A method and system for determining a power outage section in a low-voltage distribution network

Abstract

The present invention discloses a method and system for determining a power outage interval in a low-voltage distribution network, including: when there are power outage users in the current substation area of the low-voltage distribution network, query whether there is a research and judgment result or a power outage record. If so, determine and return the power outage interval based on the research and judgment result or the power outage record; otherwise, based on summoning the meter information of the power outage users, judge whether it is a fault of the power outage users themselves. If it is a fault of the power outage users themselves, return the self-fault; otherwise, based on the topological structure of the substation area and the power outage probability of users in the low-voltage distribution network, conduct an analysis of the power outage interval in the substation area to determine the power outage interval corresponding to the power outage users. This method is used to quickly and accurately determine the specific power outage interval in a substation area when it is detected that a certain user or some users in the area under the jurisdiction of a distribution transformer are powered off, and then inform the relevant operation and maintenance personnel to carry out effective emergency repairs, thereby shortening the power outage time and improving user satisfaction.

Description

Technical Field

[0001] The present invention relates to the field of operation and maintenance of smart grids, and particularly to a method and system for determining a power outage interval in a low-voltage distribution network. Background Art

[0002] The distribution network is composed of overhead lines, cables, poles, distribution transformers, disconnect switches, reactive power compensators and some auxiliary facilities, etc., and is a network that plays an important role in distributing electric energy in the power network. Classified by voltage level, it can be divided into high-voltage distribution network, medium-voltage distribution network and low-voltage distribution network. Due to the popularization of the distribution automation system, the current high-voltage distribution network and medium-voltage distribution network both have relatively complete power outage management measures. Once a fault power outage event occurs, the fault point can be immediately located and relevant measures can be taken to reduce the impact brought by the fault power outage.

[0003] However, the power outage management measures for the low-voltage distribution network are not yet perfect. Once a fault power outage event occurs, the operation and maintenance personnel cannot quickly and accurately determine the power outage interval. This will not only increase the work difficulty of the operation and maintenance personnel, resulting in the extension of the fault power outage time, but also reduce the user satisfaction and affect the image of the power grid. The existing methods for determining the power outage interval determine the power outage interval by directly judging the equipment conditions on the upstream line connected to the fault electric meter. For example, a patent named "A Smart Management Method for Low-Voltage Power Outage", but in the current distribution network, the coverage rate of the equipment used to judge the power outage is very low, making the use of this method limited; in addition, a patent named "Fault Location Method for Medium and Low-Voltage Distribution Networks" mentions determining the power outage interval based on the tripping of low-voltage branch line switches or the power failure alarm information of low-voltage collectors, but these information are very difficult to obtain in actual applications. Therefore, when a fault power outage event occurs, quickly and accurately determining the power outage interval of the low-voltage distribution network has become an urgent task in the current power outage management work of the distribution network. Summary of the Invention

[0004] In order to solve the deficiency in the prior art that after a fault power outage event occurs in the low-voltage distribution network, the operation and maintenance personnel cannot quickly and accurately determine the power outage interval, the present invention provides a method and system for determining a power outage interval in a low-voltage distribution network.

[0005] A method for determining a power outage interval in a low-voltage distribution network provided by the present invention specifically includes:

[0006] S1. When there are power outage users in the current substation area of the low-voltage distribution network, query whether there is a research and judgment result or a power outage record. If there is, determine and return the power outage interval based on the research and judgment result or the power outage record, and at the same time end the process; otherwise, execute S2;

[0007] S2. Based on the summoned meter information of the power outage user, determine whether it is a fault of the power outage user itself. If it is a fault of the power outage user itself, return the self - fault. Otherwise, execute S3;

[0008] S3. Based on the topological structure of the sub - station area and the power outage probability of users in the low - voltage distribution network, conduct a power outage interval analysis for the sub - station area, determine the power outage interval corresponding to the power outage user, and return the research and judgment result generated based on the power outage user and the power outage interval corresponding to the power outage user.

[0009] Preferably, the step of determining whether it is a fault of the power outage user itself based on the summoned meter information of the power outage user includes:

[0010] S201. Determine whether the power outage user is cut off due to arrears. If it is cut off due to arrears, return that it is cut off due to arrears. Otherwise, proceed to step S202;

[0011] S202. Summon the meter data of the power outage user. If it can be summoned, return a false power outage alarm. Otherwise, it is not a false power outage alarm;

[0012] The self - fault includes: cut - off due to arrears and false power outage alarm.

[0013] Preferably, the step of conducting a power outage interval analysis for the sub - station area based on the topological structure of the sub - station area and the power outage probability of users in the low - voltage distribution network, determining the power outage interval corresponding to the power outage user, and returning the research and judgment result generated based on the power outage user and the power outage interval corresponding to the power outage user includes:

[0014] S301. Based on the topological structure of the sub - station area and the power outage probability of users, select the users to be judged on the sub - line according to the set sub - line research and judgment rules from the sub - line where the power outage user is located; summon the meters corresponding to the users to be judged on the sub - line in sequence according to the summoning rules, and at the same time judge and record the status of the meters; when it is found that two users have power, stop summoning, determine the power outage interval as the sub - line where the power outage user is located, and return the research and judgment result generated based on the meter status, the power outage user, and the power outage interval corresponding to the power outage user; otherwise, execute S302;

[0015] S302. Based on the topological structure of the sub - station area and the power outage probability of users, select the users to be judged on the main line according to the set main - line and whole - sub - station - area research and judgment rules from the other sub - lines except the main line to which the sub - line where the power outage user is located belongs; summon the meters corresponding to the users to be judged on the main line in sequence according to the summoning rules, and at the same time judge and record the status of the meters; when it is found that two users have power, stop summoning, determine the power outage interval as the main line corresponding to the sub - line where the power outage user is located, and return the research and judgment result generated based on the meter status, the power outage user, and the power outage interval corresponding to the power outage user; otherwise, execute S303;

[0016] S303. Select the users to be judged in the power distribution area from all the sub-lines under the other main lines except the main line where the power-off user is located according to the topological structure of the power distribution area and the power-off probability of the users, and in accordance with the set judgment rules for the main line and the entire power distribution area; sequentially summon the electricity meters corresponding to the users to be judged in the power distribution area, and at the same time judge and record the status of the electricity meters; when it is found that two users have power, stop the summoning, determine that the power-off interval is not the entire power distribution area, and return the judgment result generated based on the status of the electricity meters, the power-off user, and the power-off interval corresponding to the power-off user; otherwise, the power-off interval is the entire power distribution area.

[0017] Preferably, the sub-line judgment rules include:

[0018] When the total number of users on the sub-line where the power-off user is located ≤ 8, then select all the other users on the sub-line except the power-off user according to the preset summoning rules;

[0019] When the total number of users on the sub-line where the power-off user is located > 8 and ≤ 16, then select the first 8 users with a lower power-off occurrence probability among all the other users on the sub-line except the power-off user according to the preset summoning rules;

[0020] When the total number of users on the sub-line where the power-off user is located < 16 and ≤ 50, then select the first 10 users with a lower power-off occurrence probability among all the other users on the sub-line except the power-off user according to the preset summoning rules;

[0021] When the total number of users on the sub-line where the power-off user is located > 50, then select the first 15 users with a lower power-off occurrence probability among all the other users on the sub-line except the power-off user according to the preset summoning rules.

[0022] Preferably, the judgment rules for the main line and the entire power distribution area include:

[0023] If the total number of users on the sub-line ≤ 8, then select 1 user on the sub-line according to the preset summoning rules;

[0024] If the total number of users on the sub-line > 8 and ≤ 16, then select 2 users on the sub-line according to the preset summoning rules;

[0025] If the total number of users on the sub-line < 16 and ≤ 50, then select 3 users on the sub-line according to the preset summoning rules;

[0026] If the total number of users on the sub-line > 50, then select 5 users on the sub-line according to the preset summoning rules.

[0027] Preferably, the calling rule includes:

[0028] Based on a pre-established prediction model for the probability of power outage of low-voltage users in the distribution network, predict the probability of power outage of all users in the substation area within a set future time;

[0029] Give priority to calling the electricity meters of users with a lower probability of power outage;

[0030] If the difference in the probability of power outage between users is 0.5%, then according to the importance of the users, give priority to calling the electricity meters of important users;

[0031] If the importance of users is the same, then give priority to calling the electricity meters of users with fewer historical arrears;

[0032] If there are at least two users with the same number of historical arrears, then call them in sequence according to the ID of each user.

[0033] Preferably, the prediction model for the probability of power outage of low-voltage users in the distribution network includes:

[0034] Obtain the historical data of users in all low-voltage distribution networks in the current substation area;

[0035] In the historical data, screen out the training samples that meet the set characteristics;

[0036] Based on the training samples and the logistic regression algorithm, determine the weight parameters;

[0037] Based on the weight parameters and the historical data of users, obtain the power outage probability of all users in the substation area;

[0038] Among them, the characteristics include: user ID, power outage time, temperature, whether it is a peak electricity consumption period, user importance, user historical power outage times, user historical arrears times, and whether it is a sensitive user.

[0039] Preferably, the power outage probability is calculated according to the following formula:

[0040]

[0041] In the formula: p: power outage probability; θ n : weight coefficient; x n : the nth feature of the user; y: the label of whether the user has a power outage.

[0042] Preferably, the sub-line judgment rule, the main-line and the whole substation area judgment rule also include an arrears judgment rule.

[0043] Preferably, the arrears judgment rule includes:

[0044] In the sub-line judgment rule, when the total number of users on the sub-line where the power-off user is located > 8, when summoning the users to be judged on the sub-line according to the summoning rule, or in the judgment rules of the main line and the entire power distribution area, when summoning the users to be judged on the main line or the users to be judged in the power distribution area according to the summoning rule, if there is a user with a power-off, it is judged whether the current user is a power-off due to arrears, otherwise it is not judged whether the current user is a power-off due to arrears;

[0045] If the current user belongs to a power-off due to arrears, then continue to select users with a lower probability of power-off from the sub-line corresponding to the current user to join the users to be judged on the sub-line, the users to be judged on the main line or the users to be judged in the power distribution area, and summon the electricity meter according to the summoning rule.

[0046] Preferably, the recording of the status of the electricity meter includes:

[0047] The user's electricity meter is powered on and the user's electricity meter is powered off.

[0048] Preferably, the determination that the power-off interval is not the entire power distribution area and the return of the judgment result generated based on the electricity meter status and the power-off interval further includes:

[0049] Record the users with a power-off and not due to arrears, and for the users with a power-off and not due to arrears, continue to judge according to S1 to S3, and add the newly generated judgment result to the previous judgment result.

[0050] Preferably, the power-off record includes:

[0051] Planned power-off record and fault power-off record caused by faults in lines of 10 kV and above.

[0052] Based on the same inventive concept, the present invention also provides a system for determining a power-off interval in a low-voltage distribution network, including:

[0053] The first judgment module is used to query whether there is a judgment result or a power-off record when there is a power-off user in the current power distribution area of the low-voltage distribution network. If so, determine and return the power-off interval based on the judgment result or the power-off record, and at the same time end the process. Otherwise, call the second judgment module;

[0054] The second judgment module is used to judge whether it is a fault of the power-off user itself based on the electricity meter information of the summoned power-off user. If it is a fault of the power-off user itself, return the self-fault. Otherwise, call the third judgment module;

[0055] The third judgment module is used to analyze the power-off interval of the power distribution area based on the topological structure of the power distribution area and the power-off probability of users in the low-voltage distribution network, determine the power-off interval corresponding to the power-off user, and return the judgment result generated based on the power-off user and the power-off interval corresponding to the power-off user.

[0056] Preferably, the third judgment module includes:

[0057] The 301st judgment unit is used to select sub-line users to be judged from the sub-line where the power-off user is located according to the set sub-line judgment rules based on the topological structure of the substation area and the power-off probability of the user; sequentially call the electric meters corresponding to the sub-line users to be judged according to the call rules, and at the same time judge and record the status of the electric meters; when it is found that two users have power, stop calling, determine the power-off interval as the sub-line where the power-off user is located, and return the judgment result generated based on the electric meter status, the power-off user, and the power-off interval corresponding to the power-off user; otherwise, call the 302nd judgment unit;

[0058] The 302nd judgment unit is used to select main-line users to be judged from other sub-lines except the main line to which the sub-line where the power-off user is located belongs according to the set main-line and whole-substation-area judgment rules based on the topological structure of the substation area and the power-off probability of the user; sequentially call the electric meters corresponding to the main-line users to be judged according to the call rules, and at the same time judge and record the status of the electric meters; when it is found that two users have power, stop calling, determine the power-off interval as the main line corresponding to the sub-line where the power-off user is located, and return the judgment result generated based on the electric meter status, the power-off user, and the power-off interval corresponding to the power-off user; otherwise, call the 303rd judgment unit;

[0059] The 303rd judgment unit is used to select substation-area users to be judged from all sub-lines under other main lines except the main line where the power-off user is located according to the set main-line and whole-substation-area judgment rules based on the topological structure of the substation area and the power-off probability of the user; sequentially call the electric meters corresponding to the substation-area users to be judged according to the call rules, and at the same time judge and record the status of the electric meters; when it is found that two users have power, stop calling, determine that the power-off interval is not the whole substation area, and return the judgment result generated based on the electric meter status, the power-off user, and the power-off interval corresponding to the power-off user; otherwise, the power-off interval is the whole substation area.

[0060] Preferably, the 301st judgment unit includes:

[0061] The first rule-making sub-unit is used to make sub-line judgment rules; the sub-line judgment rules include:

[0062] When the total number of users in the sub-line where the power-off user is located ≤ 8, then select all the other users in the sub-line except the power-off user according to the preset call rules;

[0063] When the total number of users in the sub-line where the power-off user is located > 8 and ≤ 16, then select the first 8 users with a lower power-off occurrence probability among all the other users in the sub-line except the power-off user according to the preset call rules;

[0064] When the total number of users on the sub-line where the power outage user is located is less than 16 and ≤ 50, the first 10 users with a lower probability of power outage among all the users except the power outage user on the sub-line are selected according to a preset summoning rule;

[0065] When the total number of users on the sub-line where the power outage user is located is > 50, the first 15 users with a lower probability of power outage among all the users except the power outage user on the sub-line are selected according to a preset summoning rule.

[0066] Preferably, the 302 research and judgment unit includes:

[0067] A second formulation sub-unit for formulating the research and judgment rules for the main line and the entire distribution area; the research and judgment rules for the main line and the entire distribution area include:

[0068] If the total number of users on the sub-line is ≤ 8, 1 user on the sub-line is selected according to a preset summoning rule;

[0069] If the total number of users on the sub-line is > 8 and ≤ 16, 2 users on the sub-line are selected according to a preset summoning rule;

[0070] If the total number of users on the sub-line is less than 16 and ≤ 50, 3 users on the sub-line are selected according to a preset summoning rule;

[0071] If the total number of users on the sub-line is > 50, 5 users on the sub-line are selected according to a preset summoning rule.

[0072] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0073] The technical solution provided by the present invention, when there is a power outage user in the current distribution area of the low-voltage distribution network, queries whether there is a research and judgment result or a power outage record. If there is, the power outage interval is determined and returned based on the research and judgment result or the power outage record. Otherwise, based on summoning the meter information of the power outage user, it is judged whether it is a fault of the power outage user itself. If it is a fault of the power outage user itself, the self-fault is returned. Otherwise, based on the topological structure of the distribution area and the power outage probability of users in the low-voltage distribution network, the power outage interval analysis of the distribution area is carried out to determine the power outage interval corresponding to the power outage user; it is used to quickly and accurately determine the specific power outage interval of the distribution area when it is detected that a certain user or some users in the distribution area under the jurisdiction of a certain distribution transformer are powered off, and then inform the relevant operation and maintenance personnel. After learning the specific power outage interval, the operation and maintenance personnel can quickly and accurately judge the fault point and implement effective emergency repairs, thereby shortening the power outage time and improving user satisfaction.

[0074] When the technical solution provided by the present invention discovers power outage users and it is necessary to determine the specific power outage section through preliminary judgment, first, the power outage probabilities of all users in the transformer substation area are obtained based on relevant historical data and the logistic regression algorithm. Then, according to the power outage probabilities, the users to be summoned are sequentially selected from the secondary line, the main line, and the transformer substation area according to the set judgment rules as needed. Among the users to be summoned, the corresponding smart meters are summoned in ascending order of the power outage probability. By judging whether other users are also powered off, the specific power outage range is determined.

[0075] In the technical solution provided by the present invention, the data of smart meters is obtained. At present, the coverage rate of smart meters is extremely high, so the practical application value of the technical solution provided by the present invention is very high.

[0076] In the process of determining the power outage section by the technical solution provided by the present invention, the summons is made sequentially according to the power outage probability of the users. Therefore, it is not necessary to summon the smart meters of all users, which improves the efficiency.

[0077] The technical solution provided by the present invention will save the historical judgment results of a certain line. When other users of this line come to consult the power outage reason, the power outage reason can be directly obtained from the historical judgment results and fed back to the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 is the main judgment flow chart of the present invention;

[0079] Figure 2 is the judgment flow chart of the analysis of the power outage section in the transformer substation area of the present invention;

[0080] Figure 3 is the user transformer substation area topology structure diagram of the present invention;

[0081] Figure 4 is the flow chart of the method for determining the power outage section in a low-voltage distribution network of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0082] To better understand the present invention, the content of the present invention will be further described below in conjunction with the accompanying drawings of the specification and examples.

[0083] Based on the historical data of users in the low-voltage distribution network, the present invention conceives a sorting algorithm for the power outage probabilities of users in the low-voltage distribution network; according to the power outage probabilities of users obtained by the sorting algorithm, in accordance with the set judgment process and judgment rules, the smart meters of the corresponding users are summoned sequentially, and the specific power outage section in the user transformer substation area is judged according to the information of the summoned meters, realizing the rapid and accurate positioning of the power outage section in the low-voltage user transformer substation area of the distribution network, thereby reducing the work difficulty of maintenance personnel, reducing the fault power outage time, and improving user satisfaction.

[0084] Such as Figure 4As shown in the figure, a method for determining a power outage interval in a low-voltage distribution network provided by the present invention includes:

[0085] S1. When there are power outage users in the current substation area of the low-voltage distribution network, query whether there is a research and judgment result or a power outage record. If there is, determine and return the power outage interval based on the research and judgment result or the power outage record, and at the same time end the process; otherwise, execute S2.

[0086] S2. Based on calling the meter information of the power outage user, judge whether it is a fault of the power outage user itself. If it is a fault of the power outage user itself, return the self-fault; otherwise, execute S3.

[0087] S3. Analyze the power outage interval of the substation area based on the topological structure of the substation area and the power outage probability of users in the low-voltage distribution network, determine the power outage interval corresponding to the power outage user, and return the research and judgment result generated based on the power outage user and the power outage interval corresponding to the power outage user.

[0088] When it is found that user i in the substation area under a certain distribution transformer is powered off, the following steps are used to conduct a power outage research and judgment on this user:

[0089] S1. When there are power outage users in the current substation area of the low-voltage distribution network, query whether there is a research and judgment result or a power outage record. If there is, determine and return the power outage interval based on the research and judgment result or the power outage record, and at the same time end the process. Specifically, it includes:

[0090] Check the historical research and judgment results or power outage records according to the ID of user i. The power outage records include: whether there is a planned power outage record of this user, and whether there is a fault power outage record caused by a 10 kV or higher voltage line fault.

[0091] S2. Based on calling the meter information of the power outage user, judge whether it is a fault of the power outage user itself. If it is a fault of the power outage user itself, return the self-fault. Specifically, it includes:

[0092] If there is no historical research and judgment result of this user in the system, check the marketing system according to the ID of user i to see if it is a power outage due to arrears.

[0093] If it is not a power outage due to arrears, we call the meter of user i to see if there is a false power outage report.

[0094] If it is found that the data of this meter cannot be called, it means that there is also no false power outage report, and then it is necessary to analyze the power outage range of user i in the substation area.

[0095] S3. Analyze the power outage interval of the substation area based on the topological structure of the substation area and the power outage probability of users in the low-voltage distribution network, determine the power outage interval corresponding to the power outage user, and return the research and judgment result generated based on the power outage user and the power outage interval corresponding to the power outage user. Specifically, it includes:

[0096] Assume that user i is located in the k-th sub-line of the main line A of a certain transformer area. First, we obtain the number of users on the sub-line corresponding to user i according to the topological relationship.

[0097] 1) When the number of users on the sub-line corresponding to user i is less than or equal to 8:

[0098] S31. Judge whether the power outage interval is the k-th sub-line corresponding to user i according to the set sub-line judgment rule

[0099] It is necessary to call the electricity meters of all other users under the k-th sub-line of the main line A corresponding to user i in turn. During the traversal process, if two users are found to have power at the same time, it means that only the k-th sub-line corresponding to user i has a fault. If all users are powered off, it is necessary to further judge whether the power outage interval is the main line A.

[0100] S32. Judge whether the power outage interval is the main line A corresponding to user i according to the set main line and the judgment rules of the entire transformer area. Traverse all sub-lines under the main line according to the set main line and the judgment rules of the entire transformer area, select the main line users to be called. During the calling process, when a user is found to be powered off, judge whether it is a power outage due to arrears according to the arrears rule. When the user is powered off and not due to arrears, it is necessary to select another user with the lowest power outage probability from the sub-line corresponding to the user and add it to the main line users to be called, and continue to call according to the calling rule;

[0101] Specifically, it includes: if there are m sub-lines under the main line A and m≥2, it is necessary to judge the other m - 1 sub-lines. When judging these m - 1 sub-lines, first, it is necessary to obtain the user with the lowest power outage probability in each sub-line, and then call the electricity meters of the users with the lower power outage probability among these m - 1 users in turn according to the calling rule. If a user is found to be powered off but due to arrears, select the electricity meter of another user with a lower power outage probability in the sub-line corresponding to the user to replace the current user. If the replaced user is also powered off and due to arrears, continue to replace until the electricity meters of all users in the sub-line are called. If two users are found to have power at the same time during the entire calling process, it means that only the main line A has a fault, otherwise it is necessary to further judge the main line B and main line C of the transformer area to judge whether the power outage interval is the entire transformer area.

[0102] S33. Judge whether the power outage interval is the entire transformer area according to the set main line and the judgment rules of the entire transformer area

[0103] If there are n sub - lines under both the main lines B and C, it is also necessary to judge these 2n sub - lines. When judging these 2n sub - lines, first, it is necessary to obtain the user with the lowest power outage probability in each sub - line, and then call the electricity meters of the users with relatively low power outage probabilities among these 2n users in turn according to the call rule. If there is a user with a power outage among these 2n users, but it is a power outage due to arrears, then continue to call the electricity meter of a user with a relatively low power outage probability in the corresponding sub - line of this user until the electricity meters of all users in this sub - line are called. If two users are found to have power during the entire calling process, it means that not the entire transformer area is powered off; otherwise, it means that the power - off area is the entire transformer area.

[0104] 2) When the number of users in the sub - line corresponding to user i is greater than 8 and less than or equal to 16, call the electricity meters of the first 8 users with relatively low power outage probabilities among all users in turn. If a user is found to have a power outage and it is a power outage due to arrears, then continue to call the electricity meter of a user with a relatively low power outage probability in the corresponding sub - line of this user until the electricity meters of all users in this sub - line are called. If two users are found to have power during the entire calling process, it means that only the k - th sub - line corresponding to user i has a problem. Otherwise, it is necessary to further judge the upstream line. The steps for judging the upstream line are the same as steps S32 and S33.

[0105] 3) When the number of users in the sub - line corresponding to user i is greater than 16 and less than or equal to 50, call the electricity meters of the first 10 users with relatively low power outage probabilities among all users in turn. If a user is found to have a power outage and it is a power outage due to arrears, then continue to call the electricity meter of a user with a relatively low power outage probability in the corresponding sub - line of this user until the electricity meters of all users in this sub - line are called. If two users are found to have power during the entire calling process, it means that only the k - th sub - line corresponding to user i has a problem. Otherwise, it is necessary to further judge the upstream line. The steps for judging the upstream line are the same as steps S32 and S33.

[0106] 4) When the number of users in the sub - line corresponding to user i is greater than 50, call the electricity meters of the first 15 users with relatively low power outage probabilities among all users in turn. If a user is found to have a power outage and it is a power outage due to arrears, then continue to call the electricity meter of a user with a relatively low power outage probability in the corresponding sub - line of this user until the electricity meters of all users in this sub - line are called. If two users are found to have power during the entire calling process, it means that only the k - th sub - line corresponding to user i has a problem. Otherwise, it is necessary to further judge the upstream line. The steps for judging the upstream line are the same as steps S32 and S33.

[0107] In the concept of the present invention, based on the historical data of low - voltage distribution network users, a sorting algorithm for the power outage probabilities of low - voltage distribution network users is proposed. According to the power outage probabilities of users obtained by the sorting algorithm, it includes:

[0108] The power outage scope analysis of the substation area is carried out by summoning the user's smart meter. When summoning the meter, there needs to be a sequence. The logistic regression algorithm of machine learning is adopted, and the historical power consumption information of any user substation area in the distribution network is combined to establish a prediction model for the probability of power outage of low-voltage users in the distribution network, so as to predict the power outage probability of all users in a certain substation area within the next 24 hours. Then, combined with the formulated research and judgment process and rules, the sequence of summoning the meter is obtained.

[0109] Features adopted: User ID + time + temperature + whether it is a peak power consumption period + user importance level + user's historical power outage times + user's historical arrears times + whether it is a sensitive user. The above features are represented in English as shown in Table 1:

[0110] Table 1 Feature Description Table

[0111]

[0112] Common linear regression models for probability prediction:

[0113] p(y=1|x)=θ0+θ1x1+θ2x2+...+θ n x n

[0114] In this application, since the probability p and the dependent variable are non-linear, in order to solve this problem, the logit transformation is introduced, so that there is a linear correlation between logit(p) and the independent variable. The logistic regression model is defined as follows:

[0115]

[0116] If a batch of data samples {x i ,y i} i＝1,2...,n is given, then x i represents the feature, y i =0 represents the negative sample, and y i =1 represents the positive sample.

[0117] The probability that a sample x belongs to the positive sample is

[0118] In this application, the corresponding relationship between variables and features or labels is shown in the following table:

[0119] Table 2 Corresponding Relationship Table between Variables and Features or Labels

[0120] Variable Corresponding feature or label Explanation <![CDATA[x1]]> ID User ID <![CDATA[x2]]> TIME Time <![CDATA[x3]]> QW Air temperature <![CDATA[x4]]> SFYDGFSQ Whether it is a peak electricity consumption period <![CDATA[x5]]> YHZYCD Importance level of the user <![CDATA[x6]]> YHLSTDCS Number of historical power outages of the user <![CDATA[x7]]> YHLSQFCS Number of historical overdue payments of the user <![CDATA[x8]]> SFMGYH Whether it is a sensitive user y SFTD Label in the sample data indicating whether there is a power outage

[0121] In practical applications, in combination with the historical electricity consumption information of a specific user substation area, training sample data that meet the above characteristics and labels are screened out. Then, based on the sample data and the logistic regression model, the weight parameters θ0, θ1, θ2, θ3, θ4, θ5, θ6, θ7, and θ8 are obtained, thereby obtaining a prediction model for the probability of power outage occurrence:

[0122]

[0123] During actual calculation, sample data that meet the above characteristics and labels can be input at a set time every day, and then the computer calculates the probability of a power outage event occurring at each time point for all users in the specified substation area within the next 24 hours. If a power outage of a certain user is detected during the calculation process, the previously calculated probability is used to analyze the power outage range of this user in the substation area.

[0124] This embodiment further illustrates the judgment process in the inventive concept of the present invention:

[0125] a. The reason why users with a lower probability of power outage are more closely connected to the upstream line is that, for example, if the upstream line fails and powers off, then all downstream users will lose power, and this power outage information will be recorded in the system. And if the upstream line does not lose power, but only a certain sub-line corresponding to it fails, then only the users of the faulty sub-line will lose power, and the users of the other sub-lines corresponding to this upstream line will not experience a power outage event. To sum up, it can be known that users who are more closely connected to the upstream line will have fewer power outage records, and users who are more closely connected to themselves will have more power outage records. Therefore, the probability of power outage for users who are more closely connected to themselves will also be higher. In other words, users with a lower probability of power outage are more closely connected to the upstream line.

[0126] b. When conducting the main judgment, the reason for first checking the historical judgment results is that if multiple users' power outages are detected simultaneously, it is very likely that similar events need to be merged.

[0127] When multiple users' power outages are detected simultaneously, first judge one of the users, and then save the judgment result. For the other users, if they lose power due to the same reason, the judgment result of the previous user can be directly used.

[0128] In addition, the reason for first judging whether it is a power outage due to arrears is that the time and cost required to recall the electricity meter information of this user are greater than the time and cost required to judge whether this user is in arrears and without power. Judging whether it is a power outage due to arrears only requires judging the flag bit related to arrears.

[0129] c. The calling rule is as follows: Calculate the probability of power outage occurrence according to the optimization algorithm, and then preferentially call users with a lower probability of power outage occurrence.

[0130] If the probability of power outage differs by only 0.5%, important users will be summoned first according to their importance.

[0131] If the importance of the users is the same, priority will be given to summoning users with fewer historical arrears.

[0132] If all the previous conditions are the same, the users will be summoned in sequence according to their user IDs.

[0133] In addition, the reason for giving priority to summoning users with fewer historical arrears is to avoid as much as possible the situation where the electricity meter information of the summoned users shows a power outage, but the power outage is caused by arrears.

[0134] d. When the number of users corresponding to a sub-line of a detected power outage user is less than or equal to 8, it is necessary to determine whether the power outage section is the sub-line corresponding to the user. At this time, it is necessary to call the meters of all other users of the sub-line in turn. In the process of calling the user's meter, if it is found that the user has a power outage, it is not necessary to determine whether the power outage is due to arrears. This is because even if the power outage of the user is caused by arrears, it is still necessary to determine whether all other users corresponding to the sub-line have a power outage in turn. Therefore, in this case, determining whether the user has a power outage due to arrears will not be of any benefit to the subsequent judgment.

[0135] e. If after judging sub-line A1 of main line A, it is found that all called users are in a power outage state, it is necessary to further judge other sub-lines of main line A. At this time, the number of users of other sub-lines may be the four situations mentioned above. Therefore, main line A is judged according to the set judgment rules for the main line and the entire substation. The judgment of main lines B and C is the same as that of main line A.

[0136] The rules for judging the main line and the entire area:

[0137] If the total number of users in the sub-line is ≤8, one user in the sub-line is selected according to a preset calling rule;

[0138] If the total number of users in the sub-line is greater than 8 and less than or equal to 16, two users in the sub-line are selected according to a preset calling rule;

[0139] If the total number of users in the sub-line is <16 and ≤50, three users in the sub-line are selected according to a preset calling rule;

[0140] If the total number of users in the sub-line is greater than 50, five users in the sub-line are selected according to a preset calling rule.

[0141] f. During the summoning process, the meter status of all summoned users (user meters with power and user meters without power) will be recorded in the analysis results, so that subsequent analysis can minimize the number of summoned meters.

[0142] g. After the summons is completed, all users affected by the power outage will be analyzed through the topological structure, and a research and judgment result will be generated for subsequent merging of similar power outage events.

[0143] h. During the summons process, if it is found that all users on the first sub-line of the main line A are without power, it is necessary to further determine whether the power outage scope is the entire main line A. Therefore, it is necessary to continue to summon the electricity meters of users in other sub-lines of the main line A according to the research and judgment rules. If it is found during the summons process that some users have power and some users are without power and it is not due to arrears. In this case, it can be confirmed that the power outage scope is not the entire main line A, but the first sub-line of the main line A. However, for the users who have just been found to be without power and it is not due to arrears, these users need to be recorded, and after the end of this research and judgment, the power outage scope analysis of these users in the transformer area will be carried out according to the normal research and judgment rules to ensure that all power outage intervals can be obtained, and the user electricity meter status, power outage users, and new power outage scope recorded during this research and judgment process will be added to the generated research and judgment result.

[0144] i. If it is detected that some users are without power at the same time, it is very likely that the power outages of these users are due to the same reason. In this case, it is necessary to conduct research and judgment on these users in turn, and after the research and judgment of the first user is completed, the subsequent research and judgment can often be merged. If they cannot be merged, continue according to the normal research and judgment rules.

[0145] j. The sub-line research and judgment rules include:

[0146] When the total number of users on the sub-line where the power outage user is located ≤ 8, then all other users on the sub-line except the power outage user are selected according to the preset summons rules;

[0147] When the total number of users on the sub-line where the power outage user is located > 8 and ≤ 16, then the first 8 users with a lower probability of power outage among all other users on the sub-line except the power outage user are selected according to the preset summons rules;

[0148] When the total number of users on the sub-line where the power outage user is located < 16 and ≤ 50, then the first 10 users with a lower probability of power outage among all other users on the sub-line except the power outage user are selected according to the preset summons rules;

[0149] When the total number of users on the sub-line where the power outage user is located > 50, then the first 15 users with a lower probability of power outage among all other users on the sub-line except the power outage user are selected according to the preset summons rules.

[0150] Next, the research and judgment process for quickly and accurately locating the power outage interval of low-voltage users in the distribution network provided by the present invention is sorted out as follows:

[0151] When it is found that user i in the area served by a certain distribution transformer is out of power, the main process of judging this user is carried out according to the following steps:

[0152] 1.1 Judge whether there are historical judgment results and power outage records of this user;

[0153] 1.2 If it does not meet the basic step 1.1, then judge whether this user is out of power due to arrears;

[0154] 1.3 If it does not meet the basic step 1.2, then judge whether it is a misreport of power outage for this user;

[0155] 1.4 If it does not meet the basic step 1.3, then judge the power outage range of the area where this user is located according to the following steps:

[0156] 1.4.1 According to the number of users on the sub-line corresponding to user i and the established judgment rules, judge whether the power outage interval is only the sub-line where this user is located by calling the smart meter;

[0157] 1.4.2 If it does not meet the basic step 1.4.1, then according to the number of sub-lines on the main line to which the sub-line corresponding to user i belongs and the established judgment rules, judge whether the power outage interval is only the main line corresponding to the sub-line where this user is located by calling the smart meter;

[0158] 1.4.3 If it does not meet the basic step 1.4.2, then according to the number of sub-lines on the other two main lines in the area where user i is located and the established judgment rules, judge whether the power outage interval is the entire area by calling the smart meter.

[0159] Further explanation of the judgment rules mentioned in the above process:

[0160] 2.1 Give priority to calling the smart meters of users with a lower probability of power outage. Here, the probability of power outage is calculated by a prediction model obtained through an optimized algorithm.

[0161] 2.2 On the premise of meeting the judgment rule 2.1, if the difference in the probability of power outage is only 0.5%, then according to the importance of the users, give priority to calling the smart meters of important users.

[0162] 2.3 On the premise of meeting the judgment rules 2.1 and 2.2, give priority to calling the smart meters of users with fewer historical arrears.

[0163] 2.4 When performing step 1.4.1, several users who meet the judgment rules 2.1, 2.2, and 2.3 are selected in the sub-circuit. When the total number of users in the sub-circuit where user i is located is less than or equal to 8, the smart meters of all other users in this sub-circuit except user i need to be summoned in sequence; when the total number of users in the sub-circuit where user i is located is greater than 8 and less than or equal to 16, the smart meters of the first 8 users with a lower probability of power outage among all other users in this sub-circuit except user i need to be summoned in sequence; when the total number of users in the sub-circuit where user i is located is greater than 16 and less than or equal to 50, the smart meters of the first 10 users with a lower probability of power outage among all other users in this sub-circuit except user i need to be summoned in sequence; when the total number of users in the sub-circuit where user i is located is greater than 50, the smart meters of the first 15 users with a lower probability of power outage among all other users in this sub-circuit except user i need to be summoned in sequence.

[0164] 2.5 When performing step 1.4.1, when the total number of users in the sub-circuit where user i is located is less than or equal to 8, during the process of summoning the smart meters of all other users in this sub-circuit except user i, if it is found that a user has a power outage, it is not judged whether it is a power outage due to arrears.

[0165] 2.6 When performing steps 1.4.1, 1.4.2, and 1.4.3, except for the scenario in judgment rule 2.5, in other scenarios, if it is found that a user has a power outage and it is judged that it is a power outage due to arrears, then continue to summon the smart meter of a user with a lower probability of power outage in the corresponding sub-circuit of this user until the smart meters of all users in this sub-circuit are summoned.

[0166] 2.7 When performing steps 1.4.2 and 1.4.3, several users who meet the judgment rules 2.1, 2.2, and 2.3 need to be selected in the other sub-circuits except the sub-circuit where user i is located or in all sub-circuits of the other two main circuits except the main circuit where user i is located. If the total number of users in the sub-circuit is less than or equal to 8, then only 1 user in this sub-circuit is selected according to the judgment rules 2.1, 2.2, and 2.3; if the total number of users in the sub-circuit is greater than 8 and less than or equal to 16, then 2 users in this sub-circuit are selected according to the judgment rules 2.1, 2.2, and 2.3; if the total number of users in the sub-circuit is greater than 16 and less than or equal to 50, then 3 users in this sub-circuit are selected according to the judgment rules 2.1, 2.2, and 2.3; if the total number of users in the sub-circuit is greater than 50, then 5 users in this sub-circuit are selected according to the judgment rules 2.1, 2.2, and 2.3.

[0167] 2.8 Before summoning the smart meter of user i, first detect whether the status information of this user is saved in the system.

[0168] 2.9 During the judgment process, record the status of all users summoned and save it for a certain period of time to meet Judgment Rule 2.8, so that the subsequent judgment can summon the smart meters of users as few times as possible, thereby shortening the overall judgment time.

[0169] 2.10 After the judgment is completed, analyze all users affected by the power outage through the topological structure, record and save it for a certain period of time for subsequent execution of Step 1.1.

[0170] 2.11 During the judgment process, if it is found that some users have a power outage and it is not due to arrears, and they do not belong to the user scope described in Judgment Rule 2.10, these users need to be recorded, and after the end of this judgment, continue to execute Step 1.1 for these users to ensure that all power outage intervals can be obtained.

[0171] Embodiment 2

[0172] This embodiment takes the topological structure as shown in Figure 3 as an example, and uses the inventive concept provided by the present invention to determine the power outage interval. There are 208 users in this substation area, and the distribution transformer in this substation area supplies power to downstream users through three main lines. These three main lines are denoted as main lines A, B, and C. Main line A is divided into sub-lines A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, and each of these sub-lines supplies power to 8 users. For example, sub-line A1 supplies power to 8 users. Main line B is divided into sub-lines B1, B2, B3, B4, B5, and each of these sub-lines supplies power to 16 users. Main line C is divided into sub-lines C1, C2, C3, C4, C5, C6, and each of these sub-lines supplies power to 8 users. It should be noted that this embodiment takes three main lines as an example, but is not limited to this topological structure.

[0173] Analyzed the historical power outage information of this substation area in 2017, and the relevant information is summarized as follows:

[0174] a. User classification

[0175] 7 important users: numbered 15, 32, 48, 49, 153, 172, 206 respectively

[0176] 5 sensitive users: numbered 21, 37, 58, 131, 43 respectively

[0177] 4 users have had arrears records but have not had power outages: numbered 98, 135, 141, 155 respectively

[0178] 9 users have had arrears records and have had power outages: numbered 41, 64, 78, 85, 91, 99, 106, 120, 160 respectively

[0179] 5 users had more than 1 power outage in history but no subsequent power outages: their numbers are 7, 77, 112, 147, and 182 respectively

[0180] 9 users had subsequent power outages on the premise that they had more than 1 power outage in history: their numbers are 42, 57, 71, 77, 85, 148, 162, 169, and 183 respectively

[0181] Among them, user No. 85 belongs to both the 4th and 6th types of the above user types. Its historical arrears record is 4. There was a total of 1 power outage event in this year. Before this power outage, its historical power outage times were 2, and after this power outage, its historical power outage times became 3. In addition, for the other 8 users with arrears records and power outages, there were no power outage events before this power outage event in this year. For the other 8 users who had subsequent power outages on the premise that their historical power outage times exceeded 1, they had no arrears records

[0182] b. All users in this distribution area had a total of 19 power outages in 2017

[0183] The power outage events of 2 users occurred during normal times: their numbers are 71 and 91 respectively

[0184] The power outage events of 8 users occurred on weekends: their numbers are 37, 41, 42, 43, 57, 85, 120, and 160 respectively

[0185] The power outage events of 2 users occurred on ordinary festivals: their numbers are 78 and 183 respectively

[0186] The power outage events of 4 users occurred during the sweltering heat: their numbers are 64, 98, 148, and 169 respectively

[0187] The power outage events of 3 users occurred during the Spring Festival: their numbers are 99, 131, and 162 respectively

[0188] c. Actual case:

[0189] At around 9 pm on March 10, 2018, the 3rd user under the sub-line A6 of line A called to report a power outage event. We recorded this user as the user numbered 43 among these 208 users. First, we based on Figure 1 the main flow chart shown for main research and judgment:

[0190] Step 1. Check if there is a historical research and judgment record

[0191] After checking the relevant historical records, it is found that there is neither a planned power outage record for this user, nor a power outage record caused by a line failure of 10 kV or above, nor any other relevant research results. Therefore, it is necessary to determine whether the reason for the user's power outage is due to overdue payment.

[0192] Step 2. Determine whether it is a power outage due to overdue payment

[0193] After checking the relevant historical records, it is found that the user was not in an overdue payment status at 9:00 p.m. on March 10, 2018. Therefore, the next step is to recall the user's electric meter to see if there is a false power outage report.

[0194] Step 3. Determine whether there is a false power outage report

[0195] After checking the smart electric meter data of this user around 9:00 p.m. on March 10, 2018, it is found that the data of the smart electric meter during this period are all 0. Therefore, there is no false power outage report for this user, and the next step is to analyze the power outage scope of the transformer area.

[0196] Step 4. Analyze the power outage scope of the transformer area

[0197] First, according to the topological structure, it is known that there are 8 users in the sub-line A6 where user No. 43 is located. Then, according to the prediction model of power outage probability, calculate the power outage occurrence probability of these 8 users.

[0198] Calculate the power outage probability:

[0199] It is statistically found that there are a total of 1,814,998 pieces of historical power consumption information in the transformer area of this user in 2017, among which the historical power outage information is 19 pieces. The difference between the positive and negative samples is too large. Therefore, the training samples are selected according to the following rules:

[0200] 1. The selected sample data are evenly sourced from these 208 users

[0201] 2. Each user provides 1 piece of sample data for normal days, weekends, ordinary festivals, sweltering heat, and the Spring Festival respectively

[0202] 3. All power outage sample data are also required

[0203] Since there may be duplicates in the above data, the finally selected sample data is 1,057 pieces, 1057 = 208×5 + 17 (originally there were 19 pieces of power outage information, but due to 2 pieces of duplicate data, it is 17 pieces).

[0204] Based on the input training samples, it is obtained that: θ0 = 1.486, θ1 = -0.007, θ2 = -0.270, θ3 = -0.024, θ4 = -0.173, θ5 = -0.145, θ6 = 1.101, θ7 = 0.359, θ8 = 0.307.

[0205] The characteristics of user No. 43 at 9:00 p.m. on March 10, 2018 are shown in the following table:

[0206] Table 3 Corresponding relationship table of user No. 43's characteristics and characteristic values

[0207] Feature Feature value Explanation ID 43 User No. 43, so the ID is 43 TIME 21 9 p.m., so TIME is 21 QW 14 The air temperature is 14°C, so QW is 14 SFYDGFSQ 1 March 10, 2018 is a weekend, so SFYDGFSQ is 1 YHZYCD 0 User No. 43 is an ordinary user, so YHZYCD is 0 LSTDCS 2 The number of historical power outages of User No. 43 is 2, so LSTDCS is 2 LSQFCS 0 The number of historical overdue payments of User No. 43 is 0, so LSQFCS is 0 SFMGYH 1 User No. 43 is a sensitive user, so SFMGYH is 1

[0208] Therefore, {x1, x2, x3, x4, x5, x6, x7, x8} = {43, 21, 14, 1, 0, 2, 0, 1}

[0209] Then,

[0210] The calculated power outage probability is as follows:

[0211] User No. 41: 2.88% (relatively high number of historical arrears)

[0212] User No. 42: 5.81% (relatively high number of historical power outages)

[0213] User No. 43: 7.69% (sensitive user)

[0214] Users No. 44 and 45: 0.67%

[0215] Users No. 46 and 47: 0.66%

[0216] User No. 48: 0.56% (important user)

[0217] Among them, user No. 41 has a relatively high historical arrears record, user No. 42 has a relatively high number of historical power outages, and user No. 43 is a sensitive user, so their power outage probabilities should be relatively high. And user No. 48 is an important user. It is found in the user electricity consumption information table in 2017 that none of the important users have experienced a single power outage event, so its power outage probability should also be relatively low. Therefore, the result obtained is consistent with our assumption.

[0218] Next, make a judgment according to Figure 2 the flow chart of substation area power outage range analysis shown below:

[0219] Step 41. Judge whether the power outage interval is the sub-line A6 where the user is located

[0220] According to the set judgment rules, user No. 48 should be summoned first, and then users No. 46, 47, 44, 45, 41, and 42 should be summoned in turn. After checking the relevant historical records, it is found that the meter records of these users were all 0 at 9:00 p.m. on March 10, 2018. Therefore, it can be determined that the sub-line A6 of the main line A failed at that time, and it is necessary to further judge the main line A.

[0221] Step 42. Determine whether the power outage area is the main line A corresponding to the sub-line A6 where the user is located

[0222] There are 10 sub-lines in the main line A. Therefore, it is necessary to determine whether the other 9 sub-lines are faulty. According to the research and judgment rules, it is necessary to select one user with the lowest power outage probability from each of the other 9 sub-lines. In this way, 9 users can be selected, and then the electricity meters of the users with relatively low power outage probabilities among these 9 users are called in turn.

[0223] Power outage probabilities of all users on the 1st sub-line A1 of line A:

[0224] No. 1: 0.89%

[0225] Nos. 2 and 3: 0.88%

[0226] No. 4: 0.87%

[0227] Nos. 5 and 6: 0.86%

[0228] No. 7: 2.52% (relatively high historical power outage times)

[0229] No. 8: 0.85%

[0230] Power outage probabilities of all users on the 2nd sub-line A2 of line A:

[0231] Nos. 9 and 10: 0.84%

[0232] Nos. 11 and 12: 0.83%

[0233] Nos. 13 and 14: 0.82%

[0234] No. 15: 0.70% (important user)

[0235] No. 16: 0.80%

[0236] Power outage probabilities of all users on the 3rd sub-line A3 of line A:

[0237] No. 17: 0.80%

[0238] Nos. 18 and 19: 0.79%

[0239] No. 20: 0.78%

[0240] No. 21: 8.79% (sensitive user)

[0241] Nos. 22 and 23: 0.77%

[0242] No. 24: 0.76%

[0243] Power outage probabilities of all users on the 4th sub-line A4 of line A:

[0244] No. 25: 0.76%

[0245] Nos. 26 and 27: 0.75%

[0246] Nos. 28 and 29: 0.74%

[0247] Nos. 30 and 31: 0.73%

[0248] No. 32: 0.63% (important users)

[0249] Power outage probability of all users on the 5th sub-line A5 of Circuit A:

[0250] No. 33: 0.72%

[0251] Nos. 34, 35 and 36: 0.71%

[0252] No. 37: 43.95% (sensitive users)

[0253] No. 38: 0.70%

[0254] Nos. 39 and 40: 0.69%

[0255] Power outage probability of all users on the 7th sub-line A7 of Circuit A:

[0256] No. 49: 0.56% (important users)

[0257] Nos. 50, 51 and 52: 0.64%

[0258] Nos. 53 and 54: 0.63%

[0259] Nos. 55 and 56: 0.62%

[0260] Power outage probability of all users on the 8th sub-line A8 of Circuit A:

[0261] No. 57: 14.39% (higher historical power outage times)

[0262] No. 58: 18.5% (sensitive users)

[0263] No. 59: 0.61%

[0264] Nos. 60 and 61: 0.60%

[0265] Nos. 62 and 63: 0.59%

[0266] No. 64: 6.95% (higher historical overdue payment times)

[0267] Power outage probability of all users on the 9th sub-line A9 of Circuit A:

[0268] Nos. 65, 66 and 67: 0.58%

[0269] Numbers 68 and 69: 0.57%

[0270] Numbers 70 and 72: 0.56%

[0271] Number 71: 4.85% (relatively high historical power outage times)

[0272] Power outage probability of all users on the 10th sub-line A10 of line A:

[0273] Numbers 73, 74 and 75: 0.55%

[0274] Number 76: 0.54%

[0275] Number 77: 1.60% (relatively high historical power outage times)

[0276] Number 78: 2.27% (relatively high historical overdue payment times)

[0277] Numbers 79 and 80: 0.53%

[0278] Based on the power outage probabilities shown above, the 9 selected users are 8, 15, 24, 32, 39, 49, 62, 70, 79. According to the summoning principle, the final summoning order is: User No. 49, 32, 15, 79, 70, 62, 39, 24, 8. If it is found during the summoning process that the summoned user has a power outage due to overdue payment, then summon a user with a lower power outage probability under the corresponding sub-line of this user. If this user still has a power outage due to overdue payment, continue to summon until all users on this sub-line have been summoned. After checking the relevant historical records, it is found that the electricity meter records of these users were all 0 at 9 pm on March 10, 2018, and these users had no overdue payment records at this time point. Therefore, it can be determined that the main line A failed at that time, and it is necessary to further judge the main line B and the main line C, and then judge whether the power outage area is the entire transformer substation area.

[0279] Step 43. Judge whether the power outage area is the entire transformer substation area

[0280] Since each sub-line of the main line B has 16 users, according to the judgment rule, two users with lower power outage probability should be selected from each sub-line. And each sub-line of the main line C has 8 users, so only one user with the lowest power outage probability needs to be selected from each sub-line. The 10 users selected from the main line B are No. 94, 95, 110, 111, 125, 126, 142, 143, 153, and 159 respectively. The 6 users selected from the main line C are No. 164, 172, 180, 191, 197, and 206 respectively. Among them, No. 153, 172, and 206 are important users. The power outage probability of No. 153 is 0.28%, the power outage probability of No. 172 is 0.25%, and the power outage probability of No. 206 is 0.20%. So the final call order is: users No. 206, 172, 153, 197, 191, 180, 164, 159, 142, 143, 125, 126, 110, 111, 94, 95. After checking the relevant historical records, we found that the electricity meter record of user No. 206 was not zero at 9 pm on March 10, 2018, that is, the user was not in a power outage state. Then we also found that the electricity meter record of user No. 172 was not zero at 9 pm on March 10, 2018. So it can be determined that the power outage interval is not the entire distribution area. Therefore, the judgment result of the fault interval is the main line A, and the judgment is over.

[0281] Compared with the existing methods for determining the power outage range, the technical solution provided by the present invention has the following advantages:

[0282] If the relevant judgment rules are not used for the analysis of the power outage range in the distribution area, once a power outage event occurs, the operation and maintenance personnel cannot timely discover the power outage event. Even if the power outage event is discovered, they cannot quickly and accurately know the specific fault point, which will reduce the user satisfaction. However, if the judgment rules set by the present invention are used, the specific power outage interval in the distribution area can be quickly and accurately judged, and this information can be provided to the operation and maintenance personnel, which helps them quickly and accurately judge the specific fault point. At the same time, if other users of line A call in to inquire about the power outage reason, the specific power outage reason can also be immediately feedback.

[0283] In addition, most of the current smart electricity meters still cannot automatically upload power outage information, so the judgment process needs to be triggered by the way that users call in to inquire about the power outage reason. In the future, when smart electricity meters can automatically upload power outage information, the judgment process can be triggered more timely, and the specific power outage interval can also be obtained more timely.

[0284] Embodiment 3

[0285] Based on the same inventive concept, the present invention also provides a system for determining the power outage interval in a low-voltage distribution network, including:

[0286] The first judgment module is used to query whether there is a judgment result or a power outage record when there are power outage users in the current area of the low-voltage distribution network. If there is, it determines and returns the power outage interval based on the judgment result or the power outage record, and ends the process at the same time. Otherwise, it calls the second judgment module;

[0287] The second judgment module is used to judge whether it is a fault of the power outage user itself based on summoning the electricity meter information of the power outage user. If it is a fault of the power outage user itself, it returns the self-fault. Otherwise, it calls the third judgment module;

[0288] The third judgment module is used to analyze the power outage interval of the area based on the topological structure of the area and the power outage probability of users in the low-voltage distribution network, determine the power outage interval corresponding to the power outage user, and return the judgment result generated based on the power outage user and the power outage interval corresponding to the power outage user.

[0289] In the embodiment, the third judgment module includes:

[0290] The 301st judgment unit is used to select the sub-line users to be judged from the sub-line where the power outage user is located according to the set sub-line judgment rule based on the topological structure of the area and the power outage probability of users; summon the electricity meters corresponding to the sub-line users to be judged in turn according to the summoning rule, and judge and record the status of the electricity meters at the same time; when it is found that two users have power, stop summoning, determine the power outage interval as the sub-line where the power outage user is located, and return the judgment result generated based on the electricity meter status, the power outage user and the power outage interval corresponding to the power outage user; otherwise, call the 302nd judgment unit;

[0291] The 302nd judgment unit is used to select the main-line users to be judged from the other sub-lines except the main line to which the sub-line where the power outage user is located belongs according to the set judgment rules of the main line and the whole area based on the topological structure of the area and the power outage probability of users; summon the electricity meters corresponding to the main-line users to be judged in turn according to the summoning rule, and judge and record the status of the electricity meters at the same time; when it is found that two users have power, stop summoning, determine the power outage interval as the main line corresponding to the sub-line where the power outage user is located, and return the judgment result generated based on the electricity meter status, the power outage user and the power outage interval corresponding to the power outage user; otherwise, call the 303rd judgment unit;

[0292] The 303rd judgment unit is used to select the users to be judged in the power distribution area from all the sub-lines under other main lines except the main line where the power-off user is located according to the set main line and the judgment rules of the entire power distribution area based on the topological structure of the power distribution area and the power-off probability of users; call the electricity meters corresponding to the users to be judged in the power distribution area in turn according to the calling rules, and at the same time judge and record the status of the electricity meters; when it is found that two users have power, stop calling, determine that the power-off interval is not the entire power distribution area, and return the judgment result generated based on the electricity meter status, the power-off user, and the power-off interval corresponding to the power-off user; otherwise, the power-off interval is the entire power distribution area.

[0293] In the embodiment, the 301st judgment unit includes:

[0294] The first rule-making sub-unit is used to make the judgment rules for sub-lines; the judgment rules for sub-lines include:

[0295] When the total number of users on the sub-line where the power-off user is located ≤ 8, then select all the other users on the sub-line except the power-off user according to the preset calling rules;

[0296] When the total number of users on the sub-line where the power-off user is located > 8 and ≤ 16, then select the first 8 users with a lower power-off probability among all the other users on the sub-line except the power-off user according to the preset calling rules;

[0297] When the total number of users on the sub-line where the power-off user is located < 16 and ≤ 50, then select the first 10 users with a lower power-off probability among all the other users on the sub-line except the power-off user according to the preset calling rules;

[0298] When the total number of users on the sub-line where the power-off user is located > 50, then select the first 15 users with a lower power-off probability among all the other users on the sub-line except the power-off user according to the preset calling rules.

[0299] In the embodiment, the 302nd judgment unit includes:

[0300] The second rule-making sub-unit is used to make the judgment rules for the main line and the entire power distribution area; the judgment rules for the main line and the entire power distribution area include:

[0301] If the total number of users on the sub-line ≤ 8, then select 1 user on the sub-line according to the preset calling rules;

[0302] If the total number of users on the sub-line > 8 and ≤ 16, then select 2 users on the sub-line according to the preset calling rules;

[0303] If the total number of users in a sub - circuit is less than 16 and ≤ 50, then 3 users in the sub - circuit are selected according to a preset calling rule;

[0304] If the total number of users in a sub - circuit is greater than 50, then 5 users in the sub - circuit are selected according to a preset calling rule.

[0305] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer - usable storage media (including but not limited to disk storage, CD - ROM, optical storage, etc.) containing computer - usable program code.

[0306] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general - purpose computer, a special - purpose computer, an embedded processor, or other programmable data - processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data - processing devices generate means for realizing the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0307] These computer program instructions can also be stored in a computer - readable memory that can direct a computer or other programmable data - processing device to work in a specific manner, so that the instructions stored in the computer - readable memory generate a manufactured article including instruction means, and the instruction means realizes the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0308] These computer program instructions can also be loaded onto a computer or other programmable data - processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer - implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0309] The above are only embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included within the scope of the claims of the present invention pending approval.

Claims

1. A method for determining a power outage section in a low-voltage distribution network, characterized in that, Including: S1. When there are power outage users in the current substation area of the low-voltage distribution network, query whether there is a research and judgment result or a power outage record. If there is, determine and return the power outage interval based on the research and judgment result or the power outage record, and end the process at the same time. Otherwise, execute S2; S2. Based on calling the meter information of the power outage user, judge whether it is a fault of the power outage user itself. If it is a fault of the power outage user itself, return the self-fault. Otherwise, execute S3; S3. Conduct a power outage interval analysis of the substation area based on the topological structure of the substation area and the power outage probability of users in the low-voltage distribution network, determine the power outage interval corresponding to the power outage user, and return the research and judgment result generated based on the power outage user and the power outage interval corresponding to the power outage user; The judging whether it is a fault of the power outage user itself based on calling the meter information of the power outage user includes: S201. Judge whether the power outage user is out of power due to arrears. When it is out of power due to arrears, return that it is out of power due to arrears. Otherwise, proceed to step S202; S202. Call the meter data of the power outage user. When it can be called, return a power outage false alarm. Otherwise, it is not a power outage false alarm; The self-fault includes: out of power due to arrears and power outage false alarm; The conducting a power outage interval analysis of the substation area based on the topological structure of the substation area and the power outage probability of users in the low-voltage distribution network, determining the power outage interval corresponding to the power outage user, and returning the research and judgment result generated based on the power outage user and the power outage interval corresponding to the power outage user includes: S301. Based on the topological structure of the substation area and the power outage probability of users, select the sub-line users to be researched and judged from the sub-line where the power outage user is located according to the set sub-line research and judgment rules; call the meters corresponding to the sub-line users to be researched and judged in sequence according to the calling rules, and judge and record the status of the meters at the same time; when it is found that two users have power, stop calling, determine the power outage interval as the sub-line where the power outage user is located, and return the research and judgment result generated based on the meter status, the power outage user, and the power outage interval corresponding to the power outage user; otherwise, execute S302; S302. Based on the topological structure of the substation area and the power outage probability of users, select the main-line users to be researched and judged from the other sub-lines except the main line to which the sub-line where the power outage user is located belongs according to the set main-line and whole-substation area research and judgment rules; call the meters corresponding to the main-line users to be researched and judged in sequence according to the calling rules, and judge and record the status of the meters at the same time; when it is found that two users have power, stop calling, determine the power outage interval as the main line corresponding to the sub-line where the power outage user is located, and return the research and judgment result generated based on the meter status, the power outage user, and the power outage interval corresponding to the power outage user; otherwise, execute S303; S303. Select the users to be judged in the power distribution area from all the sub-lines under other main lines except the main line where the power outage user is located according to the topological structure of the power distribution area and the power outage probability of the users, in accordance with the set judgment rules for the main line and the entire power distribution area; summon the electric meters corresponding to the users to be judged in the power distribution area in turn according to the summons rule, and at the same time judge and record the status of the electric meters; when it is found that two users have power, stop the summons, determine that the power outage interval is not the entire power distribution area, and return the judgment result generated based on the electric meter status, the power outage user, and the power outage interval corresponding to the power outage user; otherwise, the power outage interval is the entire power distribution area. The sub-line judgment rules include: When the total number of users in the sub-line where the power outage user is located ≤ 8, then select all the users in the sub-line except the power outage user according to the preset summons rule. When the total number of users in the sub-line where the power outage user is located > 8 and ≤ 16, then select the first 8 users with a lower power outage occurrence probability among all the users in the sub-line except the power outage user according to the preset summons rule. When the total number of users in the sub-line where the power outage user is located > 16 and ≤ 50, then select the first 10 users with a lower power outage occurrence probability among all the users in the sub-line except the power outage user according to the preset summons rule. When the total number of users in the sub-line where the power outage user is located > 50, then select the first 15 users with a lower power outage occurrence probability among all the users in the sub-line except the power outage user according to the preset summons rule. The judgment rules for the main line and the entire power distribution area include: If the total number of users in the sub-line ≤ 8, then select 1 user from the sub-line according to the preset summons rule. If the total number of users in the sub-line > 8 and ≤ 16, then select 2 users from the sub-line according to the preset summons rule. If the total number of users in the sub-line > 16 and ≤ 50, then select 3 users from the sub-line according to the preset summons rule. If the total number of users in the sub-line > 50, then select 5 users from the sub-line according to the preset summons rule.

2. The determination method according to claim 1, characterized in that The summons rule includes: Based on the pre-established prediction model of the power outage occurrence probability of low-voltage users in the distribution network, predict the power outage occurrence probability of all users in the power distribution area within the future set time. Give priority to summoning the electric meters of users with a lower power outage occurrence probability. If the difference in the power outage occurrence probability between users is 0.5%, then according to the importance of the users, give priority to summoning the electric meters of important users. If the importance of users is the same, then give priority to summoning the electric meters of users with fewer historical arrears. If there are at least 2 users with the same number of historical arrears, then summon them in turn according to the ID of each user.

3. The determination method according to claim 2, characterized in that, The prediction model of the power outage occurrence probability of low-voltage users in the distribution network includes: Obtain the historical data of users in all low-voltage distribution networks in the current power distribution area. In the historical data, screen out the training samples that meet the set characteristics. Based on the training samples and the logistic regression algorithm, determine the weight parameters. Based on the weight parameters and the historical data of users, obtain the power outage probability of all users in the power distribution area. Among them, the features include: user ID, power outage time, temperature, whether it is a peak electricity consumption period, user importance level, user's historical power outage times, user's historical arrears times, and whether the user is a sensitive user.

4. The determination method according to claim 3, characterized in that The power outage probability is calculated according to the following formula: In the formula: p: power outage probability; θ n : Weight coefficient; x n : The nth feature of the user; y: The label indicating whether the user has a power outage.

5. The determination method according to claim 1, wherein The sub-line judgment rules, the main-line judgment rules, and the judgment rules for the entire distribution area also include an arrears judgment rule.

6. The determination method according to claim 5, characterized in that, The arrears judgment rule includes: In the sub-line judgment rule, when the total number of users in the sub-line where the power outage user is located > 8, when summoning the users to be judged in the sub-line according to the summoning rule, or in the main-line and the judgment rules for the entire distribution area, when summoning the users to be judged in the main-line or the users to be judged in the distribution area according to the summoning rule, if a user has a power outage, it is judged whether the current user has an arrears power outage, otherwise it is not judged whether the current user has an arrears power outage; If the current user belongs to an arrears power outage, then continue to select users with a lower power outage probability from the sub-line corresponding to the current user to join the users to be judged in the sub-line, the users to be judged in the main-line, or the users to be judged in the distribution area, and summon the electricity meters according to the summoning rule.

7. The determination method according to claim 1, characterized in that, The recording of the status of the electricity meter includes: The user's electricity meter is powered on and the user's electricity meter is powered off.

8. The determination method according to claim 7, wherein The determination that the power outage interval is not the entire distribution area and the return of the judgment result generated based on the electricity meter status and the power outage interval also includes: Recording the users who have a power outage and are not in arrears for power outage, and for the users who have a power outage and are not in arrears for power outage, continue to judge according to S1 to S3, and add the newly generated judgment result to the previous judgment result.

9. The determination method according to claim 1, characterized in that The power outage record includes: The planned power outage record and the fault power outage record caused by faults in lines of 10 kV and above.

10. A determination system applicable to the determination method of a power outage section in a low-voltage distribution network according to claim 1, characterized in that, It includes: The first judgment module is used to, when there are power outage users in the current distribution area of the low-voltage distribution network, query whether there is a judgment result or a power outage record. If there is, determine and return the power outage interval based on the judgment result or the power outage record, and at the same time end the process; otherwise, call the second judgment module; The second judgment module is used to, based on summoning the electricity meter information of the power outage user, judge whether it is a fault of the power outage user itself. If it is a fault of the power outage user itself, return the self-fault; otherwise, call the third judgment module; The third judgment module is used to analyze the power outage interval of the distribution area based on the topological structure of the distribution area and the power outage probability of users in the low-voltage distribution network, determine the power outage interval corresponding to the power outage user, and return the judgment result generated based on the power outage user and the power outage interval corresponding to the power outage user; The third judgment module includes: The 301 judgment unit is used to select the users to be judged in the sub-line from the sub-line where the power outage user is located based on the topological structure of the distribution area and the power outage probability of the users according to the set sub-line judgment rules; summon the electricity meters corresponding to the users to be judged in the sub-line in turn, and at the same time judge and record the status of the electricity meters; when it is found that two users are powered on, stop summoning, determine the power outage interval as the sub-line where the power outage user is located, and return the judgment result generated based on the electricity meter status, the power outage user, and the power outage interval corresponding to the power outage user; otherwise, call the 302 judgment unit; The 302nd judgment unit is used to select the main line users to be judged from the other sub-lines except the main line to which the sub-line where the power outage user is located belongs according to the set main line and the judgment rules of the entire distribution area based on the topological structure of the distribution area and the power outage probability of users; call the electricity meters corresponding to the main line users to be judged in turn according to the call rules, and at the same time judge and record the status of the electricity meters; when it is found that two users have power, stop calling, determine the power outage interval as the main line corresponding to the sub-line where the power outage user is located, and return the judgment result generated based on the electricity meter status, the power outage user, and the power outage interval corresponding to the power outage user; otherwise, call the 303rd judgment unit; The 303rd judgment unit is used to select the distribution area users to be judged from all sub-lines under other main lines except the main line where the power outage user is located according to the set main line and the judgment rules of the entire distribution area based on the topological structure of the distribution area and the power outage probability of users; call the electricity meters corresponding to the distribution area users to be judged in turn according to the call rules, and at the same time judge and record the status of the electricity meters; when it is found that two users have power, stop calling, determine that the power outage interval is not the entire distribution area, and return the judgment result generated based on the electricity meter status, the power outage user, and the power outage interval corresponding to the power outage user; otherwise, the power outage interval is the entire distribution area; The 301st judgment unit includes: The first rule-making subunit is used to formulate the sub-line judgment rules; the sub-line judgment rules include: When the total number of users in the sub-line where the power outage user is located ≤ 8, then select all the other users in the sub-line except the power outage user according to the preset call rules; When the total number of users in the sub-line where the power outage user is located > 8 and ≤ 16, then select the first 8 users with a lower power outage probability among all the other users in the sub-line except the power outage user according to the preset call rules; When the total number of users in the sub-line where the power outage user is located > 16 and ≤ 50, then select the first 10 users with a lower power outage probability among all the other users in the sub-line except the power outage user according to the preset call rules; When the total number of users in the sub-line where the power outage user is located > 50, then select the first 15 users with a lower power outage probability among all the other users in the sub-line except the power outage user according to the preset call rules; The 302nd judgment unit includes: The second rule-making subunit is used to formulate the judgment rules of the main line and the entire distribution area; the judgment rules of the main line and the entire distribution area include: If the total number of users in the sub-line ≤ 8, then select 1 user from the sub-line according to the preset call rules; If the total number of users in the sub-line > 8 and ≤ 16, then select 2 users from the sub-line according to the preset call rules; If the total number of users in the sub-line > 16 and ≤ 50, then select 3 users from the sub-line according to the preset call rules; If the total number of users in the sub-line > 50, then select 5 users from the sub-line according to the preset call rules.

Images