Random forest-based medium-voltage distribution network segment change relation anomaly identification and analysis method and system

Abstract

The invention discloses a random forest-based medium-voltage distribution network segment change relation abnormity identification analysis method and system. The method comprises the steps of obtaining measurement, ledger, topology, use and collection and fusion terminal power failure event data of a distribution transformer and a switch; generating a distribution transformer power failure event set, topological segments to which the distribution transformers belong and distribution transformer information in the segments; generating 11 characteristic values such as power failure events of the distribution transformers, voltage correlation coefficient mean values of the distribution transformers and the distribution transformers in the subsections, power failure moment difference values among the power failure events of the distribution transformers and the like; forming a sample set by combining the characteristic value of each distribution transformation item and the abnormal condition of the historical segment transformation relationship, training and optimizing a random forest model, and evaluating the random forest model by using an AUC value; and carrying out batch identification on the distribution transformers in each segment by using the trained random forest model, and finally outputting the distribution transformers with abnormal segment-change relations. According to the method, distribution transformer operation and topological data change characteristics are learned through the data model, whether the distribution transformer has the segment change relation abnormity phenomenon or not is judged, and segment change relation abnormity diagnosis is achieved.

Description

Technical Field

[0001] The present invention relates to a precise diagnosis technology for abnormal situation of segmented connection position of medium-voltage equipment in a distribution network, and in particular to a method and system for identifying and analyzing abnormal relationship of medium-voltage distribution network segments based on random forest. Background Art

[0002] With the continuous growth of electricity demand and the continuous expansion of the power grid, the complexity of the distribution system is increasing. Accurate segment-transformer relationship is crucial to ensure the safe and stable operation of the power system. If the segment-transformer relationship is abnormal, it may lead to incorrect power supply scheduling, affect fault location and isolation, and even cause large-scale power outages.

[0003] Traditional distribution network management methods may rely on manual records and regular inspections to determine the segment-transformer relationship, but this method is inefficient and prone to errors and omissions. At the same time, with the development of smart grid technology, a large number of intelligent monitoring equipment and sensors are used in distribution networks, providing a data basis for the automated identification of abnormal segment-transformer relationships. However, there are currently many line-transformer relationship abnormality diagnosis methods, and there is still no technology related to segment-transformer relationship abnormality identification.

[0004] In addition, improving power supply reliability and power quality is also an important factor in promoting the development of abnormal distribution transformer identification technology. By timely discovering and correcting abnormal distribution transformer relationships, fault handling and power supply restoration can be carried out quickly and accurately, reducing power outage time and losses, improving users' power experience, and assisting in accurately calculating segment line loss rates. Therefore. Distribution transformer identification technology with abnormal distribution transformer relationships has important research significance in power systems. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a method and system for identifying and analyzing abnormal segment-transformer relationships in a medium-voltage distribution network based on random forests, which can identify abnormal segment-transformer relationship attachment devices of any scale.

[0006] Technical solution: A method for identifying and analyzing abnormal relationship between medium voltage distribution network segments based on random forests of the present invention comprises:

[0007] Obtain the measurement, ledger, topology, usage and fusion terminal power outage event data of distribution transformers and switches;

[0008] Generate a distribution transformer power outage event set, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment based on the acquired data;

[0009] Determine characteristic values ​​based on the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of operating distribution variables within the segment, the power outage distribution transformer ratio within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment;

[0010] Combine all characteristic values ​​of each distribution transformer and the abnormal situations of historical section-transformer relationships to form a sample set, train and optimize the pre-built random forest model, and finally evaluate the random forest model with the AUC value;

[0011] The trained random forest model is used to batch identify the distribution transformers in each segment, and finally the distribution transformers with abnormal segment-transformer relationships are output.

[0012] Furthermore, the measurement, ledger, topology, usage and fusion terminal power outage event data of distribution transformers and switches are obtained, including:

[0013] Obtain the measurement, ledger, topology, and power outage event data of the distribution transformer: basic information maintained by the distribution transformer in the PMS system, drawing topology path data, power outage events and measurement data sent by the equipment;

[0014] Get the fusion terminal power outage event: the fusion terminal power outage event sent by the distribution transformer in the cloud master station system;

[0015] Obtain switch measurement and inventory data: basic information inventory data and measurement data of all switches in the D5200 system distribution network.

[0016] Furthermore, a distribution transformer power outage event set, a topological segment to which the distribution transformer belongs, and information about the distribution transformer in the segment are generated based on the acquired data, including:

[0017] Distribution transformer D k The active power measurement data is P k =[p k,1 ,p k,2 ,…,p k,i ,…,p k,n ], the voltage measurement data is V k =[v k,1 ,v k,2 ,…,v k,i ,…,v k,n ], n represents the number of sampling points of active power or voltage measurement data of distribution transformer;

[0018] If the D k Active power measurement data Pk In the continuous data segment [p k,i ,…,p k,j ] is 0 or empty, and the voltage measurement data V k In the continuous data segment [v k,i ,…,v k,j ] is 0 or empty, then the distribution transformer D k The data segment [i,…,j] generates a power outage event set for the data;

[0019] Distribution transformer D k If a power outage event is collected, and the power outage time sent is earlier than the power restoration time, and the power outage duration during the power outage period is within two natural days, a power outage event is generated as the distribution transformer D k The user adopts the power outage event set sent up;

[0020] Distribution transformer D k If the power outage event of the fusion terminal is collected, and the power outage and restoration time data is complete, and the power outage duration during the power outage period is within two natural days, a power outage event is generated as the distribution transformer D k The fusion terminal sends the power outage event set;

[0021] From the obtained distribution transformer measurement data and collected power outage events, the data of each distribution transformer is generated as a unit, and the power outage event set is generated, and the power outage event set sent up and the power outage event set of the fusion terminal are used to form the primary power outage event of each distribution transformer;

[0022] The power outage data of distribution transformers on the day and in the past week are counted in units of distribution transformers, and the data of distribution transformers with frequent power outages are eliminated;

[0023] Taking distribution transformer as unit, calculate the power consumption drop ratio of distribution transformer during power outage period compared with the same period of similar days, and eliminate the distribution transformer data with low power consumption drop ratio;

[0024] Taking the line as the unit, divide the topology into segments according to the line topology and switch type, and record the start switch, end switch and distribution transformer information of the segment;

[0025] According to the obtained topological data, the power outage ratio of each segment is calculated based on the topological segment, and the power outage data of the distribution transformer in the segment with a small power outage ratio is eliminated;

[0026] According to the above steps, the distribution transformer data is fully scanned, and finally a distribution transformer power outage event set, a topological segment to which the distribution transformer belongs, and information about the distribution transformer within the segment are formed.

[0027] Further, characteristic values ​​are determined based on the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the proportion of distribution transformers with power outage within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment, including:

[0028] Based on the distribution transformer outage event set, check the distribution transformer D k Whether there is a distribution transformer power outage event. If there is a distribution transformer power outage event, the value of the distribution transformer power outage event indicator is set to yes, otherwise it is set to no;

[0029] Based on the distribution transformer segment information, the number of operating distribution transformers in the segment is counted as the internal operating distribution transformers in the segment;

[0030] Based on the distribution transformer power outage event set and segment information, obtain all distribution transformer events in the segment to which the distribution transformer belongs in the distribution transformer power outage event set, and calculate the absolute value of the time difference between the event power outage start time in minutes as the power outage time difference between the distribution transformer power outage events;

[0031] The number of in-service distribution variables with power outage events in the statistical segment is taken as the numerator, and the number of in-service distribution variables in the segment is taken as the denominator to calculate the power outage distribution variable ratio in the segment;

[0032] Based on the distribution transformer measurement and segment information, the Pearson correlation coefficient method is used to calculate the voltage correlation coefficient value of the distribution transformer and all distribution transformers in the segment on the same day and calculate the average value, which is used as the mean value of the voltage correlation coefficient between the distribution transformer and the distribution transformer in the segment. The Pearson correlation coefficient calculation formula is as follows:

[0033]

[0034] Among them, R k,f Indicates distribution transformer D k About any other distribution transformer D in the segment f The correlation coefficient index value of k,i Indicates distribution transformer D k The data value of the voltage measurement data at the i-th sampling point; v f,i Indicates any other distribution transformer D in the segment f The data value of the voltage measurement data at the i-th sampling point; Indicates distribution transformer D k The average value of voltage measurement data; Indicates any other distribution transformer D in the segment f The average value of voltage measurement data, n is the number of distribution transformer sampling points on that day.

[0035] Further, characteristic values ​​are determined based on the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the proportion of distribution transformers with power outage within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment, including:

[0036] Based on the acquired switch measurement and topology data, two characteristic values ​​are generated: the available state of the switch measurement at the beginning of the segment and the available state of the switch measurement at the end of the segment:

[0037] First, obtain the power measurement value of the switch on the day, then obtain the distribution transformer information downstream of the switch, and calculate the ratio of the power value collected by the switch at each time point on the day to the power sum of the distribution transformer downstream of the switch. The calculation formula is as follows:

[0038]

[0039] Where q is the total variable in operation downstream of the switch; B x,i is the power value of the switch at time point i; P x,i For distribution transformer D k At the power value at time point i, if the ratio U is within 1 plus or minus 0.2, the power measurement value collected by the switch at time point i is considered to be available, otherwise it is unavailable. The available status of the switch data at all sampling time points of the day is counted. When the power measurement at 80% or more of the time points of the day is in the available state, the switch measurement is recorded as available, otherwise it is unavailable;

[0040] The available state of the start switch measurement is determined according to the ratio U, and the available state of the start switch measurement of the segment is obtained;

[0041] According to the ratio U, the measurement availability status of the end switches in the segment is judged in turn. When the measurement availability status of any end switch in the segment is unavailable, the measurement availability status of the segment end switch is judged to be unavailable, otherwise it is available.

[0042] Further, characteristic values ​​are determined based on the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the proportion of distribution transformers with power outage within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment, including:

[0043] Based on the acquired switch measurement data, the power ratio of the switch to the distribution transformer in the section is calculated. The calculation formula is as follows:

[0044]

[0045] Where m is the number of power outage time points of the power outage event, q is the number of distribution transformers in operation downstream of the distribution transformer section switch, P x,t For distribution transformer D k The power value at time t, B x,t is the power value of the switch at time t;

[0046] According to the above formula, the power ratio of the switch to the distribution transformer on the statistical day is calculated respectively. The power ratio of the switch to the distribution transformer on the statistical day minus 7 days is taken as a similar day and the power ratio of the switch to the distribution transformer on the similar day and the same period is calculated.

[0047]

[0048] Then use the above formula to get the power reduction ratio BE of the segment start switch;

[0049] According to the above formula, the power ratio of each end switch to the distribution transformer on the statistical day B1 is obtained in turn, and the power ratio of the switch to the distribution transformer on the statistical day minus 7 days is taken as a similar day and the power ratio of the switch to the distribution transformer in the same period of the similar day B2 is calculated. If the number of end switches is greater than 1, B1 and B2 are the averages, and then the above formula is used to calculate the power reduction ratio of the segmented end switch;

[0050] According to the distribution transformer measurement, the section to which the distribution transformer belongs and the section switch measurement data, the voltage measurement data of the distribution transformer and the section start switch on the day are obtained, and the Pearson correlation coefficient is used to calculate the voltage correlation coefficient between the distribution transformer and the section start switch to obtain the voltage correlation coefficient between the distribution transformer and the section start switch;

[0051] According to the measurement data of the distribution transformer, the section to which the distribution transformer belongs and the section switch, the voltage measurement data of the distribution transformer and the section end switch on that day are obtained. The Pearson correlation coefficient is used to calculate the voltage correlation coefficient values ​​of the distribution transformer and the section end switch in turn, and the average is calculated to obtain the voltage correlation coefficient of the distribution transformer and the section end switch.

[0052] Furthermore, a sample set is formed by combining all the characteristic values ​​of the above distribution transformers and the abnormal situations of the historical section transformer relationship, and the pre-built random forest model is trained and optimized. Finally, the random forest model is evaluated by the AUC value, including:

[0053] A sample set is formed by combining all the characteristic values ​​of the above distribution transformers and the abnormal situations of the historical section transformer relationship. The sample set is divided into a training set and a validation set in proportion. The training set is used to train the pre-built random forest model, and the validation set is used to test the training effect of the random forest model and tune the random forest model. Finally, the AUC value is calculated to evaluate the random forest model. The calculation formula of the AUC value is as follows:

[0054]

[0055] Among them, M is the number of positive samples; N is the number of negative samples; ∑i∈positive sample ranki is the sum of the serial numbers of positive samples.

[0056] Based on the same inventive concept, a medium voltage distribution network segment transformer relationship anomaly recognition and analysis system based on random forest of the present invention comprises:

[0057] Data acquisition module, used to obtain the measurement, ledger, topology, usage and fusion terminal power outage event data of distribution transformers and switches;

[0058] A topology segmentation module is used to generate a distribution transformer power outage event set, a topology segment to which the distribution transformer belongs, and information about the distribution transformer within the segment based on the acquired data;

[0059] A characteristic value determination module is used to determine characteristic values ​​according to the acquired data and the distribution transformer power outage event set, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the voltage correlation coefficient mean value between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the power outage distribution transformer ratio within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment;

[0060] The model building and training module is used to combine all the characteristic values ​​of the above distribution transformers and the abnormal situations of the historical section transformer relationship to form a sample set, train and optimize the pre-built random forest model, and finally evaluate the random forest model with the AUC value;

[0061] The identification module is used to use the trained random forest model to batch identify the distribution transformers in each segment, and finally output the distribution transformers with abnormal segment-transformer relationships.

[0062] Based on the same inventive concept, a medium-voltage distribution network segment transformer relationship anomaly identification and analysis device based on random forest of the present invention comprises a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the electronic device implements the steps of the above-mentioned medium-voltage distribution network segment transformer relationship anomaly identification and analysis method based on random forest.

[0063] Based on the same inventive concept, the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for identifying and analyzing abnormal relationships of medium-voltage distribution network segments based on random forests.

[0064] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are:

[0065] The present invention learns the trend of massive historical data changes in a data-driven manner, establishes a total of 11 relevant characteristic values ​​related to distribution transformers, and uses the AUC value to accurately evaluate and determine whether the model is highly accurate, thereby avoiding human input and device collection errors.

[0066] The present invention comprehensively considers various data features such as voltage characteristics of each distribution transformer, topological aggregation, switch measurement drop ratio, etc., gradually eliminates the distribution transformer with normal section transformer relationship, and has high accuracy.

[0067] The present invention has simple calculation and clear principle, can help distribution network operation personnel to timely discover distribution transformers with abnormal section-transformer relationships and adjust their position records, and has good application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a flow chart of a method for identifying and analyzing abnormal relationship of medium voltage distribution network segment changes based on random forests disclosed in an embodiment of the present invention;

[0069] Figure 2 It is a structural schematic diagram of a medium voltage distribution network segment change relationship anomaly identification and analysis system based on random forest disclosed in an embodiment of the present invention;

[0070] Figure 3 It is a structural schematic diagram of a medium-voltage distribution network segment change relationship anomaly identification and analysis device based on random forest disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It will be understood by those skilled in the art that the purposes and advantages that can be achieved with the present invention are not limited to the specific description of the above beneficial effects, and the above and other purposes that can be achieved by the present invention will be more clearly understood based on the following detailed description.

[0072] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed in the present invention can be implemented in hardware, software or a combination of the two. Whether it is implemented in hardware or software depends on the specific application and design of the technical solution. Professional and technical personnel can use different methods to implement the functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0073] Reference to "embodiments" in the present invention means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0074] Example 1

[0075] See also Figure 1 , Figure 1 1 is a flow chart of a method for identifying and analyzing abnormal relationship of medium voltage distribution network segments based on random forests disclosed in an embodiment of the present invention. Figure 1 The described method for identifying and analyzing abnormal relationship between medium-voltage distribution network segments is applied to power systems, such as identifying abnormal situation of segmented connection positions of medium-voltage equipment in distribution networks, etc., and the embodiments of the present invention are not limited thereto. Figure 1 As shown, the medium voltage distribution network segment transformer relationship anomaly identification and analysis method based on random forest may include the following operations:

[0076] S1. Obtain the measurement, ledger, topology, usage and fusion terminal power outage event data of distribution transformers and switches.

[0077] Obtain measurement, ledger, topology, power outage event data: basic information maintained by the distribution transformer in the PMS system (power consumption information collection system), drawing topology path data, power outage events and measurement data sent by the equipment.

[0078] Select the distribution transformer to be processed in the power consumption information collection system (i.e., the PMS system is a system existing in the prior art), and read the basic data of the distribution transformer stored in the existing system, including basic ledger data, various equipment connection relationship data, and the sent distribution transformer power outage event data;

[0079] The power consumption information collection system exports the distribution transformer active power, electricity meter bottom, and voltage measurement data sampled every 15 minutes. The aforementioned sampling frequency of sampling every 15 minutes can be modified according to actual conditions.

[0080] Get the fusion terminal power outage event: the fusion terminal power outage event sent by the distribution transformer in the cloud master station system;

[0081] Obtain switch measurement and inventory data: basic information inventory data and measurement data of all switches in the D5200 system (distribution automation system) distribution network.

[0082] S2. Generate a distribution transformer power outage event set, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment based on the acquired data.

[0083] Considering that when the active power and voltage measurement data of the distribution transformer drop to 0 or are empty at the same time, it may be a real power outage, or it may be a false power outage caused by signal transmission, data storage, etc., it is necessary to use data verification methods to effectively and reasonably screen out real power outage events.

[0084] In this embodiment, step S2 specifically includes the following steps:

[0085] S2.1, a distribution transformer D k The active power measurement data is P k =[p k,1 ,p k,2 ,…,p k,i ,…,p k,n ], the voltage measurement data is V k =[v k,1 ,v k,2 ,…,v k,i ,…,v k,n ], n represents the number of sampling points of active power or voltage measurement data of distribution transformer;

[0086] S2.2, if equipped with transformer D k Active power measurement data P k In the continuous data segment [p k,i ,…,p k,j ] is 0 or empty, and the voltage measurement data V k In the continuous data segment [v k,i ,…,v k,j ] is 0 or empty, then the distribution transformer D k The data segment [i,…,j] generates a power outage event set for the data;

[0087] S2.3, a distribution transformer D k If a power outage event is collected, and the power outage time sent is earlier than the power restoration time, and the power outage duration during the power outage period is within two natural days, a power outage event is generated as the distribution transformer D k The user adopts the power outage event set sent up;

[0088] In this embodiment, a distribution transformer D kIf a power outage event is collected, and the power outage time sent is earlier than the power restoration time, and the power outage duration during the power outage period is within two natural days, a power outage event is generated as the distribution transformer D k The user adopts the power outage event set.

[0089] S2.4, a distribution transformer D k If the power outage event of the fusion terminal is collected, and the power outage and restoration time data is complete, and the power outage duration during the power outage period is within two natural days, a power outage event is generated as the distribution transformer D k The fusion terminal sends the power outage event set;

[0090] In this embodiment, a distribution transformer D k If the power outage event of the fusion terminal is collected, and the power outage and restoration time data is complete, and the power outage duration during the power outage period is within two natural days, a power outage event is generated as the distribution transformer D k The fusion terminal sends the power outage event set.

[0091] S2.5. From the obtained distribution transformer measurement data and collected power outage events, generate a power outage event set for each distribution transformer as a unit, use the power outage event set sent up and the power outage event set of the integrated terminal to form the primary power outage event of each distribution transformer;

[0092] S2.6. Count the power outage data of distribution transformers on the day and in the past week, and exclude the data of distribution transformers with frequent power outages;

[0093] In this embodiment, the power outage data of the distribution transformer on the day and in the past week are counted as a unit. If the distribution transformer has more than 10 power outage events on the day or there are power outage events with exactly the same power outage time for more than three days in the past week, it is considered that the distribution transformer has frequent power outages, and the distribution transformer data with frequent power outages are finally eliminated.

[0094] S2.7, using distribution transformers as units, calculate the power consumption drop ratio of distribution transformers during power outage periods compared to the same period on similar days, and exclude distribution transformer data with low power consumption drop ratios;

[0095] In this embodiment, the power consumption drop ratio of the distribution transformer during the power outage period is similar to that during the same period of the day, and the calculation formula is as follows:

[0096]

[0097] Where m is the number of power outage time points of the power outage event; E x,m E is the power consumption value of the distribution transformer at a certain time point during the power outage period on a similar power outage day (the default is before the 7th day); m It is the power consumption value of the distribution transformer at a certain time during the power outage period on the day of the power outage. When the power consumption drop ratio of the distribution transformer is above 150%, it is considered that the power consumption drop is more obvious. Finally, the distribution transformer data with insignificant power consumption drop ratio are eliminated.

[0098] S2.8. Taking the line as the unit, divide the topology into segments according to the line topology and switch type, and record the start switch, end switch and distribution transformer information of the segment;

[0099] It is worth noting that in this segment, there is only one start switch and the end switch can be one or more.

[0100] S2.9. Based on the acquired topological data, calculate the power outage ratio of each segment by taking the topological segment as the unit, and exclude the power outage data of the distribution transformer in the segment with a small power outage ratio;

[0101] S2.10. Scan the distribution transformer data in full according to the above steps, and finally form a distribution transformer power outage event set, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment.

[0102] S3. Determine characteristic values ​​according to the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the proportion of distribution transformers with power outage within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment. Specifically, the following steps are included:

[0103] S3.1. Based on the distribution transformer outage event set, check the distribution transformer D k Whether there is a distribution transformer power outage event. If there is a distribution transformer power outage event, the value of the distribution transformer power outage event indicator is set to yes, otherwise it is set to no;

[0104] S3.2. Based on the distribution transformer segment information, the number of distribution transformers in operation in the segment is counted as the number of distribution transformers in operation in the segment;

[0105] S3.3. Based on the distribution transformer power outage event set and segment information, obtain all distribution transformer events in the segment to which the distribution transformer belongs in the distribution transformer power outage event set, and calculate the absolute value of the time difference between the event power outage start time in minutes as the power outage time difference between the distribution transformer power outage events; count the number of operating distribution transformers with distribution transformer power outage events in the segment as the numerator, and the number of operating distribution transformers in the segment as the denominator, and calculate the power outage distribution transformer ratio in the segment;

[0106] S3.4. Based on the distribution transformer measurement and segment information, the Pearson correlation coefficient method is used to calculate the voltage correlation coefficient value of the distribution transformer and all distribution transformers in the segment on the same day and the average value is taken as the mean value of the voltage correlation coefficient between the distribution transformer and the distribution transformer in the segment. The Pearson correlation coefficient calculation formula is as follows:

[0107]

[0108] Among them, R k,f Indicates distribution transformer D k About any other distribution transformer D in the segment f The correlation coefficient index value of k,i Indicates distribution transformer D k The data value of the voltage measurement data at the i-th sampling point; v f,i Indicates any other distribution transformer D in the segment f The data value of the voltage measurement data at the i-th sampling point; Indicates distribution transformer D k The average value of voltage measurement data; Indicates any other distribution transformer D in the segment f The average value of voltage measurement data, n is the number of distribution transformer sampling points on that day.

[0109] It is worth noting that the distribution transformer measurement includes active power, voltage, and current, and the sampling points of these measurements are consistent, so the distribution transformer sampling point on that day = distribution transformer active power measurement sampling point = distribution transformer voltage measurement sampling point.

[0110] S3.5. Based on the acquired switch measurement and topology data, two characteristic values ​​are generated: the available state of the switch measurement at the beginning of the segment and the available state of the switch measurement at the end of the segment:

[0111] First, obtain the power measurement value of the switch on the day, then obtain the distribution transformer information downstream of the switch, and calculate the ratio of the power value collected by the switch at each time point on the day to the power sum of the distribution transformer downstream of the switch. The calculation formula is as follows:

[0112]

[0113] Where q is the total variable in operation downstream of the switch; B x,i is the power value of the switch at time point i; P x,i For distribution transformer D k At the power value at time point i, if the ratio U is within 1 plus or minus 0.2, the power measurement value collected by the switch at time point i is considered to be available, otherwise it is unavailable. The available status of the switch data at all sampling time points of the day is counted. When the power measurement at 80% or more of the time points of the day is in the available state, the switch measurement is recorded as available, otherwise it is unavailable;

[0114] The available state of the start switch measurement is determined according to the ratio U, and the available state of the start switch measurement of the segment is obtained;

[0115] According to the ratio U, the measurement availability status of the end switches in the segment is judged in turn. When the measurement availability status of any end switch in the segment is unavailable, the measurement availability status of the segment end switch is judged to be unavailable, otherwise it is available.

[0116] S3.6. Based on the acquired switch measurement data, calculate the power ratio of the switch to the distribution transformer in the section. The calculation formula is as follows:

[0117]

[0118] Where m is the number of power outage time points of the power outage event, q is the number of distribution transformers in operation downstream of the distribution transformer section switch, P x,t For distribution transformer D k The power value at time t, B x,t is the power value of the switch at time t;

[0119] According to the above formula, the power ratio of the switch to the distribution transformer on the statistical day is calculated respectively. The power ratio of the switch to the distribution transformer on the statistical day minus 7 days is taken as a similar day and the power ratio of the switch to the distribution transformer on the similar day and the same period is calculated.

[0120]

[0121] Then use the above formula to get the power reduction ratio of the segment start switch;

[0122] According to the above formula, the power ratio of each ending switch to the distribution transformer on the statistical day B1 is obtained in turn, and the power ratio of the switch to the distribution transformer on the statistical day minus 7 days is taken as a similar day and the power ratio of the switch to the distribution transformer in the same period of the similar day B2 is calculated. If the number of ending switches is greater than 1, B1 and B2 are the averages, and then the above formula is used to calculate the power reduction ratio of the segmented ending switches.

[0123] It is worth noting that the start switch and the end switch are both switches on the line, and the calculation formula for the power reduction ratio characteristics of the two is the same. They are only marked as the start or end switches of the segment due to different switch topological positions.

[0124] S3.7. According to the distribution transformer measurement data, the section to which the distribution transformer belongs and the section switch measurement data, obtain the voltage measurement data of the distribution transformer and the section start switch on the day, calculate the voltage correlation coefficient value of the distribution transformer and the section start switch using the Pearson correlation coefficient, and obtain the voltage correlation coefficient of the distribution transformer and the section start switch;

[0125] S3.8. According to the distribution transformer measurement data, the section to which the distribution transformer belongs and the section switch measurement data, obtain the voltage measurement data of the distribution transformer and the section end switch on the day, use the Pearson correlation coefficient to calculate the voltage correlation coefficient values ​​of the distribution transformer and each section end switch in turn, and calculate the average to obtain the voltage correlation coefficient of the distribution transformer and the section end switch.

[0126] In step S3, a power outage event index for the distribution transformer is generated according to the distribution transformer power outage event set; a mean characteristic value of the voltage correlation coefficient between the distribution transformer and the distribution transformer in the segment is generated according to the line segmentation information and the distribution transformer voltage measurement; two characteristic values, namely, the power outage time difference between the power outage events of the distribution transformer in the segment and the power outage distribution transformer ratio in the segment, are generated according to the power outage events sent by all the distribution transformers in the segment; a characteristic value of the number of distribution variables in operation in the segment is generated according to the line segmentation information; two characteristic values, namely, the available state of switch measurement at the beginning of the segment and the available state of switch measurement at the end of the segment, are generated according to the switch measurement data; two characteristic values, namely, the power reduction ratio of the switch at the beginning of the segment and the power reduction ratio of the switch at the end of the segment, are generated according to the switch measurement data and the power outage event of the distribution transformer; two characteristic values, namely, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment, are generated according to the voltage measurement of the distribution transformer and the switch in the segment; and finally, a total of 11 characteristic values ​​are generated.

[0127] S4. Combining all characteristic values ​​of each distribution transformer obtained in step S3 and the abnormal situations of historical section-transformer relationships to form a sample set, training and optimizing the pre-built random forest model, and finally evaluating the random forest model with the AUC value.

[0128] A sample set is formed by combining the 11 characteristic values ​​of each distribution transformer and the abnormal situation of the historical section transformer relationship. The sample set is divided into a training set and a validation set in a ratio of 7:3. The training set is used to train the pre-built random forest model, and the validation set is used to test the training effect of the random forest model and tune the random forest model. Finally, the AUC value is calculated to evaluate the random forest model. The calculation formula of the AUC value is as follows:

[0129]

[0130] Among them, M is the number of positive samples; N is the number of negative samples; ∑i∈positive sample ranki is the sum of the serial numbers of positive samples. The AUC value of the model is finally calculated to be 0.938, and the model effect is good.

[0131] It is worth noting that the random forest model is an existing open source model that can be called directly.

[0132] S5. Use the trained random forest model to batch identify the distribution transformers in each segment, and finally output the distribution transformers with abnormal segment-transformer relationships.

[0133] After the model prediction accuracy meets the requirements, the trained model is used to batch identify the feature data of each distribution transformer, and finally the abnormal section-transformer relationship corresponding to each distribution transformer is output. That is, the power outage events of each distribution transformer are batch generated and the required feature values ​​are extracted, which are sent to the model for calculation, and finally the section-transformer relationship abnormal distribution transformer data is output.

[0134] Aiming at the problem that the current traditional distribution network segment transformer relationship abnormality management relies on manual inspection, and it is difficult to fully and timely grasp the abnormal information of distribution network terminal connection between different sections of the same line, the present invention proposes a medium-voltage distribution network segment transformer relationship abnormality identification method based on random forest, which can be used to identify segment transformer relationship abnormal connection devices of any scale. The calculation is simple and the principle is clear, which can help distribution network operation personnel to timely discover abnormalities in ledger maintenance and make adjustments.

[0135] The present invention uses a data model to learn the characteristics of distribution transformer operation and topological data changes, and determines whether there is any abnormal section-transformer relationship in the distribution transformer, thereby realizing the abnormal section-transformer relationship diagnosis.

[0136] Example 2

[0137] See also Figure 2 , Figure 2 The present invention discloses a structural diagram of a medium-voltage distribution network segment transformer relationship anomaly identification and analysis system based on random forest, which can realize the identification of abnormal situation of the connection segment position of medium-voltage equipment in the distribution network, and specifically includes:

[0138] Data acquisition module, used to obtain the measurement, ledger, topology, usage and fusion terminal power outage event data of distribution transformers and switches;

[0139] A topology segmentation module is used to generate a distribution transformer power outage event set, a topology segment to which the distribution transformer belongs, and information about the distribution transformer within the segment based on the acquired data;

[0140] A characteristic value determination module is used to determine characteristic values ​​according to the acquired data and the distribution transformer power outage event set, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the voltage correlation coefficient mean value between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the power outage distribution transformer ratio within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment;

[0141] The model building and training module is used to combine all the characteristic values ​​of the above configurations and the abnormal historical segment change relationship to form a sample set, train and optimize the pre-built random forest model, and finally evaluate the random forest model with the AUC value;

[0142] The identification module is used to use the trained random forest model to batch identify the distribution transformers in each segment, and finally output the distribution transformers with abnormal segment-transformer relationships.

[0143] In an optional embodiment, a method for identifying and analyzing abnormalities in the relationship between medium-voltage distribution network segments and transformers based on random forests includes: a) obtaining measurement, ledger, topology, procurement and fusion terminal power outage event data of distribution transformers and switches; b) generating a set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and information on the distribution transformers within the segment; c) generating 11 characteristic values, such as the existence of power outage events in the distribution transformer, the mean value of the voltage correlation coefficient between the distribution transformer and the distribution transformers within the segment, and the difference in power outage times between distribution transformer power outage events; d) combining the 11 characteristic values ​​of each distribution transformer and the historical abnormalities in the relationship between segment transformers to form a sample set, training and optimizing a pre-built random forest model, and finally evaluating the random forest model with an AUC value; e) using the trained random forest model to batch identify the distribution transformers within each segment, and finally outputting the distribution transformers with abnormal segment-transformer relationships.

[0144] Example 3

[0145] See also Figure 3 , Figure 3 The structure diagram of a medium voltage distribution network segment change relationship anomaly recognition and analysis device based on random forest is disclosed in an embodiment of the present invention. Figure 3 The described device can be applied in power systems, such as for identifying abnormal situations of the connection segment positions of medium-voltage equipment in a distribution network, etc., and the embodiments of the present invention are not limited thereto.

[0146] like Figure 3 As shown, the device may include a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the electronic device implements the steps of the method described in the above embodiment and can achieve the technical effect consistent with the above method.

[0147] The memory may include computer system readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory. The device may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the memory may be used to read and write non-removable, non-volatile magnetic media (commonly referred to as a "hard drive"). A program / utility having a set (at least one) of program modules may be stored in, for example, the memory, such program modules including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include the implementation of a network environment. The program modules typically perform the functions and / or methods in the embodiments described herein.

[0148] The processor executes various functional applications and data processing by running the program stored in the memory, such as implementing the method provided in the first embodiment of the present invention.

[0149] Example 4

[0150] Embodiment 4 of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the method described in the above embodiment are implemented and the technical effect consistent with the above method can be achieved.

[0151] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a device or a device or used in combination with it.

[0152] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0153] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0154] Computer program code for performing the operation of the present invention may be written in one or more programming languages ​​or combinations thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0155] Of course, the storage medium containing computer executable instructions provided by an embodiment of the present invention, whose computer executable instructions are not limited to the above method operations, can also execute related operations in the method provided by any embodiment of the present invention.

[0156] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying and analyzing abnormal relationship between medium voltage distribution network segments based on random forest, characterized in that: include: Obtain the measurement, ledger, topology, usage and fusion terminal power outage event data of distribution transformers and switches; Generate a distribution transformer power outage event set, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment based on the acquired data; Determine characteristic values ​​based on the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of operating distribution variables within the segment, the power outage distribution transformer ratio within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment; Combining all the characteristic values ​​of the above distribution transformers and the abnormal situations of historical section transformer relationships to form a sample set, the pre-built random forest model is trained and optimized, and finally the random forest model is evaluated using the AUC value; The trained random forest model is used to batch identify the distribution transformers in each segment, and finally the distribution transformers with abnormal segment-transformer relationships are output.

2. The method for identifying and analyzing abnormal relationship between medium voltage distribution network segments based on random forest according to claim 1 is characterized in that: Obtain the measurement, ledger, topology, usage and fusion terminal power outage event data of distribution transformers and switches, including: Obtain the measurement, ledger, topology, and power outage event data of the distribution transformer: basic information maintained by the distribution transformer in the PMS system, drawing topology path data, power outage events and measurement data sent by the equipment; Get the fusion terminal power outage event: the fusion terminal power outage event sent by the distribution transformer in the cloud master station system; Obtain switch measurement and inventory data: basic information inventory data and measurement data of all switches in the D5200 system distribution network.

3. The method for identifying and analyzing abnormal relationship of medium voltage distribution network segments based on random forest according to claim 1 is characterized in that: Generate a distribution transformer power outage event set, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment based on the acquired data, including: Distribution transformer D k The active power measurement data is P k =[p k,1 ,p k,2 ,…,p k,i ,…,p k,n ], the voltage measurement data is V k =[v k,1 ,v k,2 ,…,v k,i ,…,v k,n ], n represents the number of sampling points of active power or voltage measurement data of distribution transformer; If the D k Active power measurement data P k In the continuous data segment [p k,i ,…,p k,j ] is 0 or empty, and the voltage measurement data V k In the continuous data segment [v k,i ,…,v k,j ] is 0 or empty, then the distribution transformer D k The data segment [i,…,j] generates a power outage event set for the data; Distribution transformer D k If a power outage event is collected, and the power outage time sent is earlier than the power restoration time, and the power outage duration during the power outage period is within two natural days, a power outage event is generated as the distribution transformer D k The user adopts the power outage event set sent up; Distribution transformer D k If the power outage event of the fusion terminal is collected, and the power outage and restoration time data is complete, and the power outage duration during the power outage period is within two natural days, a power outage event is generated as the distribution transformer D k The fusion terminal sends the power outage event set; From the obtained distribution transformer measurement data and collected power outage events, the data of each distribution transformer is generated as a unit, and the power outage event set is generated, and the power outage event set sent up and the power outage event set of the fusion terminal are used to form the primary power outage event of each distribution transformer; The power outage data of distribution transformers on the day and in the past week are counted in units of distribution transformers, and the data of distribution transformers with frequent power outages are eliminated; Taking distribution transformer as unit, calculate the power consumption drop ratio of distribution transformer during power outage period compared with the same period of similar days, and eliminate the distribution transformer data with low power consumption drop ratio; Taking the line as the unit, divide the topology into segments according to the line topology and switch type, and record the start switch, end switch and distribution transformer information of the segment; According to the obtained topological data, the power outage ratio of each segment is calculated based on the topological segment, and the power outage data of the distribution transformer in the segment with a small power outage ratio is eliminated; According to the above steps, the distribution transformer data is fully scanned, and finally a distribution transformer power outage event set, a topological segment to which the distribution transformer belongs, and information about the distribution transformer within the segment are formed.

4. The method for identifying and analyzing abnormal relationship of medium voltage distribution network segments based on random forest according to claim 1 is characterized in that: The characteristic values ​​are determined based on the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the proportion of distribution transformers with power outage within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment, including: Based on the distribution transformer outage event set, check the distribution transformer D k Whether there is a distribution transformer power outage event. If there is a distribution transformer power outage event, the value of the distribution transformer power outage event indicator is set to yes, otherwise it is set to no; Based on the distribution transformer segment information, the number of operating distribution transformers in the segment is counted as the number of operating distribution transformers in the segment; Based on the distribution transformer power outage event set and segment information, obtain all distribution transformer events in the segment to which the distribution transformer belongs in the distribution transformer power outage event set, and calculate the absolute value of the time difference between the event power outage start time in minutes as the power outage time difference between the distribution transformer power outage events; The number of in-service distribution variables with power outage events in the statistical segment is taken as the numerator, and the number of in-service distribution variables in the segment is taken as the denominator to calculate the power outage distribution variable ratio in the segment; Based on the distribution transformer measurement and segment information, the Pearson correlation coefficient method is used to calculate the voltage correlation coefficient value of the distribution transformer and all distribution transformers in the segment on the same day and calculate the average value, which is used as the mean value of the voltage correlation coefficient between the distribution transformer and the distribution transformer in the segment. The Pearson correlation coefficient calculation formula is as follows: Among them, R k,f Indicates distribution transformer D k About any other distribution transformer D in the segment f The correlation coefficient index value of k,i Indicates distribution transformer D k The data value of the voltage measurement data at the i-th sampling point; v f,i Indicates any other distribution transformer D in the segment f The data value of the voltage measurement data at the i-th sampling point; Indicates distribution transformer D k The average value of voltage measurement data; Indicates any other distribution transformer D in the segment f The average value of voltage measurement data, n is the number of distribution transformer sampling points on that day.

5. The method for identifying and analyzing abnormal relationship of medium voltage distribution network segments based on random forest according to claim 1 is characterized in that: The characteristic values ​​are determined based on the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the proportion of distribution transformers with power outage within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment, including: Based on the acquired switch measurement and topology data, two characteristic values ​​are generated: the available state of the switch measurement at the beginning of the segment and the available state of the switch measurement at the end of the segment: First, obtain the power measurement value of the switch on the day, then obtain the distribution transformer information downstream of the switch, and calculate the ratio of the power value collected by the switch at each time point on the day to the power sum of the distribution transformer downstream of the switch. The calculation formula is as follows: Where q is the total variable in operation downstream of the switch; B x,i is the power value of the switch at time point i; P x,i For distribution transformer D k At the power value at time point i, if the ratio U is within 1 plus or minus 0.2, the power measurement value collected by the switch at time point i is considered to be available, otherwise it is unavailable. The available status of the switch data at all sampling time points of the day is counted. When the power measurement at 80% or more of the time points of the day is in the available state, the switch measurement is recorded as available, otherwise it is unavailable; The available state of the start switch measurement is determined according to the ratio U, and the available state of the start switch measurement of the segment is obtained; According to the ratio U, the measurement availability status of the end switches in the segment is judged in turn. When the measurement availability status of any end switch in the segment is unavailable, the measurement availability status of the segment end switch is judged to be unavailable, otherwise it is available.

6. The method for identifying and analyzing abnormal relationship of medium voltage distribution network segments based on random forest according to claim 1 is characterized in that: The characteristic values ​​are determined based on the acquired data and the set of distribution transformer power outage events, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the average value of the voltage correlation coefficient between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the proportion of distribution transformers with power outage within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment, including: Based on the acquired switch measurement data, the power ratio of the switch to the distribution transformer in the section is calculated. The calculation formula is as follows: Where m is the number of power outage time points of the power outage event, q is the number of distribution transformers in operation downstream of the distribution transformer section switch, P x,t For distribution transformer D k The power value at time t, B x,t is the power value of the switch at time t; According to the above formula, the power ratio of the switch to the distribution transformer on the statistical day is calculated respectively. The power ratio of the switch to the distribution transformer on the statistical day minus 7 days is taken as a similar day and the power ratio of the switch to the distribution transformer on the similar day and the same period is calculated. Then use the above formula to get the power reduction ratio BE of the segment start switch; According to the above formula, the power ratio of each end switch to the distribution transformer on the statistical day B1 is obtained in turn, and the power ratio of the switch to the distribution transformer on the statistical day minus 7 days is taken as a similar day and the power ratio of the switch to the distribution transformer in the same period of the similar day B2 is calculated. If the number of end switches is greater than 1, B1 and B2 are the averages, and then the above formula is used to calculate the power reduction ratio of the segmented end switch; According to the distribution transformer measurement, the section to which the distribution transformer belongs and the section switch measurement data, the voltage measurement data of the distribution transformer and the section start switch on the day are obtained, and the Pearson correlation coefficient is used to calculate the voltage correlation coefficient between the distribution transformer and the section start switch to obtain the voltage correlation coefficient between the distribution transformer and the section start switch; According to the measurement data of the distribution transformer, the section to which the distribution transformer belongs and the section switch, the voltage measurement data of the distribution transformer and the section end switch on that day are obtained. The Pearson correlation coefficient is used to calculate the voltage correlation coefficient values ​​of the distribution transformer and the section end switch in turn, and the average is calculated to obtain the voltage correlation coefficient of the distribution transformer and the section end switch.

7. The method for identifying and analyzing abnormal relationship of medium voltage distribution network segments based on random forest according to claim 1 is characterized in that: Combine all the characteristic values ​​of each distribution transformer and the abnormal situations of historical section transformer relationships to form a sample set, train and optimize the pre-built random forest model, and finally use the AUC value to evaluate the random forest model, including: A sample set is formed by combining all characteristic values ​​of each distribution transformer and the abnormal situations of historical section-transformer relationships. The sample set is divided into a training set and a validation set in proportion. The training set is used to train the pre-built random forest model. The validation set is used to test the training effect of the random forest model and to tune the random forest model. Finally, the AUC value is calculated to evaluate the random forest model. The calculation formula of the AUC value is as follows: Among them, M is the number of positive samples; N is the number of negative samples; ∑i∈positive sample ranki is the sum of the serial numbers of positive samples.

8. A medium voltage distribution network segment change relationship anomaly recognition and analysis system based on random forest, characterized in that: include: Data acquisition module, used to obtain the measurement, ledger, topology, usage and fusion terminal power outage event data of distribution transformers and switches; A topology segmentation module is used to generate a distribution transformer power outage event set, a topology segment to which the distribution transformer belongs, and information about the distribution transformer within the segment based on the acquired data; A characteristic value determination module is used to determine characteristic values ​​according to the acquired data and the distribution transformer power outage event set, the topological segment to which the distribution transformer belongs, and the distribution transformer information within the segment. The characteristic values ​​include the distribution transformer power outage event index, the voltage correlation coefficient mean value between the distribution transformer and the distribution transformer within the segment, the power outage time difference between the distribution transformer power outage events, the available state of the switch measurement at the beginning of the segment, the available state of the switch measurement at the end of the segment, the power reduction ratio of the switch at the beginning of the segment, the power reduction ratio of the switch at the end of the segment, the number of distribution variables in operation within the segment, the power outage distribution transformer ratio within the segment, the voltage correlation coefficient between the distribution transformer and the switch at the beginning of the segment, and the voltage correlation coefficient between the distribution transformer and the switch at the end of the segment; The model building and training module is used to combine all the characteristic values ​​of the above distribution transformers and the abnormal situations of the historical section transformer relationship to form a sample set, train and optimize the pre-built random forest model, and finally evaluate the random forest model with the AUC value; The identification module is used to use the trained random forest model to batch identify the distribution transformers in each segment, and finally output the distribution transformers with abnormal segment-transformer relationships.

9. A medium voltage distribution network segment change relationship anomaly recognition and analysis device based on random forest, characterized in that: The electronic device comprises a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the electronic device implements the steps of the medium-voltage distribution network segment transformer relationship anomaly identification and analysis method based on random forest as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for identifying and analyzing abnormal relationships between medium-voltage distribution network segments based on random forests as described in any one of claims 1 to 7.

Images