Electricity stealing clue screening method based on long-period electricity utilization information

Abstract

The invention relates to an electricity stealing clue screening method based on long-period electricity utilization information. The first part is to add enterprise name keyword filtering; firstly, relevant information of suspected users is imported into a secondary study and judgment auxiliary tool, and keyword filtering is carried out by adding a screening rule, so that power utilization suspected users are reduced. And the second part is to check the ultra-long time span power utilization data. According to the method, power utilization data of suspected users and basic archive information of the users are acquired, a visual interface of power utilization data such as current and voltage of the suspected users with a super-long time span is realized, a power utilization data curve of 3 months can be displayed, and a mode of reading the power utilization data of the users day by day by using a power utilization information acquisition system for analysis, study and judgment is abandoned; and the searching, studying and judging time of the abnormal power consumption data of the suspected user is shortened.

Description

Technical Field

[0001] The present invention belongs to the technical field of electricity theft clue screening, and specifically relates to an electricity theft clue screening method based on long-term electricity usage information. Background Art

[0002] Modify the current coil of the short-circuit metering device. Usually, the current coil is short-circuited with a wire inside or outside the energy meter. The more common practice is to insert a wire or a parallel resistor into the phase line input and output of the energy meter to play a shunt role.

[0003] If the wire is short-circuited and the wire resistance is almost zero, most of the current will pass through the short-circuited wire, and there will be almost no current in the current coil of the energy meter, causing the energy meter to stop rotating; if a resistor smaller than the resistance value of the current coil is connected in parallel, the current coil and the parallel resistor form a parallel circuit. According to the shunt principle of the parallel circuit, most of the current will pass through the parallel resistor, and only a small current will pass through the current coil, causing the energy meter to rotate slowly at a certain ratio, thereby achieving the purpose of stealing electricity. This kind of electricity stealing method is called undercurrent theft.

[0004] Anti-theft and illegal investigation work is an important guarantee for protecting the legitimate operating rights and interests of power supply companies and maintaining the normal order of power supply and use in accordance with the law. The Inspection and Anti-Theft Center has established a provincial electricity inspection team. Just as the existing technology patent number is "CN201611025302.X" and the patent name is "Regional Distributed Photovoltaic Electricity Theft Supervision System", it uses the analyzed illegal theft clues to actively carry out provincial direct investigations. In order to ensure that the company's electricity bills are returned to the warehouse, the threshold of the anti-theft diagnosis model is moderately relaxed, and the number of suspected users output is increased. After the model is used to conduct a preliminary screening of abnormal users, the power consumption information collection system and the marketing business application system are manually applied to analyze the monthly historical data of the suspected users output by the model in detail, and the suspected illegal users are locked in to ensure that the anti-theft investigation is not missed. For the suspected illegal users output by the model, it is necessary to comprehensively analyze their user files, electricity consumption data and other information to determine whether they have illegal theft behavior. At present, this process is time-consuming and inefficient.

[0005] That is to say, Figure 1 As shown in the figure, the current method of screening and analyzing electricity theft clues is to ensure that the provincial direct investigation is carried out every week. The anti-electricity theft model sets wide parameters and outputs a large number of suspected users of abnormal electricity consumption every week. Among them, public schools and hospitals are mainly for lighting, and there is an imbalance in electricity consumption, or large state-owned enterprises such as banks and mobile phones lack electric motors for theft, so there is no suspicion of electricity theft. Through investigation, it was found that the proportion of misjudged users output by the model reached 34.73%, and analysts also need to spend time to identify a large amount of useless information, which greatly increases the time for clue screening.

[0006] Through investigation and statistics, we have analyzed the time span of the anti-electricity theft personnel's analysis of theft clues, and the time it takes for the analysts to query data in the marketing business system and the electricity information collection system. In particular, the electricity data needs to be checked on a daily basis to analyze the start point of the user's abnormal electricity consumption, infer the user's actual electricity load, and confirm whether there is an abnormality. This process consumes a lot of working time, accounting for more than 90% of the total time. Summary of the invention

[0007] In order to solve the defects in the prior art, the present invention proposes a method for screening clues of electricity theft based on long-term electricity consumption information, the first part of which is to add enterprise name keyword filtering. First, the relevant information of the suspected user is imported into the secondary analysis and judgment auxiliary tool, and keyword filtering is performed by adding screening rules to reduce the number of suspected electricity users. The second part is to view the electricity consumption data over a very long time span. The electricity consumption data and basic profile information of the suspected user are obtained, and a visualization interface of the current, voltage and other electricity consumption data of the suspected user over a very long time span is realized. The electricity consumption data curve for 3 months can be displayed, and the method of using the electricity collection information system to flip through the user's electricity consumption data on a daily basis for analysis and judgment is abandoned, thereby shortening the time for finding and judging abnormal electricity consumption data of suspected users.

[0008] The present invention uses the following technical solutions.

[0009] A method for screening electricity theft clues based on long-term electricity usage information, comprising:

[0010] Step 1: Obtain the electricity consumption data of the suspected user;

[0011] Step 2: Combine the electricity consumption data of suspected users to filter the electricity consumption data of suspected users according to keywords;

[0012] Step 3: Set the start and end time and data category of the electricity consumption details of the remaining suspected users after deleting the misjudged suspected users, so as to obtain the corresponding electricity consumption data curve;

[0013] Step 4: judging whether the corresponding suspected user is a suspected electricity theft user on the electricity consumption data curve;

[0014] Step 5: End the electricity theft clue screening method based on long-term electricity usage information.

[0015] Preferably, step 1 specifically includes:

[0016] Obtain the electricity usage data and basic profile information of suspected users.

[0017] Preferably, the suspected user's electricity usage data is electricity usage data of suspected users who may steal electricity obtained through an anti-electricity theft intelligent diagnosis model, and the electricity usage data includes the voltage, current and amount of electricity used by the suspected user.

[0018] Preferably, the basic profile information of the suspected user includes the user's category, and the user's category includes the user's belonging to factories, enterprises, companies, households, tenants, governments, middle schools, elementary schools, universities, hospitals, the military, municipal administration, management offices, urban management, high schools, street lights, construction bureaus, kindergartens, power grids, troops, state organs, construction offices, subways, mobile, Unicom, telecommunications, law enforcement bureaus, water supply, heating, electricity supply, banks, public security, transportation bureaus, substations, post offices, police, brigades, civil affairs, police stations, transportation bureaus, management bureaus, Communist Party organs, county committees, and petition bureaus.

[0019] Preferably, step 2 specifically includes:

[0020] In the secondary analysis software, keywords are input to combine with the electricity consumption data of suspicious users to filter the electricity consumption data of suspicious users according to the keywords.

[0021] Preferably, in the secondary judgment software, the method of filtering the electricity usage data of the suspected users according to the keywords by entering the keywords and combining them with the electricity usage data of the suspected users includes:

[0022] In the secondary analysis software, a keyword filtering link is added, that is, the attribute names of users such as government agencies, public institutions, and communication base stations that are unlikely to steal electricity are collected, and a keyword thesaurus is established. The basic file information of the suspected users in the suspected users' electricity usage data can be automatically filtered out, and the basic file information of the suspected users whose basic file information matches the keywords in the keyword library can be used to eliminate the suspected users who were misjudged in the initial screening of clues by the anti-electricity theft intelligent diagnosis model.

[0023] Preferably, the method for establishing a keyword database includes:

[0024] By typing the keywords in the keyword column of the secondary analysis software, the keywords are then stored in the keyword library.

[0025] Preferably, the attribute names of such users who are unlikely to steal electricity, such as government agencies, public institutions, and communication base stations, specifically include:

[0026] Such attributes include government, middle school, elementary school, university, hospital, military, municipal administration, management office, urban management, high school, street lights, construction bureau, kindergarten, power grid, troops, state organs, construction office, subway, mobile, Unicom, telecommunications, law enforcement bureau, water supply, heating, power supply, bank, public security, transportation bureau, substation, post office, police, brigade, civil affairs, police station, transportation bureau, management bureau, Communist Party of China organs, county party committee, and petition bureau.

[0027] Preferably, step 3 specifically includes:

[0028] By calling the electricity consumption collection system, the voltage, current, and electricity consumption data of the remaining suspect users in the past three months as data types are extracted as the start and end time of electricity consumption details after deleting the misjudged suspect users, and the least squares method is used to derive and draw the electricity consumption data curve of the suspect users over time, so as to realize the visualization interface of the current, voltage, and electricity consumption data of the suspect users over a very long time span, which can display the electricity consumption data curve of three months;

[0029] The method of extracting the voltage, current, and power consumption data of the remaining suspect users in the past three months as data types after deleting the misjudged suspect users as the start and end time of the electricity consumption details by calling the electricity consumption collection system includes:

[0030] The electricity consumption collection system is communicatively connected to a corresponding device that runs a method for screening electricity theft clues based on long-term electricity consumption information;

[0031] The electricity consumption collection system transmits the electricity consumption data of the remaining suspect users in the past three months as data types, such as voltage, current, and power, stored therein as the start and end time of electricity consumption details after deleting the misjudged suspect users, to the corresponding device;

[0032] After the electricity consumption collection system transmits the electricity consumption data of the remaining suspect users in the past three months as data types, such as voltage, current, and power, stored in the electricity consumption collection system as the start and end time of electricity consumption details after deleting the misjudged suspect users, to the corresponding device, the method executed in the corresponding device further includes:

[0033] Step 3-1: The corresponding device receives and stores the electricity consumption data transmitted by the electricity consumption collection system;

[0034] Step 3-2: Perform anti-modification isolation on the power consumption data through the coding isolation mode;

[0035] Step 3-3: Regularly detect whether a modification operation occurs in the power consumption data through a detection method;

[0036] Step 3-4: Perform replacement restoration on the modification operations existing in the power usage data via the copied power usage data;

[0037] Step 3-5: Perform enhanced isolation for power consumption data via system isolation mode. The system here can be a power consumption collection system.

[0038] A device for screening electricity theft clues based on long-term electricity usage information, comprising:

[0039] An acquisition module, which is used to acquire the electricity consumption data of suspected users;

[0040] A filtering module, which is used to filter the suspected user's electricity usage data according to keywords in combination with the suspected user's electricity usage data;

[0041] A setting module, which is used to set the start and end time and data category of the electricity consumption details of the remaining suspected users after deleting the misjudged suspected users, so as to obtain the corresponding electricity consumption data curve;

[0042] A judgment module, which is used to judge whether the corresponding suspected user is a suspected user of electricity theft on the electricity consumption data curve;

[0043] The end module is used to end the electricity theft clue screening method based on long-term electricity consumption information.

[0044] The beneficial effect of the present invention is that, compared with the prior art, the present invention mainly solves the problem of time-consuming manual screening of clues of electricity theft, and reduces the number of suspected electricity users by adding a whitelist of suspected electricity users and expanding the query time span, thereby effectively saving the working time consumed in screening clues of electricity theft. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic diagram of a prior art method for screening electricity theft clues;

[0046] Figure 2 It is a schematic diagram of an interface for filtering the electricity consumption data of suspicious users according to the keywords by entering keywords in the secondary judgment software of the present invention;

[0047] Figure 3 It is a flow chart of the method for screening electricity theft clues based on long-term electricity usage information described in the present invention;

[0048] Figure 4 It is a structural diagram of the electricity theft clue screening device based on long-term electricity usage information described in the present invention;

[0049] Figure 5 It is a schematic diagram of the interface of the electricity consumption data curve described in the present invention;

[0050] Figure 6 It is a schematic diagram of the software interface in the present invention. DETAILED DESCRIPTION

[0051] Modify the current coil of the short-circuit metering device. Usually, the current coil is short-circuited with a wire inside or outside the energy meter. The more common practice is to insert a wire or a parallel resistor into the phase line input and output of the energy meter to play a shunt role.

[0052] If the wire is short-circuited and the wire resistance is almost zero, most of the current will pass through the short-circuited wire, and there will be almost no current in the current coil of the energy meter, causing the energy meter to stop rotating; if a resistor smaller than the resistance value of the current coil is connected in parallel, the current coil and the parallel resistor form a parallel circuit. According to the shunt principle of the parallel circuit, most of the current will pass through the parallel resistor, and only a small current will pass through the current coil, causing the energy meter to rotate slowly at a certain ratio, thereby achieving the purpose of stealing electricity. This kind of electricity stealing method is called undercurrent theft.

[0053] The current method of screening and analyzing electricity theft clues is to ensure that the provincial direct investigation is carried out every week. The anti-electricity theft model sets wide parameters and outputs a large number of suspected users of abnormal electricity consumption every week. Among them, public schools and hospitals are mainly lighting-based, and there is an imbalance in electricity consumption, or large state-owned enterprises such as banks and mobile phones lack electric theft motors, and there is no suspicion of electricity theft. Through investigation, it was found that the proportion of misjudged users output by the model reached 34.73%, and analysts also need to spend time to identify a large amount of useless information, which greatly increases the time for clue screening. Through investigation and statistics, the time span of anti-electricity theft personnel's analysis of electricity theft clues was counted. It takes time for analysts to query data in the marketing business system and the electricity information collection system, especially the electricity data needs to be checked on a daily basis, analyze the starting point of the user's abnormal electricity consumption, infer the user's actual electricity load, and confirm whether there is an abnormality. This process consumes a lot of working time, accounting for more than 90% of the total time.

[0054] The software of the present invention is divided into two parts. The first part is to add enterprise name keyword filtering. First, the relevant information of the suspected user is imported into the secondary judgment auxiliary tool, and keyword filtering is performed by adding screening rules, thereby reducing the number of suspected electricity users.

[0055] The second part is to view the electricity consumption data over a long period of time. The electricity consumption data and basic profile information of suspected users are obtained, and a visualization interface of the current, voltage and other electricity consumption data of suspected users over a long period of time is realized. The electricity consumption data curve for three months can be displayed, and the method of using the electricity consumption information collection system to review the user's electricity consumption data on a daily basis for analysis and judgment is abandoned, thus shortening the time for finding and judging abnormal electricity consumption data of suspected users.

[0056] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely expressed in combination with the drawings in the embodiments of the present invention. The embodiments expressed in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0057] like Figure 3 As shown, the present invention provides a method for screening electricity theft clues based on long-term electricity usage information, comprising:

[0058] Step 1: Obtain the electricity consumption data of the suspected user;

[0059] In a preferred but non-limiting embodiment of the present invention, step 1 specifically comprises:

[0060] Obtain the electricity usage data and basic profile information of suspected users.

[0061] In a preferred but non-limiting embodiment of the present invention, the suspected user's electricity consumption data is the electricity consumption data of the suspected user who may steal electricity obtained by the anti-electricity theft intelligent diagnosis model, and the electricity consumption data includes the voltage, current and amount of electricity used by the suspected user. The anti-electricity theft intelligent diagnosis model can be such a model in the prior art.

[0062] In a preferred but non-limiting embodiment of the present invention, the basic profile information of the suspected user includes the user's category, and the user's category includes the user's belonging to factories, enterprises, companies, households, tenants, governments, middle schools, elementary schools, universities, hospitals, the military, municipal administration, management offices, urban management, high schools, street lights, construction bureaus, kindergartens, power grids, troops, state organs, construction offices, subways, mobile, Unicom, telecommunications, law enforcement bureaus, water supply, heating, power supply, banks, public security, transportation bureaus, substations, post offices, police, brigades, civil affairs, police stations, transportation bureaus, management bureaus, Communist Party organs, county committees, and petition bureaus.

[0063] Step 2: Combine the electricity consumption data of suspected users to filter the electricity consumption data of suspected users according to keywords;

[0064] In a preferred but non-limiting embodiment of the present invention, step 2 specifically comprises:

[0065] In the secondary analysis software, keywords are input to combine with the electricity consumption data of suspicious users to filter the electricity consumption data of suspicious users according to the keywords.

[0066] like Figure 2 As shown, in a preferred but non-limiting embodiment of the present invention, a method for filtering the electricity consumption data of suspicious users according to keywords by entering keywords in the secondary judgment software in combination with the electricity consumption data of suspicious users includes:

[0067] In the secondary analysis software, a keyword filtering link is added, that is, the attribute names of users such as government agencies, public institutions, and communication base stations that are unlikely to steal electricity are collected, and a keyword thesaurus is established. The basic file information of the suspected users in the suspected users' electricity usage data can be automatically filtered out, and the basic file information of the suspected users whose basic file information matches the keywords in the keyword library can be used to eliminate the suspected users who were misjudged in the initial screening of clues by the anti-electricity theft intelligent diagnosis model.

[0068] In a preferred but non-limiting embodiment of the present invention, the method for establishing a keyword library comprises:

[0069] By typing the keywords in the keyword column of the secondary analysis software, the keywords are then stored in the keyword library.

[0070] In a preferred but non-limiting embodiment of the present invention, the attribute names of such users who are unlikely to steal electricity, such as government agencies, public institutions, and communication base stations, specifically include:

[0071] Such attributes include government, middle school, elementary school, university, hospital, military, municipal administration, management office, urban management, high school, street lights, construction bureau, kindergarten, power grid, troops, state organs, construction office, subway, mobile, Unicom, telecommunications, law enforcement bureau, water supply, heating, power supply, bank, public security, transportation bureau, substation, post office, police, brigade, civil affairs, police station, transportation bureau, management bureau, Communist Party of China organs, county party committee, and petition bureau.

[0072] By typing in keywords, the secondary analysis software can eliminate users with keywords in their user information, such as government, middle school, elementary school, university, etc. These users have not stolen electric motors and can be considered not to be suspected of electricity theft. After eliminating interference factors, the test screening clues dropped from 2991 to 1335, and the secondary analysis clues dropped by 55.4%, improving the efficiency of abnormal user data analysis and reducing the time to identify a large amount of useless information. At the same time, in the secondary manual screening process, through manual analysis of electricity customer categories, users who are not suspected of electricity theft, such as transportation bureaus, law enforcement bureaus, public security bureaus, etc., are expanded, and the keyword thesaurus is continuously enriched to achieve the best filtering effect.

[0073] Step 3: Set the start and end time and data category of the electricity consumption details of the remaining suspected users after deleting the misjudged suspected users, so as to obtain the corresponding electricity consumption data curve;

[0074] In a preferred but non-limiting embodiment of the present invention, step 3 specifically comprises:

[0075] like Figure 5 As shown, by calling the electricity consumption collection system, the voltage, current, and electricity consumption data of the remaining suspect users in the past three months as data types are extracted as the start and end time of the electricity consumption details after deleting the misjudged suspect users, and the least squares method is used to derive and draw the electricity consumption data curve of the suspect users changing over time, so as to realize the visualization interface of the current, voltage, and electricity consumption data of the suspect users with a very long time span, which can display the electricity consumption data curve for three months.

[0076] After improvement, in a preferred but non-limiting embodiment of the present invention, by calling the electricity collection system, a method for extracting the voltage, current, and electricity consumption data of the remaining suspicious users in the past three months as data types after deleting the misjudged suspected users as the start and end time of electricity consumption details includes:

[0077] The electricity consumption collection system is connected to a corresponding device (such as a computer) for running an electricity theft clue screening method based on long-term electricity consumption information;

[0078] The electricity consumption collection system transmits the electricity consumption data of the voltage, current and power quantity as data types for the last three months of the remaining suspect users after deleting the misjudged suspect users, which are stored therein as the start and end time of electricity consumption details, to the corresponding device.

[0079] After the electricity consumption data collection system transmits the voltage, current, and power consumption data of the remaining suspect users in the past three months as data types, which are stored as the start and end time of electricity consumption details after deleting the misjudged suspect users, to the corresponding device, the electricity consumption data is often stored. When unexpected situations such as corresponding device failure and hacker attack occur, the electricity consumption data will be cleared. This simple storage method cannot ensure the availability and reliability of electricity consumption data; electricity consumption data is often modified, which is not conducive to subsequent processing; currently, modification detection is performed on electricity consumption data based on the method of calculating hash values, but the existing hash value detection method calculates the hash value of a program segment based on self-detection rules, and forms feedback based on the value; however, the retrieved program segment is often detected and deciphered by hackers, that is, the method is not well hidden.

[0080] In a preferred but non-limiting implementation method of the present invention, after the electricity consumption collection system transmits the electricity consumption data of the remaining suspect users in the past three months as data types, such as voltage, current, and electricity, which are stored in the electricity consumption details as the start and end time after deleting the misjudged suspect users, to the corresponding device, the method executed in the corresponding device further includes:

[0081] Step 3-1: The corresponding device receives and stores the electricity consumption data transmitted by the electricity consumption collection system;

[0082] Step 3-2: Perform anti-modification isolation on the power consumption data through the coding isolation mode;

[0083] Step 3-3: Regularly detect whether a modification operation occurs in the power consumption data through a detection method;

[0084] Step 3-4: Perform replacement restoration on the modification operations existing in the power usage data via the copied power usage data;

[0085] Step 3-5: Perform enhanced isolation for power consumption data via system isolation mode. The system here can be a power consumption collection system.

[0086] In a preferred but non-restrictive implementation method of the present invention, in step 3-1, a copy and storage is performed on the electricity consumption data transmitted by the electricity consumption collection system.

[0087] In a preferred but non-restrictive implementation method of the present invention, in step 3-2, the coding blocking mode segments the electricity usage data, and encodes the segmented electricity usage data separately, and stores them together through several processes.

[0088] In a preferred but non-limiting implementation method of the present invention, the method of encoding the blocking mode comprises:

[0089] The corresponding device randomly selects p values ​​in the prime number field X(r) as the private code n j , j∈{1,2...p}, r satisfies r>p, and by selecting l-1 factors in X(r), the following p equations are constructed:

[0090] g(y j )=b0y0+b1y1+...+b l-1 y l-1 ;

[0091] In the equation, g(y j ) represents the amount of private code, g(y j ) is n j , put p g(y j ) are assigned to the set {i1,i2...i p}, and publish the variable parameter {y j};

[0092] The collected power consumption data is V, the encoding parameter is L, the decoding parameter is S, I = {i1, i2...i s}, s<p is used to illustrate the read-write mode. If the encoding value indicators after encoding the power consumption data are D1 and D2, all characteristic indicators are combined into a mode code in the read-write mode, and then the encoding value after encoding the power consumption data is constructed into the following format:

[0093]

[0094]

[0095] In the format, D n represents the encoded value after encoding the power consumption data, p1 and p2 are both primitive roots, k∈[1,s] represents the number of power consumption data features, and DU represents the derived encoded value after encoding the power consumption data. The feature can be the type of power consumption data, and the power consumption data type can be the type of power consumption voltage, current, or power.

[0096] In a preferred but non-limiting implementation method of the present invention, a decoding method of a coding blocking mode comprises:

[0097] After the corresponding device obtains the coded value of the power consumption data, the coded value DU and the decoding parameter S derived from the coding blocking mode method are introduced to obtain the calculation index W:

[0098]

[0099] Wherein, E1 and E2 represent the pre-set calculation weights; S1 and S2 represent the decoding parameters; D1 and D2 represent the coding value indicators;

[0100] p1 and p2 represent the original roots;

[0101] The collected electricity consumption data is obtained through the following equation:

[0102]

[0103] In the equation, V represents the collected power consumption data obtained by decoding.

[0104] In the detection method of step 3-3, a hash value is obtained by combining the command in the program and the command in the main memory. When the hash value is confirmed to be a normal hash value, it is determined that the power usage data has not been modified. When it is confirmed to be an abnormal hash value, it is confirmed that the power usage data has been modified.

[0105] In a preferred but non-limiting embodiment of the present invention, the equation of the detection method is:

[0106] I(U)=I(Z,N,Q,R);

[0107] In the equation, I(U) represents the hash value obtained by the program operation; I represents the operation hash value; U represents the program corresponding to the power consumption data; Z represents the power consumption data command Z = {z0, z1...z m}; N represents the storage unit queue; Q and R are the starting address of the command and the main memory unit respectively. The identification of the power data command is as follows: during the transmission of the power data command, the transmitted value z is hashed by the hash method. j Execute the operation to generate the hash value IW j1 , then decode the parameter S to IW j1 Execute the code to obtain the numerical mark TJHO, that is, TJHO = IW j1 e npeo, finally put z j It is transmitted to the receiving process together with TJHO, and the encoding parameter L is published; after the receiving process obtains and, the encoding parameter L of the transmitting process is used to decode TJHO, that is, IW j2 =TJHO f NPEO, the collection process uses the same hashing method to j Execute the operation to obtain IWj3 , and finally compared with IW j2 With IW j3 Are they the same? If they are the same, it indicates that the value has not been modified, otherwise it indicates that the value has been modified; here the encoding parameter is L = (f, o), and the decoding parameter is S = (e, o), where the greatest common divisor of f and e is one, and o is the amount obtained by multiplying two prime numbers q and r. In the replacement restoration method in step 3-4, after decoding the power consumption data encoded by the coding barrier mode, the power consumption data read out based on the copy is then replaced and restored.

[0108] It should be noted that NPEO represents a remainder operation performed on o in the encoding parameter or the decoding parameter.

[0109] In step 3-5, the system blocking mode detects the user's confidence and sets the confidence threshold of the electricity consumption collection system, and adds users whose confidence is less than the confidence threshold to the blacklist.

[0110] In a preferred but non-limiting implementation method of the present invention, the method of the system barrier mode comprises:

[0111] During the period of public use of electricity data, the identification shall be carried out separately according to the system conditions;

[0112] If the system is a new system, the confidence J1 of the new system in the communication network is calculated:

[0113] J1=χN+δO;

[0114] Among them, χ and δ represent the pre-set criticality factors, N represents the overall index of the system when the power consumption data is shared, and O represents the stability of the communication network;

[0115] If the system is an old system and no modification or harassment occurs after the old system is incorporated into the communication network, the confidence J2 of the old system in the communication network is increased:

[0116] J2 = ηJ1 + B;

[0117] Among them, γ represents the loss factor; B represents the increased confidence of the system that has not been attacked;

[0118] If the system is an old system and a modification harassment situation occurs after the old system is incorporated into the communication network, the confidence J3 of the old system in the communication network is reduced:

[0119] J3 = ηJ1-C;

[0120] Among them, C represents the reduced confidence of the system after being attacked.

[0121] The power consumption data collection system transmits power consumption data, stores and copies the transmitted power consumption data, and performs anti-modification isolation on the stored power consumption data through the coding isolation mode. The method of segmenting and encoding the power consumption data is a critical quantity marking method. If the power consumption data is to be modified during the public use period, the number of processes is set to 0. At this time, more than L processes must intervene to modify the power consumption data.

[0122] Data modification protection is implemented by performing segmentation and encoding on the electricity usage data to improve the reliability of the electricity usage data. The hash value is calculated through the detection method based on the program and command stored in the electricity usage data. The availability of the hash value determines whether it has been attacked by modification. It has good concealment, and the system confidence is detected through the system isolation mode to perform enhanced isolation for the electricity usage data to prevent the electricity usage data from being modified due to system problems.

[0123] The least squares method is a mathematical tool that is widely used in many disciplines of data processing such as error estimation, uncertainty, system identification, prediction, and forecasting.

[0124] The least squares method is a commonly used mathematical method for fitting experimental data and solving the best estimate. It minimizes the sum of squared errors between the model's predicted values ​​and the actual observed values ​​by finding a function or model.

[0125] The least squares method is a mathematical optimization technique used to deal with data fitting problems. It finds the parameter values ​​of a mathematical model so that the sum of squared errors between the model and the actual observed data is minimized. The least squares method has a wide range of applications, including statistics, economics, engineering, etc. It is often used in various data analysis problems such as curve fitting, regression analysis, signal processing, image processing, etc. The basic principle of the least squares method is to determine the best parameter value by minimizing the sum of squares of the residuals. The residual refers to the difference between the observed value and the model predicted value. The least squares method adjusts the model parameters so that the sum of squares of these differences is minimized. The least squares method can be applied to both linear and nonlinear models. For linear models, the least squares method can obtain the optimal solution by solving the normal equations or using matrix operations. For nonlinear models, the least squares method usually needs to be solved using an iterative optimization algorithm, such as the Levenberg-Marquardt algorithm. The advantages of the least squares method are that it is easy to understand, computationally efficient, and has a solid mathematical foundation. It can obtain the optimal parameter estimation and provide statistical indicators such as parameter confidence interval and goodness of fit to evaluate the fit and reliability of the model.

[0126] Step 4: judging whether the corresponding suspected user is a suspected electricity theft user on the electricity consumption data curve;

[0127] The judgment method is a method of using electricity consumption data in the existing technology to judge suspected users of electricity theft. By calling the electricity consumption collection system, the voltage, current, power consumption and other electricity consumption data of the suspected user in the past three months are extracted, and the electricity consumption data curve of the suspected user over time is drawn, so as to realize the visualization interface of the current, voltage and other electricity consumption data of the suspected user for an ultra-long time span, and the electricity consumption data curve for three months can be displayed. The method of using the electricity consumption collection information system to review the user's electricity consumption data on a daily basis for analysis and judgment is abandoned, so as to shorten the time for finding and judging abnormal electricity consumption data of suspected users, and efficiently carry out the abnormal user data analysis work.

[0128] Step 5: End the electricity theft clue screening method based on long-term electricity usage information.

[0129] like Figure 4 As shown, the present invention provides a device for screening electricity theft clues based on long-term electricity usage information, comprising:

[0130] An acquisition module, which is used to acquire the electricity consumption data of suspected users;

[0131] A filtering module, which is used to filter the suspected user's electricity usage data according to keywords in combination with the suspected user's electricity usage data;

[0132] A setting module, which is used to set the start and end time and data category of the electricity consumption details of the remaining suspected users after deleting the misjudged suspected users, so as to obtain the corresponding electricity consumption data curve;

[0133] A judgment module, which is used to judge whether the corresponding suspected user is a suspected user of electricity theft on the electricity consumption data curve;

[0134] The end module is used to end the electricity theft clue screening method based on long-term electricity consumption information.

[0135] The beneficial effect of the present invention is that, compared with the prior art, the present invention mainly solves the problem of time-consuming manual screening of clues of electricity theft, and reduces the number of suspected electricity users by adding a whitelist of suspected electricity users and expanding the query time span, thereby effectively saving the working time consumed in screening clues of electricity theft.

[0136] The present invention is applied to the secondary screening of clues of electricity theft in anti-electricity theft work, providing more clues of illegal theft for electricity inspectors, solving the problems of heavy workload, long time consumption and low efficiency for front-line employees in screening illegal electricity users, effectively reducing the burden on grassroots employees, greatly reducing the operating costs of enterprises, effectively cracking down on illegal theft, curbing the spread of illegal theft, and ensuring safe and orderly electricity use and the safety of power facilities.

[0137] The present invention can be extended to the field of electricity consumption auditing, and electricity consumption auditing work can be carried out efficiently by observing information flows over very long time spans.

[0138] Method for judging whether the corresponding suspected user is a suspected electricity theft user based on the electricity consumption data curve:

[0139] The arithmetic average is performed on the difference between the peak value and the trough value on the electricity consumption data curve, and then the mean of the difference obtained after the arithmetic average is used for machine learning to form an 80% training set and a 20% test set. If the mean obtained after machine learning exceeds the pre-set threshold, it indicates that the user is suspected of electricity theft. Otherwise, the user is not suspected of electricity theft.

[0140] The present disclosure can be a system, method and / or computer program product. The computer program product can include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to achieve each aspect of the present disclosure.

[0141] The computer readable storage medium can be a tangible network capable of holding and storing instructions for use by the instruction execution network. The computer readable storage medium can be, but is not limited to, an electrical storage network, a magnetic storage network, an optical storage network, an electromagnetic storage network, a semiconductor storage network, or any suitable combination of the above. Further examples of computer readable storage media (non-exhaustive list) include: portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read only memory (HD-ROM), digital versatile disk (DXD), memory stick, floppy disk, mechanically encoded network, such as punch cards or raised structures in grooves with instructions stored thereon, and any suitable combination of the above. The computer readable storage medium used herein is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (such as, light pulses through a transmission line cable), or an electrical signal transmitted through a wire.

[0142] The computer readable program instructions expressed herein can be downloaded from a computer readable storage medium to each inference / processing grid line, or downloaded to an external computer or external storage grid line through a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, power transmission line transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge service devices. The network adapter card or network interface in each inference / processing grid line receives the computer readable program instructions from the network and forwards the computer readable program instructions for storage in the computer readable storage medium in each inference / processing grid line.

[0143] The computer program instructions for performing the operations of the present disclosure can be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, conditional setting values, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Sdaletal A, H++, etc., and conventional procedural programming languages ​​such as "H" language or similar programming languages. The computer-readable program instructions can be executed entirely on the client computer, partially on the client computer, as a separate software package, partially on the client computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the client computer through any other network, including a local area network (LAb) or a wide area network (WAb), or can be connected to an external computer (just like connecting through the Internet using an Internet service provider). In some embodiments, the electronic circuit is customized by using the status values ​​of computer-readable program instructions, such as a programmable logic circuit, a field programmable gate array (FPGA) or a programmable logic array (PLA), and the electronic circuit can execute the computer-readable program instructions to achieve each aspect of the present disclosure.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not deviate from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for screening electricity theft clues based on long-term electricity usage information, characterized in that: include: Step 1: Obtain the electricity consumption data of the suspected user; Step 2: Combine the electricity consumption data of suspected users to filter the electricity consumption data of suspected users according to keywords; Step 3: Set the start and end time and data category of the electricity consumption details of the remaining suspected users after deleting the misjudged suspected users, so as to obtain the corresponding electricity consumption data curve; Step 4: judging whether the corresponding suspected user is a suspected electricity theft user on the electricity consumption data curve; Step 5: End the electricity theft clue screening method based on long-term electricity usage information.

2. The method for screening electricity theft clues based on long-term electricity usage information according to claim 1 is characterized in that: Step 1 specifically includes: Obtain the electricity usage data and basic profile information of suspected users.

3. The method for screening electricity theft clues based on long-term electricity usage information according to claim 2 is characterized in that: The suspected user electricity consumption data is the electricity consumption data of the suspected user who may steal electricity obtained through the anti-electricity theft intelligent diagnosis model, and the electricity consumption data includes the voltage, current and amount of electricity used by the suspected user.

4. The method for screening electricity theft clues based on long-term electricity usage information according to claim 2 is characterized in that: The basic file information of suspected users includes the user's category, which includes the user's belonging to factories, enterprises, companies, households, tenants, governments, middle schools, elementary schools, universities, hospitals, the military, municipal administration, management offices, urban management, high schools, street lights, construction bureaus, kindergartens, power grids, troops, state organs, construction offices, subways, mobile, Unicom, telecommunications, law enforcement bureaus, water supply, heating, electricity supply, banks, public security, transportation bureaus, substations, post offices, police, brigades, civil affairs, police stations, transportation bureaus, management bureaus, Communist Party organs, county committees, and petition bureaus.

5. The method for screening electricity theft clues based on long-term electricity usage information according to claim 1 is characterized in that: Step 2 specifically includes: In the secondary analysis software, keywords are input to combine with the electricity consumption data of suspicious users to filter the electricity consumption data of suspicious users according to the keywords.

6. The method for screening electricity theft clues based on long-term electricity usage information according to claim 5 is characterized in that: The method of filtering the electricity consumption data of the suspected user according to the keyword by entering the keyword in the secondary judgment software includes: In the secondary analysis software, a keyword filtering link is added, that is, the attribute names of users such as government agencies, public institutions, and communication base stations that are unlikely to steal electricity are collected, and a keyword thesaurus is established. The basic file information of the suspected users in the suspected users' electricity usage data can be automatically filtered out, and the basic file information of the suspected users whose basic file information matches the keywords in the keyword library can be used to eliminate the suspected users who were misjudged in the initial screening of clues by the anti-electricity theft intelligent diagnosis model.

7. The method for screening electricity theft clues based on long-term electricity usage information according to claim 6 is characterized in that: Methods for building a keyword database include: By typing the keywords in the keyword column of the secondary analysis software, the keywords are then stored in the keyword library.

8. The method for screening electricity theft clues based on long-term electricity usage information according to claim 6 is characterized in that: The specific attributes of users such as government agencies, public institutions, and communication base stations that are unlikely to steal electricity include: Such attributes include government, middle school, elementary school, university, hospital, military, municipal administration, management office, urban management, high school, street lights, construction bureau, kindergarten, power grid, troops, state organs, construction office, subway, mobile, Unicom, telecommunications, law enforcement bureau, water supply, heating, power supply, bank, public security, transportation bureau, substation, post office, police, brigade, civil affairs, police station, transportation bureau, management bureau, Communist Party of China organs, county party committee, and petition bureau.

9. The method for screening electricity theft clues based on long-term electricity usage information according to claim 1 is characterized in that: Step 3 specifically includes: By calling the electricity consumption collection system, the voltage, current, and electricity consumption data of the remaining suspect users in the past three months as data types are extracted as the start and end time of electricity consumption details after deleting the misjudged suspect users, and the least squares method is used to derive and draw the electricity consumption data curve of the suspect users over time, so as to realize the visualization interface of the current, voltage, and electricity consumption data of the suspect users over a very long time span, which can display the electricity consumption data curve of three months; The method of extracting the voltage, current, and power consumption data of the remaining suspect users in the past three months as data types after deleting the misjudged suspect users as the start and end time of the electricity consumption details by calling the electricity consumption collection system includes: The electricity consumption collection system is communicatively connected to a corresponding device that runs a method for screening electricity theft clues based on long-term electricity consumption information; The electricity consumption collection system transmits the electricity consumption data of the remaining suspect users in the past three months as data types, such as voltage, current, and power, stored therein as the start and end time of electricity consumption details after deleting the misjudged suspect users, to the corresponding device; After the electricity consumption collection system transmits the electricity consumption data of the remaining suspect users in the past three months as data types, such as voltage, current, and power, stored in the electricity consumption collection system as the start and end time of electricity consumption details after deleting the misjudged suspect users, to the corresponding device, the method executed in the corresponding device further includes: Step 3-1: The corresponding device receives and stores the electricity consumption data transmitted by the electricity consumption collection system; Step 3-2: Perform anti-modification isolation on the power consumption data through the coding isolation mode; Step 3-3: Regularly detect whether a modification operation occurs in the power consumption data through a detection method; Step 3-4: Perform replacement restoration on the modification operations existing in the power usage data via the copied power usage data; Step 3-5: Perform enhanced isolation for power consumption data via system isolation mode. The system here can be a power consumption collection system.

10. A device for screening electricity theft clues based on long-term electricity usage information, characterized in that: include: An acquisition module, which is used to acquire the electricity consumption data of suspected users; A filtering module, which is used to filter the suspected user's electricity usage data according to keywords in combination with the suspected user's electricity usage data; A setting module, which is used to set the start and end time and data category of the electricity consumption details of the remaining suspected users after deleting the misjudged suspected users, so as to obtain the corresponding electricity consumption data curve; A judgment module, which is used to judge whether the corresponding suspected user is a suspected user of electricity theft on the electricity consumption data curve; The end module is used to end the electricity theft clue screening method based on long-term electricity consumption information.

Images