A dry-type transformer partial discharge on-line monitoring and positioning system

Abstract

This invention provides an online monitoring and location system for partial discharge in dry-type transformers, relating to the field of transformer technology. Based on the spatial distribution of the high-voltage winding, low-voltage winding, and core in a dry-type transformer, the system divides the transformer interior into multiple discharge structure regions and establishes a corresponding signal propagation reference path length range for each region. By setting multiple partial discharge signal sensing units at different external installation locations, the system synchronously collects and uniformly corrects the electromagnetic pulse signals generated by partial discharge, and generates candidate discharge events based on this. The system further combines the arrival sequence and pulse arrival time difference characteristics of the electromagnetic pulses in the candidate discharge events with the propagation path constraints corresponding to the discharge structure regions, filtering out discharge events that do not meet the structural propagation conditions. When a unique matching condition is met, the system determines the region where the partial discharge occurs, thereby achieving online monitoring and location of partial discharge in dry-type transformers.

Description

Technical Field

[0001] This invention relates to the field of transformer technology, and in particular to an online monitoring and location system for partial discharge in dry-type transformers. Background Technology

[0002] Existing online monitoring technologies for partial discharge in dry-type transformers typically involve placing current transformers (HFCTs), ultra-high frequency sensors, or acoustic emission sensors on the transformer body or bushings to collect pulse signals generated by partial discharges and transmit these signals to a centralized monitoring host. The monitoring host identifies partial discharges based on amplitude, phase, or time difference of arrival from multiple sensors and, to some extent, infers the discharge location, thus achieving online status monitoring.

[0003] In practical applications of dry-type transformer substations, the aforementioned technologies often employ single-sided or limited fixed-position sensors, such as installing HFCTs only on the incoming line side. When the transformer has a multi-winding parallel structure or complex electromagnetic coupling between the core and windings, partial discharge signals are prone to reflection and superposition during propagation. The arrival time of the pulses received by the monitoring host no longer corresponds one-to-one with the actual discharge point, potentially misjudging the discharge on the low-voltage side winding as an insulation defect on the high-voltage side, thus affecting the accuracy of maintenance decisions. Summary of the Invention

[0004] The purpose of this invention is to provide an online monitoring and location system for partial discharge in dry-type transformers, aiming to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] A dry-type transformer partial discharge online monitoring and location system, the system comprising:

[0007] The discharge structure path definition module is used to divide the interior of the dry-type transformer into multiple discharge structure regions based on the spatial distribution relationship of the high-voltage winding, low-voltage winding and core in the dry-type transformer, and to determine the corresponding signal propagation reference path length range for each discharge structure region, thereby generating discharge structure path constraint information.

[0008] The differentiated measurement point signal acquisition module is used to set up partial discharge signal sensing units at multiple different external installation positions corresponding to the discharge structure path constraint information, and simultaneously acquire multiple original electromagnetic pulse signal data generated by partial discharge.

[0009] The multi-channel signal unified correction module is used to unify the time base and correct the amplitude scale of multiple raw electromagnetic pulse signal data to generate corrected electromagnetic pulse signal data.

[0010] The associated discharge event construction module is used to generate the time correlation interval of the combined electromagnetic pulse based on the correction electromagnetic pulse signal data, and to combine multiple electromagnetic pulses with the same polarity change direction to form candidate discharge events.

[0011] The structure path matching calculation module is used to calculate the pulse arrival time difference of each local discharge signal sensing unit for each candidate discharge event, and compare the time difference with the discharge structure path constraint information to generate path matching results.

[0012] The interference event discrimination and elimination module is used to eliminate a candidate discharge event when no path matching result that meets the path matching conditions is generated in any discharge structure region.

[0013] The target discharge region determination module is used to determine the discharge structure region as the region where partial discharge occurs when a candidate discharge event corresponds to only a single discharge structure region and generates a path matching result that satisfies the path matching condition.

[0014] Preferably, the discharge structure path definition module includes:

[0015] The structural relationship analysis submodule is used to analyze the relative spatial positions between structural units based on the radial covering relationship and axial stacking relationship of the high voltage winding, low voltage winding and iron core in the dry-type transformer, and generate structural relative position description information.

[0016] The structural region generation submodule is used to divide the inside of the transformer into multiple non-overlapping discharge structural regions based on the structural relative position description information, and generate a corresponding structural region identifier for each discharge structural region.

[0017] The direct path filtering submodule is used to determine the direct signal propagation path from each discharge structure area to the external installation location without passing through other discharge structure areas, based on the structure area identifier, and generate a set of direct paths;

[0018] The path length limiting submodule is used to limit the geometric length of each direct propagation path in the set of direct paths, and generate the corresponding signal propagation reference path length range;

[0019] The path constraint formation submodule is used to associate the length range of the signal propagation reference path with the corresponding structural region identifier to form discharge structure path constraint information.

[0020] Preferably, the associated discharge event construction module includes:

[0021] The pulse interval statistics submodule is used to statistically analyze the time intervals between adjacent electromagnetic pulses in the correction electromagnetic pulse signal data and generate pulse time interval distribution results.

[0022] The associated window generation submodule is used to determine the time association interval for combining electromagnetic pulses based on the pulse time interval distribution results.

[0023] The polarity consistency screening submodule is used to compare the polarity change direction of multiple electromagnetic pulses within the same time correlation interval based on the time correlation interval, and generate a set of polarity consistent pulses.

[0024] The quantity determination submodule is used to determine whether electromagnetic pulses from a preset number of different partial discharge signal sensing units are included based on the identifier of the corresponding partial discharge signal sensing unit in the set of pulses with consistent polarity, and to generate a quantity determination result.

[0025] The event result generation submodule is used to combine the corresponding electromagnetic pulses into candidate discharge events when the quantity determination result is true, and to record the identifiers of the partial discharge signal sensing units participating in the combination.

[0026] Preferably, the structure path matching calculation module includes:

[0027] The arrival order generation submodule is used to generate the electromagnetic pulse arrival order result based on the time sequence of electromagnetic pulses received by each partial discharge signal sensing unit in the candidate discharge event.

[0028] The structure sequence comparison submodule is used to compare the electromagnetic pulse arrival sequence with the direct propagation path sequence corresponding to each discharge structure region, and generate a set of discharge structure regions that match the sequence.

[0029] The time difference interval formation submodule is used to form the pulse arrival time difference interval between each partial discharge signal sensing unit based on the electromagnetic pulse arrival sequence.

[0030] The path matching determination submodule is used to compare the pulse arrival time difference interval with the signal propagation reference path length range corresponding to the discharge structure region with the correct sequence, and generate a path matching determination result.

[0031] Preferably, the direct path filtering submodule includes:

[0032] The path segmentation result generation unit is used to divide the signal propagation path into multiple continuous spatial path segments based on the spatial connection relationship between the discharge structure area and the external installation position, and generate path segmentation results.

[0033] The path segment coverage determination unit is used to determine whether each spatial path segment overlaps with the spatial range corresponding to other discharge structure regions based on the path segmentation results, and to generate path segment coverage determination results.

[0034] The non-direct path exclusion unit is used to identify and exclude the corresponding signal propagation path as a non-direct propagation path when any path segment overlaps with other discharge structure regions, based on the path segment coverage determination result.

[0035] The direct path confirmation unit is used to confirm the corresponding signal propagation path as a direct propagation path and generate a direct path confirmation result when all path segments do not overlap in the path segment coverage determination result.

[0036] The direct path set construction unit is used to aggregate confirmed direct propagation paths and generate a direct path set based on the direct path confirmation results.

[0037] Preferably, the path length limiting submodule includes:

[0038] The path length benchmark generation unit is used to obtain the spatial path length corresponding to each direct propagation path based on the set of direct paths, and generate the path length benchmark result.

[0039] The multi-path length alignment unit is used to align the lengths of each path according to the path length benchmark result when there are multiple direct propagation paths in the same discharge structure region, and generate a range with consistent path lengths.

[0040] The propagation length boundary determination unit is used to determine the shortest and longest propagation length boundaries corresponding to the discharge structure region based on the consistent path length interval, and generate the propagation length boundary results.

[0041] The path range generation unit is used to combine the propagation length boundary results to generate a signal propagation reference path length range that corresponds one-to-one with the discharge structure region.

[0042] Preferably, the associated window generation submodule includes:

[0043] The time interval sequence generation unit is used to construct a continuous pulse time interval sequence in chronological order based on the pulse time interval distribution results, and generate the time interval sequence result.

[0044] The continuous segment identification unit is used to identify time segments in which adjacent time intervals occur consecutively based on the time interval sequence results, and generate a set of continuous time segments.

[0045] The isolated segment elimination unit is used to eliminate isolated time segments that do not form a continuous relationship with other time segments based on the set of continuous time segments, and generate a set of valid time segments.

[0046] The window boundary determination unit is used to determine the start and end boundaries of the time-related interval based on the start and end time intervals of each time segment in the set of effective time segments, and to generate the time-related interval.

[0047] Preferably, the polarity consistency screening submodule includes:

[0048] The polarity sequence generation unit is used to extract the polarity change direction of each electromagnetic pulse in time sequence according to the electromagnetic pulse within the time correlation interval, and generate a polarity change sequence.

[0049] The continuous and consistent segment determination unit is used to determine whether the polarity change direction of adjacent electromagnetic pulses is consistent based on the polarity change sequence, and to generate a continuous and consistent polarity segment.

[0050] The non-uniform pulse elimination unit is used to eliminate electromagnetic pulses that do not form a continuous and consistent polarity segment based on the continuous and consistent polarity segment, and generate a uniform pulse retention result;

[0051] The uniform pulse set generation unit is used to collect electromagnetic pulses that form a continuous and uniform polarity segment based on the uniform pulse retention result, and generate a uniform polarity pulse set.

[0052] Preferably, the structural sequence comparison submodule includes:

[0053] The structure path sequence generation unit is used to generate the expected propagation sequence from the discharge structure area to each external installation position based on the direct propagation path corresponding to each discharge structure area.

[0054] Arrival sequence alignment unit is used to align the arrival sequence of electromagnetic pulses corresponding to candidate discharge events with the expected propagation sequence to generate an alignment result;

[0055] The sequence consistency determination unit is used to determine whether the actual arrival sequence of the electromagnetic pulse is consistent with the expected propagation sequence of the corresponding discharge structure region based on the sequence alignment result, and to generate the sequence consistency determination result.

[0056] The structural region filtering unit is used to filter out the corresponding discharge structural regions from the candidate discharge structural regions when the order consistency determination results are inconsistent, thereby generating a set of discharge structural regions that meet the order.

[0057] Preferably, the time difference interval forming submodule includes:

[0058] The time difference sequence generation unit is used to construct the pulse arrival time difference sequence between each partial discharge signal sensing unit based on the electromagnetic pulse arrival sequence result, and generate the time difference sequence result.

[0059] The interval candidate generation unit is used to generate multiple time difference interval candidates based on the time difference sequence results and the relationship of consecutive occurrence of adjacent time differences;

[0060] The interval consistency filtering unit is used to filter candidates for time difference intervals, retaining the time difference intervals in which the time difference intervals corresponding to each partial discharge signal sensing unit in the same candidate discharge event are consistent, and generating the pulse arrival time difference interval.

[0061] The above-described solution of the present invention has at least the following beneficial effects:

[0062] By introducing a discharge structure path definition mechanism in the partial discharge monitoring process of dry-type transformers, the internal structural regions of the transformer are divided according to the spatial distribution relationship of the high-voltage winding, low-voltage winding and iron core. Corresponding signal propagation reference path length ranges are established for different discharge structure regions. This makes the propagation analysis of partial discharge signals no longer rely solely on a single measuring point or simple time difference relationship, but has a clear structural constraint basis. This reduces the problem of propagation path confusion caused by multi-winding parallel structure or electromagnetic coupling between iron core and winding from the source.

[0063] Based on this, by acquiring differentiated measurement point signals and uniformly correcting multiple signals, electromagnetic pulse signals from different external installation locations are synchronously acquired and uniformly processed. Furthermore, associated discharge events are constructed, so that multiple pulses belonging to the same discharge process can be effectively merged in terms of time and polarity. This reduces the possibility of reflected signals, superimposed signals, or single-point interference pulses being misidentified as real discharge events, providing a stable event basis for subsequent location analysis.

[0064] Furthermore, by comprehensively comparing the arrival sequence and pulse arrival time difference of electromagnetic pulses in candidate discharge events with the propagation path constraints corresponding to the discharge structure region, the discharge occurrence region is determined only when both the structure propagation sequence and propagation interval conditions are met simultaneously. This ensures that the discharge location results are consistent with the actual internal structure of the dry-type transformer. In complex application scenarios such as power distribution rooms, this effectively avoids misjudging low-voltage side winding discharge as high-voltage side insulation defects, thereby improving the reliability and pertinence of maintenance decision-making. Attached Figure Description

[0065] Figure 1 This is an architectural diagram of an online monitoring and location system for partial discharge of a dry-type transformer provided by an embodiment of the present invention. Detailed Implementation

[0066] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0067] like Figure 1 As shown, an embodiment of the present invention proposes an online monitoring and location system for partial discharge of a dry-type transformer, the system comprising:

[0068] The discharge structure path definition module is used to divide the interior of the dry-type transformer into multiple discharge structure regions based on the spatial distribution relationship of the high-voltage winding, low-voltage winding and core in the dry-type transformer, and to determine the corresponding signal propagation reference path length range for each discharge structure region, thereby generating discharge structure path constraint information.

[0069] The differentiated measurement point signal acquisition module is used to set up partial discharge signal sensing units at multiple different external installation positions corresponding to the discharge structure path constraint information, and simultaneously acquire multiple original electromagnetic pulse signal data generated by partial discharge.

[0070] The multi-channel signal unified correction module is used to unify the time base and correct the amplitude scale of multiple raw electromagnetic pulse signal data to generate corrected electromagnetic pulse signal data.

[0071] The associated discharge event construction module is used to generate the time correlation interval of the combined electromagnetic pulse based on the correction electromagnetic pulse signal data, and to combine multiple electromagnetic pulses with the same polarity change direction to form candidate discharge events.

[0072] The structure path matching calculation module is used to calculate the pulse arrival time difference of each local discharge signal sensing unit for each candidate discharge event, and compare the time difference with the discharge structure path constraint information to generate path matching results.

[0073] The interference event discrimination and elimination module is used to eliminate a candidate discharge event when no path matching result that meets the path matching conditions is generated in any discharge structure region.

[0074] The target discharge region determination module is used to determine the discharge structure region as the region where partial discharge occurs when a candidate discharge event corresponds to only a single discharge structure region and generates a path matching result that satisfies the path matching condition.

[0075] In this embodiment of the invention, by introducing a discharge structure path definition module, and combining the spatial distribution relationship of the high-voltage winding, low-voltage winding and core in the dry-type transformer, the internal structure region of the transformer is divided, and a corresponding signal propagation reference path length range is established for different discharge structure regions. This gives the propagation analysis of partial discharge signals a clear structural constraint basis, thereby avoiding the structural confusion problem caused by relying solely on single time difference information for positioning, and providing stable prior conditions for subsequent discharge position determination.

[0076] During partial discharge monitoring, a differentiated measurement point signal acquisition module and a multi-channel signal unified correction module are used to synchronously acquire and uniformly correct electromagnetic pulse signals from different external installation locations, and based on this, correlated discharge events are constructed. By generating time-correlated intervals and filtering electromagnetic pulses with consistent polarity change directions and meeting the measurement point quantity conditions, the actual discharge events are effectively merged, reducing the impact of single-point interference signals or sporadic pulses on the positioning process. Simultaneously, combined with a structural path matching calculation module, the pulse arrival sequence and pulse arrival time difference in candidate discharge events are analyzed and compared with the corresponding structural path constraint information, ensuring that the discharge location judgment simultaneously meets the dual conditions of structural sequence and propagation interval, thereby improving the reliability of discharge area judgment.

[0077] For example, during the operation of a dry-type transformer installed in an indoor power distribution facility, multiple partial discharge signal sensing units are installed at different external locations on the transformer. When a partial discharge occurs in the high-voltage winding area inside the transformer at a certain moment, the electromagnetic pulse signals collected by each sensing unit are first time-corrected and combined into a candidate discharge event. Then, the pulse arrival sequence is compared with the direct propagation path sequence corresponding to the high-voltage winding area, and the pulse arrival time difference is combined with the propagation reference path length range of the area to determine the final location of the discharge event corresponding to the high-voltage winding area, thus providing maintenance personnel with a clear reference for the location of the discharge.

[0078] In a preferred embodiment of the present invention, the differential measurement point signal acquisition module includes the following steps:

[0079] First, based on the discharge structure region formed by the discharge structure path definition module and its corresponding signal propagation reference path length range, multiple installation positions are selected outside the dry-type transformer at intervals to create differentiated propagation path relationships between each installation position and different discharge structure regions.

[0080] Subsequently, partial discharge signal sensing units are installed at each of the aforementioned installation locations, and each partial discharge signal sensing unit is numbered to distinguish different acquisition locations.

[0081] When partial discharge occurs during the operation of a dry-type transformer, each partial discharge signal sensing unit synchronously acquires the corresponding electromagnetic pulse signal and outputs the acquired electromagnetic pulse signal in the form of a raw signal, forming multiple raw electromagnetic pulse signal data, which provides the basic input for subsequent signal correction and event construction.

[0082] In a preferred embodiment of the present invention, the multi-channel signal unified correction module includes the following implementation steps:

[0083] First, the original electromagnetic pulse signals from different partial discharge signal sensing units are time-stamped and organized, with the sampling start time of each signal acquisition channel as the initial reference.

[0084] Subsequently, by aligning the time stamps of each acquisition channel, the representation of the same electromagnetic pulse on the time axis in different channels has a unified time reference, thereby eliminating the time deviation between the acquisition channels.

[0085] Next, the amplitude of the electromagnetic pulses acquired by each channel is scaled to ensure that the electromagnetic pulses acquired by different partial discharge signal sensing units are within the same scale range in terms of amplitude representation, so as to eliminate the influence of the difference in sensitivity of the sensing units.

[0086] Finally, the corrected electromagnetic pulse signal data, after time base unification and amplitude scale correction, is output for subsequent construction of associated discharge events and path matching analysis.

[0087] In a preferred embodiment of the present invention, the path matching condition is used to compare the pulse arrival characteristics in the candidate discharge event with the discharge structure path constraint information to determine whether the candidate discharge event meets the propagation conditions of the corresponding discharge structure region, including:

[0088] First, based on the correction electromagnetic pulse signals collected by each partial discharge signal sensing unit in the candidate discharge event, the arrival order of each electromagnetic pulse is determined, and the pulse arrival time difference interval between each sensing unit is formed accordingly.

[0089] Subsequently, for each discharge structure region, the signal propagation reference path length range corresponding to that discharge structure region is read, and the correspondence between the path length range and the pulse arrival time difference interval is determined.

[0090] When the pulse arrival sequence corresponding to a candidate discharge event is consistent with the direct propagation path sequence of the discharge structure region, and the pulse arrival time difference interval falls within the signal propagation reference path length range corresponding to the discharge structure region, the candidate discharge event is determined to satisfy the path matching condition of the discharge structure region.

[0091] When a candidate discharge event satisfies the path matching condition in a certain discharge structure region but not in other discharge structure regions, the discharge structure region is taken as the discharge occurrence region corresponding to the candidate discharge event.

[0092] In a preferred embodiment of the present invention, the discharge structure path definition module includes:

[0093] The structural relationship analysis submodule is used to analyze the relative spatial positions between structural units based on the radial covering relationship and axial stacking relationship of the high voltage winding, low voltage winding and iron core in the dry-type transformer, and generate structural relative position description information.

[0094] The structural region generation submodule is used to divide the inside of the transformer into multiple non-overlapping discharge structural regions based on the structural relative position description information, and generate a corresponding structural region identifier for each discharge structural region.

[0095] The direct path filtering submodule is used to determine the direct signal propagation path from each discharge structure area to the external installation location without passing through other discharge structure areas, based on the structure area identifier, and generate a set of direct paths;

[0096] The path length limiting submodule is used to limit the geometric length of each direct propagation path in the set of direct paths, and generate the corresponding signal propagation reference path length range;

[0097] The path constraint formation submodule is used to associate the length range of the signal propagation reference path with the corresponding structural region identifier to form discharge structure path constraint information.

[0098] In this embodiment of the invention, by analyzing the radial coverage and axial stacking relationships between the high-voltage winding, low-voltage winding, and core, the interior of the dry-type transformer is divided into multiple non-overlapping discharge structure regions. Furthermore, direct propagation paths from each discharge structure region to the external installation location that do not pass through other discharge structure regions are selected, ensuring that the propagation analysis of partial discharge signals is based on a clearly defined spatial structure. Simultaneously, by limiting the geometric length of the direct propagation paths and forming corresponding path constraint information, the signal propagation ranges corresponding to different structure regions have distinguishable boundary conditions, thereby avoiding the problem of unclear discharge region identification in cases of complex internal transformer structures and intersecting signal propagation paths.

[0099] In a preferred embodiment of the present invention, the structural relationship parsing submodule includes the following steps:

[0100] First, obtain the structural layout information of the high-voltage winding, low-voltage winding and core of the dry-type transformer. This structural layout information includes the covering layer relationship of each structural unit in the radial direction and the vertical arrangement relationship in the axial direction.

[0101] Subsequently, based on the radial covering hierarchy, the order of the inner and outer positions of each structural unit in the radial direction is determined, and the covering correspondence between each structural unit is recorded;

[0102] Simultaneously, based on the axial arrangement relationship, the relative positional order of each structural unit in the axial direction is determined, and the stacking correspondence between adjacent structural units is recorded;

[0103] Finally, the radial envelopment relationship and axial stacking relationship are integrated and organized to form structural relative position description information describing the relative spatial position of each structural unit, which is used for subsequent division of the discharge structure region.

[0104] In a preferred embodiment of the present invention, the structural region generation sub-module includes the following implementation steps:

[0105] First, based on the structural relative position description information, the boundaries of the spatial ranges corresponding to the high-voltage winding, low-voltage winding and iron core are determined so that the spatial ranges corresponding to different structural units can be distinguished from each other.

[0106] Subsequently, based on the spatial range corresponding to each structural unit, the interior of the transformer is divided into multiple non-overlapping discharge structural regions, so that each discharge structural region corresponds to a unique structural unit or combination of structural units.

[0107] Next, a unique structural region identifier is assigned to each discharge structure region to distinguish different discharge structure regions during subsequent path analysis and event matching.

[0108] Finally, the output includes the region division results containing each discharge structure region and its corresponding structure region identifier, providing a basis for generating discharge structure path constraint information.

[0109] In a preferred embodiment of the present invention, the associated discharge event construction module includes:

[0110] The pulse interval statistics submodule is used to statistically analyze the time intervals between adjacent electromagnetic pulses in the correction electromagnetic pulse signal data and generate pulse time interval distribution results.

[0111] The associated window generation submodule is used to determine the time association interval for combining electromagnetic pulses based on the pulse time interval distribution results.

[0112] The polarity consistency screening submodule is used to compare the polarity change direction of multiple electromagnetic pulses within the same time correlation interval based on the time correlation interval, and generate a set of polarity consistent pulses.

[0113] The quantity determination submodule is used to determine whether electromagnetic pulses from a preset number of different partial discharge signal sensing units are included based on the identifier of the corresponding partial discharge signal sensing unit in the set of pulses with consistent polarity, and to generate a quantity determination result.

[0114] The event result generation submodule is used to combine the corresponding electromagnetic pulses into candidate discharge events when the quantity determination result is true, and to record the identifiers of the partial discharge signal sensing units participating in the combination.

[0115] In this embodiment of the invention, by statistically analyzing the time intervals of the corrected electromagnetic pulse signals and generating time-related intervals, multiple electromagnetic pulses belonging to the same partial discharge process can be effectively aggregated in the time dimension. Based on this, by screening electromagnetic pulses within the same time-related interval for consistency in polarity change direction, and combining this with a determination of the number of sensing units from different partial discharge signals, it is ensured that the constructed candidate discharge events simultaneously meet the conditions of time consistency, polarity consistency, and multi-measurement point consistency. This reduces the probability of occasional noise pulses or single-point interference being misidentified as discharge events, allowing subsequent localization analysis to be based on stable and reliable discharge events.

[0116] In a preferred embodiment of the present invention, the quantity determination submodule includes the following steps:

[0117] First, the electromagnetic pulses contained in the set of pulses with consistent polarity are traversed, and the partial discharge signal sensing unit identifier corresponding to each electromagnetic pulse is read.

[0118] Subsequently, the identifiers of the partial discharge signal sensing units are deduplicated, and the number of different partial discharge signal sensing units actually participating in the set of pulses with consistent polarity is counted.

[0119] Next, the number of the different partial discharge signal sensing units is compared with the preset minimum number of measurement points to determine the result.

[0120] When the number of different partial discharge signal sensing units obtained by statistics reaches the minimum number of measuring points required, the determination result of the generation quantity is valid; when the minimum number of measuring points required is not reached, the determination result of the generation quantity is invalid.

[0121] In a preferred embodiment of the present invention, the event result generation submodule includes the following implementation steps:

[0122] First, receive the quantity validity determination result output by the quantity validity determination submodule, and determine whether the quantity validity determination result is valid.

[0123] When the quantity determination result is true, the electromagnetic pulses in the polarity consistent pulse set are arranged according to their corresponding time sequence.

[0124] Subsequently, the processed electromagnetic pulses are combined as components of the same discharge process to form a candidate discharge event.

[0125] Meanwhile, the identifiers of each partial discharge signal sensing unit participating in the candidate discharge event are recorded in the candidate discharge event so that they can be used for arrival order analysis and time difference interval formation in the subsequent structural path matching calculation process;

[0126] When the quantity determination result is invalid, no candidate discharge event is generated, and the current electromagnetic pulse combination process ends.

[0127] In a preferred embodiment of the present invention, the structural path matching calculation module includes:

[0128] The arrival order generation submodule is used to generate the electromagnetic pulse arrival order result based on the time sequence of electromagnetic pulses received by each partial discharge signal sensing unit in the candidate discharge event.

[0129] The structure sequence comparison submodule is used to compare the electromagnetic pulse arrival sequence with the direct propagation path sequence corresponding to each discharge structure region, and generate a set of discharge structure regions that match the sequence.

[0130] The time difference interval formation submodule is used to form the pulse arrival time difference interval between each partial discharge signal sensing unit based on the electromagnetic pulse arrival sequence.

[0131] The path matching determination submodule is used to compare the pulse arrival time difference interval with the signal propagation reference path length range corresponding to the discharge structure region with the correct sequence, and generate a path matching determination result.

[0132] In this embodiment of the invention, by analyzing the arrival order of each electromagnetic pulse in a candidate discharge event and comparing this arrival order with the direct propagation path order corresponding to different discharge structure regions, discharge structure regions that do not conform to spatial propagation logic can be eliminated in advance. Simultaneously, by forming pulse arrival time difference intervals between each partial discharge signal sensing unit and comparing these intervals with the propagation reference path length range in the discharge structure path constraint information, the determination of the discharge region simultaneously satisfies the dual constraints of propagation order and propagation interval. This improves the certainty and consistency of the discharge region matching results in the presence of multiple measurement points and multiple propagation paths.

[0133] In a preferred embodiment of the present invention, the path conformity determination submodule includes the following implementation steps:

[0134] First, the set of discharge structure regions whose output order matches the structure sequence comparison submodule is received, and the corresponding signal propagation reference path length range is read for each discharge structure region in the set.

[0135] Subsequently, the pulse arrival time difference intervals formed between the partial discharge signal sensing units in the candidate discharge event are obtained and organized into interval data corresponding one-to-one with each sensing unit.

[0136] Next, the pulse arrival time difference intervals are compared item by item with the signal propagation reference path length range corresponding to the current discharge structure region to determine whether each pulse arrival time difference interval falls within the corresponding signal propagation reference path length range.

[0137] When all pulse arrival time difference intervals corresponding to a candidate discharge event satisfy the above-mentioned comparison relationship, the path matching judgment result corresponding to the discharge structure region is satisfied.

[0138] When any pulse arrival time difference interval does not fall within the corresponding signal propagation reference path length range, the path matching determination result corresponding to the generated discharge structure region is not satisfied.

[0139] In a preferred embodiment of the present invention, the direct path filtering submodule includes:

[0140] The path segmentation result generation unit is used to divide the signal propagation path into multiple continuous spatial path segments based on the spatial connection relationship between the discharge structure area and the external installation position, and generate path segmentation results.

[0141] The path segment coverage determination unit is used to determine whether each spatial path segment overlaps with the spatial range corresponding to other discharge structure regions based on the path segmentation results, and to generate path segment coverage determination results.

[0142] The non-direct path exclusion unit is used to identify and exclude the corresponding signal propagation path as a non-direct propagation path when any path segment overlaps with other discharge structure regions, based on the path segment coverage determination result.

[0143] The direct path confirmation unit is used to confirm the corresponding signal propagation path as a direct propagation path and generate a direct path confirmation result when all path segments do not overlap in the path segment coverage determination result.

[0144] The direct path set construction unit is used to aggregate confirmed direct propagation paths and generate a direct path set based on the direct path confirmation results.

[0145] In this embodiment of the invention, the signal propagation path between the discharge structure region and the external installation location is divided into multiple continuous spatial path segments. Each segment is then assessed to determine whether it spatially overlaps with other discharge structure regions. This eliminates reliance on overall empirical judgment in determining the signal propagation path; instead, analysis is based on refined spatial segmentation results. When any path segment overlaps with another discharge structure region, the corresponding propagation path is excluded, retaining only the direct propagation path that does not overlap with other structure regions. This effectively distinguishes between direct and indirect propagation under complex internal structural conditions, reducing interference from reflection or detour paths in discharge location determination.

[0146] In a preferred embodiment of the present invention, the path segmentation result generation unit includes the following steps:

[0147] First, determine the spatial boundaries of the current discharge structure area and the spatial coordinates of the corresponding external installation location;

[0148] Subsequently, a continuous spatial line is established between the discharge structure region and the external installation location to indicate the possible propagation path direction of the signal;

[0149] Next, the spatial connection is divided according to the boundary of the discharge structure region, the boundary of the adjacent structure region, and the spatial location of the external installation position, so that the spatial connection is divided into multiple spatial path segments that are connected end to end.

[0150] Finally, the start and end positions of each spatial path segment are recorded to form the path segmentation results used for subsequent coverage determination.

[0151] In a preferred embodiment of the present invention, the path segment coverage determination unit includes the following steps:

[0152] First, read the start and end positions of each spatial path segment in the path segmentation result, and obtain the spatial range information of other discharge structure regions besides the current discharge structure region;

[0153] Subsequently, for each spatial path segment, it is determined whether the spatial path segment enters or crosses the spatial range of other discharge structure regions in the spatial extension direction.

[0154] When any part of a spatial path segment is located within the spatial range of other discharge structure regions, the path segment coverage determination result corresponding to that spatial path segment is determined to be overlapping.

[0155] When the entire spatial path segment is located outside the spatial range of other discharge structure regions, the path segment coverage determination result corresponding to the generated spatial path segment is that no overlap has occurred.

[0156] Finally, the path segment coverage determination results of each spatial path segment are collected to form a set of path segment coverage determination results for subsequent non-direct path exclusion determination.

[0157] In a preferred embodiment of the present invention, the path length limiting submodule includes:

[0158] The path length benchmark generation unit is used to obtain the spatial path length corresponding to each direct propagation path based on the set of direct paths, and generate the path length benchmark result.

[0159] The multi-path length alignment unit is used to align the lengths of each path according to the path length benchmark result when there are multiple direct propagation paths in the same discharge structure region, and generate a range with consistent path lengths.

[0160] The propagation length boundary determination unit is used to determine the shortest and longest propagation length boundaries corresponding to the discharge structure region based on the consistent path length interval, and generate the propagation length boundary results.

[0161] The path range generation unit is used to combine the propagation length boundary results to generate a signal propagation reference path length range that corresponds one-to-one with the discharge structure region.

[0162] In this embodiment of the invention, the spatial path lengths of multiple direct propagation paths corresponding to the same discharge structure region are obtained, and the path lengths are aligned to form a consistent path length interval, so that the establishment of path constraints does not depend on a single path result. Based on this, the corresponding shortest and longest propagation length boundaries are determined according to the consistent path length interval, ensuring that the signal propagation reference path length range can truly reflect the spatial propagation characteristics of the discharge structure region. This provides stable and distinguishable propagation range constraints for different structure regions during subsequent path matching.

[0163] In a preferred embodiment of the present invention, the multi-path length alignment unit includes the following steps:

[0164] First, the spatial path lengths corresponding to multiple direct propagation paths output by the path length reference generation unit are received, and these path lengths are classified according to their respective discharge structure regions.

[0165] Subsequently, within the same discharge structure region, the spatial path lengths of each direct propagation path are collected and organized to identify the set of path lengths with similar length distributions.

[0166] Next, the minimum and maximum path lengths covered by this set of path lengths are used as a unified reference range to make the multiple direct propagation paths within the discharge structure region form a consistent interval in terms of length representation.

[0167] Finally, the consistent interval is output as the basic input for determining the subsequent propagation length boundary of the discharge structure region.

[0168] In a preferred embodiment of the present invention, the associated window generation submodule includes:

[0169] The time interval sequence generation unit is used to construct a continuous pulse time interval sequence in chronological order based on the pulse time interval distribution results, and generate the time interval sequence result.

[0170] The continuous segment identification unit is used to identify time segments in which adjacent time intervals occur consecutively based on the time interval sequence results, and generate a set of continuous time segments.

[0171] The isolated segment elimination unit is used to eliminate isolated time segments that do not form a continuous relationship with other time segments based on the set of continuous time segments, and generate a set of valid time segments.

[0172] The window boundary determination unit is used to determine the start and end boundaries of the time-related interval based on the start and end time intervals of each time segment in the set of effective time segments, and to generate the time-related interval.

[0173] In this embodiment of the invention, by constructing the pulse time interval distribution results into a continuous time interval sequence and identifying time segments in which adjacent time intervals occur consecutively, while eliminating isolated segments that do not form a continuous relationship with other time segments, the generation process of the time-related interval can eliminate the interference of scattered pulses. Based on this, the starting and ending boundaries of the time-related interval are determined according to the effective time segments, ensuring that the electromagnetic pulses participating in the construction of the discharge event have continuity and consistency in the time dimension, thereby providing a reliable time basis for subsequent discharge event identification and location analysis.

[0174] In a preferred embodiment of the present invention, the continuous segment identification unit includes the following steps:

[0175] First, the pulse time interval sequence output by the time interval sequence generation unit is received and arranged in chronological order;

[0176] Subsequently, the arrangement of adjacent time intervals in the sequence is compared one by one. When adjacent time intervals are directly adjacent in time order, they are grouped into the same time segment.

[0177] Next, at the point where the time intervals are no longer adjacent or there is a break, the previous time segment ends and a new time segment begins to be recorded;

[0178] Finally, the identified time segments are aggregated to form a set of continuous time segments, which are used for subsequent isolated segment removal processing.

[0179] In a preferred embodiment of the present invention, the polarity consistency screening submodule includes:

[0180] The polarity sequence generation unit is used to extract the polarity change direction of each electromagnetic pulse in time sequence according to the electromagnetic pulse within the time correlation interval, and generate a polarity change sequence.

[0181] The continuous and consistent segment determination unit is used to determine whether the polarity change direction of adjacent electromagnetic pulses is consistent based on the polarity change sequence, and to generate a continuous and consistent polarity segment.

[0182] The non-uniform pulse elimination unit is used to eliminate electromagnetic pulses that do not form a continuous and consistent polarity segment based on the continuous and consistent polarity segment, and generate a uniform pulse retention result;

[0183] The uniform pulse set generation unit is used to collect electromagnetic pulses that form a continuous and uniform polarity segment based on the uniform pulse retention result, and generate a uniform polarity pulse set.

[0184] In this embodiment of the invention, the polarity change direction of electromagnetic pulses is extracted sequentially within a time-related interval, and a polarity change sequence is constructed. The continuity and consistency of the polarity change directions of adjacent electromagnetic pulses are determined, ensuring that the electromagnetic pulses participating in the construction of candidate discharge events maintain continuity in their polarity characteristics. Electromagnetic pulses that do not form a continuous and consistent polarity change segment are discarded, and only electromagnetic pulses that meet the continuity and consistency condition are retained to generate a set of polarity-consistent pulses. This reduces the possibility of incorrectly merging different discharge sources or intermittent interference pulses that are accidentally identical in polarity, thus making the polarity characteristics of discharge events more stable.

[0185] In a preferred embodiment of the present invention, the continuous and consistent segment determination unit includes the following steps:

[0186] First, the polarity change sequence output by the polarity sequence generation unit is received and arranged in chronological order of the electromagnetic pulses;

[0187] Subsequently, the polarity change direction of adjacent electromagnetic pulses is compared one by one. When the polarity change direction of adjacent electromagnetic pulses is the same, they are classified into the same continuous and consistent segment.

[0188] When the polarity change direction of an adjacent electromagnetic pulse changes, the current continuous and consistent segment ends, and a new continuous and consistent segment begins to be recorded.

[0189] Finally, the resulting continuous and consistent segments are sorted to generate polarity continuous and consistent segment results, which are used for subsequent non-consistent pulse removal processing.

[0190] In a preferred embodiment of the present invention, the structural sequence comparison submodule includes:

[0191] The structure path sequence generation unit is used to generate the expected propagation sequence from the discharge structure area to each external installation position based on the direct propagation path corresponding to each discharge structure area.

[0192] Arrival sequence alignment unit is used to align the arrival sequence of electromagnetic pulses corresponding to candidate discharge events with the expected propagation sequence to generate an alignment result;

[0193] The sequence consistency determination unit is used to determine whether the actual arrival sequence of the electromagnetic pulse is consistent with the expected propagation sequence of the corresponding discharge structure region based on the sequence alignment result, and to generate the sequence consistency determination result.

[0194] The structural region filtering unit is used to filter out the corresponding discharge structural regions from the candidate discharge structural regions when the order consistency determination results are inconsistent, thereby generating a set of discharge structural regions that meet the order.

[0195] In this embodiment of the invention, an expected propagation order is generated based on the direct propagation path corresponding to each discharge structure region. The actual arrival order of electromagnetic pulses in candidate discharge events is then aligned with and consistent with this expected propagation order. This allows discharge structure regions that do not conform to spatial propagation logic to be pre-screened, retaining only those with consistent propagation orders for subsequent judgment. By introducing propagation order constraints, the judgment of discharge regions depends not only on the time difference but also on the structural propagation direction, thereby improving the rationality of the discharge region screening results under multi-path propagation conditions.

[0196] In a preferred embodiment of the present invention, the structural path sequence generation unit includes the following steps:

[0197] First, read the direct propagation path information between the determined discharge structure area and each external installation location in the discharge structure path definition module;

[0198] Subsequently, based on the relative spatial distances of each direct propagation path, the order in which the signal propagates from the discharge structure area to each external installation location is determined;

[0199] Next, the sequence is organized into the expected propagation sequence result corresponding to the discharge structure region, which is used to compare with the actual electromagnetic pulse arrival sequence in the candidate discharge event;

[0200] Finally, the expected propagation order of each discharge structure region is output as a reference input for structure order comparison.

[0201] In a preferred embodiment of the present invention, the time difference interval forming submodule includes:

[0202] The time difference sequence generation unit is used to construct the pulse arrival time difference sequence between each partial discharge signal sensing unit based on the electromagnetic pulse arrival sequence result, and generate the time difference sequence result.

[0203] The interval candidate generation unit is used to generate multiple time difference interval candidates based on the time difference sequence results and the relationship of consecutive occurrence of adjacent time differences;

[0204] The interval consistency filtering unit is used to filter candidates for time difference intervals, retaining the time difference intervals in which the time difference intervals corresponding to each partial discharge signal sensing unit in the same candidate discharge event are consistent, and generating the pulse arrival time difference interval.

[0205] In this embodiment of the invention, a pulse arrival time difference sequence is constructed based on the arrival order of electromagnetic pulses, and multiple candidate time difference intervals are generated based on this sequence. Then, consistency screening is performed on each candidate time difference interval, retaining only those intervals where the time difference intervals corresponding to the partial discharge signal sensing units in the same candidate discharge event remain consistent. This ensures that the time difference characteristics corresponding to the discharge event have a stable interval form. By using time differences in interval form for structural path matching, rather than a single numerical comparison, the positioning judgment has better adaptability to minute time fluctuations, thereby improving the reliability of the discharge region matching results.

[0206] In a preferred embodiment of the present invention, the interval candidate generation unit includes the following steps:

[0207] First, the pulse arrival time difference sequence output by the time difference sequence generation unit is received and arranged in the order of its formation;

[0208] Subsequently, the arrangement relationship of adjacent time differences in the sequence is judged in turn. When adjacent time differences appear consecutively in the sequence, they are classified into the same time difference interval candidate.

[0209] When time differences are no longer continuous, the current time difference interval candidate ends and a new time difference interval candidate begins to be formed;

[0210] Finally, the multiple time difference interval candidates are aggregated and used as input for subsequent interval consistency screening units.

[0211] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A system for online monitoring and locating partial discharge in dry-type transformers, characterized in that, The system includes: The discharge structure path definition module is used to divide the interior of the dry-type transformer into multiple discharge structure regions based on the spatial distribution relationship of the high-voltage winding, low-voltage winding and core in the dry-type transformer, and to determine the corresponding signal propagation reference path length range for each discharge structure region, thereby generating discharge structure path constraint information. The differentiated measurement point signal acquisition module is used to set up partial discharge signal sensing units at multiple different external installation positions corresponding to the discharge structure path constraint information, and simultaneously acquire multiple original electromagnetic pulse signal data generated by partial discharge. The multi-channel signal unified correction module is used to unify the time base and correct the amplitude scale of multiple raw electromagnetic pulse signal data to generate corrected electromagnetic pulse signal data. The associated discharge event construction module is used to generate the time correlation interval of the combined electromagnetic pulse based on the correction electromagnetic pulse signal data, and to combine multiple electromagnetic pulses with the same polarity change direction to form candidate discharge events. The structure path matching calculation module is used to calculate the pulse arrival time difference of each local discharge signal sensing unit for each candidate discharge event, and compare the time difference with the discharge structure path constraint information to generate path matching results. The interference event discrimination and elimination module is used to eliminate a candidate discharge event when no path matching result that meets the path matching conditions is generated in any discharge structure region. The target discharge region determination module is used to determine the discharge structure region as the region where a path matching result that satisfies the path matching condition is generated when a candidate discharge event corresponds to only a single discharge structure region. The discharge structure path definition module includes a direct path filtering submodule, which includes: The path segmentation result generation unit is used to divide the signal propagation path into multiple continuous spatial path segments based on the spatial connection relationship between the discharge structure area and the external installation position, and generate path segmentation results. The path segment coverage determination unit is used to determine whether each spatial path segment overlaps with the spatial range corresponding to other discharge structure regions based on the path segmentation results, and to generate path segment coverage determination results. The non-direct path exclusion unit is used to identify and exclude the corresponding signal propagation path as a non-direct propagation path when any path segment overlaps with other discharge structure regions, based on the path segment coverage determination result. The direct path confirmation unit is used to confirm the corresponding signal propagation path as a direct propagation path and generate a direct path confirmation result when all path segments do not overlap in the path segment coverage determination result. The direct path set construction unit is used to aggregate the confirmed direct propagation paths and generate a direct path set based on the direct path confirmation results. The set of direct paths is used to determine the range of signal propagation reference path lengths corresponding to the discharge structure region.

2. The online monitoring and location system for partial discharge of a dry-type transformer according to claim 1, characterized in that, The discharge structure path definition module also includes: The structural relationship analysis submodule is used to analyze the relative spatial positions between structural units based on the radial covering relationship and axial stacking relationship of the high voltage winding, low voltage winding and iron core in the dry-type transformer, and generate structural relative position description information. The structural region generation submodule is used to divide the inside of the transformer into multiple non-overlapping discharge structural regions based on the structural relative position description information, and generate a corresponding structural region identifier for each discharge structural region. The path length limiting submodule is used to limit the geometric length of each direct propagation path in the set of direct paths, and generate the corresponding signal propagation reference path length range; The path constraint formation submodule is used to associate the length range of the signal propagation reference path with the corresponding structural region identifier to form discharge structure path constraint information.

3. The online monitoring and location system for partial discharge of a dry-type transformer according to claim 1, characterized in that, The associated discharge event construction module includes: The pulse interval statistics submodule is used to statistically analyze the time intervals between adjacent electromagnetic pulses in the correction electromagnetic pulse signal data and generate pulse time interval distribution results. The associated window generation submodule is used to determine the time association interval for combining electromagnetic pulses based on the pulse time interval distribution results. The polarity consistency screening submodule is used to compare the polarity change direction of multiple electromagnetic pulses within the same time correlation interval based on the time correlation interval, and generate a set of polarity consistent pulses. The quantity determination submodule is used to determine whether electromagnetic pulses from a preset number of different partial discharge signal sensing units are included based on the identifier of the corresponding partial discharge signal sensing unit in the set of pulses with consistent polarity, and to generate a quantity determination result. The event result generation submodule is used to combine the corresponding electromagnetic pulses into candidate discharge events when the quantity determination result is true, and to record the identifiers of the partial discharge signal sensing units participating in the combination.

4. The online monitoring and location system for partial discharge of a dry-type transformer according to claim 1, characterized in that, The structural path matching calculation module includes: The arrival order generation submodule is used to generate the electromagnetic pulse arrival order result based on the time sequence of electromagnetic pulses received by each partial discharge signal sensing unit in the candidate discharge event. The structure sequence comparison submodule is used to compare the electromagnetic pulse arrival sequence with the direct propagation path sequence corresponding to each discharge structure region, and generate a set of discharge structure regions that match the sequence. The time difference interval formation submodule is used to form the pulse arrival time difference interval between each partial discharge signal sensing unit based on the electromagnetic pulse arrival sequence. The path matching determination submodule is used to compare the pulse arrival time difference interval with the signal propagation reference path length range corresponding to the discharge structure region with the correct sequence, and generate a path matching determination result.

5. The online monitoring and location system for partial discharge of a dry-type transformer according to claim 2, characterized in that, The path length limiting submodule includes: The path length benchmark generation unit is used to obtain the spatial path length corresponding to each direct propagation path based on the set of direct paths, and generate the path length benchmark result. The multi-path length alignment unit is used to align the lengths of each path according to the path length benchmark result when there are multiple direct propagation paths in the same discharge structure region, and generate a range with consistent path lengths. The propagation length boundary determination unit is used to determine the shortest and longest propagation length boundaries corresponding to the discharge structure region based on the consistent path length interval, and generate the propagation length boundary results. The path range generation unit is used to combine the propagation length boundary results to generate a signal propagation reference path length range that corresponds one-to-one with the discharge structure region.

6. The online monitoring and location system for partial discharge of a dry-type transformer according to claim 3, characterized in that, The associated window generation submodule includes: The time interval sequence generation unit is used to construct a continuous pulse time interval sequence in chronological order based on the pulse time interval distribution results, and generate the time interval sequence result. The continuous segment identification unit is used to identify time segments in which adjacent time intervals occur consecutively based on the time interval sequence results, and generate a set of continuous time segments. The isolated segment elimination unit is used to eliminate isolated time segments that do not form a continuous relationship with other time segments based on the set of continuous time segments, and generate a set of valid time segments. The window boundary determination unit is used to determine the start and end boundaries of the time-related interval based on the start and end time intervals of each time segment in the set of effective time segments, and to generate the time-related interval.

7. The online monitoring and location system for partial discharge of a dry-type transformer according to claim 3, characterized in that, The polarity consistency screening submodule includes: The polarity sequence generation unit is used to extract the polarity change direction of each electromagnetic pulse in time sequence according to the electromagnetic pulse within the time correlation interval, and generate a polarity change sequence. The continuous and consistent segment determination unit is used to determine whether the polarity change direction of adjacent electromagnetic pulses is consistent based on the polarity change sequence, and to generate a continuous and consistent polarity segment. The non-uniform pulse elimination unit is used to eliminate electromagnetic pulses that do not form a continuous and consistent polarity segment based on the continuous and consistent polarity segment, and generate a uniform pulse retention result; The uniform pulse set generation unit is used to collect electromagnetic pulses that form a continuous and uniform polarity segment based on the uniform pulse retention result, and generate a uniform polarity pulse set.

8. The online monitoring and location system for partial discharge of a dry-type transformer according to claim 4, characterized in that, The structural sequence comparison submodule includes: The structure path sequence generation unit is used to generate the expected propagation sequence from the discharge structure area to each external installation position based on the direct propagation path corresponding to each discharge structure area. Arrival sequence alignment unit is used to align the arrival sequence of electromagnetic pulses corresponding to candidate discharge events with the expected propagation sequence to generate an alignment result; The sequence consistency determination unit is used to determine whether the actual arrival sequence of the electromagnetic pulse is consistent with the expected propagation sequence of the corresponding discharge structure region based on the sequence alignment result, and to generate the sequence consistency determination result. The structural region filtering unit is used to filter out the corresponding discharge structural regions from the candidate discharge structural regions when the order consistency determination results are inconsistent, thereby generating a set of discharge structural regions that meet the order.

9. The online monitoring and location system for partial discharge of a dry-type transformer according to claim 4, characterized in that, The time difference interval forming submodule includes: The time difference sequence generation unit is used to construct the pulse arrival time difference sequence between each partial discharge signal sensing unit based on the electromagnetic pulse arrival sequence result, and generate the time difference sequence result. The interval candidate generation unit is used to generate multiple time difference interval candidates based on the time difference sequence results and the relationship of consecutive occurrence of adjacent time differences; The interval consistency filtering unit is used to filter candidates for time difference intervals, retaining the time difference intervals in which the time difference intervals corresponding to each partial discharge signal sensing unit in the same candidate discharge event are consistent, and generating the pulse arrival time difference interval.

Images