Anti-electricity-stealing inspection monitoring method and system based on big data

Abstract

The invention provides an anti-electricity-stealing inspection monitoring method and system based on big data, and the method comprises the steps: obtaining the real-time electricity utilization data of a plurality of users, and carrying out the preprocessing of the real-time electricity utilization data, and obtaining the electricity utilization processing data; judging whether the user has a suspected electricity stealing behavior or not according to the electricity utilization processing data; if yes, adjusting the acquisition frequency of the real-time power consumption data; otherwise, continuously detecting the power utilization processing data; acquiring power utilization data of the corresponding user according to the acquisition frequency; analyzing whether the user has an electricity stealing behavior or not according to the electricity utilization data; if yes, analyzing the type of the electricity stealing behavior; otherwise, the suspicion of the suspected electricity stealing behavior of the corresponding user is relieved; and generating and sending early warning information to a power grid manager based on the electricity stealing behavior type of the user so as to process the electricity stealing behavior of the user. According to the invention, the technical problem that the electricity stealing behavior of a user cannot be found in time due to poor real-time performance of electricity data monitoring is solved.

Description

Technical Field

[0001] The present invention relates to the field of power grid monitoring, and in particular to an anti-electricity theft inspection and monitoring method and system based on big data. Background Art

[0002] Electricity theft not only causes economic losses to power companies, but also affects the safe and stable operation of the power grid and even threatens public safety; therefore, anti-electricity theft work is of great significance; anti-electricity theft refers to the process of preventing, discovering and combating illegal electricity theft through a series of measures and technical means; electricity theft is a direct theft of power company assets, which will cause huge economic losses to the company; through anti-electricity theft inspection and monitoring, electricity theft can be discovered and stopped in time, thereby protecting the legitimate rights and interests of power companies and reducing economic losses; electricity theft may also affect the normal operation and power supply efficiency of the power grid. Anti-electricity theft inspection and monitoring can help maintain the stability and reliability of the power grid and improve power supply efficiency by timely discovering and handling electricity theft problems.

[0003] The existing invention patent application document "Anti-electricity theft inspection and monitoring method" with publication number CN113239087A, the existing method includes: obtaining user electricity consumption history data, and pre-processing the user electricity consumption history data to obtain user electricity processing data; determining an abnormal event text record table based on different types of electricity theft means, and determining the user's characteristic data based on the abnormal behavior record table and the user's electricity processing data; constructing an anti-electricity theft inspection and monitoring machine learning model based on a random forest algorithm, and using the user's characteristic data to train and optimize the model to determine the final anti-electricity theft inspection and monitoring machine learning model; using the final anti-electricity theft inspection and monitoring machine learning model to judge the user's electricity consumption data to monitor the user's electricity theft behavior; however, the above-mentioned prior art uses the historical electricity consumption data to train the model, and then uses the model to judge the user's electricity consumption data and monitor the user's electricity theft behavior; it does not take into account the randomness and uncertainty of the user's electricity theft behavior, and does not monitor the user's electricity consumption data in real time, which may result in the failure to timely discover the user's electricity theft behavior, causing economic losses to the power company.

[0004] The existing invention patent application document "A method, device, equipment and storage medium for determining electricity theft" with publication number CN116777485A, the existing method includes: obtaining electricity consumption data and weather information of the substation during a detection period, and then configuring the initial power supply model of the substation in a simulation environment based on the electricity consumption data and weather information to obtain a simulated power supply model of the substation; then performing a simulated power supply simulation on the simulated power supply model of the substation to obtain a simulated line loss amount of the substation during the detection period, wherein the simulated power supply model of the substation is configured with a time ratio greater than 1; determining whether there is electricity theft based on the simulated line loss amount and the loss amount, if the difference between the simulated line loss amount and the loss amount is greater than an error threshold, it is determined that there is electricity theft.

[0005] The existing invention patent application document "A real-time prediction system for electricity theft of smart meters based on data decision library" with publication number CN118446733A, the existing method includes: a data acquisition module, a data preprocessing module, a data extraction module, a user analysis module, a data storage module, an electricity theft prediction module, a detection module and a response module of a smart meter are arranged in the data decision library; the data acquisition module collects data and transmits it to the data preprocessing module for classification, the data extraction module removes interference data from the classified data, the user analysis module establishes a user electricity consumption curve, the electricity theft prediction module predicts electricity consumption according to the user electricity consumption curve, and predicts electricity theft through real-time data comparison of the smart meter detection module, and sends the prediction result to the smart meter response module.

[0006] The existing invention patent application document "A low-voltage substation electricity theft monitoring system based on line loss and user electricity consumption" with publication number CN114094705A, the existing method includes: a data acquisition function module: collecting user profile information, user electricity consumption data, line loss data and substation information from the metering automation system; a data processing function module: calculating the "line loss threshold" of each substation, if the statistical line loss rate of the substation is higher than the calculated line loss threshold, marking the substation; analyzing and calculating the marked substation to determine whether there is electricity theft; an electricity theft warning module: warning users who steal electricity in the system, and reminding the system administrator through message reminders and text messages in the system; an early warning processing module: realizing on-site inspection of suspected electricity theft users.

[0007] However, the aforementioned existing technologies do not fully utilize the real-time electricity consumption data of users, and effective real-time electricity consumption data is difficult to retain. There is a lot of invalid data, which leads to low efficiency of subsequent data processing. At the same time, the aforementioned existing technologies do not judge the user's suspected electricity theft behavior based on the electricity consumption data, and the frequency of collecting the user's real-time electricity consumption data is fixed, which makes it impossible to accurately monitor users who may have electricity theft behaviors, resulting in incomplete monitoring problems, and the user's electricity theft behaviors cannot be discovered in time, causing economic losses to the power company.

[0008] The aforementioned traditional scheme does not fully analyze the user's suspected electricity theft behavior, nor does it analyze the type of the user's electricity theft behavior, which restricts the audit effect; at the same time, the early warning method of the aforementioned existing technology does not have strong pertinence in the treatment measures taken for different types of electricity theft behavior, making it difficult to reduce the losses of the power company.

[0009] In summary, the existing technology has the technical problem that the real-time monitoring of electricity consumption data is poor, resulting in the delay in discovering the user's electricity theft behavior. Summary of the invention

[0010] The technical problem to be solved by the present invention is: how to solve the technical problem in the prior art that the real-time monitoring of electricity consumption data is poor, resulting in untimely discovery of the user's electricity theft behavior.

[0011] The present invention adopts the following technical solutions to solve the above technical problems: A big data-based anti-electricity theft inspection and monitoring method comprises:

[0012] S1. Collect real-time electricity consumption data from users, pre-process the real-time electricity consumption data, and obtain electricity consumption processing data;

[0013] S2. Determine whether the user is suspected of stealing electricity based on the electricity processing data; if so, adjust the frequency of collecting real-time electricity data; if not, continue to detect the electricity processing data;

[0014] S3. Collect the real-time electricity consumption data of suspected electricity theft users according to the collection frequency; analyze whether the suspected electricity theft users have committed electricity theft behavior according to the real-time electricity consumption data; if so, retrieve the real-time electricity consumption data of the suspected electricity theft users within a preset time period, and integrate the real-time electricity consumption data into a type input sequence according to the collection order; retrieve the preset electricity theft type recognition model, input the type input sequence into the electricity theft type recognition model, obtain the corresponding type label, and match the corresponding electricity theft behavior type according to the type label; if not, screen out the suspected electricity theft users;

[0015] S4. Generate warning information based on the electricity theft behavior type of the electricity theft user to handle the electricity theft behavior.

[0016] The present invention fully considers the randomness and uncertainty of the user's electricity theft behavior, monitors the user's electricity consumption data in real time, can timely discover the user's electricity theft behavior, and reduce the economic losses of the power company.

[0017] The present invention analyzes the electricity usage data of users suspected of electricity theft. When a user is found to be engaged in electricity theft, the type of the user's electricity theft behavior is analyzed and early warning information is generated. The early warning information is sent to management personnel, who handle the user's electricity theft behavior based on the early warning information. Different handling measures can be taken for different types of electricity theft behavior, thereby reducing the losses of the power company.

[0018] In a more specific technical solution, in S1, the real-time electricity consumption data includes: collection time, real-time voltage, real-time current and real-time power consumption.

[0019] In a more specific technical solution, S1 includes:

[0020] S11, retrieve and determine whether the real-time power consumption data contains sub-data of all preset data types;

[0021] S13: If yes, retain the current real-time power consumption data;

[0022] S14, if not, discard the current real-time power consumption data;

[0023] S15, determining whether the collection time in the real-time power consumption data is the same;

[0024] S16: If yes, remove errors and duplicates from the real-time electricity consumption data;

[0025] S17, if not, retain the current real-time power consumption data;

[0026] S18. Mark the real-time electricity consumption data as electricity consumption processing data.

[0027] The present invention collects real-time electricity consumption data of users and processes the real-time electricity consumption data, can retain valid real-time electricity consumption data, and eliminate invalid data, which is beneficial to improving the efficiency of subsequent data processing.

[0028] In a more specific technical solution, S2 includes:

[0029] S21, retrieving the real-time voltage and real-time current in the power consumption processing data, and processing to obtain the preset normal fluctuation range of the real-time voltage and real-time current;

[0030] S22, within a set time period, analyzing the real-time voltage and the real-time current to obtain a voltage fluctuation curve and a current fluctuation curve;

[0031] S23, respectively obtaining the maximum value and the minimum value in the voltage fluctuation curve and the current fluctuation curve;

[0032] S25, calculating the boundary difference between the maximum value, the minimum value and the preset normal fluctuation range;

[0033] S26, determining whether the boundary differences are all less than a preset boundary difference threshold;

[0034] S27: If yes, it is determined that the current user is suspected of stealing electricity;

[0035] S28, if not, analyzing the current real-time power consumption of the user;

[0036] S29, retrieve the real-time power from the power consumption processing data, fit the real-time power in the set time period into the power function F(t), and calculate the first-order derivative f(t) of the power function F(t);

[0037] S210, analyzing the first-order derivative f(t), obtaining the first-order derivative minimum value, and determining whether the first-order derivative minimum value is less than a preset minimum value threshold;

[0038] S212: If yes, determine that the current user is suspected of stealing electricity;

[0039] S213: If not, it is determined that the current user does not have any suspected behavior of stealing electricity.

[0040] In a more specific technical solution, S2 also includes:

[0041] S21', retrieving the voltage fluctuation difference and the current fluctuation difference from the power consumption processing data;

[0042] S22', retrieve the minimum value of the first-order derivative of the preset power function, calculate the difference ZC between the minimum value of the first-order derivative and the preset minimum threshold value; obtain the original acquisition frequency YC;

[0043] S23', from the voltage fluctuation difference and the current fluctuation difference, the voltage fluctuation maximum value difference VCmax, the minimum value difference VCmin, the current fluctuation maximum value difference ICmax and the minimum value difference ICmin are calculated to obtain the acquisition frequency adjustment coefficient TX;

[0044] S24', according to the collection frequency adjustment coefficient TX, the collection frequency of the real-time power consumption data is processed.

[0045] The present invention judges the user's suspected electricity theft behavior based on the electricity consumption data and adjusts the collection frequency of the corresponding user's real-time electricity consumption data, which can strengthen the monitoring of users who may have electricity theft behavior and avoid the failure to timely discover the user's electricity theft behavior due to incomplete monitoring, causing economic losses to the power company.

[0046] In a more specific technical solution, in S23', the acquisition frequency adjustment coefficient TX is obtained using the following logic:

[0047] TX=α×(VCmax+VCmin) / 2+β×(ICmax+ICmin) / 2+γ×ZC

[0048] Where α, β and γ are weight coefficients greater than 0.

[0049] In a more specific technical solution, in S24', the following logic is used to process and obtain the acquisition frequency:

[0050] CP=YC-θ×TX

[0051] Wherein, θ is a proportional coefficient greater than 0.

[0052] In a more specific technical solution, S3 includes:

[0053] S31, retrieving the real-time voltage and real-time current in the power consumption data within a preset time period;

[0054] S32, analyzing the real-time voltage and the real-time current to obtain a voltage fluctuation curve and a current fluctuation curve;

[0055] S33, determining whether the voltage fluctuation and the current fluctuation are both within the normal range;

[0056] S34: If yes, analyze the user's real-time power consumption;

[0057] S35, if not, it is determined that the user has committed electricity theft;

[0058] S36, retrieve the real-time power consumption in the power consumption data within a preset time period, refit the real-time power consumption into a power consumption function H(t), and solve the first-order derivative h(t) of the power consumption function H(t);

[0059] S37, determining whether the minimum value of the first-order derivative h(t) is less than the minimum value threshold;

[0060] S38: If yes, it is determined that the user has committed electricity theft;

[0061] S39. If not, it is determined that the user does not engage in electricity theft.

[0062] In a more specific technical solution, in S3, the operation of building a power theft type identification model also includes:

[0063] S31', obtaining and dividing a standard data set to obtain a training set and a validation set, and using the training set to train a preset artificial intelligence model;

[0064] S32', using the verification set to verify the artificial intelligence model until a power theft type recognition model is obtained.

[0065] In a more specific technical solution, a big data-based anti-electricity theft inspection and monitoring system includes:

[0066] The power consumption data preprocessing module is used to collect real-time power consumption data from users, preprocess the real-time power consumption data, and obtain power consumption processing data;

[0067] A suspected electricity theft identification module is used to determine whether the user has suspected electricity theft behavior based on the electricity processing data; if so, adjust the frequency of collecting real-time electricity data; if not, continue to detect the electricity processing data. The suspected electricity theft identification module is connected to the electricity data preprocessing module;

[0068] The electricity theft user and type determination module is used to collect the real-time electricity consumption data of suspected electricity theft users according to the collection frequency; analyze whether the suspected electricity theft user has committed electricity theft behavior based on the real-time electricity consumption data; if so, retrieve the real-time electricity consumption data of the suspected electricity theft user within a preset time period, and integrate the real-time electricity consumption data into a type input sequence according to the collection order; retrieve the preset electricity theft type recognition model, input the type input sequence into the electricity theft type recognition model, obtain the corresponding type label, and match the corresponding electricity theft behavior type according to the type label; if not, screen out the suspected electricity theft user, and the electricity theft user and type determination module is connected to the suspected electricity theft recognition module;

[0069] The electricity theft behavior processing module is used to generate warning information based on the electricity theft behavior type of the electricity theft user to handle the electricity theft behavior. The electricity theft behavior processing module is connected to the electricity theft user and the type determination module.

[0070] Compared with the prior art, the present invention has the following advantages:

[0071] The present invention fully considers the randomness and uncertainty of the user's electricity theft behavior, monitors the user's electricity consumption data in real time, can timely discover the user's electricity theft behavior, and reduce the economic losses of the power company.

[0072] The present invention analyzes the electricity usage data of users suspected of electricity theft. When a user is found to be engaged in electricity theft, the type of the user's electricity theft behavior is analyzed and early warning information is generated. The early warning information is sent to management personnel, who handle the user's electricity theft behavior based on the early warning information. Different handling measures can be taken for different types of electricity theft behavior, thereby reducing the losses of the power company.

[0073] The present invention collects real-time electricity consumption data of users and processes the real-time electricity consumption data, can retain valid real-time electricity consumption data, and eliminate invalid data, which is beneficial to improving the efficiency of subsequent data processing.

[0074] The present invention judges the user's suspected electricity theft behavior based on the electricity consumption data and adjusts the collection frequency of the corresponding user's real-time electricity consumption data, which can strengthen the monitoring of users who may have electricity theft behavior and avoid the failure to timely discover the user's electricity theft behavior due to incomplete monitoring, causing economic losses to the power company.

[0075] The present invention collects the real-time electricity usage data of users and processes the real-time electricity usage data, can retain valid real-time electricity usage data, and eliminate invalid data, which is beneficial to improving the efficiency of subsequent data processing; the user's suspected electricity theft behavior is judged according to the electricity usage data, and the collection frequency of the corresponding user's real-time electricity usage data is adjusted, which can strengthen the monitoring of users who may have electricity theft behavior, avoid incomplete monitoring and failure to timely discover the user's electricity theft behavior, causing economic losses to the power company.

[0076] The present invention analyzes the electricity usage data of users suspected of electricity theft. When a user is found to be engaged in electricity theft, the type of the user's electricity theft behavior is analyzed and early warning information is generated. The early warning information is sent to management personnel, who handle the user's electricity theft behavior based on the early warning information. Different handling measures can be taken for different types of electricity theft behavior, thereby reducing the losses of the power company.

[0077] The present invention solves the technical problem in the prior art that the real-time monitoring of electricity consumption data is poor, resulting in untimely discovery of the user's electricity theft behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 A schematic diagram of the basic steps of an anti-electricity theft inspection and monitoring method based on big data according to Embodiment 1 of the present invention;

[0079] Figure 2 This is a schematic diagram of specific steps of preprocessing real-time power consumption data according to Embodiment 1 of the present invention;

[0080] Figure 3 This is a schematic diagram of specific steps for determining suspected electricity theft behavior according to Embodiment 1 of the present invention;

[0081] Figure 4 This is a schematic diagram of specific steps for adjusting the frequency of collecting real-time power consumption data according to Embodiment 1 of the present invention;

[0082] Figure 5 This is a schematic diagram of specific steps for analyzing whether a user has committed electricity theft in Example 1 of the present invention;

[0083] Figure 6 This is a schematic diagram of specific steps for building an electricity theft type identification model based on an artificial intelligence model according to Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0084] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] Example 1

[0086] like Figure 1 As shown, the present invention provides an anti-electricity theft inspection and monitoring method based on big data, which includes the following basic steps:

[0087] S1. Acquire real-time electricity consumption data of several users, and pre-process the real-time electricity consumption data to obtain processed electricity consumption data;

[0088] In this embodiment, the real-time power consumption data includes but is not limited to: collection time, real-time voltage, real-time current and real-time power consumption;

[0089] like Figure 2 As shown, in this embodiment, the step S1 of preprocessing the real-time power consumption data further includes the following specific steps:

[0090] S11, retrieve real-time electricity consumption data;

[0091] S12, determining whether the real-time power consumption data contains sub-data of all data types;

[0092] S13: If yes, retain the corresponding real-time power consumption data;

[0093] S14, if not, discard the corresponding real-time power consumption data;

[0094] S15, determining whether the collection time in the real-time power consumption data is the same;

[0095] S16: If yes, remove the wrong or duplicate real-time electricity consumption data;

[0096] S17. If not, retain the corresponding real-time power consumption data;

[0097] S18. Mark the retained real-time electricity consumption data as electricity consumption processing data.

[0098] S2. Determine whether the user is suspected of stealing electricity based on the electricity processing data; if so, adjust the frequency of collecting real-time electricity data; if not, continue to detect the electricity processing data;

[0099] like Figure 3As shown, in this embodiment, the operation of judging suspected electricity theft behavior also includes the following specific steps:

[0100] S21, retrieve the real-time voltage and real-time current in the power consumption processing data;

[0101] S22, obtaining normal fluctuation ranges of real-time voltage and real-time current;

[0102] S23, analyzing the real-time voltage and real-time current within the set time period to obtain a voltage fluctuation curve and a current fluctuation curve;

[0103] S24, respectively obtaining the maximum value and the minimum value in the voltage fluctuation curve and the current fluctuation curve;

[0104] S25, calculating the difference between the corresponding maximum and minimum values ​​and the boundary of the normal fluctuation range;

[0105] S26, determining whether the boundary differences are all less than the difference threshold;

[0106] S27: If yes, it is determined that the user is suspected of stealing electricity;

[0107] S28. If not, analyzing the user's real-time power consumption;

[0108] S29, retrieve the real-time power in the power consumption processing data, fit the real-time power in the set time period into the power function F(t), and calculate the first-order derivative f(t) of the power function;

[0109] In this embodiment, the set time period is set based on the frequency of acquisition experience. When the acquisition frequency is fast, the set time period can be set shorter, and when the acquisition frequency is slow, the set time period can be set longer.

[0110] S210, analyzing the first-order derivative f(t) to obtain a minimum value;

[0111] S211, determining whether the minimum value is less than the minimum value threshold;

[0112] S212: If yes, determine that the user is suspected of stealing electricity;

[0113] S213: If not, it is determined that the user does not have any suspected behavior of stealing electricity.

[0114] like Figure 4 As shown, in this embodiment, the process of adjusting the collection frequency of real-time power consumption data also includes the following specific steps:

[0115] S21', obtaining the voltage fluctuation difference and the current fluctuation difference;

[0116] S22', marking the difference of the maximum value of voltage fluctuation, the difference of the minimum value, the difference of the maximum value of current fluctuation and the difference of the minimum value as VCmax, VCmin, ICmax and ICmin respectively;

[0117] S23', retrieve the minimum value of the first-order derivative of the electric quantity function, calculate the difference ZC between the minimum value of the first-order derivative and the minimum threshold value; obtain the original acquisition frequency YC;

[0118] S24', through the following formula:

[0119] TX=α×(VCmax+VCmin) / 2+β×(ICmax+ICmin) / 2+γ×ZC

[0120] Calculate and obtain the adjustment coefficient of the acquisition frequency;

[0121] S25', through the following formula:

[0122] CP=YC-θ×TX

[0123] Calculate the frequency of collecting real-time electricity consumption data;

[0124] Among them, α, β and γ are weight coefficients greater than 0; θ is a proportional coefficient greater than 0.

[0125] S3. Collect the electricity consumption data of the corresponding user according to the collection frequency; analyze whether the user has committed electricity theft according to the electricity consumption data; if so, analyze the type of electricity theft; if not, remove the suspicion of the corresponding user's suspected electricity theft;

[0126] like Figure 5 As shown, in this embodiment, the operation of analyzing whether the user has committed electricity theft also includes the following specific steps:

[0127] S31, retrieving the real-time voltage and real-time current in the power consumption data within a preset time period;

[0128] In this embodiment, the preset time period is set based on experience. During peak power consumption, the preset time period can be set shorter, and during low power consumption, the preset time period can be set longer. The preset time period can be dynamically adjusted manually.

[0129] S32, analyzing the real-time voltage and the real-time current to obtain a voltage fluctuation curve and a current fluctuation curve;

[0130] S33, determining whether the voltage fluctuation and the current fluctuation are both within the normal range;

[0131] S34: If yes, analyze the user's real-time power consumption;

[0132] S35, if not, it is determined that the user has committed electricity theft;

[0133] S36, retrieve the real-time power consumption in the power consumption data within a preset time period, refit the real-time power consumption into a power consumption function H(t), and solve the first-order derivative h(t) of the power consumption function H(t);

[0134] S37, determining whether the minimum value of the first-order derivative h(t) is less than the minimum value threshold;

[0135] S38: If yes, it is determined that the user has committed electricity theft;

[0136] S39. If not, it is determined that the user does not engage in electricity theft.

[0137] In this embodiment, in the operation of analyzing the type of electricity theft behavior, the electricity usage data within the preset time period of the corresponding user is retrieved, and the electricity usage data is integrated into a type input sequence according to the collection order; the electricity theft type identification model is retrieved, and the type input sequence is input into the electricity theft type identification model to obtain the corresponding type label, and the corresponding electricity theft behavior type is matched according to the type label; specifically, the electricity theft type identification model is constructed based on the artificial intelligence model.

[0138] like Figure 6 As shown, in this embodiment, the operation of building a power theft type identification model based on an artificial intelligence model also includes the following specific steps:

[0139] S31', obtain standard data set;

[0140] In this embodiment, the standard data set includes standard input data consistent with the type input sequence content attribute, and standard output data consistent with the type tag content attribute;

[0141] S32', dividing the standard data set into a training set and a validation set, and using the training set to train the artificial intelligence model;

[0142] S33', using the validation set to validate the artificial intelligence model and obtain validation indicators;

[0143] S34', judging whether all verification indicators have reached the verification standard;

[0144] S35', if yes, marking the artificial intelligence model as an electricity theft type identification model;

[0145] S36', if not, retrain the artificial intelligence model;

[0146] In this embodiment, the artificial intelligence model includes but is not limited to: a recurrent neural network model, or a long short-term memory neural network model.

[0147] In this embodiment, the types of electricity theft include, but are not limited to: unauthorized wiring and electricity use, privately increasing the electricity capacity of the light users, and using technical means to interfere with the normal operation of the energy meter so that it cannot accurately measure the electricity.

[0148] S4, generating warning information based on the type of electricity theft behavior of the user, and sending the warning information to the power grid management personnel, who will handle the electricity theft behavior of the user according to the warning information;

[0149] In this embodiment, the warning information includes: the electricity theft user, the type of electricity theft behavior and the time of electricity theft.

[0150] In summary, the present invention fully considers the randomness and uncertainty of users' electricity theft behaviors, monitors users' electricity consumption data in real time, and can timely discover users' electricity theft behaviors, thus reducing the economic losses of power companies.

[0151] The present invention analyzes the electricity usage data of users suspected of electricity theft. When a user is found to be engaged in electricity theft, the type of the user's electricity theft behavior is analyzed and early warning information is generated. The early warning information is sent to management personnel, who handle the user's electricity theft behavior based on the early warning information. Different handling measures can be taken for different types of electricity theft behavior, thereby reducing the losses of the power company.

[0152] The present invention collects real-time electricity consumption data of users and processes the real-time electricity consumption data, can retain valid real-time electricity consumption data, and eliminate invalid data, which is beneficial to improving the efficiency of subsequent data processing.

[0153] The present invention judges the user's suspected electricity theft behavior based on the electricity consumption data and adjusts the collection frequency of the corresponding user's real-time electricity consumption data, which can strengthen the monitoring of users who may have electricity theft behavior and avoid the failure to timely discover the user's electricity theft behavior due to incomplete monitoring, causing economic losses to the power company.

[0154] The present invention solves the technical problem in the prior art that the real-time monitoring of electricity consumption data is poor, resulting in untimely discovery of the user's electricity theft behavior.

[0155] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. Anti-electricity theft inspection and monitoring method based on big data, It is characterized in that The method comprises: S1. Collecting real-time electricity consumption data from users, preprocessing the real-time electricity consumption data, and obtaining electricity consumption processing data; S2. Determine whether the user is suspected of stealing electricity based on the electricity processing data; if so, adjust the frequency of collecting the real-time electricity data; if not, continue to detect the electricity processing data; S3. Collect the real-time electricity consumption data of the suspected electricity theft user according to the collection frequency; analyze whether the suspected electricity theft user has committed electricity theft behavior according to the real-time electricity consumption data; if so, retrieve the real-time electricity consumption data of the suspected electricity theft user within the preset time period, and integrate the real-time electricity consumption data into a type input sequence according to the collection order; retrieve the preset electricity theft type recognition model, input the type input sequence into the electricity theft type recognition model, obtain the corresponding type label, and match the corresponding electricity theft behavior type according to the type label; if not, screen out the suspected electricity theft user; S4. Generate warning information based on the electricity theft behavior type of the electricity theft user to handle the electricity theft behavior.

2. According to the big data-based anti-electricity theft inspection and monitoring method of claim 1, It is characterized in that In S1, the real-time power consumption data includes: collection time, real-time voltage, real-time current and real-time power consumption.

3. According to the big data-based anti-electricity theft inspection and monitoring method of claim 1, It is characterized in that The S1 includes: S11, retrieving and determining whether the real-time power consumption data contains sub-data of all preset data types; S13: If yes, retain the current real-time power consumption data; S14, if not, discarding the current real-time electricity consumption data; S15, determining whether the collection time of the real-time power consumption data is the same; S16: If yes, then remove errors and duplicates from the real-time electricity consumption data; S17, if not, retaining the current real-time power consumption data; S18: Mark the real-time power consumption data as the power consumption processed data.

4. According to the big data-based anti-electricity theft inspection and monitoring method of claim 1, It is characterized in that The S2 includes: S21, retrieving the real-time voltage and real-time current in the power consumption processing data, and processing to obtain the preset normal fluctuation range of the real-time voltage and the real-time current; S22, within a set time period, analyzing the real-time voltage and the real-time current to obtain a voltage fluctuation curve and a current fluctuation curve; S23, respectively obtaining the maximum value and the minimum value of the voltage fluctuation curve and the current fluctuation curve; S25, calculating the boundary difference between the maximum value, the minimum value and the preset normal fluctuation range; S26, determining whether the boundary differences are all less than a preset boundary difference threshold; S27: If yes, determining that the current user has the suspected electricity theft behavior; S28. If not, analyzing the current real-time power consumption of the user; S29, extracting the real-time power from the power consumption processing data, fitting the real-time power within the set time period into a power function F(t), and calculating the first-order derivative f(t) of the power function F(t); S210, analyzing the first-order derivative f(t) to obtain a first-order derivative minimum value, and determining whether the first-order derivative minimum value is less than a preset minimum value threshold; S212: If yes, determining that the current user has the suspected electricity theft behavior; S213: If not, it is determined that the current user does not have the suspected electricity theft behavior.

5. According to the big data-based anti-electricity theft inspection and monitoring method of claim 1, It is characterized in that The S2 further includes: S21', retrieving a voltage fluctuation difference and a current fluctuation difference from the power consumption processing data; S22', retrieve the minimum value of the first-order derivative of the preset power function, calculate the difference ZC between the minimum value of the first-order derivative and the preset minimum threshold value; obtain the original acquisition frequency YC; S23', calculating the acquisition frequency adjustment coefficient TX from the voltage fluctuation difference and the current fluctuation difference, the voltage fluctuation maximum value difference VCmax, the minimum value difference VCmin, the current fluctuation maximum value difference ICmax and the minimum value difference ICmin; S24', obtaining the collection frequency of the real-time power consumption data according to the collection frequency adjustment coefficient TX.

6. According to the big data-based anti-electricity theft inspection and monitoring method of claim 5, It is characterized in that In S23', the acquisition frequency adjustment coefficient TX is obtained using the following logic: TX=α×(VCmax+VCmin) / 2+β×(ICmax+ICmin) / 2+γ×ZC Where α, β and γ are weight coefficients greater than 0.

7. According to the big data-based anti-electricity theft inspection and monitoring method of claim 5, It is characterized in that In S24', the acquisition frequency is obtained by processing using the following logic: CP=YC-θ×TX Wherein, θ is a proportional coefficient greater than 0.

8. According to the big data-based anti-electricity theft inspection and monitoring method of claim 1, It is characterized in that The S3 includes: S31, retrieving the real-time voltage and real-time current in the power consumption data within a preset time period; S32, analyzing the real-time voltage and the real-time current to obtain a voltage fluctuation curve and a current fluctuation curve; S33, determining whether the voltage fluctuation and the current fluctuation are both within a normal range; S34: If yes, analyze the user's real-time power consumption; S35, if not, it is determined that the user has committed electricity theft; S36, retrieve the real-time power consumption in the power consumption data within a preset time period, refit the real-time power consumption into a power consumption function H(t), and solve the first-order derivative h(t) of the power consumption function H(t); S37, determining whether the minimum value of the first-order derivative h(t) is less than the minimum value threshold; S38: If yes, it is determined that the user has committed electricity theft; S39. If not, it is determined that the user does not engage in electricity theft.

9. According to the big data-based anti-electricity theft inspection and monitoring method of claim 1, It is characterized in that In S3, the operation of constructing the electricity theft type identification model further includes: S31', obtaining and dividing a standard data set to obtain a training set and a validation set, and using the training set to train a preset artificial intelligence model; S32', using the verification set to verify the artificial intelligence model until the electricity theft type identification model is obtained. 10.An anti-electricity theft inspection and monitoring system based on big data, It is characterized in that The system comprises: An electricity consumption data preprocessing module is used to collect real-time electricity consumption data from users, preprocess the real-time electricity consumption data, and obtain electricity consumption processed data; A suspected electricity theft identification module is used to determine whether the user has suspected electricity theft behavior based on the electricity consumption processing data; if so, adjust the frequency of collecting the real-time electricity consumption data; if not, continue to detect the electricity consumption processing data, and the suspected electricity theft identification module is connected to the electricity consumption data preprocessing module; The electricity theft user and type determination module is used to collect the real-time electricity consumption data of the suspected electricity theft user according to the collection frequency; analyze whether the suspected electricity theft user has committed electricity theft behavior according to the real-time electricity consumption data; if so, retrieve the real-time electricity consumption data of the suspected electricity theft user within the preset time period, and integrate the real-time electricity consumption data into a type input sequence according to the collection order; retrieve the preset electricity theft type identification model, input the type input sequence into the electricity theft type identification model, obtain the corresponding type label, and match the corresponding electricity theft behavior type according to the type label; if not, screen out the suspected electricity theft user, and the electricity theft user and type determination module is connected to the suspected electricity theft identification module; The electricity theft behavior processing module is used to generate warning information based on the electricity theft behavior type of the electricity theft user to handle the electricity theft behavior. The electricity theft behavior processing module is connected to the electricity theft user and the type determination module.

Images