leakage current test result of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence based on the variation trend of the residual current in the sub-sequence of the circuit includes: The theoretical capacitance current of each sampling point in the subsequence corresponding to each target circuit is superimposed to obtain the total theoretical capacitance current of the mine combined low-voltage power supply system. The residual current of each sampling point in the subsequence corresponding to each target circuit is superimposed to obtain the total residual current of the mine-use combined low-voltage power supply system. The total theoretical capacitance current and the total residual current are superimposed to obtain the reconfigured total zero-sequence current of the mine-use combined low-voltage power supply system. Based on the deviation between the total zero-sequence current and the reconstructed total zero-sequence current, the leakage current test result of the mine-use combined low-voltage power supply system is determined, where the total zero-sequence current is the zero-sequence current at each sampling point in the data sequence of the mine-use combined low-voltage power supply system.

9. The leakage current testing method for a mine-use combined low-voltage power supply system according to claim 8, characterized in that, After determining that the leakage current test result of the mining combined low-voltage power supply system indicates the presence of a leakage current fault, the method further includes: In the case where only one of the target circuits in the mining combined low-voltage power supply system exhibits an increasing trend in residual current, it is determined whether the residual current of the target circuit is greater than a first preset value. If the residual current of the target circuit is greater than the first preset value, determine whether the growth trend of the residual current of the target circuit is synchronized with the growth trend of the total residual current. If the growth trend is synchronized, determine that the leakage test result of the target circuit is a leakage fault. In the case where the residual current of multiple target circuits in the mining combined low-voltage power supply system shows an increasing trend, the average value of the residual current of each target circuit is determined. Determine the ratio of the average residual current of the target circuit to the total residual current, and identify the target circuit corresponding to the largest ratio as the primary leakage circuit, and the target circuits corresponding to other ratios as secondary leakage circuits.

10. The leakage current testing method for a mine-use combined low-voltage power supply system according to claim 1, characterized in that, When the target scenario is a shutdown scenario, the step of performing theoretical capacitance current separation on the zero-sequence current contained in the sub-sequence of the circuit based on the target scenario to obtain the residual current at each sampling point in the sub-sequence of the circuit includes: The zero-sequence current contained in the subsequence of the circuit is decomposed in the frequency domain to obtain the first current containing the low-frequency component; The first current is correlated with environmental parameters to obtain a correlation coefficient. If the correlation coefficient is greater than a second preset value, the first current is taken as the theoretical capacitance current, and the part of the zero-sequence current other than the first current is taken as the residual current. The step of determining the leakage current test result of the mine-use combined low-voltage power supply system in the slice window corresponding to the sub-sequence based on the variation trend of the residual current in the sub-sequence of the circuit includes: If the absolute value of the residual current in the subsequence of the circuit is greater than a third preset value and the fluctuation amplitude is less than a fourth preset value, the fluctuation amplitude of the environmental parameter in the time interval corresponding to the subsequence is determined. If the fluctuation range of the environmental parameters is less than the fifth preset value, the leakage test result of the circuit is determined to be that leakage exists.

Images