User side electricity consumption abnormity monitoring system and method based on machine learning

Abstract

The invention relates to the technical field of electricity consumption abnormity monitoring of household AC distribution lines, and discloses a user side electricity consumption abnormity monitoring system and method based on machine learning, and the system comprises a user side electricity consumption abnormity analysis module, a user side electricity consumption abnormity object analysis module, and a user side electricity consumption abnormity feedback module. Based on the instantaneous power consumption parameter of the power consumption object of the power distribution line, combining with an intelligent search algorithm and a scientifically preset instantaneous rated power consumption parameter of the power consumption object of the power distribution line, accurately identifying the power consumption abnormal state of the power consumption object of the power distribution line of the user side; and according to the power consumption abnormal state identification parameters of the power consumption object of the power distribution line, combining a machine learning intelligent algorithm with the standard stored power consumption object identity text information of the power distribution line, carrying out efficient search on the identity information of the power consumption abnormal state object in the power distribution line of the user side, and realizing intelligent monitoring on the power consumption equipment of the power consumption abnormal state of the user side.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormal power consumption monitoring of household AC distribution lines, and specifically to a system and method for abnormal power consumption monitoring on a user side based on machine learning. Background Art

[0002] Electricity consumption refers to the amount of active electric energy consumed by an electricity-consuming object. It is the product of the active power of electricity consumption and time, and is usually measured by an electric energy meter. According to the different electricity-consuming objects, it can be divided into electricity consumption of a single electrical equipment, electricity consumption of product production, electricity consumption of enterprises and electricity consumption of industries. Electricity consumption is mainly used for energy consumption statistics, trade settlement, and energy consumption evaluation of equipment and products. Electricity consumption is a very important characteristic quantity in the process of electricity consumption and an important energy consumption indicator. By monitoring the electricity consumption parameters on the user side, the operating status of household distribution lines can be safely and reliably monitored. However, the existing user-side electricity consumption monitoring cannot accurately identify abnormal electricity consumption objects in household distribution lines, nor can it realize visual feedback of characteristic information of abnormal electricity consumption objects in household distribution lines, which reduces the safety and reliability of electricity consumption on the user side.

[0003] A Chinese invention patent with announcement number CN102237681B discloses a method, system and corresponding equipment for controlling the power consumption of household appliances. The method collects first working information and second working information of the monitored household appliances and transmits them to the power control system, scientifically analyzes the working status of the power control system and controls the working status of the corresponding household appliances. However, the above technical solution cannot realize intelligent identification and positioning of household appliances with abnormal power consumption and accurate and efficient feedback of abnormal household appliance information, which reduces the safety of household electricity use. Summary of the invention

[0004] 1. Technical issues to be resolved In order to solve the problem that some user-side electricity consumption monitoring cannot accurately identify abnormal electricity consumption objects in household distribution lines, and cannot realize visual feedback of abnormal electricity consumption objects in household distribution lines, which reduces the safety and reliability of electricity consumption on the user side, the above-mentioned purposes are achieved: accurately collect the total instantaneous electricity consumption parameters of the user-side distribution lines, scientifically analyze the abnormal electricity consumption status of the user side, collect the instantaneous electricity consumption parameters of the distribution line electricity users online, intelligently identify the abnormal electricity consumption status of the distribution line electricity users, efficiently identify the characteristic information of the abnormal electricity consumption objects of the distribution lines, accurately measure the distribution line loss electricity consumption parameters, accurately evaluate the abnormal electricity consumption status of the distribution lines, and visually and accurately prompt and feedback the characteristic information of the abnormal electricity consumption objects on the user side.

[0005] (II) Technical solution The present invention is implemented by the following technical scheme: a method for monitoring abnormal power consumption at a user side, the method comprising the following steps: S1, collect instantaneous power consumption data of distribution lines on the user side; S2, analyzing and processing the abnormal state of power consumption on the user side according to the instantaneous power consumption data of the user side distribution line and the abnormal instantaneous power consumption interval of the user side distribution line, generating abnormal state analysis data of power consumption on the user side, and ending the abnormal power consumption monitoring operation on the user side when it is normal; S3. When it is abnormal, collect instantaneous power consumption data of power users in the distribution line; S4, based on the instantaneous power consumption data of the power consumption objects in the distribution line and the instantaneous rated power consumption data of the power consumption objects in the distribution line on the user side, an abnormal power consumption state identification process is performed on the power consumption objects in the distribution line on the user side, and abnormal power consumption state identification data of the power consumption objects in the distribution line is constructed; S5, performing identification processing of the identity information of the power user in the abnormal power consumption state in the user-side distribution line according to the power user abnormal power consumption state identification data of the distribution line power user and the power user identity text data of the distribution line power user, and generating the power user identity text data of the distribution line abnormal power consumption; S6, measuring the power consumption data of the distribution line loss and performing abnormal power consumption state evaluation processing on the user-side distribution line loss with the normal power consumption interval of the distribution line loss, and generating abnormal power consumption state evaluation data of the distribution line loss; S7. Construct user-side power consumption abnormality monitoring data and execute user-side power consumption abnormality monitoring information feedback operation.

[0006] Preferably, the operation steps of collecting instantaneous power consumption data of the user-side distribution line are as follows: S11. Collect the instantaneous total power consumption data of the household distribution line on the user side online through the power meter, and generate the instantaneous power consumption data of the user side distribution line , the instantaneous power consumption data of the user-side distribution line represents the total power consumption data of the user-side distribution line in unit time, where The unit is kilowatt-second.

[0007] Preferably, the abnormal power consumption state of the user side is analyzed and processed according to the instantaneous power consumption data of the user side distribution line and the abnormal instantaneous power consumption interval of the user side distribution line, and the abnormal power consumption state analysis data of the user side is generated. When it is normal, the operation steps of ending the abnormal power consumption monitoring operation of the user side are as follows: S21. Establish abnormal instantaneous power consumption interval of user-side distribution lines ,in Indicates the abnormal instantaneous power consumption range of the user-side distribution line Abnormal minimum instantaneous power consumption data of the distribution line on the user side, Indicates the abnormal instantaneous power consumption range of the user-side distribution line Abnormal maximum instantaneous power consumption data of the distribution line on the user side, and The unit is kilowatt-second; the abnormal instantaneous power consumption range of the user-side distribution line indicates the total power consumption range value per unit time when the user-side distribution line is in an abnormal power consumption state. The abnormal power consumption state includes distribution line short circuit, distribution line overload, distribution line overload and distribution line leakage; S22: The instantaneous power consumption data of the user-side power distribution line The abnormal instantaneous power consumption interval of the user-side distribution line Abnormal minimum instantaneous power consumption data of the user-side distribution line described in , the abnormal maximum instantaneous power consumption data of the user-side distribution line Compare the power consumption values ​​and generate abnormal power consumption status analysis data on the user side based on the power consumption value comparison results ; when ∈ When the power distribution line on the user side is in an abnormal power consumption state, the abnormal power consumption state analysis data on the user side is output. is abnormal; when ∉ When the power distribution line on the user side is in a normal power consumption state, the abnormal power consumption state analysis data on the user side is output. If the result is normal, the abnormal power consumption monitoring operation on the user side will be terminated directly.

[0008] Preferably, when an abnormality occurs, the operation steps for collecting instantaneous power consumption data of power consumption objects of the power distribution line are as follows: S31, when the user side power consumption abnormal state analysis data When the power consumption is abnormal, the instantaneous power consumption data of the power users connected to the distribution line on the user side is collected online through the power meter, and the instantaneous power consumption data set of the power users of the distribution line is generated. , ;in Indicates the collected The instantaneous power consumption data of the power distribution line users corresponding to the power distribution line users. Indicates the maximum value of the number of power consumption objects in the distribution line; the instantaneous power consumption data of the power consumption objects in the distribution line indicates the total power consumption data of the power consumption objects in the distribution line within a unit time, where The unit is kilowatt-second. The objects that consume electricity in the distribution line include refrigerators, air conditioners, televisions, electric fans, humidifiers, rice cookers, induction cookers and electric kettles.

[0009] Preferably, based on the instantaneous power consumption data of the power consumption objects in the distribution line and the instantaneous rated power consumption data of the power consumption objects in the distribution line at the user side, the abnormal power consumption state identification processing of the power consumption objects in the distribution line is performed, and the operation steps of constructing the abnormal power consumption state identification data of the power consumption objects in the distribution line are as follows: S41. Establishing instantaneous rated power consumption data set of power users in distribution lines ,in Indicates The instantaneous rated power consumption data of the power consumption objects of the distribution lines corresponding to the power consumption objects of the distribution lines represent the power consumption data required for the power consumption objects of the distribution lines to operate normally per unit time under the rated voltage condition. The unit of is kilowatt-second; S42, using XGBoost algorithm to collect the instantaneous power consumption data of the power consumption objects of the distribution line Instantaneous power consumption data of power consumption objects in the distribution line The instantaneous rated power consumption data set of the power consumption objects of the distribution line is ordered according to the power consumption object number of the distribution line Instantaneous rated power consumption data of power distribution line power users described in Compare the power consumption values ​​and build a data set for identifying abnormal power consumption status of power users in distribution lines based on the power consumption value comparison results. ,in Indicates Identification data of abnormal power consumption status of power consumption objects of distribution lines corresponding to power consumption objects of distribution lines; when > When If the instantaneous power consumption of a power consumption object of a distribution line exceeds the instantaneous rated power consumption, and the power consumption object of the distribution line is in an abnormal power consumption state, the power consumption abnormal state identification data of the power consumption object of the distribution line is output. is abnormal; when ≤ When If the instantaneous power consumption of the power consumption object of the distribution line does not exceed the instantaneous rated power consumption, the power consumption object of the distribution line is in a normal state, and the power consumption abnormal state identification data of the power consumption object of the distribution line is output. is normal.

[0010] Preferably, the identification process of the power user identity information of the power user in the power distribution line with abnormal power consumption in the user side power distribution line is performed according to the power distribution line power user abnormal power consumption state identification data and the power distribution line power user identity text data, and the operation steps of generating the power distribution line power consumption abnormal power consumption object identity text data are as follows: S51. Establishing a text data set of the identity of the power users of the distribution line ,in Indicates The power distribution line user object identity text data corresponding to the power distribution line user object includes the product name identity text information, product model identity text information, product production number identity text information and manufacturer identity text information of the power distribution line user object; S52: Collecting the abnormal power consumption status identification data of the power consumption objects of the distribution line The abnormal power consumption state identification data of the power consumption object of the distribution line whose abnormal power consumption state analysis result is abnormal A text data set related to the identity of the power user of the distribution line The identity text data of the power user of the distribution line described in Match the number of the power consumption object of the distribution line, and search for the power consumption abnormal state identification data of the power consumption object of the distribution line whose power consumption abnormal state analysis result is abnormal The corresponding distribution line electricity user identity text data And construct a text data set of the identity of the abnormal power consumption objects of the distribution line , execute to generate the identity text data set of the abnormal power consumption object of the distribution line The specific steps are as follows: S521, propagation: the distribution line electricity user identity text data set The identity text data of the power user of the distribution line described in Synchronization is set to water wave, each of the distribution line power object identity text data The water wave is an independent individual, so the identity text data of each power distribution line power object Water waves will have three properties: position, wavelength and wave height ; In each iteration, the identity text data of each power distribution line power object The water waves will propagate the identity text data set of the power consumption object of the distribution line The power distribution line electricity user identity text data in the search space The abnormal power consumption state identification data of the power consumption object of the distribution line whose abnormal power consumption state analysis result is abnormal Search and match the number of the power object of the distribution line, and at the same time It will decrease by 1, and its position update formula is as follows: ,in Indicates that in the propagation stage After iterations, water waves The power distribution line electricity user identity text data set Position in the search space; Indicates that in the propagation stage Water wave before iteration The power distribution line electricity user identity text data set Position in the search space; , They are respectively the identity text data sets of the electricity users in the distribution line The upper and lower bounds in the search space, Indicates value Random function of S522, Refraction: a text data of the identity of the power user in the distribution line After the water wave propagates, the identity text data of the power user of the distribution line Water waves are refracted and the identity text data set of the power distribution line power object is collected The power distribution line electricity user identity text data in the search space The abnormal power consumption state identification data of the power consumption object of the distribution line whose abnormal power consumption state analysis result is abnormal Perform a search and match on the distribution line electricity object number, and each time transmit the distribution line electricity object identity text data water wave height will decrease by 1 when When it decreases to 0, the power distribution line electricity user identity text data The water waves will be refracted, and their height and wavelength will change; the position after refraction is normally distributed in the text data of the current power distribution line power user identity Water wave and optimal distribution line electricity user identity text data The midpoint of the water wave is the mean value, and the current distribution line power user identity text data Water wave and optimal distribution line electricity user identity text data The water wave distance is the position of the variance, and the identity text data of the power distribution line power object after refraction The wave height will be reinitialized to the maximum wave height After refraction, recalculate the identity text data of the power distribution line power user Wavelength of water waves ; S523, breaking waves: the identity text data of the power user of the distribution line The text data set of the identity of the power users of the distribution line described by Shuibo After propagation in the search space, the distribution line power user identity text data that is better than the current optimal one is reached. The location of the water wave, i.e., the identity text data set of the power user in the distribution line Search the search space for the abnormal power consumption state identification data of the power consumption object of the distribution line that is abnormal in the power consumption abnormal state analysis result The matching text data of the power distribution line power user identity The position of the power distribution line power object identity text data The water waves will break and send the current optimal distribution line power user identity text data The water wave propagates to the location where the breaking waves are generated, that is, the power consumption abnormal state identification data of the power consumption object of the distribution line whose power consumption abnormal state analysis result is abnormal is output. The matching text data of the power distribution line power user identity The formula for generating the breaking wave position is as follows: ,in Indicates that in the breaking wave stage After iterations, water waves The power distribution line electricity user identity text data set The position in the search space, Indicates value The random function of to Indicates that each wave randomly selects a spatial dimension number, represents the specific spatial dimension that is actually randomly selected for each breaking wave, to The specific spatial dimension parameters randomly selected for each breaking wave, is a constant; S524: Outputting the abnormal power consumption state identification data of the power consumption object of the distribution line that is abnormal according to the abnormal power consumption state analysis result The matching text data of the power distribution line power user identity ; S525: the power distribution line electricity user identity text data output in step S524 After data identification, the identity text data set of the abnormal power consumption objects of the distribution line is constructed ,in ,in Indicates The identity text data of the distribution line power consumption abnormal object corresponding to the distribution line power consumption abnormal object, Indicates The identity text data of the distribution line power consumption abnormal object corresponding to the distribution line power consumption abnormal object.

[0011] Preferably, the power loss and power consumption data of the distribution line is measured and the power loss and power consumption abnormal state evaluation processing of the user-side distribution line is performed with the normal power loss and power consumption interval of the distribution line, and the operation steps of generating the power loss and power consumption abnormal state evaluation data of the distribution line are as follows: S61, based on the instantaneous power consumption data set of the power consumption objects of the distribution line Instantaneous power consumption data of power consumption objects in the distribution line Combined with the summation formula, the total instantaneous power consumption of all power users in the user-side distribution line is measured and processed to generate the total instantaneous power consumption data of the power users in the distribution line. ,in , The unit of is kilowatt-second; S62: The instantaneous power consumption data of the user-side distribution line Total instantaneous power consumption data of the power consumption objects on the distribution line Perform differential value processing on instantaneous power consumption data to measure power consumption data of power distribution line loss ,in , The unit is kilowatt-second, and the power consumption data of the power distribution line loss represents the power consumption data of the power distribution line at the user side in unit time; S63. Establish normal power consumption range of distribution lines ,in Indicates the normal power consumption range of the distribution line The minimum normal power loss data of the distribution line, Indicates the normal power consumption range of the distribution line The maximum normal power consumption data of the distribution line, and The unit is kilowatt-second; the normal power consumption range of the distribution line indicates the range of line power consumption per unit time when the user-side distribution line is normally powered on; S64, the power consumption data of the power distribution line loss The normal power consumption range of the distribution line The minimum normal power consumption data of the distribution line described in , the abnormal maximum instantaneous power consumption data of the user-side distribution line Compare the power loss values ​​and generate the distribution line power loss abnormal state assessment data based on the power loss values ​​comparison results ; when ∈ When , it means that the power loss of the user-side distribution line is in normal state, then the power loss of the distribution line abnormal state assessment data is output. is normal; when ∉ When , it means that the power loss of the user-side distribution line is in an abnormal state, and the power loss abnormal state evaluation data of the distribution line is output. is abnormal.

[0012] Preferably, the operation steps of constructing the user-side power consumption abnormality monitoring data and performing the user-side power consumption abnormality monitoring information feedback operation are as follows: S71: Analyze the abnormal state of power consumption at the user side , the identity text data set of the power distribution line power consumption abnormal object , the distribution line loss power consumption abnormal state assessment data Construct abnormal power consumption monitoring data on the user side through data combination ,in ; S72: The abnormal power consumption monitoring data of the user side The abnormal power consumption monitoring information feedback operation on the user side is pushed to the user monitoring terminal through the Internet of Things communication network. The user monitoring terminal includes any one of a smart phone, a tablet computer, and a smart watch.

[0013] A user-side power consumption abnormality monitoring system based on machine learning, used to implement the user-side power consumption abnormality monitoring method, the system includes a user-side power consumption abnormality analysis module, a user-side power consumption abnormality object analysis module, and a user-side power consumption abnormality feedback module; The user-side power consumption abnormality analysis module includes a user-side distribution line instantaneous power consumption acquisition unit, a user-side distribution line abnormal instantaneous power consumption interval storage unit, and a user-side power consumption abnormality state analysis unit; The user-side distribution line instantaneous power consumption collection unit collects instantaneous power consumption data of the user-side distribution line through an electric energy meter; the user-side distribution line abnormal instantaneous power consumption interval storage unit is used to store the abnormal instantaneous power consumption interval of the user-side distribution line; the user-side power consumption abnormal state analysis unit performs power consumption abnormal state analysis processing on the user side according to the user-side distribution line instantaneous power consumption data and the user-side distribution line abnormal instantaneous power consumption interval, and generates user-side power consumption abnormal state analysis data; The user-side power consumption abnormal object analysis module includes a distribution line power consumption object instantaneous power consumption collection unit, a distribution line power consumption object instantaneous rated power consumption storage unit, a distribution line power consumption object power consumption abnormal state identification unit, a distribution line power consumption object identity information storage unit, a distribution line power consumption abnormal object identification unit, a distribution line loss power consumption metering unit, a distribution line normal loss power consumption interval storage unit, and a distribution line loss power consumption abnormal state evaluation unit; The instantaneous power consumption collection unit of the power consumption object of the distribution line collects the instantaneous power consumption data of the power consumption object of the distribution line through the electric energy meter; the instantaneous rated power consumption storage unit of the power consumption object of the distribution line is used to store the instantaneous rated power consumption data of the power consumption object of the distribution line; the power consumption abnormal state identification unit of the power consumption object of the distribution line performs abnormal power consumption state identification processing of the power consumption object in the user-side distribution line based on the instantaneous power consumption data of the power consumption object of the distribution line and the instantaneous rated power consumption data of the power consumption object of the distribution line, and constructs the power consumption abnormal state identification data of the power consumption object of the distribution line; the power consumption object identity information storage unit of the distribution line is used to store the identity text data of the power consumption object of the distribution line; the power consumption abnormal object identification unit of the distribution line, according to the distribution line The abnormal power consumption state identification data of the power line power user object is combined with the machine learning intelligent algorithm and the distribution line power user object identity text data to perform power user object identity information identification processing of the abnormal power consumption state in the user-side distribution line, and generate the distribution line power consumption abnormal object identity text data; the distribution line loss power consumption metering unit is used to measure the distribution line loss power consumption data; the distribution line normal loss power consumption interval storage unit is used to store the distribution line normal loss power consumption interval; the distribution line loss power consumption abnormal state evaluation unit performs the loss power consumption abnormal state evaluation processing of the user-side distribution line based on the distribution line loss power consumption data and the distribution line normal loss power consumption interval, and generates the distribution line loss power consumption abnormal state evaluation data; The user-side power consumption abnormality feedback module includes a user-side power consumption abnormality monitoring information acquisition unit and a user-side power consumption abnormality monitoring information feedback unit; The user-side power consumption abnormality monitoring information acquisition unit is used to construct the user-side power consumption abnormality monitoring data; the user-side power consumption abnormality monitoring information feedback unit pushes the user-side power consumption abnormality monitoring data to the user monitoring terminal through the Internet of Things communication network to perform the user-side power consumption abnormality monitoring information feedback operation.

[0014] (III) Beneficial effects The present invention provides a system and method for monitoring abnormal power consumption at the user side based on machine learning. It has the following beneficial effects: 1. Accurately collect the instantaneous power consumption parameters of the user-side distribution line online through the electric energy meter to provide reliable data support for the accurate assessment of the abnormal power consumption status on the user side; based on numerical analysis, the instantaneous power consumption parameters of the user-side distribution line and the abnormal instantaneous power consumption interval of the user-side distribution line stored in the standard are used to conduct efficient and accurate analysis of the abnormal power consumption status on the user side, so as to realize digital intelligent monitoring of the abnormal power consumption status on the user side.

[0015] 2. The instantaneous power consumption parameters of the power users of the distribution lines are collected autonomously and efficiently through the electric energy meters, providing real data support for the scientific analysis of the abnormal power consumption status of the power equipment of the distribution lines; based on the instantaneous power consumption parameters of the power users of the distribution lines, combined with the intelligent search algorithm and the scientifically preset instantaneous rated power consumption parameters of the power users of the distribution lines, the abnormal power consumption status of the power users of the distribution lines on the user side is accurately identified, and dynamic and accurate monitoring of the abnormal power consumption status of the power users of the power equipment on the user side is achieved, thereby improving the safety of power consumption on the user side; based on the abnormal power consumption status identification parameters of the power users of the distribution lines, combined with the machine learning intelligent algorithm and the standard stored identity text information of the power users of the distribution lines, the identity information of the objects with abnormal power consumption status in the distribution lines on the user side is efficiently searched, and intelligent monitoring of the power users with abnormal power consumption status on the user side is achieved; the loss power consumption parameters of the distribution lines are autonomously measured, combined with numerical analysis and the scientifically preset normal loss power consumption interval of the distribution lines, to perform intelligent evaluation of the abnormal loss power consumption status of the distribution lines on the user side, so as to accurately monitor the abnormal loss power consumption status of the distribution lines on the user side, and improve the scientificity and accuracy of the abnormal power consumption monitoring results on the user side.

[0016] 3. By combining data processing with the abnormal power consumption status analysis results on the user side, the abnormal power consumption status analysis results of the power users of the distribution line, and the abnormal power consumption status assessment results of the distribution line loss, the abnormal power consumption monitoring information on the user side is independently and efficiently constructed, and the abnormal power consumption monitoring information on the user side is efficiently and accurately collected, which improves the efficiency of the feedback of the abnormal power consumption monitoring results on the user side; the abnormal power consumption monitoring information on the user side is efficiently and timely pushed to the user monitoring terminal through the Internet of Things communication network, which realizes intuitive and visual feedback of the abnormal power consumption monitoring information on the user side, and improves the applicability of the abnormal power consumption monitoring system on the user side. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a module of a user-side power consumption anomaly monitoring system based on machine learning provided by the present invention; Figure 2 This is a flow chart of the method for monitoring abnormal power consumption at the user side provided by the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] The embodiments of the user-side power consumption abnormality monitoring system and method based on machine learning are as follows: Example

[0020] See also Figure 1 - Figure 2 , a method for monitoring abnormal power consumption at the user side, the method comprising the following steps: S1, collect instantaneous power consumption data of distribution lines on the user side; S2. Analyze and process the abnormal power consumption state of the user side according to the instantaneous power consumption data of the user side distribution line and the abnormal instantaneous power consumption interval of the user side distribution line, generate abnormal power consumption state analysis data of the user side, and end the abnormal power consumption monitoring operation of the user side when it is normal; S3. When it is abnormal, collect instantaneous power consumption data of power users in the distribution line; S4, based on the instantaneous power consumption data of the power consumption objects in the distribution line and the instantaneous rated power consumption data of the power consumption objects in the distribution line on the user side, an abnormal power consumption state identification process is performed on the power consumption objects in the distribution line on the user side, and abnormal power consumption state identification data of the power consumption objects in the distribution line is constructed; S5. Perform identification processing on the identity information of the power user with abnormal power consumption in the distribution line on the user side according to the power user abnormal power consumption state identification data of the power user in the distribution line and the power user identity text data of the power user in the distribution line, and generate the power user identity text data of the power user with abnormal power consumption in the distribution line; S6, measuring the power consumption data of the distribution line loss and performing abnormal power consumption state evaluation processing on the user-side distribution line loss with the normal power consumption interval of the distribution line loss, and generating abnormal power consumption state evaluation data of the distribution line loss; S7. Construct user-side power consumption abnormality monitoring data and execute user-side power consumption abnormality monitoring information feedback operation.

[0021] For further information, see Figure 1 - Figure 2 , the steps for collecting instantaneous power consumption data of the user-side distribution line are as follows: S11. Collect the instantaneous total power consumption data of the household distribution line on the user side online through the power meter, and generate the instantaneous power consumption data of the user side distribution line , the instantaneous power consumption data of the user-side distribution line represents the total power consumption data of the user-side distribution line in unit time, where The unit is kilowatt-second.

[0022] According to the instantaneous power consumption data of the user-side distribution line and the abnormal instantaneous power consumption interval of the user-side distribution line, the abnormal power consumption state of the user side is analyzed and processed to generate the abnormal power consumption state analysis data of the user side. When it is normal, the operation steps to end the abnormal power consumption monitoring operation of the user side are as follows: S21. Establish abnormal instantaneous power consumption interval of user-side distribution lines ,in Indicates the abnormal instantaneous power consumption range of the user-side distribution line Abnormal minimum instantaneous power consumption data of the distribution line on the user side, Indicates the abnormal instantaneous power consumption range of the user-side distribution line Abnormal maximum instantaneous power consumption data of the distribution line on the user side, and The unit is kilowatt-second; the abnormal instantaneous power consumption range of the user-side distribution line indicates the total power consumption range value per unit time when the user-side distribution line is in an abnormal power consumption state. The abnormal power consumption state includes distribution line short circuit, distribution line overload, distribution line overload and distribution line leakage; S22, the instantaneous power consumption data of the user-side distribution line Abnormal instantaneous power consumption interval of the user-side distribution line Abnormal minimum instantaneous power consumption data of distribution lines on the user side , Abnormal maximum instantaneous power consumption data of user-side distribution lines Compare the power consumption values ​​and generate abnormal power consumption status analysis data on the user side based on the power consumption value comparison results ; when ∈ When the power distribution line on the user side is in an abnormal power consumption state, the abnormal power consumption state analysis data on the user side is output. is abnormal; when ∉ When the power distribution line on the user side is in a normal power consumption state, the abnormal power consumption state analysis data on the user side is output. If the result is normal, the abnormal power consumption monitoring operation on the user side will be terminated directly.

[0023] Through the user-side distribution line instantaneous power consumption collection unit, the electric energy meter is used to accurately collect the instantaneous power consumption parameters of the user-side distribution line online, providing reliable data support for the accurate evaluation of the abnormal power consumption status on the user side; the user-side power consumption abnormal status analysis unit, based on numerical analysis, compares the instantaneous power consumption parameters of the user-side distribution line with the standard stored abnormal instantaneous power consumption interval of the user-side distribution line to conduct efficient and accurate analysis of the abnormal power consumption status on the user side, thereby realizing digital intelligent monitoring of the abnormal power consumption status on the user side.

[0024] For further information, see Figure 1 - Figure 2 When it is abnormal, the steps to collect the instantaneous power consumption data of the power consumption objects of the distribution line are as follows: S31. Analyze data when power consumption on the user side is abnormal When the power consumption is abnormal, the instantaneous power consumption data of the power users connected to the distribution line on the user side is collected online through the power meter, and the instantaneous power consumption data set of the power users of the distribution line is generated. , ;in Indicates the collected The instantaneous power consumption data of the power distribution line users corresponding to the power distribution line users. Indicates the maximum value of the number of power consumption objects in the distribution line; the instantaneous power consumption data of the power consumption objects in the distribution line indicates the total power consumption data of the power consumption objects in the distribution line within a unit time, where The unit is kilowatt-second. The objects that consume electricity in the distribution line include refrigerators, air conditioners, televisions, electric fans, humidifiers, rice cookers, induction cookers and electric kettles.

[0025] Based on the instantaneous power consumption data of the power consumption objects in the distribution line and the instantaneous rated power consumption data of the power consumption objects in the distribution line on the user side, the abnormal power consumption state identification processing of the power consumption objects in the distribution line on the user side is performed, and the operation steps for constructing the abnormal power consumption state identification data of the power consumption objects in the distribution line are as follows: S41. Establishing instantaneous rated power consumption data set of power users in distribution lines ,in Indicates The instantaneous rated power consumption data of the power consumption objects of the distribution line corresponding to the power consumption objects of the distribution line, the instantaneous rated power consumption data of the power consumption objects of the distribution line represent the power consumption data required for the normal operation of the power consumption objects of the distribution line under the rated voltage condition per unit time, The unit of is kilowatt-second; S42, using XGBoost algorithm to collect instantaneous power consumption data of distribution line power users Instantaneous power consumption data of power users in power distribution lines The data set of instantaneous rated power consumption of the power consumption objects of the distribution line is sorted according to the power consumption object number of the distribution line Instantaneous rated power consumption data of power users in power distribution lines Compare the power consumption values ​​and build a data set for identifying abnormal power consumption status of power users in distribution lines based on the power consumption value comparison results. ,in Indicates Identification data of abnormal power consumption status of power consumption objects of distribution lines corresponding to power consumption objects of distribution lines; when > When If the instantaneous power consumption of a power consumption object of a distribution line exceeds the instantaneous rated power consumption, and the power consumption object of the distribution line is in an abnormal power consumption state, the power consumption abnormal state identification data of the power consumption object of the distribution line is output. is abnormal; when ≤ When If the instantaneous power consumption of the power consumption object of the distribution line does not exceed the instantaneous rated power consumption, the power consumption object of the distribution line is in a normal state, and the power consumption abnormal state identification data of the power consumption object of the distribution line is output. is normal.

[0026] The steps for identifying the identity information of the power user in the power distribution line with abnormal power consumption status and the power user identity text data of the power distribution line are as follows: S51. Establishing a text data set of the identity of the power users of the distribution line ,in Indicates The distribution line power user object identity text data corresponding to the distribution line power user object, the distribution line power user object identity text data includes the product name identity text information, product model identity text information, product production number identity text information and manufacturer identity text information of the distribution line power user object; S52, collecting the abnormal power consumption status identification data of the power consumption objects of the distribution line Abnormal power consumption state identification data of power consumption objects in distribution lines whose abnormal power consumption state analysis results are abnormal A collection of text data about the identity of the power users of distribution lines The identity text data of the power user in the power distribution line Match the number of the power consumption object of the distribution line, and search for the power consumption abnormality status identification data of the power consumption object of the distribution line whose power consumption abnormality status analysis result is abnormal The corresponding distribution line power user identity text data And construct a text data set of the identity of the abnormal power consumption objects of the distribution line , execute to generate the distribution line power consumption abnormal object identity text data set The specific steps are as follows: S521, dissemination: collect the identity text data of the power users of the distribution lines The identity text data of the power user in the power distribution line Synchronization is set to water wave, and each distribution line power object identity text data The water wave is an independent individual, so the identity text data of each distribution line power user Water waves will have three properties: position, wavelength and wave height ; In each iteration, the identity text data of each distribution line power user is Water waves will propagate through the distribution line electricity user identity text data set Distribution line electricity user identity text data in search space The abnormal power consumption state identification data of the power consumption object of the distribution line whose abnormal power consumption state analysis result is abnormal Search and match the number of the power object of the distribution line, and at the same time It will decrease by 1, and its position update formula is as follows: ,in Indicates that in the propagation stage After iterations, water waves A collection of text data on the identity of electricity users in distribution lines Position in the search space; Indicates that in the propagation stage Water wave before iteration A collection of text data on the identity of electricity users in distribution lines Position in the search space; , They are the identity text data sets of the electricity users in the distribution lines. The upper and lower bounds in the search space, Indicates value Random function of S522, Refraction: Text data of the identity of an electricity user in a distribution line After the water wave propagates, the identity text data of the power user of the distribution line Water wave refraction and distribution line power user identity text data collection Distribution line electricity user identity text data in search space The abnormal power consumption state identification data of the power consumption object of the distribution line whose abnormal power consumption state analysis result is abnormal Search and match the distribution line power object number, and transmit the distribution line power object identity text data each time water wave height will decrease by 1 when When it decreases to 0, the identity text data of the power user of the distribution line The water waves will be refracted, and their height and wavelength will change; the position after refraction is normally distributed in the text data of the current distribution line power user identity Water wave and optimal distribution line electricity user identity text data The midpoint of the water wave is the mean value, and the current distribution line power user identity text data Water waves and optimal distribution line electricity user identity text data The distance of the water wave is the position of the variance, and the identity text data of the power user of the distribution line after refraction The wave height will be reinitialized to the maximum wave height ; After refraction, recalculate the identity text data of the power user of the distribution line Wavelength of water waves ; S523, Breaking Waves: Distribution Line Power User Identity Text Data Water wave distribution line electricity user identity text data set After propagation in the search space, the user identity text data of the distribution line is better than the current optimal one. The location of the water wave, i.e. the identity text data set of the power user in the distribution line Search the search space to find the abnormal power consumption state identification data of the power consumption objects of the distribution line that are abnormal in the power consumption abnormal state analysis result Matching distribution line power user identity text data The location of the power distribution line power user identity text data The water waves will break and send the current optimal distribution line electricity user identity text data The water waves propagate to the location where the breaking waves are generated, that is, the abnormal power consumption state identification data of the power consumption objects of the distribution line whose power consumption abnormal state analysis results are abnormal is output. Matching distribution line power user identity text data The formula for generating the breaking wave position is as follows: ,in Indicates that in the breaking wave stage After iterations, water waves A collection of text data on the identity of electricity users in distribution lines The position in the search space, Indicates value The random function of to Indicates that each wave randomly selects a spatial dimension number, represents the specific spatial dimension that is actually randomly selected for each breaking wave, to The specific spatial dimension parameters randomly selected for each breaking wave, is a constant; S524, outputting abnormal power consumption state identification data of the power consumption object of the distribution line whose abnormal power consumption state analysis result is abnormal Matching distribution line power user identity text data ; S525: The power distribution line electricity user identity text data output in step S524 is After data identification, the identity text data set of the abnormal power consumption objects of the distribution line is constructed ,in ,in Indicates The identity text data of the distribution line power consumption abnormal object corresponding to the distribution line power consumption abnormal object, Indicates The identity text data of the distribution line power consumption abnormal object corresponding to the distribution line power consumption abnormal object.

[0027] Measure the power consumption data of the distribution line loss and perform abnormal power consumption status evaluation of the user-side distribution line loss in the normal power consumption interval of the distribution line. The steps for generating abnormal power consumption status evaluation data of the distribution line loss are as follows: S61. Data collection based on instantaneous power consumption of power users on distribution lines Instantaneous power consumption data of power users in power distribution lines Combined with the summation formula, the total instantaneous power consumption of all power users in the user-side distribution line is measured and processed to generate the total instantaneous power consumption data of the power users in the distribution line. ,in , The unit of is kilowatt-second; S62, the instantaneous power consumption data of the user side distribution line Total instantaneous power consumption data of power users in distribution lines Perform differential value processing on instantaneous power consumption data to measure power consumption data of power distribution line loss ,in , The unit is kilowatt-second. The power consumption data of the distribution line loss indicates the power consumption data of the user-side distribution line loss per unit time. S63. Establish normal power consumption range of distribution lines ,in Indicates the normal power consumption range of the distribution line The minimum normal power loss data of the distribution line, Indicates the normal power consumption range of the distribution line The maximum normal power consumption data of the distribution line, and The unit is kilowatt-second; the normal power consumption range of the distribution line indicates the range of line power consumption per unit time when the user-side distribution line is normally powered on; S64, the power consumption data of the distribution line loss The normal power consumption range of the distribution line Minimum normal power loss data of power distribution lines , Abnormal maximum instantaneous power consumption data of user-side distribution lines Compare the power loss values ​​and generate the distribution line power loss abnormal state assessment data based on the power loss values ​​comparison results ; when ∈ When , it means that the power loss of the user-side distribution line is in normal state, then the power loss of the distribution line abnormal state assessment data is output. is normal; when ∉ When , it means that the power loss of the user-side distribution line is in an abnormal state, and the power loss abnormal state evaluation data of the distribution line is output. is abnormal.

[0028] Through the distribution line power consumption object instantaneous power consumption collection unit, the electric energy meter is used to autonomously and efficiently collect the instantaneous power consumption parameters of the distribution line power consumption objects, providing real data support for the scientific analysis of the abnormal power consumption status of the distribution line power equipment; the distribution line power consumption object abnormal state identification unit accurately identifies the abnormal power consumption status of the user-side distribution line power consumption object based on the instantaneous power consumption parameters of the distribution line power consumption object combined with the intelligent search algorithm and the scientifically preset instantaneous rated power consumption parameters of the distribution line power consumption object, realizes dynamic and accurate monitoring of the abnormal power consumption status of the user-side power consumption equipment, and improves the safety of power consumption on the user side; the distribution line power consumption abnormal object identification unit, according to the abnormal power consumption of the distribution line power consumption object The normal state recognition parameters are combined with the machine learning intelligent algorithm and the standard stored distribution line power user identity text information to efficiently search for the identity information of the objects with abnormal power consumption in the user-side distribution line, so as to realize the intelligent monitoring of the power consumption equipment with abnormal power consumption on the user side; the distribution line loss power consumption metering unit and the distribution line loss power consumption abnormal state evaluation unit cooperate with each other to independently measure the distribution line loss power consumption parameters and combine them with numerical analysis and scientifically preset distribution line normal loss power consumption intervals to perform intelligent evaluation of the abnormal power consumption state of the distribution line on the user side, so as to realize accurate monitoring of the abnormal power consumption state of the distribution line on the user side and improve the scientificity and accuracy of the abnormal power consumption monitoring results on the user side.

[0029] For further information, see Figure 1 - Figure 2 The steps for constructing the abnormal power consumption monitoring data on the user side and performing the abnormal power consumption monitoring information feedback operation on the user side are as follows: S71, analyzing the abnormal state of power consumption on the user side , Distribution line power consumption abnormal object identity text data set , Distribution line loss power consumption abnormal status assessment data Construct abnormal power consumption monitoring data on the user side through data combination ,in ; S72, abnormal power consumption monitoring data on the user side The abnormal power consumption monitoring information feedback operation on the user side is pushed to the user monitoring terminal through the Internet of Things communication network. The user monitoring terminal includes any one of a smart phone, a tablet computer, and a smart watch.

[0030] Through the user-side power consumption abnormality monitoring information acquisition unit, based on the user-side power consumption abnormality status analysis results, the distribution line power consumption object abnormality status analysis results, the distribution line loss power consumption abnormality status assessment results combined with data processing, the user-side power consumption abnormality monitoring information is independently and efficiently constructed, so as to realize efficient and accurate collection of user-side power consumption abnormality monitoring information, and improve the efficiency of user-side power consumption abnormality monitoring result feedback; the user-side power consumption abnormality monitoring information feedback unit efficiently and timely pushes the user-side power consumption abnormality monitoring information to the user monitoring terminal through the Internet of Things communication network, realizes intuitive and visual feedback of the user-side power consumption abnormality monitoring information, and improves the applicability of the user-side power consumption abnormality monitoring system. Example

[0031] See also Figure 1 - Figure 2 , a user-side power consumption abnormality monitoring system based on machine learning, used to implement a user-side power consumption abnormality monitoring method, the system includes a user-side power consumption abnormality analysis module, a user-side power consumption abnormality object analysis module, and a user-side power consumption abnormality feedback module; The user-side power consumption abnormality analysis module includes a user-side distribution line instantaneous power consumption collection unit, a user-side distribution line abnormal instantaneous power consumption interval storage unit, and a user-side power consumption abnormality state analysis unit; The user-side distribution line instantaneous power consumption collection unit collects the user-side distribution line instantaneous power consumption data through the electric energy meter; the user-side distribution line abnormal instantaneous power consumption interval storage unit is used to store the user-side distribution line abnormal instantaneous power consumption interval; the user-side power consumption abnormal state analysis unit performs power consumption abnormal state analysis and processing on the user side according to the user-side distribution line instantaneous power consumption data and the user-side distribution line abnormal instantaneous power consumption interval, and generates user-side power consumption abnormal state analysis data; The user-side power consumption abnormal object analysis module includes a distribution line power consumption object instantaneous power consumption collection unit, a distribution line power consumption object instantaneous rated power consumption storage unit, a distribution line power consumption object power consumption abnormal state identification unit, a distribution line power consumption object identity information storage unit, a distribution line power consumption abnormal object identification unit, a distribution line loss power consumption metering unit, a distribution line normal loss power consumption interval storage unit, and a distribution line loss power consumption abnormal state evaluation unit; The instantaneous power consumption collection unit of the power consumption object of the distribution line collects the instantaneous power consumption data of the power consumption object of the distribution line through the electric energy meter; the instantaneous rated power consumption storage unit of the power consumption object of the distribution line is used to store the instantaneous rated power consumption data of the power consumption object of the distribution line; the power consumption abnormal state identification unit of the power consumption object of the distribution line performs abnormal power consumption state identification processing of the power consumption object in the user-side distribution line based on the instantaneous power consumption data of the power consumption object of the distribution line and the instantaneous rated power consumption data of the power consumption object of the distribution line, and constructs the power consumption abnormal state identification data of the power consumption object of the distribution line; the power consumption object identity information storage unit of the distribution line is used to store the identity text data of the power consumption object of the distribution line; the power consumption abnormal object identification unit of the distribution line The power consumption abnormal state identification data of the power consumption object is combined with the machine learning intelligent algorithm and the power consumption object identity text data of the distribution line to perform the power consumption object identity information identification processing of the abnormal power consumption state in the user-side distribution line, and generate the power consumption abnormal state object identity text data of the distribution line; the power consumption metering unit of the distribution line loss is used to measure the power consumption data of the distribution line loss; the power consumption interval storage unit of the distribution line normal loss is used to store the power consumption interval of the distribution line normal loss; the power consumption abnormal state evaluation unit of the distribution line loss is used to evaluate the abnormal power consumption state of the user-side distribution line based on the power consumption data of the distribution line loss and the normal power consumption interval of the distribution line loss, and generate the power consumption abnormal state evaluation data of the distribution line loss; The user-side power consumption abnormality feedback module includes a user-side power consumption abnormality monitoring information acquisition unit and a user-side power consumption abnormality monitoring information feedback unit; The user-side power consumption abnormality monitoring information acquisition unit is used to construct the user-side power consumption abnormality monitoring data; the user-side power consumption abnormality monitoring information feedback unit pushes the user-side power consumption abnormality monitoring data to the user monitoring terminal through the Internet of Things communication network to execute the user-side power consumption abnormality monitoring information feedback operation.

[0032] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring abnormal power consumption at the user side, characterized in that: The method comprises the following steps: S1, collect instantaneous power consumption data of distribution lines on the user side; S2, analyzing and processing the abnormal state of power consumption on the user side according to the instantaneous power consumption data of the user side distribution line and the abnormal instantaneous power consumption interval of the user side distribution line, generating abnormal state analysis data of power consumption on the user side, and ending the abnormal power consumption monitoring operation on the user side when it is normal; S3. When it is abnormal, collect instantaneous power consumption data of power users in the distribution line; S4, based on the instantaneous power consumption data of the power consumption objects in the distribution line and the instantaneous rated power consumption data of the power consumption objects in the distribution line on the user side, an abnormal power consumption state identification process is performed on the power consumption objects in the distribution line on the user side, and abnormal power consumption state identification data of the power consumption objects in the distribution line is constructed; S5, performing identification processing of the identity information of the power user in the abnormal power consumption state in the user-side distribution line according to the power user abnormal power consumption state identification data of the distribution line power user and the power user identity text data of the distribution line power user, and generating the power user identity text data of the distribution line abnormal power consumption; S6, measuring the power consumption data of the distribution line loss and performing abnormal power consumption state evaluation processing on the user-side distribution line loss with the normal power consumption interval of the distribution line loss, and generating abnormal power consumption state evaluation data of the distribution line loss; S7. Construct user-side power consumption abnormality monitoring data and execute user-side power consumption abnormality monitoring information feedback operation.

2. The method for monitoring abnormal power consumption at the user side according to claim 1, characterized in that: The S1 comprises the following steps: S11. Collect the instantaneous total power consumption data of the household distribution line on the user side online through the power meter, and generate the instantaneous power consumption data of the user side distribution line ,in The unit is kilowatt-second.

3. The method for monitoring abnormal power consumption at the user side according to claim 2, characterized in that: The S2 comprises the following steps: S21. Establish abnormal instantaneous power consumption interval of user-side distribution lines ,in Indicates the abnormal instantaneous power consumption range of the user-side distribution line Abnormal minimum instantaneous power consumption data of the distribution line on the user side, Indicates the abnormal instantaneous power consumption range of the user-side distribution line Abnormal maximum instantaneous power consumption data of the distribution line on the user side, and The unit of kW is kilowatt-second; S22, the With the As stated in , Compare the power consumption values ​​and generate abnormal power consumption status analysis data on the user side based on the power consumption value comparison results ; when ∈ When the abnormal state analysis data of power consumption at the user side is output is abnormal; when ∉ When the abnormal state analysis data of power consumption at the user side is output If the result is normal, the abnormal power consumption monitoring operation on the user side will be terminated directly.

4. The method for monitoring abnormal power consumption at the user side according to claim 3 is characterized in that: The S3 comprises the following steps: S31, when the When the power consumption is abnormal, the instantaneous power consumption data of the power users connected to the distribution line on the user side is collected online through the power meter, and the instantaneous power consumption data set of the power users of the distribution line is generated. , ;in Indicates the collected The instantaneous power consumption data of the power distribution line users corresponding to the power distribution line users. Indicates the maximum number of power consumption objects in the distribution line; The unit is kilowatt-second.

5. The method for monitoring abnormal power consumption at the user side according to claim 4 is characterized in that: The S4 comprises the following steps: S41. Establishing instantaneous rated power consumption data set of power users in distribution lines ,in Indicates The instantaneous rated power consumption data of the power consumption objects of the distribution lines corresponding to the power consumption objects of the distribution lines, The unit of is kilowatt-second; S42, using XGBoost algorithm to As stated in According to the number of the power users of the distribution line, they are arranged in order As stated in Compare the power consumption values ​​and build a data set for identifying abnormal power consumption status of power users in distribution lines based on the power consumption value comparison results. ,in Indicates Identification data of abnormal power consumption status of power consumption objects of distribution lines corresponding to power consumption objects of distribution lines; when > When the power consumption of the power consumption object of the distribution line is abnormal, the power consumption status identification data is output. is abnormal; when ≤ When the power consumption of the power consumption object of the distribution line is abnormal, the power consumption status identification data is output. is normal.

6. The method for monitoring abnormal power consumption at the user side according to claim 5, characterized in that: The S5 comprises the following steps: S51. Establishing a text data set of the identity of the power users of the distribution line ,in Indicates The power distribution line electricity user identity text data corresponding to each power distribution line electricity user; S52, the The abnormal power consumption status analysis result is abnormal With the As stated in Match the power consumption object number of the distribution line, and search for the abnormal power consumption status analysis result. The corresponding And construct a text data set of the identity of the abnormal power consumption objects of the distribution line , execute to generate the identity text data set of the abnormal power consumption object of the distribution line The specific steps are as follows: S521, dissemination: As stated in Synchronize to water waves, each of the The water wave is an independent individual, so each of the Water waves will have three properties: position, wavelength and wave height ; In each iteration, each Water waves will propagate through the The search space The abnormal power consumption state analysis result is abnormal Search and match the number of the power object of the distribution line, and at the same time will decrease by 1; S522, refraction: in a After the water wave propagates, the Water waves refract and The search space The abnormal power consumption state analysis result is abnormal Search and match the distribution line power object number, each time the transmission, the water wave height will decrease by 1 when When it decreases to 0, the The water waves will be refracted, and their height and wavelength will change; the position after refraction is normally distributed in the current Water waves and optimal The midpoint of the water wave is the mean value. Water waves and optimal The distance of the water wave is the position of the variance, after refraction The wave height will be reinitialized to the maximum wave height ; After refraction, recalculate the Wavelength of water waves ; S523, Breaking Waves: Water waves After propagating in the search space, the optimal The position of the water wave, that is, Search the search space for the abnormal power consumption state analysis result. Matching the ; then the The waves will break and the current optimal The water wave propagates to the location where the breaking waves are generated, that is, the output and the power consumption abnormal state analysis result are abnormal. Matching the location; S524: the output and power consumption abnormal state analysis result is abnormal Matching the ; S525, the output in step S524 After data identification, the identity text data set of the abnormal power consumption objects of the distribution line is constructed ,in ,in Indicates The identity text data of the distribution line power consumption abnormal object corresponding to the distribution line power consumption abnormal object, Indicates The identity text data of the distribution line power consumption abnormal object corresponding to the distribution line power consumption abnormal object.

7. The method for monitoring abnormal power consumption at the user side according to claim 6, characterized in that: The S6 comprises the following steps: S61. According to the As stated in Combined with the summation formula, the total instantaneous power consumption of all power users in the user-side distribution line is measured and processed to generate the total instantaneous power consumption data of the power users in the distribution line. ,in The unit of is kilowatt-second; S62, the With the Perform differential value processing on instantaneous power consumption data to measure power consumption data of power distribution line loss ,in The unit of is kilowatt-second; S63. Establish normal power consumption range of distribution lines ,in Indicates the normal power consumption range of the distribution line The minimum normal power loss data of the distribution line, Indicates the normal power consumption range of the distribution line The maximum normal power consumption data of the distribution line, and The unit of kW is kilowatt-second; S64, the With the As stated in , Compare the power loss values ​​and generate the distribution line power loss abnormal state assessment data based on the power loss values ​​comparison results ; when ∈ When the power loss and power consumption abnormal status evaluation data of the distribution line is output is normal; when ∉ When the power loss and power consumption abnormal status evaluation data of the distribution line is output is abnormal.

8. The method for monitoring abnormal power consumption at the user side according to claim 7, characterized in that: The S7 comprises the following steps: S71, the , , Construct abnormal power consumption monitoring data on the user side through data combination ,in ; S72, the The abnormal power consumption monitoring information on the user side is fed back through the IoT communication network to the user monitoring terminal.

9. A user-side power consumption abnormality monitoring system based on machine learning, used to implement the user-side power consumption abnormality monitoring method according to any one of claims 1 to 8, characterized in that: The system comprises a user-side power consumption anomaly analysis module, a user-side power consumption anomaly object analysis module, and a user-side power consumption anomaly feedback module.

Images