Intelligent on-off control system for transformer in power distribution area

Abstract

The invention discloses an intelligent switching control system for a transformer in a power distribution area, and relates to the technical field of transformer switching control. A transformer detection terminal is arranged to periodically collect load data of the transformer in the power distribution area in a target area in a target power grid; a user side processing unit periodically collects user side load data in a target area, the user side processing unit processes the user side load data, a basic sequence and a recursive sequence are determined based on frequencies of different arrays to be verified in the user side load data in the processing process, a first compression mapping chain and a second compression mapping chain are constructed, and the first compression mapping chain and the second compression mapping chain are compressed. A mapping relation is established by using a shorter character sequence based on the basic sequence and the first and second compression mapping chains thereof, so that the capacity of user side load data is greatly compressed, the transmission rate of the user side load data in a network is further shortened, a switching strategy terminal can receive the data more quickly and input the data into a model, and the user side load data transmission efficiency is improved. And the generation rate of the commissioning and quitting strategy is accelerated.

Description

Technical Field

[0001] The invention relates to the technical field of transformer switching control, and in particular to an intelligent switching control system for transformers in a distribution station area. Background Art

[0002] In the stable operation of the power system, the reasonable switching on and off of the transformers in the distribution area is undoubtedly a key link to ensure the quality and efficiency of power supply, and its importance cannot be underestimated. At present, in the planning and formulation process of the switching on and off strategy of the distribution area transformer, an advanced distribution transformer load forecasting model has been introduced. Specifically, the LSTM-RNN architecture combining the long short-term memory network (LSTM) and the recurrent neural network (RNN) is used to construct a short-term load forecasting model for the distribution transformer area. The core function of this model is to accurately predict the maximum load condition of the distribution transformer area in a short period of time in the future, so that a scientific and reasonable switching on and off strategy can be formulated based on this accurate maximum load data.

[0003] However, to ensure the accuracy of load forecasting, a large amount of data on related parameters must be collected to assist the model in making accurate forecasting and analysis. Among the many parameters, there is a crucial parameter data that needs to be collected from the user side. However, the actual situation is that the amount of data on the user side is extremely large, and even the data generated in a short period of time is quite considerable. The current processing method for this type of data is to first use compression technology to transmit it, which takes into account the efficiency of the data transmission process, because a large amount of data transmission will take up more network resources and time. At the same time, since the user-side data may contain sensitive information such as user privacy, the transmitted data is encrypted using a key to ensure data security. However, this operation method of compression and encryption, respectively, inevitably increases the complexity and time cost of data processing. In actual operation, the cumbersome data processing process makes the whole process take longer, which further leads to a delay in the generation time of the investment and withdrawal strategy. This may have a certain impact on the timely response and efficient operation of the power system. For example, in emergency situations such as peak power consumption, the investment and withdrawal status of the transformer may not be adjusted in time, thereby affecting the stability and reliability of power supply.

[0004] In order to solve the above problems, the present invention proposes a solution. Summary of the invention

[0005] The purpose of the present invention is to provide a distribution substation transformer intelligent switching control system to solve the problems raised in the above background technology;

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] An intelligent switching control system for transformers in distribution stations, characterized in that:

[0008] At each load collection period, the transformer monitoring terminal generates the device-side load data of the load collection period according to all load data monitored in the load collection period;

[0009] At each load collection period, the user-side collection unit collects the power load data of the target area within the load collection period, and performs binary conversion on the collected data to generate the user-side load data of the load collection period;

[0010] After receiving the user side load data of a load collection period, the user side processing unit processes the data in accordance with the preset preprocessing rules to obtain a number of user side data packets corresponding to the load collection period, and the first and second compression mapping chains;

[0011] After receiving a plurality of user-side data packets of the load collection period and the first and second compression mapping chains, the investment and withdrawal strategy terminal restores the plurality of user-side data packets according to the first and second compression mapping chains to obtain the user-side load data of the load collection period;

[0012] The restored user-side load data of the load collection cycle and the received device-side load data of the load collection cycle are input into the distribution transformer load prediction model to obtain the maximum load of the target area of ​​the next load collection cycle, and the corresponding investment and withdrawal strategy is matched according to the maximum load.

[0013] Furthermore, the preprocessing rules of several user-side data packets, the first compression mapping chain and the second compression mapping chain in a load collection cycle are obtained as follows:

[0014] S11: specifying a cutting step length of 8, cutting the user-side load data in order from left to right to obtain a plurality of groups of cutting arrays, and then marking all the obtained groups of cutting arrays as A1, A2...Aa, a≥1, from left to right according to the position of each group of cutting arrays in the user-side load data before cutting;

[0015] S12: According to the preset first screening rule, a number of cutting arrays are selected from the cutting arrays A1, A2...Aa as the arrays to be reviewed C1, C2...Cb, and the frequency quantity of the corresponding arrays to be reviewed is calculated during the selection process:

[0016] S13: Extract the first 4 characters and the last 4 characters of the array C1 to be examined from left to right as a set of disassembly sequences of the array C1 to be examined. According to the above, disassemble the arrays C2, C3, ..., Cb to be examined in turn to obtain two sets of disassembly sequences corresponding thereto;

[0017] S14: selecting basic sequences F1, F2, F3, and F4 from two groups of disassembly sequences of the arrays C1, C2, ..., Cb to be examined according to a preset second screening rule;

[0018] S15: Screening in sequence according to a preset third screening rule to obtain a first progressive sequence and a second progressive sequence of the basic sequences F1, F2, F3, and F4;

[0019] S16: traverse the basic sequences F1, F2, F3, F4 and the first progressive sequence and the second progressive sequence thereof, remove the numbers corresponding to the four-bit binary numbers that are respectively consistent with the basic sequences F1, F2, F3, F4 and the first progressive sequence and the second progressive sequence thereof from the numbers 0 to 15, and then mark the four-bit binary numbers of the four numbers that are not removed from the numbers 0 to 15 as recursive sequences, and mark them as F5, F6, F7, F8 respectively;

[0020] S17: construct a first compression mapping chain according to the basic sequences F1, F2, F3, F4 and their first progressive sequence and second progressive sequence; construct a second compression mapping chain according to the recursive sequences F5, F6, F7, F8; after the second compression mapping chain is constructed, use character 1 as the chain identifier of the first compression mapping chain, and use character 0 as the chain identifier of the second compression mapping chain;

[0021] S18: Generate a compressed security sequence of the to-be-examined array C1 according to a preset generation rule;

[0022] S19: according to S18, the compressed safety sequences of the arrays C2, C3, ..., Cb to be reviewed are calculated in sequence, and then b load processing packages of the load collection cycle are generated according to the compressed safety sequences of the arrays C1, C2, ..., Cb to be reviewed.

[0023] Furthermore, the first screening rule for selecting several cut arrays as the arrays to be examined C1, C2...Cb and calculating their frequency quantities is as follows:

[0024] S121: De-duplicate the cut arrays A1, A2...Aa, and re-mark all the remaining cut arrays after de-duplicate as arrays to be reviewed, marked as B1, B2...Bb, 1≤b≤a;

[0025] S122: Obtain the total number of the cut arrays that are consistent with the array to be examined B1 from the cut arrays A1, A2...Aa, and mark the total number as the frequency quantity of the array to be examined B1, wherein consistent means that the order and content of the characters constituting the cut arrays are the same as the characters constituting the array to be examined;

[0026] S123: According to S122, the frequency quantities of the pending arrays B2, B3, ... Bb are obtained in sequence, and then the corresponding pending arrays are re-labeled as C1, C2, ... Cb in descending order of the frequency quantities.

[0027] Beneficial effects of the present invention:

[0028] (1) The present invention sets a transformer detection terminal to periodically collect the load data of the transformer in the distribution area of ​​the target area in the target power grid, and the user-side processing unit periodically collects the user-side load data in the target area, and the user-side processing unit processes the user-side load data. During the processing, the basic sequence and the recursive sequence are determined based on the frequency of different arrays to be reviewed in the user-side load data, and the first and second compression mapping chains are constructed. The mapping relationship is constructed based on the constructed first and second compression mapping chains. The mapping relationship is selected from the characters 0, 1, 00, 01, 10, and 11. At the same time, each basic sequence in the first compression mapping chain corresponds to the first and second recursive sequences. Based on the basic sequence and its first and second compression mapping chains, 1 or 0 can directly replace the first and second compression mapping chains corresponding to the basic sequence. In this way, the capacity of the user-side load data is greatly compressed, and the transmission rate of the user-side load data in the network is further shortened, so that the investment and withdrawal strategy terminal can receive data faster and input it into the model, thereby accelerating the generation rate of the investment and withdrawal strategy;

[0029] (2) The present invention introduces the first and second compression mapping chains in the process of compressing the user-side load data, and obtains corresponding matching identifiers based on the relationship between the two groups of disassembled series in the array to be reviewed in the first and second compression mapping chains. In this way, the security of user-side load data transmission is guaranteed. A third party cannot know the clear mapping relationship and will not be able to crack the user-side load data. In addition, this method combines the compression and encryption operation mode, reduces the complexity and time cost of data processing, and further shortens the generation time of the investment and withdrawal strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will be further described below in conjunction with the accompanying drawings.

[0031] Figure 1 is a system block diagram of the present invention;

[0032] Figure 2 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] like Figure 1 , 2 As shown, a distribution substation area transformer intelligent switching control system includes a transformer monitoring terminal, a switching strategy terminal and a load side terminal;

[0035] The transformer monitoring terminal is used to monitor the load data of the target transformer in the target power grid in real time;

[0036] The transformer monitoring terminal generates the device-side load data of the load collection period according to all the load data monitored in the load collection period at intervals of one load collection period, and transmits the device-side load data to the input / output strategy terminal, wherein the load data includes real-time current value, average current value, active power, reactive power, apparent power and load rate;

[0037] A load-side terminal, used to periodically collect power load data on the user side in the target area, wherein the load measurement terminal includes a user-side collection unit and a user-side processing unit;

[0038] The user-side collection unit collects power load data of the target area within the load collection period at intervals of one load collection period, and performs binary conversion on the collected data to generate user-side load data of the load collection period, wherein the power load data includes real-time power, current, voltage, load rate, power factor and power load, and the user side includes all power users and power enterprises in the target area that are under the control of the target power grid;

[0039] Each time user-side load data of the load collection cycle is generated, the user-side load data is transmitted to a user-side processing unit;

[0040] After receiving the user side load data of a load collection period, the user side processing unit processes the data in accordance with the preset preprocessing rules to obtain a number of user side data packets corresponding to the load collection period, and the first and second compression mapping chains;

[0041] The preprocessing rules of several user-side data packets, the first and second compression mapping chains in a load collection cycle are as follows:

[0042] S11: specifying a cutting step length of 8, cutting the user-side load data in order from left to right to obtain a plurality of groups of cutting arrays, and then marking all the obtained groups of cutting arrays as A1, A2...Aa, a≥1, from left to right according to the position of each group of cutting arrays in the user-side load data before cutting;

[0043] It should be noted here that, in the process of cutting the user-side load data, if the total number of remaining characters is less than 8, all the remaining characters are temporarily stored as a group of supplementary arrays;

[0044] S12: According to the preset first screening rule, a number of cutting arrays are selected from the cutting arrays A1, A2...Aa as the arrays to be reviewed C1, C2...Cb. During the selection process, the frequency of the corresponding arrays to be reviewed is calculated. The screening rule is as follows:

[0045] S121: De-duplicate the cut arrays A1, A2...Aa, and re-mark all the remaining cut arrays after de-duplicate as arrays to be reviewed, marked as B1, B2...Bb, 1≤b≤a;

[0046] S122: Obtain the total number of the cut arrays that are consistent with the array to be examined B1 from the cut arrays A1, A2...Aa, and mark the total number as the frequency quantity of the array to be examined B1, wherein consistent means that the order and content of the characters constituting the cut arrays are the same as the characters constituting the array to be examined;

[0047] For example, the cut array A1 is 11000101, and the pending array B1 is 11000101. The characters constituting these two arrays are consistent in both character order and content.

[0048] S123: according to S122, the frequency quantities of the arrays to be reviewed B2, B3, ... Bb are obtained in sequence, and then the corresponding arrays to be reviewed are re-marked as C1, C2, ... Cb in descending order of the frequency quantities;

[0049] S13: extracting the first 4 characters and the last 4 characters of the array C1 to be examined in order from left to right as a set of disassembly sequences of the array C1 to be examined. In other words, extracting the first 4 characters of the array C1 to be examined in order from left to right as a set of disassembly sequences of the array C1 to be examined, and taking the remaining 4 characters not extracted as another set of disassembly sequences of the array C1 to be examined;

[0050] According to the above, the arrays C2, C3, ..., Cb to be reviewed are disassembled in turn to obtain two corresponding disassembly sequences;

[0051] S14: According to the preset second screening rule, basic sequences F1, F2, F3, and F4 are selected from the two groups of disassembly sequences of the arrays C1, C2, ..., Cb to be reviewed. The second screening rule is as follows:

[0052] S141: According to the two groups of disassembly sequences corresponding to the arrays C1, C2, ..., Cb to be reviewed, four groups of disassembly sequences are randomly selected as a group of disassembly combinations to obtain a plurality of groups of disassembly combinations, and a plurality of groups of disassembly combinations that do not meet the preset selection conditions are eliminated, and all the remaining groups of disassembly combinations are re-calibrated as standard disassembly combinations;

[0053] The selection conditions are as follows: any two disassembly sequences in a set of standard disassembly combinations are inconsistent;

[0054] S142: Mark the two groups of disassembly sequences in the pending array C1 as D1 and D2 respectively, obtain the frequency quantities of all pending arrays that contain the disassembly sequence D1 and do not contain the disassembly sequence D2 from the pending arrays C2, C3, ..., Cb, and calculate the multidimensional frequency quantity F1 of the disassembly sequence D1 using the formula F1=Ee×ɑe+E2×ɑ2+...+Ee×ɑe, where E1, E2, ..., Ee refer to all the pending arrays that contain the disassembly sequence D1 and do not contain the disassembly sequence D2, and ɑ1, ɑ2, ..., ɑe are preset proportion coefficients of the corresponding pending arrays;

[0055] It should be noted here that the subscripts marked by the management personnel according to the pending arrays C1, C2...Cb correspond to the preset proportion coefficients set for each pending array. The smaller the subscript, the larger the value of the corresponding preset proportion coefficient. The preset proportion coefficients corresponding to the pending arrays C1, C2...Cb are all decimals, and the sum of their addition is 1.

[0056] Similarly, according to the frequency quantities of all the arrays to be reviewed that contain the disassembly sequence D2 and do not contain the disassembly sequence D1 obtained in the arrays to be reviewed C2, C3, ..., Cb, the multi-dimensional frequency quantity F2 of the disassembly sequence D2 is calculated;

[0057] S143: According to S142, the multi-dimensional frequency quantities of the two groups of disassembly sequences corresponding to the arrays C2, C3, ..., Cb to be reviewed are calculated in sequence;

[0058] S144: for any set of standard decomposition combinations obtained in S141, the multi-dimensional frequency quantities of all the group decomposition sequences contained therein are added to obtain the basic selected evaluation index of the corresponding standard decomposition combination, and the standard decomposition combination corresponding to the basic selected evaluation index with the largest value is selected from the basic selected evaluation indexes of all the groups of standard decomposition combinations obtained by calculation as the basic decomposition combination;

[0059] The four groups of disassembly sequences included in the basic disassembly combination are re-labeled as basic sequences, and the corresponding basic sequences are labeled F1, F2, F3, and F4 in descending order of multi-dimensional frequency;

[0060] S14: The first progressive sequence and the second progressive sequence of the basic sequences F1, F2, F3, and F4 are sequentially screened according to a preset third screening rule. The third screening rule is as follows:

[0061] S141: extracting all arrays to be reviewed that contain the basic sequence F1 but do not contain the basic sequences F2, F3, and F4 from the arrays to be reviewed C1, C2, ..., Cb, and then selecting the arrays to be reviewed with the largest frequency and the second largest frequency from all the extracted arrays to be reviewed;

[0062] For the extracted array to be reviewed with the largest frequency, the disassembly sequence that is inconsistent with the basic sequence F1 is used as the first progressive sequence of the basic sequence F1 from the two groups of disassembly sequences of the array to be reviewed;

[0063] For the extracted array to be reviewed with the second largest frequency, the disassembly sequence that is inconsistent with the basic sequence F1 is used as the second progressive sequence of the basic sequence F1 from the two groups of disassembly sequences of the array to be reviewed;

[0064] S142: Calculate and obtain the first progressive sequence and the second progressive sequence of the basic sequences F2, F3, and F4 in sequence according to S141;

[0065] S15: traverse the basic sequences F1, F2, F3, F4 and the first progressive sequence and the second progressive sequence thereof, remove the numbers corresponding to the four-bit binary numbers that are respectively consistent with the basic sequences F1, F2, F3, F4 and the first progressive sequence and the second progressive sequence thereof from the numbers 0 to 15, and then mark the four-bit binary numbers of the four numbers that are not removed from the numbers 0 to 15 as recursive sequences, and mark them as F5, F6, F7, F8 respectively;

[0066] S16: constructing a first compression mapping chain according to the basic sequences F1, F2, F3, F4 and their first progressive sequence and second progressive sequence, wherein the first compression mapping chain has 12 nodes arranged from left to right, and the 12 nodes are mapped from left to right to the basic sequence F1, the first progressive sequence of the basic sequence F1, the second progressive sequence of the basic sequence F1, the basic sequence F2, the first progressive sequence of the basic sequence F2, the second progressive sequence of the basic sequence F2, ..., the basic sequence F4, the first progressive sequence of the basic sequence F4, and the second progressive sequence of the basic sequence F4;

[0067] A second compression mapping chain is constructed according to the recursive sequences F5, F6, F7, and F8, wherein four nodes are arranged in sequence from left to right in the second mapping compression chain, and the four nodes are mapped to the recursive sequences F5, F6, F7, and F8 in sequence from left to right;

[0068] After the first and second compression mapping chains are constructed, character 1 is used as the chain identifier of the first compression mapping chain, and character 0 is used as the chain identifier of the second compression mapping chain;

[0069] In the first compression mapping chain, the two-digit binary number 00 is used as the sequence identifier of the basic sequence F1, 01 is used as the sequence identifier of the basic sequence F2, ..., and 11 is used as the sequence identifier of the basic sequence F4. In other words, the two-digit binary numbers 00, 01, 10, and 11 are used as the sequence identifiers of the basic sequences F1, F2, F3, and F4 in sequence;

[0070] In the second compression mapping chain, character 0 is used as the sequence identifier of the recursive sequence F5, character 1 is used as the sequence identifier of the recursive sequence F6, two-digit binary number 00 is used as the sequence identifier of the basic sequence F7, and two-digit binary number 01 is used as the sequence identifier of the basic sequence F8;

[0071] At this time, the 16 nodes are associated with the sequence identifiers of the basic sequence, the first progressive sequence, the second progressive sequence, or the recursive sequence mapped thereto respectively;

[0072] S17: Generate a compressed security sequence of the to-be-examined array C1 according to a preset generation rule, the generation rule is as follows:

[0073] S171: Generate a matching identifier of the disassembly sequence D1 of the to-be-approved array C1 according to a preset first generation rule. The first generation rule is as follows:

[0074] SS11: If there is a basic sequence of a node mapping in the first compression mapping chain that is consistent with the disassembly sequence D1, obtain the sequence identifier of the node, use the sequence identifier as the matching identifier of the disassembly sequence D1, and use the chain identifier of the first compression mapping chain as the chain identifier of the disassembly sequence D1;

[0075] SS12: If there is a first progressive sequence of a node mapping in the first compression mapping chain that is consistent with the disassembly sequence D1, then sequentially obtain the sequence identifier of the base sequence corresponding to the first progressive sequence and the sequence identifier of the node, and concatenate the two obtained sequence identifiers in the order of acquisition to obtain the matching identifier of the disassembly sequence D1, and use the chain identifier of the first compression mapping chain as the chain identifier of the disassembly sequence D1:

[0076] SS13: If there is a second progressive sequence of a node mapping in the first compression mapping chain that is consistent with the disassembly sequence D1, then sequentially obtain the sequence identifier of the base sequence corresponding to the second progressive sequence and the sequence identifier of the node, and concatenate the two obtained sequence identifiers in the order of acquisition to obtain the matching identifier of the disassembly sequence D1, and use the chain identifier of the first compression mapping chain as the chain identifier of the disassembly sequence D1;

[0077] SS14: If there is a recursive sequence of a node mapping in the second compression mapping chain that is consistent with the disassembly sequence D1, obtain the sequence identifier of the node, use the sequence identifier as the matching identifier of the disassembly sequence D1, and use the chain identifier of the second compression mapping chain as the chain identifier of the disassembly sequence D1:

[0078] S172: Compare the disassembled sequences D1 and D2 for consistency. If the disassembled sequences D1 and D2 are consistent, then concatenate the chain identifier of the disassembled sequence D1 and the matching identifier of the disassembled sequence D1 to obtain the compressed safety sequence of the array C1 to be examined. Otherwise, recursively generate the compressed safety sequence of the array C1 to be examined according to the preset recursive generation rule. The recursive judgment rule is as follows:

[0079] SS21: If the node consistent with the disassembly sequence D1 is the basic sequence, and the node consistent with the disassembly sequence D2 corresponds to the first progressive sequence of the basic sequence, character 0 is used as the matching identifier of the disassembly sequence D2, and the chain identifier and matching identifier of the disassembly sequence D1 and the matching identifier of the disassembly sequence D2 are concatenated to obtain the compressed security sequence of the array C1 to be examined;

[0080] SS22: If the node consistent with the disassembly sequence D1 is the basic sequence, and the node consistent with the disassembly sequence D2 corresponds to the second progressive sequence of the basic sequence, character 1 is used as the matching identifier of the disassembly sequence D2, and the chain identifier and matching identifier of the disassembly sequence D1 and the matching identifier of the disassembly sequence D2 are concatenated to obtain the compressed security sequence of the array C1 to be examined;

[0081] SS23: If the node consistent with the disassembly sequence D1 is the basic sequence, and the node consistent with the disassembly sequence D2 is not the first or second progressive sequence of the basic sequence, then the matching identifier and chain identifier of the disassembly sequence D2 are obtained according to steps SS11 to SS14;

[0082] Then the chain identifiers of the disassembly sequences D1 and D2 are compared for consistency. If they are consistent, the chain identifier and its matching identifier of the disassembly sequence D1 and the matching identifier of the disassembly sequence D2 are concatenated to obtain the compressed security sequence of the array C1 to be examined. The concatenation order is the order of the chain identifier of the disassembly sequence D1, the matching identifier, and the disassembly sequence D2.

[0083] If they are inconsistent, the chain identifier and its matching identifier of the disassembled sequence D1, and the chain identifier and its matching identifier of the disassembled sequence D2 are concatenated to obtain the compressed security sequence of the array C1 to be examined. The concatenation order is the chain identifier and matching identifier of the disassembled sequence D1, and the chain identifier and matching identifier of the disassembled sequence D2.

[0084] SS23: If the sequence that is consistent with the disassembly sequence D1 is not the basic sequence, then the matching identifier and chain identifier of the disassembly sequence D2 are obtained according to the steps from SS11 to SS14. At this time, the chain identifiers of the disassembly sequences D1 and D2 are compared for consistency. If they are consistent, the chain identifier and its matching identifier of the disassembly sequence D1 and the matching identifier of the disassembly sequence D2 are concatenated to obtain the compressed security sequence of the array C1 to be examined. The concatenation order is the order of the chain identifier of the disassembly sequence D1, the matching identifier, and the disassembly sequence D2.

[0085] If they are inconsistent, the chain identifier and its matching identifier of the disassembled sequence D1, and the chain identifier and its matching identifier of the disassembled sequence D2 are concatenated to obtain the compressed security sequence of the array C1 to be examined. The concatenation order is the chain identifier and matching identifier of the disassembled sequence D1, and the chain identifier and matching identifier of the disassembled sequence D2.

[0086] S18: according to S17, the compressed safety sequences of the arrays C2, C3, ..., Cb to be reviewed are calculated in sequence, and then b load processing packages of the load collection period are generated according to the compressed safety sequences of the arrays C1, C2, ..., Cb to be reviewed, and one load processing package corresponds to a compressed safety sequence of the array to be reviewed; the b load processing packages of the load collection period are transmitted to the investment and withdrawal strategy terminal, and the first compression mapping chain and the second compression mapping chain are synchronously transmitted to the investment and withdrawal strategy terminal;

[0087] After receiving the transmitted b load processing packets and the first compression mapping chain and the second compression mapping chain, the investment and withdrawal strategy terminal restores the b load processing packets to obtain the load collection data of the load collection period;

[0088] The switch-in / out strategy terminal pre-stores a trained distribution transformer load prediction model, and the distribution transformer load prediction model is established based on a long short-term memory recurrent neural network (LSTM-RNN);

[0089] The projection evaluation terminal pre-stores several load areas of the target power grid based on the target transformer and their corresponding switching strategies; the switching strategies include the capacity of the standby transformer, the number of standby transformers put into use, and a detailed switching process;

[0090] The restored user-side load data of the load collection period and the received device-side load data of the load collection period are input into the distribution transformer load prediction model to obtain the maximum load of the target area of ​​the next load collection period, and the corresponding load area is matched according to the maximum load, and the launch and withdrawal strategy of the load area is obtained, and the launch and withdrawal strategy is displayed to the management personnel of the target power grid;

[0091] In the description of the specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0092] The above contents are merely examples and explanations of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the invention or exceed the scope defined by the claims, they shall all fall within the protection scope of the present invention.

[0093] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An intelligent switching control system for transformers in distribution areas, characterized in that ； At each load collection period, the transformer monitoring terminal generates the device-side load data of the load collection period according to all load data monitored in the load collection period; At each load collection period, the user-side collection unit collects the power load data of the target area within the load collection period, and performs binary conversion on the collected data to generate the user-side load data of the load collection period; After receiving the user side load data of a load collection period, the user side processing unit processes the data in accordance with the preset preprocessing rules to obtain a number of user side data packets corresponding to the load collection period, and the first and second compression mapping chains; After receiving a plurality of user-side data packets of the load collection period and the first and second compression mapping chains, the investment and withdrawal strategy terminal restores the plurality of user-side data packets according to the first and second compression mapping chains to obtain the user-side load data of the load collection period; The restored user-side load data of the load collection cycle and the received device-side load data of the load collection cycle are input into the distribution transformer load prediction model to obtain the maximum load of the target area of ​​the next load collection cycle, and the corresponding investment and withdrawal strategy is matched according to the maximum load.

2. According to claim 1, a distribution station area transformer intelligent switching control system is characterized in that: The preprocessing rules of several user-side data packets, the first and second compression mapping chains in a load collection cycle are as follows: S11: specifying a cutting step length of 8, cutting the user-side load data in order from left to right to obtain a plurality of groups of cutting arrays, and then marking all the obtained groups of cutting arrays as A1, A2...Aa, a≥1, from left to right according to the position of each group of cutting arrays in the user-side load data before cutting; S12: According to the preset first screening rule, a number of cutting arrays are selected from the cutting arrays A1, A2...Aa as the arrays to be reviewed C1, C2...Cb, and the frequency quantity of the corresponding arrays to be reviewed is calculated during the selection process: S13: Extract the first 4 characters and the last 4 characters of the array C1 to be examined from left to right as a set of disassembly sequences of the array C1 to be examined. According to the above, disassemble the arrays C2, C3, ..., Cb to be examined in turn to obtain two sets of disassembly sequences corresponding thereto; S14: selecting basic sequences F1, F2, F3, and F4 from two groups of disassembly sequences of the arrays C1, C2, ..., Cb to be examined according to a preset second screening rule; S15: Screening in sequence according to a preset third screening rule to obtain a first progressive sequence and a second progressive sequence of the basic sequences F1, F2, F3, and F4; S16: traverse the basic sequences F1, F2, F3, F4 and the first progressive sequence and the second progressive sequence thereof, remove the numbers corresponding to the four-bit binary numbers that are respectively consistent with the basic sequences F1, F2, F3, F4 and the first progressive sequence and the second progressive sequence thereof from the numbers 0 to 15, and then mark the four-bit binary numbers of the four numbers that are not removed from the numbers 0 to 15 as recursive sequences, and mark them as F5, F6, F7, F8 respectively; S17: construct a first compression mapping chain according to the basic sequences F1, F2, F3, F4 and their first progressive sequence and second progressive sequence; A second compression mapping chain is constructed according to the recursive sequence F5, F6, F7, and F8; after the second compression mapping chain is constructed, character 1 is used as the chain identifier of the first compression mapping chain, and character 0 is used as the chain identifier of the second compression mapping chain; S18: Generate a compressed security sequence of the to-be-examined array C1 according to a preset generation rule; S19: according to S18, the compressed safety sequences of the arrays C2, C3, ..., Cb to be reviewed are calculated in sequence, and then b load processing packages of the load collection cycle are generated according to the compressed safety sequences of the arrays C1, C2, ..., Cb to be reviewed.

3. According to claim 2, a distribution area transformer intelligent switching control system is characterized in that: The first screening rule for selecting several cut arrays as the arrays to be examined C1, C2...Cb and calculating their frequency quantities is as follows: S121: De-duplicate the cut arrays A1, A2, ...Aa, and re-mark all the remaining cut arrays after de-duplicate as arrays to be reviewed, marked as B1, B2, ...Bb, 1≤b≤a; S122: Obtain the total number of the cut arrays that are consistent with the array to be examined B1 from the cut arrays A1, A2...Aa, and mark the total number as the frequency quantity of the array to be examined B1, wherein consistent means that the order and content of the characters constituting the cut arrays are the same as the characters constituting the array to be examined; S123: According to S122, the frequency quantities of the pending arrays B2, B3, ... Bb are obtained in sequence, and then the corresponding pending arrays are re-labeled as C1, C2, ... Cb in descending order of the frequency quantities.

4. According to claim 2, a distribution station area transformer intelligent switching control system is characterized in that: In S14, the second screening rule for selecting the basic sequences F1, F2, F3, and F4 is as follows: S141: According to the two groups of disassembly sequences corresponding to the arrays C1, C2, ..., Cb to be reviewed, four groups of disassembly sequences are randomly selected as a group of disassembly combinations to obtain a plurality of groups of disassembly combinations, and a plurality of groups of disassembly combinations that do not meet the preset selection conditions are eliminated, and all the remaining groups of disassembly combinations are re-calibrated as standard disassembly combinations; The selection conditions are as follows: any two disassembly sequences in a set of standard disassembly combinations are inconsistent; S142: Mark the two groups of disassembly sequences in the pending array C1 as D1 and D2 respectively, obtain the frequency quantities of all pending arrays that contain the disassembly sequence D1 and do not contain the disassembly sequence D2 from the pending arrays C2, C3, ..., Cb, and calculate the multidimensional frequency quantity F1 of the disassembly sequence D1 using the formula F1=Ee×ɑe+E2×ɑ2+...+Ee×ɑe, where E1, E2, ..., Ee refer to all the pending arrays that contain the disassembly sequence D1 and do not contain the disassembly sequence D2, and ɑ1, ɑ2, ..., ɑe are preset proportion coefficients of the corresponding pending arrays; Similarly, according to the frequency quantities of all the arrays to be reviewed that contain the disassembly sequence D2 and do not contain the disassembly sequence D1 obtained in the arrays to be reviewed C2, C3, ..., Cb, the multi-dimensional frequency quantity F2 of the disassembly sequence D2 is calculated; S143: According to S142, the multi-dimensional frequency quantities of the two groups of disassembly sequences corresponding to the arrays C2, C3, ..., Cb to be reviewed are calculated in sequence; S144: for any set of standard decomposition combinations obtained in S141, the multi-dimensional frequency quantities of all the group decomposition sequences contained therein are added to obtain the basic selected evaluation index of the corresponding standard decomposition combination, and the standard decomposition combination corresponding to the basic selected evaluation index with the largest value is selected from the basic selected evaluation indexes of all the groups of standard decomposition combinations obtained by calculation as the basic decomposition combination; The four groups of disassembly sequences included in the basic disassembly combination are re-labeled as basic sequences, and the corresponding basic sequences are labeled F1, F2, F3, and F4 in descending order of multi-dimensional frequency.

5. According to claim 2, a distribution station area transformer intelligent switching control system is characterized in that: S15, the third screening rule for the first progressive sequence and the second progressive sequence of the basic sequences F1, F2, F3, and F4 is as follows: S151: extracting all arrays to be reviewed that contain the basic sequence F1 but do not contain the basic sequences F2, F3, and F4 from the arrays to be reviewed C1, C2, ..., Cb, and then selecting the arrays to be reviewed with the largest frequency and the second largest frequency from all the extracted arrays to be reviewed; For the extracted array to be reviewed with the largest frequency, the disassembly sequence that is inconsistent with the basic sequence F1 is used as the first progressive sequence of the basic sequence F1 from the two groups of disassembly sequences of the array to be reviewed; For the extracted array to be reviewed with the second largest frequency, the disassembly sequence that is inconsistent with the basic sequence F1 is used as the second progressive sequence of the basic sequence F1 from the two groups of disassembly sequences of the array to be reviewed; S152: Calculate and obtain the first progressive sequence and the second progressive sequence of the basic sequences F2, F3, and F4 in sequence according to S151.

6. The intelligent switching control system for transformers in distribution stations according to claim 2 is characterized in that: In S17, 12 nodes are arranged in sequence from left to right in the first compression mapping chain, and the 12 nodes are mapped in sequence from left to right to the basic sequence F1, the first progressive sequence of the basic sequence F1, the second progressive sequence of the basic sequence F1, the basic sequence F2, the first progressive sequence of the basic sequence F2, the second progressive sequence of the basic sequence F2, ..., the basic sequence F4, the first progressive sequence of the basic sequence F4, and the second progressive sequence of the basic sequence F4.

7. The intelligent switching control system for transformers in distribution stations according to claim 2 is characterized in that: In S17, four nodes are sequentially arranged from left to right in the second mapping compression chain, and these four nodes are sequentially mapped to the recursive sequences F5, F6, F7, and F8 from left to right.

8. The intelligent switching control system for transformers in distribution stations according to claim 6 is characterized in that: In the first compression mapping chain, the two-digit binary number 00 is used as the sequence identifier of the basic sequence F1, 01 is used as the sequence identifier of the basic sequence F2, ..., 11 is used as the sequence identifier of the basic sequence F4. In other words, the two-digit binary numbers 00, 01, 10, 11 are used as the sequence identifiers of the basic sequences F1, F2, F3, F4 respectively.

9. The intelligent switching control system for transformers in distribution stations according to claim 7 is characterized in that: In the second compression mapping chain, character 0 is used as the sequence identifier of recursive sequence F5, character 1 is used as the sequence identifier of recursive sequence F6, two-digit binary number 00 is used as the sequence identifier of basic sequence F7, and two-digit binary number 01 is used as the sequence identifier of basic sequence F8.

Images