Voltage sag homologous identification method and system based on association rule mining

Abstract

A voltage sag homologous identification method and system based on association rule mining, relating to the technical field of power quality mitigation. Transactions are formed by means of screened candidate homologous sag events, a transaction table is formed, a plurality of frequent itemsets are screened on the basis of the support of each itemset in the transaction table, so as to construct a canonical-order tree corresponding to the transaction table, thereby reducing the number of misleading transactions and improving the effectiveness of voltage sag homologous identification. In addition, association rules for candidate homologous sag events corresponding to every two frequent itemsets are determined by means of the canonical-order tree, waveform similarity comparison is performed on voltage waveforms corresponding to the candidate homologous sag events, and whether the two candidate homologous sag events are homologous sag events is determined, thereby improving the accuracy of voltage sag event homologous identification.

Description

Voltage sag same source identification method and system based on association rule mining

[0001] The present application claims priority to the Chinese patent application No. 202411147815.2, filed on August 21, 2024, and entitled "Voltage sag same source identification method and system based on association rule mining", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to the field of power quality management, and particularly relates to a voltage sag same source identification method and system based on association rule mining. BACKGROUND

[0003] With the increase in the number of voltage sag monitoring devices, voltage sag monitoring data presents a massive characteristic. However, this does not mean that a corresponding number of voltage sag events occur in the power grid. Since voltage sag occurs and propagates in the power grid, multiple monitoring devices may record the same voltage sag event in the propagation time slice. This not only causes a large amount of data redundancy, but also may lead to overestimation of the severity of the regional power grid voltage sag. Therefore, it is very meaningful to detect the same source of voltage sag monitoring data.

[0004] By identifying the repeatedly recorded voltage sag events, the source of the voltage sag event is found, which also helps to analyze the scope of the voltage sag influence, so as to improve the governance efficiency of the voltage sag problem.

[0005] The existing voltage sag same source identification algorithm mainly measures the similarity between multiple voltage sag waveform data through the Wasserstein distance, and performs clustering analysis on the measurement results based on the density-based spatial clustering of applications with noise (DBSAN) clustering algorithm.

[0006] However, the existing voltage sag same source identification algorithm has poor recognition effectiveness, which easily leads to low accuracy of voltage sag event same source identification. SUMMARY

[0007] The present application provides a voltage sag same source identification method and system based on association rule mining, which solves the technical problem of poor recognition effectiveness of the existing voltage sag same source identification algorithm, which easily leads to low accuracy of voltage sag event same source identification.

[0008] Therefore, the first aspect of the present application provides a voltage sag same source identification method based on association rule mining, comprising:

[0009] According to the event groups, each event group corresponds to a transaction, and a transaction table is formed, wherein the event groups include two different candidate homologous temporary drop events;

[0010] According to the support degree of each item set in the transaction table, a plurality of frequent item sets are screened out, and the item set is a candidate homologous temporary drop event;

[0011] According to the support degree of a plurality of the frequent item sets, a standard order tree corresponding to the transaction table is constructed, and the standard order tree includes a plurality of the frequent item sets;

[0012] According to the standard order tree, the association rule of the candidate homologous temporary drop events corresponding to each two frequent item sets is determined, and the association rule is used to represent that two candidate homologous temporary drop events are in a mapping relationship;

[0013] According to the voltage waveforms corresponding to two candidate homologous temporary drop events in a mapping relationship, waveform similarity comparison is performed, and whether two candidate homologous temporary drop events are homologous temporary drop events is determined according to the waveform similarity comparison result.

[0014] Preferably, before the step of according to the event groups, each event group corresponds to a transaction, and a transaction table is formed, wherein the event groups include two different candidate homologous temporary drop events, the method further comprises:

[0015] A plurality of voltage temporary drop events are sorted according to occurrence time, and a plurality of the voltage temporary drop events are selected as temporary drop reference events according to the sorting;

[0016] For each temporary drop reference event, the time difference between the temporary drop reference event and each voltage temporary drop event except the temporary drop reference event is calculated;

[0017] The temporary drop reference event and the voltage temporary drop event with a time difference less than a preset time difference threshold value are screened out as candidate homologous temporary drop events, and the candidate homologous temporary drop events are grouped into an event group.

[0018] Preferably, the step of according to the support degree of each item set in the transaction table, a plurality of frequent item sets are screened out, and the item set is a candidate homologous temporary drop event, comprises:

[0019] The candidate homologous temporary drop events in each transaction in the transaction table are taken as item sets, and the frequency of occurrence of each item set in the transaction table is determined as the support degree of each item set;

[0020] For the support degree of each item set, the size of the support degree and a preset support degree threshold value is compared, and the item set with a support degree greater than the preset support degree threshold value is screened out as a frequent item set.

[0021] Preferably, the preset support threshold is determined according to the number of association rules mined under a plurality of different support thresholds; the step of determining the preset support threshold according to the number of association rules mined under a plurality of different support thresholds comprises:

[0022] initializing a support threshold initial value;

[0023] taking the support threshold initial value as a current support threshold, and determining the number of association rules mined under the current support threshold;

[0024] taking the product of the current support threshold and the number of association rules mined under the current support threshold as an optimization target, and constructing an objective function;

[0025] iteratively optimizing the objective function, and iteratively updating the current support threshold until convergence, and outputting the converged current support threshold as the preset support threshold.

[0026] Preferably, the step of constructing a standard order tree corresponding to the transaction table according to the support of a plurality of the frequent item sets, the standard order tree containing a plurality of the frequent item sets, comprises:

[0027] arranging a plurality of the frequent item sets in descending order according to the support of a plurality of the frequent item sets;

[0028] constructing a multi-level standard order tree according to the frequent item sets arranged in descending order, the standard order tree including a plurality of paths, each path including a plurality of nodes, the standard order tree arranging each node in descending order into a multi-level structure, each path being a group of transactions sharing the same prefix node, each node corresponding to a frequent item set, and the prefix node being a frequent item set with the highest support.

[0029] Preferably, the step of determining, according to the standard order tree, the association rules of candidate homologous temporary drop events respectively corresponding to each two of the frequent item sets, comprises:

[0030] combining, for each path in the standard order tree, candidate homologous temporary drop events respectively corresponding to each two of the frequent item sets in each path to obtain association rules.

[0031] Preferably, the step of comparing waveform similarity according to voltage waveforms respectively corresponding to two candidate homologous temporary drop events in a mapping relationship, and determining whether the two candidate homologous temporary drop events are homologous temporary drop events through the waveform similarity comparison result, comprises:

[0032] determining whether the temporary drop types respectively corresponding to the two candidate homologous temporary drop events in a mapping relationship are matched;

[0033] If it is judged that the voltage sag types corresponding to the two candidate homologous sag events in the mapping relationship respectively do not match, then the transaction constituted by the two candidate homologous sag events is filtered out from the transaction table, and the step of screening a plurality of frequent item sets according to the support degree of each item set in the transaction table is performed, the item set being a candidate homologous sag event;

[0034] If it is judged that the voltage sag types corresponding to the two candidate homologous sag events in the mapping relationship respectively match, then the voltage waveforms corresponding to the two candidate homologous sag events in the mapping relationship are transformed to the same side of the transformer;

[0035] Waveform similarity calculation is performed on the two transformed voltage waveforms based on a constrained differential dynamic time warping algorithm, and the waveform similarity is represented by a distance.

[0036] The waveform similarity of the two transformed voltage waveforms is compared with a preset waveform similarity threshold value.

[0037] If the waveform similarity of the two transformed voltage waveforms is less than or equal to the preset waveform similarity threshold value, then it is determined that the two candidate homologous sag events are homologous sag events.

[0038] If the waveform similarity of the two transformed voltage waveforms is greater than the preset waveform similarity threshold value, then it is determined that the two candidate homologous sag events are not homologous sag events, the transaction constituted by the two candidate homologous sag events is filtered out from the transaction table, and the step of screening a plurality of frequent item sets according to the support degree of each item set in the transaction table is performed, the item set being a candidate homologous sag event.

[0039] Preferably, the step of transforming the voltage waveforms corresponding to the two candidate homologous sag events in the mapping relationship to the same side of the transformer comprises:

[0040] It is judged whether the two candidate homologous sag events in the mapping relationship occur on the same side of the transformer.

[0041] If it is judged that the two candidate homologous sag events in the mapping relationship do not occur on the same side of the transformer, then the transformer type involved in the voltage sag propagation process is determined according to the voltage sag types corresponding to the two candidate homologous sag events respectively.

[0042] A voltage waveform transformation matrix is determined according to the transformer type.

[0043] The voltage waveform of one of the candidate homologous sag events is transformed by using the voltage waveform transformation matrix, and the transformed voltage waveform and the voltage waveform of the other candidate homologous sag event are on the same side of the transformer.

[0044] Preferably, the step of performing waveform similarity calculation on the two transformed voltage waveforms by the constrained differential dynamic time warping algorithm, wherein the waveform similarity is characterized by a distance, comprises:

[0045] performing first-order derivation on the two transformed voltage waveforms respectively to obtain first-order derivatives corresponding to the two voltage waveforms respectively;

[0046] determining Euclidean distances between points in the two voltage waveforms according to the first-order derivatives corresponding to the two voltage waveforms respectively;

[0047] constructing a cost matrix according to the Euclidean distances between the points in the two voltage waveforms;

[0048] constructing a cumulative cost matrix in a preset constrained search region according to the cost matrix;

[0049] determining a global cumulative cost of the two transformed voltage waveforms according to the cumulative cost matrix in the preset constrained search region;

[0050] determining a waveform similarity of the two voltage waveforms according to the global cumulative cost and a maximum time sequence length of the two voltage waveforms.

[0051] In a second aspect, the present application further provides a voltage sag same-source identification system based on association rule mining, comprising:

[0052] a transaction construction module, configured to sequentially group a plurality of event groups into transactions corresponding to the event groups respectively, and form a transaction table, wherein the event groups contain two different candidate same-source sag events;

[0053] an item set screening module, configured to screen a plurality of frequent item sets according to support sizes of item sets in the transaction table, wherein the item sets are candidate same-source sag events;

[0054] a tree construction module, configured to construct a standard order tree corresponding to the transaction table according to support sizes of a plurality of the frequent item sets, wherein the standard order tree contains a plurality of the frequent item sets;

[0055] an association mining module, configured to determine association rules of candidate same-source sag events corresponding to each two of the frequent item sets according to the standard order tree, wherein the association rules are used to represent that the two candidate same-source sag events are in a mapping relationship;

[0056] a same-source identification module, configured to perform waveform similarity comparison on voltage waveforms corresponding to two candidate same-source sag events in a mapping relationship, and determine whether the two candidate same-source sag events are same-source sag events according to a waveform similarity comparison result.

[0057] From the above technical scheme can be seen, the present application has the following advantages:

[0058] The application forms a transaction table by screening out candidate homologous voltage sag events to constitute a transaction, and screens out a plurality of frequent item sets by using the support degree of each item set in the transaction table, so as to construct a standard sequential tree corresponding to the transaction table, thereby reducing the number of misleading transactions and improving the effectiveness of voltage sag homologous identification. Meanwhile, the association rules of the candidate homologous voltage sag events corresponding to each two frequent item sets are determined by the standard sequential tree, the voltage waveforms corresponding to the candidate homologous voltage sag events are compared in waveform similarity, and it is judged whether the two candidate homologous voltage sag events are homologous voltage sag events, thereby improving the accuracy of voltage sag event homologous identification. BRIEF DESCRIPTION OF DRAWINGS

[0059] Fig. 1 is a flow chart of a voltage sag homologous identification method based on association rule mining provided by an embodiment of the application;

[0060] Fig. 2 is a judgment flow chart of a candidate homologous voltage sag event provided by an embodiment of the application;

[0061] Fig. 3 is a flow chart of screening a plurality of frequent item sets according to the support degree of each item set in a transaction table provided by an embodiment of the application;

[0062] Fig. 4 is a structural schematic diagram of a standard sequential tree provided by an embodiment of the application;

[0063] Fig. 5 is a flow chart of judging whether two candidate homologous voltage sag events are homologous voltage sag events provided by an embodiment of the application;

[0064] Fig. 6 is a flow chart of transforming the voltage waveforms corresponding to two candidate homologous voltage sag events in a mapping relationship to the same side of a transformer provided by an embodiment of the application;

[0065] Fig. 7 is a flow chart of calculating the waveform similarity of two transformed voltage waveforms based on a constrained differential dynamic time warping algorithm provided by an embodiment of the application;

[0066] Fig. 8 is a structural schematic diagram of a voltage sag homologous identification system based on association rule mining provided by an embodiment of the application;

[0067] Fig. 9 is a structural schematic diagram of an electronic device provided by an embodiment of the application. DETAILED DESCRIPTION

[0068] In order to make the personnel in the technical field better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0069] The voltage sag source identification method based on association rule mining provided by the embodiments of the present application is applicable to the case of voltage sag source identification. The method can be executed by a voltage sag source identification device, which can be realized in the form of hardware and / or software and can be configured in a computer device.

[0070] As shown in FIG. 1, the voltage sag source identification method based on association rule mining provided by the embodiments of the present application includes steps S1-S5.

[0071] In step S1, a plurality of event groups are sequentially formed into transactions corresponding to the respective event groups, and a transaction table is formed. The event groups include two different candidate source sag events.

[0072] The two different candidate source sag events refer to voltage sag events possibly caused by the same sag source after preliminary screening. The voltage sag events can be monitored by a power quality monitoring device.

[0073] In order to identify the candidate source sag events and form event groups using the two different candidate source sag events, as shown in FIG. 2, the embodiments of the present application establish a judgment rule for the candidate source sag events, which specifically includes steps A1-A3.

[0074] In step A1, a plurality of voltage sag events are sorted according to the occurrence time, and a plurality of voltage sag events are selected as sag reference events according to the sorting.

[0075] The sag reference event refers to a selected voltage sag event. Each voltage sag event can be a sag reference event.

[0076] In step A2, for each sag reference event, the time difference between the sag reference event and each voltage sag event other than the sag reference event is calculated.

[0077] In step A3, the sag reference event and the voltage sag event with a time difference less than a preset time difference threshold are selected as the candidate source sag events, and the candidate source sag events are formed into an event group.

[0078] Exemplarily, N voltage sag events obtained from the power quality monitoring device are sorted by occurrence time, the i-th voltage sag event is selected as a voltage sag reference event, it is assumed that the occurrence time of the voltage sag reference event i is t i , the occurrence time of the voltage sag event j is t j , the candidate homologous voltage sag event of the voltage sag reference event i is selected according to the comparison function in formula (1), if the difference between the occurrence time of the voltage sag event j and the occurrence time of the voltage sag reference event i is less than the set occurrence time difference threshold value, then the voltage sag reference event i and the voltage sag event j are likely to be caused by the same voltage sag source.

[0079] In the formula, δ is the occurrence time difference threshold value.

[0080] If the voltage sag reference event i and the voltage sag event j are selected as the candidate homologous voltage sag events, and the candidate homologous voltage sag events are grouped into the transaction TR k according to formula (2), in this way, different transactions are sequentially established by selecting different voltage sag reference events and corresponding candidate homologous voltage sag events. k = {sequence number (i), sag type (i), sequence number (j), sag type (j)}; (2)

[0081] In the formula, sequence number (i) refers to the sequence number of the power quality monitoring device that monitors the i-th voltage sag event, sag type (i) refers to the type of the voltage sag waveform monitored by the power quality monitoring device of the i-th voltage sag event, sequence number (j) refers to the sequence number of the power quality monitoring device that monitors the j-th voltage sag event, and sag type (j) refers to the type of the voltage sag waveform monitored by the power quality monitoring device of the j-th voltage sag event.

[0082] The power quality monitoring device refers to a device for monitoring changes in grid voltage, and the types of voltage sag waveforms include but are not limited to symmetric voltage sag, asymmetric voltage sag, sudden drop voltage sag, slow drop voltage sag, oscillation voltage sag, pulse voltage sag and phase jump voltage sag.

[0083] All transactions generated by combining any basic event and its corresponding candidate homologous voltage sag event can form a transaction table G, as shown in formula (3). G = {TR1, TR2, …, TR M}; (3)

[0084] In the formula, M is the number of transactions contained in the transaction table.

[0085] Step S2, screening a plurality of frequent item sets from the support degree of each item set in the transaction table, the item set being a candidate homologous temporary drop event.

[0086] Wherein, the item set refers to the candidate homologous temporary drop event in each transaction in the transaction table, and each transaction generally contains two different item sets.

[0087] It should be noted that in the transaction table, some transactions are generated by two non-homologous events, which should be eliminated in association rule mining.

[0088] For an item set X in the transaction table G, the support number of the item set X can be determined by the number of transactions containing the item set X in the transaction table G, denoted as count(X).

[0089] Then, the frequency of the item set X appearing in the transaction table is determined as the support degree of the item set X according to the support number of the item set X and the number of all transactions in the transaction table G.

[0090] Specifically, as shown in FIG. 3, the process of screening a plurality of frequent item sets according to the support degree of each item set in the transaction table includes steps S201-S202:

[0091] Step S201, taking the candidate homologous temporary drop event in each transaction in the transaction table as an item set, and determining the frequency of each item set appearing in the transaction table as the support degree of each item set.

[0092] Wherein, the support degree is:

[0093] In the formula, S(X) is the support degree of the item set X, G A is the number of transactions of the transaction table.

[0094] Step S202, comparing the support degree of each item set with the size of the preset support degree threshold, and screening the item set with the support degree greater than the preset support degree threshold as a frequent item set.

[0095] Wherein, when the support degree of the item set X is greater than the preset support degree threshold, the item set X is determined as a frequent item set, otherwise, it is a non-frequent item set.

[0096] Association rule mining mainly extracts rules with higher credibility from frequent item sets, so the screening of frequent item sets is very important. If the support degree threshold is set too high, the frequent item set may be eliminated in the process of mining association rules, which may result in too few matching rules. If the support degree threshold is set too low, too many rules may be generated, which may conflict with each other, resulting in inaccurate results. Therefore, an adaptive and optimized support degree threshold needs to be determined to improve the accuracy of the mined association rules.

[0097] In the embodiment of the present application, the preset support threshold is determined according to the number of association rules mined under a plurality of different support thresholds; the step of determining the preset support threshold according to the number of association rules mined under a plurality of different support thresholds comprises steps S21-S24:

[0098] Step S21, initialize the support threshold initial value.

[0099] In a general example, the support threshold initial value can be set according to an empirical value.

[0100] Step S22, take the support threshold initial value as the current support threshold, and determine the number of association rules mined under the current support threshold.

[0101] Wherein, for the number of association rules, after the frequent item sets are screened out according to the current support threshold, the frequent item sets screened out using the current support threshold are used to execute steps S3 and S4 to obtain the association rules of the candidate homologous temporary drop events and the number thereof.

[0102] Step S23, take the product of the current support threshold and the number of association rules mined under the current support threshold maximization as the optimization target, and construct the objective function.

[0103] Wherein, the objective function is: maxf(minsup) = minsup x ln(n r ); (5)

[0104] In the formula, maxf(minsup) is the updated support threshold, minsup is the support threshold before updating, n r is the number of association rules mined under the support threshold before updating.

[0105] Step S24, iteratively optimize the objective function, and iteratively update the current support threshold until convergence, and output the converged current support threshold as the preset support threshold.

[0106] Wherein, the mathematical solver is used to iteratively optimize the objective function, and iteratively update the current support threshold until convergence, and the convergence condition can be that the maximum number of convergences is reached or the change difference of the updated support threshold is less than an error value, and the converged current support threshold is output as the preset support threshold.

[0107] Step S3, construct a standard order tree corresponding to the transaction table according to the support of a plurality of frequent item sets, and the standard order tree contains a plurality of frequent item sets.

[0108] In some embodiments, a canonical-order tree corresponding to the transaction table is constructed, including steps S301-S302:

[0109] Step S301, arranging the plurality of frequent item sets in descending order according to their support degrees.

[0110] Step S302, constructing a multi-level canonical-order tree according to the frequent item sets arranged in descending order, the canonical-order tree including a plurality of paths, each path including a plurality of nodes, the canonical-order tree arranging each node in descending order into a multi-level structure, each path being a group of transactions sharing the same prefix node, each node corresponding to a frequent item set, and the prefix node being the frequent item set with the highest support degree.

[0111] The canonical-order tree (CanTree) structure includes a plurality of paths, each path including a plurality of nodes. Each path represents a group of transactions sharing the same prefix, and each node corresponds to an item set.

[0112] For example, a multi-level canonical-order tree is constructed using the transactions shown in Table 1, where the frequent item sets contained in the transaction table are item sets a-j.

[0113] Table 1

[0114] The support degrees of the frequent item sets a-j are arranged in descending order as follows:

[0115] a: 5; b: 3; c: 3; d: 3; e: 2; f: 2; g: 2; h: 1; i: 1; j: 1.

[0116] Then, the frequent item sets in the transactions shown in Table 1 are sorted according to their support degrees, and the transaction sorting result is shown in Table 2.

[0117] Table 2

[0118] According to the transaction sorting result, the item sets in each transaction are inserted to construct a corresponding multi-level canonical-order tree, as shown in FIG. 4. FIG. 4 shows the structure of the canonical-order tree, which has 5 paths, the first of which contains four nodes corresponding to item sets a, c, f, and h. As can be seen from FIG. 4, the five paths all contain item set a, which is therefore the prefix of the canonical-order tree. All transactions in the transaction table are traversed one by one, and the item set of each transaction is compared with the child nodes in the canonical-order tree structure. If the same item set exists in both the transaction and the child node, the item set is merged with the node. Otherwise, the item set of the transaction is supplemented as a new branch to the last merged node.

[0119] Step S4, determining the association rules of the candidate homologous temporary drop events corresponding to each two frequent item sets respectively according to the standard sequential tree, the association rules being used to represent that the two candidate homologous temporary drop events are in a mapping relationship.

[0120] In the formula, for each path in the standard sequential tree, the candidate homologous temporary drop events corresponding to each two frequent item sets in each path are combined to obtain the association rules.

[0121] The association rules are used to represent that the two candidate homologous temporary drop events are in a mapping relationship, as shown in the following formula (6). k k → F k}; (6)

[0122] In the formula, R k is a mapping relationship, E k ={sequence number (i), temporary drop type (i)}, and F k ={sequence number (j), temporary drop type (j)}.

[0123] It can be understood that the mapping relationship can cut the waveform characteristics of a temporary drop event between the two monitoring points i and j.

[0124] Step S5, performing waveform similarity comparison on the voltage waveforms corresponding to the two candidate homologous temporary drop events in the mapping relationship, and determining whether the two candidate homologous temporary drop events are homologous temporary drop events through the waveform similarity comparison result.

[0125] Specifically, as shown in FIG. 5, the waveform similarity comparison is performed on the voltage waveforms corresponding to the two candidate homologous temporary drop events to determine whether the two candidate homologous temporary drop events are homologous temporary drop events, including steps S501-S507.

[0126] Step S501, determining whether the temporary drop types corresponding to the two candidate homologous temporary drop events in the mapping relationship are matched.

[0127] It can be understood that if the two candidate homologous temporary drop events are caused by the same voltage temporary drop source, the temporary drop types of the two candidate homologous temporary drop events are consistent. The present embodiment determines whether the temporary drop types of the two candidate homologous temporary drop events are consistent by determining whether the temporary drop types corresponding to the two candidate homologous temporary drop events in the mapping relationship are matched.

[0128] Step S502, if it is determined that the temporary drop types corresponding to the two candidate homologous temporary drop events in the mapping relationship are not matched, filtering the transaction constituted by the two candidate homologous temporary drop events from the transaction table, and proceeding to step S2. ​

[0129] Wherein, the transaction constituted by two candidate homologous temporary events in the transaction table is filtered, and the support of each set is updated, and then the support of each set is re-determined in step S2.

[0130] In step S503, if it is judged that the temporary types corresponding to the two candidate homologous temporary events in the mapping relationship match, the voltage waveforms corresponding to the two candidate homologous temporary events in the mapping relationship are transformed to the same side of the transformer.

[0131] The transformer includes a primary side and a secondary side, and the voltage levels of different sides are not the same, so the voltage waveforms on both sides are difficult to form comparability, and therefore, the voltage waveforms corresponding to the two candidate homologous temporary events in the mapping relationship need to be transformed to the same side of the transformer.

[0132] In some embodiments, as shown in FIG. 6, the process of transforming the voltage waveforms corresponding to the two candidate homologous temporary events in the mapping relationship to the same side of the transformer includes steps S5031-S5034:

[0133] In step S5031, it is judged whether the two candidate homologous temporary events in the mapping relationship occur on the same side of the transformer.

[0134] In step S5032, if it is judged that the two candidate homologous temporary events in the mapping relationship do not occur on the same side of the transformer, the transformer type involved in the temporary propagation process is determined according to the voltage temporary types corresponding to the two candidate homologous temporary events.

[0135] For the candidate homologous temporary events and the transformer type have a mapping relationship, or the transformer type can be determined through the power grid data system, such as the transformer is divided into Y0 / Y0-12, Y / Y0-12, Y0 / Y-12, Y / Y-12, Y / -11 according to the connection form of the winding.

[0136] Y0 / Y0-12 means that both windings are connected in star (Y) form and have a neutral point (marked as 0), and there is no phase difference between the two windings (12 o'clock position, i.e. 0 degree phase difference).

[0137] Y / Y0-12 means that the high-voltage side adopts star connection, the low-voltage side adopts star connection and has a neutral point, and there is no phase difference between the two windings.

[0138] Y0 / Y-12 means that the high-voltage side is star-connected and has a neutral point, and the low-voltage side is star-connected but has no neutral point, and there is no phase difference between the two windings.

[0139] Y / Y-12 means that both sides are star-connected and have no neutral point, and there is no phase difference between the two windings.

[0140] Y / -11 means high-voltage side star connection, low-voltage side delta (△) connection, and there is a phase difference of 30 degrees between the two windings (-11 o'clock position).

[0141] Step S5033, determining the voltage waveform transformation matrix according to the transformer type.

[0142] wherein, for the voltage sag type of the connection form of the transformer winding being Y0 / Y0-12, the voltage does not change after being conducted through the first type of transformer, that is, the high-voltage and low-voltage sides of the transformer have the same voltage, and the voltage waveform transformation matrix T1 thereof is:

[0143] For the connection form of the transformer winding being Y / Y0-12, Y0 / Y-12 or Y / Y-12, the zero sequence component is not allowed to pass through, and the voltage waveform transformation matrix T2 thereof is:

[0144] For the connection form of the transformer winding being Y / -11, the voltage will change by 30° after being transmitted through the transformer, and the voltage waveform transformation matrix T3 thereof is:

[0145] Step S5034, transforming the voltage waveform of one of the candidate homogenous sag events by using the voltage waveform transformation matrix, wherein the transformed voltage waveform and the voltage waveform of the other candidate homogenous sag event are on the same side of the transformer.

[0146] wherein, the transformation is as formula (10):

[0147] In the formula, Ua, Ub, Uc are the a, b, c phase voltages of the primary side of the transformer, U'a, U'b, U'c are the a, b, c phase voltages of the secondary side of the transformer, and T is the voltage waveform transformation matrix corresponding to the type of the transformer. a , U b , Uc respectively are the a, b, c phase voltages of the primary side of the transformer, U'a, U'b, U'c respectively are the a, b, c phase voltages of the secondary side of the transformer, and T is the voltage waveform transformation matrix corresponding to the type of the transformer. b , U c , Uc respectively are the a, b, c phase voltages of the primary side of the transformer, U'a, U'b, U'c respectively are the a, b, c phase voltages of the secondary side of the transformer, and T is the voltage waveform transformation matrix corresponding to the type of the transformer.

[0148] Step S504, calculating the waveform similarity of the two transformed voltage waveforms based on the constrained differential dynamic time warping algorithm, and the waveform similarity is represented by distance.

[0149] Considering the possibility of sampling frequency variation between power quality monitoring devices, which may lead to waveforms of different lengths, a constraint differential-dynamic time warping (CD-DTW) algorithm is proposed. Compared with the traditional DTW algorithm, the algorithm can handle singular points, i.e., a single point on one waveform corresponds to multiple points on another waveform, and can solve the problem of pathological alignment of waveforms, and better evaluate the similarity of waveforms.

[0150] Specifically, as shown in FIG. 7, the process of calculating the similarity of the two transformed voltage waveforms based on the constraint differential-dynamic time warping algorithm includes steps S5041-S5046:

[0151] Step S5041, first-order derivatives are respectively performed on the two transformed voltage waveforms to obtain the first-order derivatives corresponding to the two voltage waveforms respectively.

[0152] Wherein, for the recorded time series length m waveform C = [c1, c2, …, c i ,…,c m ] and the time series length n waveform D = [d1, d2, …, d j ,…,d n ], the first-order derivatives thereof are calculated by using the following formula:

[0153] In the formula, c′ i is the first-order derivative of the i-th sampling point in the waveform C, and d′ j is the first-order derivative of the j-th sampling point in the waveform D.

[0154] Step S5042, the Euclidean distance between each point in the two voltage waveforms is determined according to the first-order derivatives corresponding to the two voltage waveforms respectively.

[0155] Step S5043, a cost matrix is constructed according to the Euclidean distance between each point in the two voltage waveforms.

[0156] Wherein, the cost matrix A is:

[0157] In the formula, e(c′ m ,d′ n ) = |c m -d n |, which refers to the Euclidean distance between two points.

[0158] Step S5044, a cumulative cost matrix in a preset constraint search region is constructed according to the cost matrix.

[0159] Wherein, the cumulative cost is calculated using a dynamic programming method. For each cell B(i,j) in the cumulative cost matrix B, the cumulative cost is determined by the following recursive relationship, resulting in:

[0160] In equation (14), the indices i and j should satisfy the following constraint condition: max{|(i-1)-j|,|i-j|,|i-(j-1)|}≤λ; (15)

[0161] In the equation, λ is the width of the constraint search area, and its value is generally set to half a period, i.e. 10 ms.

[0162] Step S5045, determining the global cumulative cost of the two transformed voltage waveforms according to the cumulative cost matrix within the preset constraint search area.

[0163] The global cumulative cost is generally used to describe the cumulative cost from one point to another point in the global. After calculating the cumulative cost matrix within the constraint search area, the cumulative cost of the cell (m,n) where the last row and the last column intersect is easily derived as the global cumulative cost by recording the time sequence length m of the waveform C and the time sequence length n of the waveform D, denoted as B(m,n).

[0164] Step S5046, determining the waveform similarity of the two voltage waveforms according to the global cumulative cost and the maximum time sequence length in the two voltage waveforms.

[0165] Wherein, the waveform similarity is calculated as: S(C,D)=B(m,n) / max(m,n); (16)

[0166] In the equation, S(C,D) is the waveform similarity between the waveforms C and D, B(m,n) is the global cumulative cost, and max(m,n) is the maximum time sequence length in the waveforms C and D.

[0167] Step S505, comparing the waveform similarity of the two transformed voltage waveforms with the size of the preset waveform similarity threshold.

[0168] The embodiment of the present application uses the 3σ-criterion to determine whether the waveforms C and D are homologous, i.e. the determination condition for determining whether the waveforms C and D are homologous is: S(C,D)≤μ+3σ; (17)

[0169] In the equation, μ and σ are the average value and the standard deviation of the similarity index of all homologous event waveform pairs, respectively.

[0170] Step S506, if the waveform similarity of the two transformed voltage waveforms is less than or equal to the preset waveform similarity threshold, the two candidate homologous voltage sag events are determined as homologous voltage sag events.

[0171] Step S507, if the waveform similarity of the two transformed voltage waveforms is greater than the preset waveform similarity threshold, the two candidate homologous voltage sag events are determined as not homologous voltage sag events, the transaction composed of the two candidate homologous voltage sag events is filtered out from the transaction table, and the process goes to step S2.

[0172] It should be noted that the embodiment of the present application considers the correlation between the occurrence time of the voltage sag event and the type of the voltage sag event and the waveform characteristics thereof, establishes relevant standards from multiple angles to verify the accuracy of the voltage sag homologous detection, ensures the matching accuracy, and uses the method of association rule mining to analyze the monitoring data, so that the voltage sag homologous detection method can be applied to a complex scenario in which the parameters or topology of the power grid are unknown.

[0173] It should be noted that if the two candidate homologous voltage sag events are determined as not homologous voltage sag events, other candidate homologous voltage sag events are selected in turn to perform waveform similarity evaluation with the reference voltage sag event to determine whether they are caused by the same voltage sag source.

[0174] It should be noted that considering that a new voltage sag event occurs or the daily operation reconstruction of the power distribution network changes the propagation rule of the voltage sag, the association rule can change, and the previously mined rule can become inaccurate. Therefore, the transaction composed of the two candidate homologous voltage sag events is filtered out from the transaction table, the support of each set in the transaction table is updated, the support of each set is determined again in step S2, the association rule of the voltage sag propagation is re-mined, the association rule data is updated, then the waveform similarity evaluation of the two candidate homologous voltage sag events is performed, and the voltage sag homologous detection is performed.

[0175] It should be noted that the present application screens out candidate homologous voltage sag events to form a transaction table, uses the support of each set in the transaction table to screen out a plurality of frequent item sets, constructs a standard order tree corresponding to the transaction table, thereby reducing the number of misleading transactions and improving the effectiveness of the voltage sag homologous identification. At the same time, the association rule of the candidate homologous voltage sag events corresponding to each two frequent item sets is determined through the standard order tree, the waveform similarity of the voltage waveforms corresponding to the candidate homologous voltage sag events is compared, and it is determined whether the two candidate homologous voltage sag events are homologous voltage sag events, thereby improving the accuracy of the voltage sag event homologous identification.

[0176] Based on the same inventive concept, the present application also provides an association rule mining-based voltage sag homologous identification system for implementing the association rule mining-based voltage sag homologous identification method.

[0177] The solution provided by the system is similar to the solution described in the above method, and therefore the specific definitions in one or more embodiments of the voltage sag source identification system based on association rule mining provided below can refer to the definitions of the method of voltage sag source identification based on association rule mining in the above, which will not be described here again.

[0178] As shown in FIG. 8, the voltage sag source identification system based on association rule mining provided by the embodiment of the application comprises:

[0179] The transaction construction module 100 is configured to sequentially group the multiple event groups into respective transactions corresponding to the event groups, and form a transaction table, wherein the event groups comprise two different candidate source sag events.

[0180] The item set screening module 200 is configured to screen multiple frequent item sets from the transaction table according to the support degree of each item set, wherein the item set is a candidate source sag event.

[0181] The tree construction module 300 is configured to construct a standard order tree corresponding to the transaction table according to the support degree of the multiple frequent item sets, wherein the standard order tree comprises the multiple frequent item sets.

[0182] The association mining module 400 is configured to determine the association rule of the candidate source sag events corresponding to each two frequent item sets according to the standard order tree, wherein the association rule is used to represent that the two candidate source sag events are in a mapping relationship.

[0183] The source identification module 500 is configured to perform waveform similarity comparison on the voltage waveforms corresponding to the two candidate source sag events in the mapping relationship, and determine whether the two candidate source sag events are source sag events according to the waveform similarity comparison result.

[0184] In some embodiments, the system comprises:

[0185] The reference sorting module is configured to sort the multiple voltage sag events according to the occurrence time, and select the multiple voltage sag events as sag reference events according to the sorting.

[0186] The time difference calculation module is configured to calculate the time difference between each sag reference event and each voltage sag event other than the sag reference event for each sag reference event.

[0187] The candidate source screening module is configured to screen the sag reference events and the voltage sag events with a time difference less than a preset time difference threshold as candidate source sag events, and group the candidate source sag events into an event group.

[0188] In some embodiments, the item set screening module 200 is specifically configured to take the candidate homologous temporary reduction events in each transaction in the transaction table as item sets, and determine the frequency of each item set appearing in the transaction table as the support degree of each item set; and is further configured to compare the support degree of each item set with the size of the preset support degree threshold, and screen out the item set with the support degree greater than the preset support degree threshold as the frequent item set.

[0189] In some embodiments, the preset support degree threshold is determined according to the number of association rules mined under a plurality of different support degree thresholds; and the preset support degree threshold is determined according to the number of association rules mined under a plurality of different support degree thresholds, comprising:

[0190] initially setting the support degree threshold initial value;

[0191] taking the support degree threshold initial value as the current support degree threshold, and determining the number of association rules mined under the current support degree threshold;

[0192] taking the product of the current support degree threshold and the number of association rules mined under the current support degree threshold as the optimization target, and constructing the objective function;

[0193] iteratively optimizing the objective function, and iteratively updating the current support degree threshold until convergence, and outputting the converged current support degree threshold as the preset support degree threshold.

[0194] In some embodiments, the tree construction module 300 is specifically configured to arrange the plurality of frequent item sets in descending order according to the support degrees of the plurality of frequent item sets; and construct a multi-level standard order tree according to the frequent item sets arranged in descending order, the standard order tree comprising a plurality of paths, each path comprising a plurality of nodes, each node being arranged in a multi-level structure in descending order, each path being a group of transactions sharing the same prefix node, each node corresponding to a frequent item set, and the prefix node being the frequent item set with the highest support degree.

[0195] In some embodiments, the association mining module 400 is specifically configured to, for each path in the standard order tree, combine the candidate homologous temporary reduction events respectively corresponding to each two frequent item sets in each path to obtain the association rule.

[0196] In some embodiments, the homologous identification module 500 is specifically configured to judge whether the temporary reduction types respectively corresponding to the two candidate homologous temporary reduction events in the mapping relationship match;

[0197] If it is judged that the temporary reduction types respectively corresponding to the two candidate homologous temporary reduction events in the mapping relationship do not match, then the transactions composed of the two candidate homologous temporary reduction events are filtered out from the transaction table, and the step of screening the plurality of frequent item sets according to the support degrees of each item set in the transaction table is turned to, and the item set is the candidate homologous temporary reduction event.

[0198] if it is judged that the types of the two candidate homologous sag events corresponding to the mapping relationship match, then the voltage waveforms corresponding to the two candidate homologous sag events are transformed to the same side of the transformer;

[0199] The waveform similarity of the two transformed voltage waveforms is calculated based on the constrained differential dynamic time warping algorithm, and the waveform similarity is represented by a distance.

[0200] The waveform similarity of the two transformed voltage waveforms is compared with a preset waveform similarity threshold.

[0201] If the waveform similarity of the two transformed voltage waveforms is less than or equal to the preset waveform similarity threshold, then the two candidate homologous sag events are determined to be homologous sag events.

[0202] If the waveform similarity of the two transformed voltage waveforms is greater than the preset waveform similarity threshold, then the two candidate homologous sag events are determined not to be homologous sag events, the transaction composed of the two candidate homologous sag events is filtered out from the transaction table, and a plurality of frequent item sets are selected according to the support degree of each item set in the transaction table, and the item set is a candidate homologous sag event.

[0203] The voltage waveforms corresponding to the two candidate homologous sag events in the mapping relationship are transformed to the same side of the transformer, including:

[0204] It is judged whether the two candidate homologous sag events in the mapping relationship occur on the same side of the transformer.

[0205] If it is judged that the two candidate homologous sag events in the mapping relationship do not occur on the same side of the transformer, then the type of the transformer involved in the voltage sag propagation process is determined according to the voltage sag types corresponding to the two candidate homologous sag events.

[0206] The voltage waveform transformation matrix is determined according to the type of the transformer.

[0207] The voltage waveform of one of the candidate homologous sag events is transformed using the voltage waveform transformation matrix, and the transformed voltage waveform and the voltage waveform of the other candidate homologous sag event are on the same side of the transformer.

[0208] The waveform similarity of the two transformed voltage waveforms is calculated based on the constrained differential dynamic time warping algorithm, and the waveform similarity is represented by a distance.

[0209] The first-order derivatives of the two transformed voltage waveforms are calculated, respectively, to obtain the first-order derivatives corresponding to the two voltage waveforms.

[0210] determine the Euclidean distance between each point in the two voltage waveforms according to the first derivative corresponding to the two voltage waveforms respectively;

[0211] construct the cost matrix according to the Euclidean distance between each point in the two voltage waveforms;

[0212] construct the accumulated cost matrix in the preset constraint search region according to the cost matrix;

[0213] determine the global accumulated cost of the two transformed voltage waveforms according to the accumulated cost matrix in the preset constraint search region;

[0214] determine the waveform similarity of the two voltage waveforms according to the global accumulated cost and the maximum time sequence length in the two voltage waveforms.

[0215] As shown in FIG. 9, the embodiment of the present application further provides an electronic device, the electronic device 10 comprises a memory 20 and a processor 30;

[0216] The memory 20 is used for storing a program;

[0217] The processor 30 executes the program to realize the steps of the voltage sag same source identification method based on the association rule mining in any one of the above-mentioned embodiments.

[0218] The embodiment of the present application further provides a computer readable storage medium, the storage medium stores a computer program, and the computer program is executed by the processor to realize the steps of the voltage sag same source identification method based on the association rule mining in any one of the above-mentioned embodiments.

[0219] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-mentioned system, electronic device and computer storage medium can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here.

[0220] It should be noted that the terms "include" and "have" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover the non-exclusive inclusion, for example, the process, method, system, product or device including a series of steps or units does not have to be limited to the clearly listed steps or units, but can include other steps or units that are not clearly listed or inherent to the process, method, product or device.

[0221] In several embodiments provided by the present application, it can be understood that each block in the flowchart or block diagram can represent a module, a segment or a portion of code which includes one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or in the reverse order, depending on the functionality involved.

[0222] In several embodiments provided by the present application, it can be understood that the disclosed system, electronic device, computer storage medium and method can be implemented in other manners. For example, the described device embodiments are merely illustrative, and the division of units can be different from the embodiment described above. For example, the units can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0223] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e. can be located in one place or distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments.

[0224] In addition, each functional unit in the embodiments of the present application can be integrated in a processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware, or in the form of a software functional unit.

[0225] If the integrated unit is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for executing all or part of the steps of the embodiments of the present application by a computer device (which can be a personal computer, a server, or a network device, etc.). The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (English full name: Read-Only Memory, English abbreviation: ROM), a random access memory (English full name: Random Access Memory, English abbreviation: RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0226] The above embodiments are only used to illustrate the technical solutions of the present application, not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying the same origin of voltage sags based on association rule mining, characterized in that, The method comprises the following steps: According to the support degree of each item set in the transaction table, a plurality of frequent item sets are screened out, and the item set is a candidate homologous temporary drop event; According to the support degree of each item set in the transaction table, a plurality of frequent item sets are screened out, and the item set is a candidate homologous temporary drop event; According to the support degree of each item set in the transaction table, a plurality of frequent item sets are screened out, and the item set is a candidate homologous temporary drop event; According to the support degree of each item set in the transaction table, a plurality of frequent item sets are screened out, and the item set is a candidate homologous temporary drop event; According to the support degree of each item set in the transaction table, a plurality of frequent item sets are screened out, and the item set is a candidate homologous temporary drop event. 2.The voltage sag common source identification method based on association rule mining according to claim 1, wherein, Before the step of sequentially forming a transaction corresponding to each event group according to a plurality of event groups, and forming a transaction table, the event group comprising two different candidate homologous temporary drop events, the method further comprises the steps of: Sorting a plurality of voltage sag events according to occurrence time, and selecting a plurality of the voltage sag events as sag reference events according to the sorting; For each sag reference event, calculating the time difference between the sag reference event and each voltage sag event other than the sag reference event; Screening out the sag reference event and the voltage sag event with a time difference less than a preset time difference threshold value as a candidate homologous temporary drop event, and grouping the candidate homologous temporary drop events into an event group. 3.The voltage sag common source identification method based on association rule mining according to claim 1, characterized in that, The step of screening out a plurality of frequent item sets according to the support degree of each item set in the transaction table, and the item set being a candidate homologous temporary drop event, comprises the steps of: Taking the candidate homologous temporary drop event in each transaction in the transaction table as an item set, and determining the frequency of occurrence of each item set in the transaction table as the support degree of each item set; For the support degree of each item set, comparing the support degree with a preset support degree threshold value, and screening out the item set with a support degree greater than the preset support degree threshold value as a frequent item set.

4. The voltage sag common cause identification method based on association rule mining according to claim 3, characterized in that, The preset support degree threshold value is determined according to the number of association rules mined under a plurality of different support degree threshold values; the step of determining the preset support degree threshold value according to the number of association rules mined under a plurality of different support degree threshold values comprises the steps of: Initializing the support degree threshold value initial value; Taking the support degree threshold value initial value as the current support degree threshold value, and determining the number of association rules mined under the current support degree threshold value; Taking the product of the current support degree threshold value and the number of association rules mined under the current support degree threshold value as the optimization target, and constructing the objective function; Iteratively optimizing the objective function, and iteratively updating the current support degree threshold value until convergence, and outputting the converged current support degree threshold value as the preset support degree threshold value.

5. The method of claim 1, wherein the method is characterized by, The step of constructing a standard sequential tree corresponding to the transaction table according to the support degree of a plurality of the frequent item sets, and the standard sequential tree comprising a plurality of the frequent item sets, comprises the steps of: arranging the plurality of frequent item sets in descending order according to the support degrees of the plurality of frequent item sets; constructing a multi-level standard sequential tree according to the frequent item sets arranged in descending order, the standard sequential tree comprising a plurality of paths, each path comprising a plurality of nodes, each node being arranged in descending order as a multi-level structure, each path being a group of transactions sharing the same prefix node, each node corresponding to a frequent item set, and the prefix node being a frequent item set with the highest support degree. 6.The voltage sag common cause identification method based on association rule mining according to claim 1, wherein, The step of determining the association rules of the candidate homologous temporary drop events corresponding to each two frequent item sets respectively according to the standard sequential tree comprises: combining the candidate homologous temporary drop events corresponding to each two frequent item sets respectively in each path of the standard sequential tree to obtain the association rules. 7.The voltage sag common origin identification method based on association rule mining according to claim 1, wherein, The step of comparing the voltage waveforms corresponding to the two candidate homologous temporary drop events respectively in a mapping relationship through waveform similarity to determine whether the two candidate homologous temporary drop events are homologous temporary drop events comprises: determining whether the temporary drop types corresponding to the two candidate homologous temporary drop events respectively in a mapping relationship are matched; if the temporary drop types corresponding to the two candidate homologous temporary drop events respectively in a mapping relationship are not matched, filtering out the transactions formed by the two candidate homologous temporary drop events from the transaction table, and proceeding to the step of screening a plurality of frequent item sets from the transaction table according to the support degrees of the item sets in the transaction table, wherein the item set is a candidate homologous temporary drop event; if the temporary drop types corresponding to the two candidate homologous temporary drop events respectively in a mapping relationship are matched, transforming the voltage waveforms corresponding to the two candidate homologous temporary drop events respectively in a mapping relationship to the same side of the transformer; calculating the waveform similarity of the two transformed voltage waveforms based on the constrained differential dynamic time warping algorithm, wherein the waveform similarity is represented by distance; comparing the waveform similarity of the two transformed voltage waveforms with a preset waveform similarity threshold value; if the waveform similarity of the two transformed voltage waveforms is less than or equal to the preset waveform similarity threshold value, determining that the two candidate homologous temporary drop events are homologous temporary drop events; if the waveform similarity of the two transformed voltage waveforms is greater than the preset waveform similarity threshold value, determining that the two candidate homologous temporary drop events are not homologous temporary drop events, filtering out the transactions formed by the two candidate homologous temporary drop events from the transaction table, and proceeding to the step of screening a plurality of frequent item sets from the transaction table according to the support degrees of the item sets in the transaction table, wherein the item set is a candidate homologous temporary drop event.

8. The voltage sag common cause identification method based on association rule mining according to claim 7, characterized in that, The step of transforming the voltage waveforms corresponding to the two candidate homologous temporary drop events respectively in a mapping relationship to the same side of the transformer comprises: determining whether the two candidate homologous temporary drop events in a mapping relationship occur on the same side of the transformer; if the two candidate homologous temporary drop events in a mapping relationship do not occur on the same side of the transformer, determining the transformer type involved in the temporary drop propagation process according to the voltage temporary drop types corresponding to the two candidate homologous temporary drop events respectively. determining a voltage waveform transformation matrix according to the transformer type; transforming a voltage waveform of one of the candidate homologous sag events using the voltage waveform transformation matrix, wherein the transformed voltage waveform and a voltage waveform of another of the candidate homologous sag events are on the same side of a transformer. 9.The voltage sag common cause identification method based on association rule mining according to claim 7, wherein, The step of performing waveform similarity calculation on the two transformed voltage waveforms using the constrained differential dynamic time warping algorithm includes: performing first-order derivation on the two transformed voltage waveforms respectively to obtain first-order derivatives corresponding to the two voltage waveforms respectively; determining Euclidean distances between points in the two voltage waveforms according to the first-order derivatives corresponding to the two voltage waveforms respectively; constructing a cost matrix according to the Euclidean distances between the points in the two voltage waveforms; constructing a cumulative cost matrix in a preset constraint search region according to the cost matrix; determining a global cumulative cost of the two transformed voltage waveforms according to the cumulative cost matrix in the preset constraint search region; determining the waveform similarity of the two voltage waveforms according to the global cumulative cost and a maximum time sequence length of the two voltage waveforms.

10. A voltage sag common source identification system based on association rule mining, characterized by, The method includes: a transaction construction module configured to sequentially group a plurality of event groups into transactions corresponding to the event groups respectively and form a transaction table, wherein the event groups include two different candidate homologous sag events; an item set screening module configured to screen a plurality of frequent item sets from the transaction table according to support sizes of item sets in the transaction table, wherein the item sets are candidate homologous sag events; a tree construction module configured to construct a standard order tree corresponding to the transaction table according to support sizes of the frequent item sets, wherein the standard order tree includes the frequent item sets; an association mining module configured to determine association rules of candidate homologous sag events corresponding to each two of the frequent item sets according to the standard order tree, wherein the association rules are used to represent that the two candidate homologous sag events are in a mapping relationship; a homologous identification module configured to perform waveform similarity comparison on voltage waveforms corresponding to the two candidate homologous sag events in the mapping relationship, and determine whether the two candidate homologous sag events are homologous sag events according to a waveform similarity comparison result.

Images