Method and system for detecting and analyzing abnormal electric quantity of user

Abstract

The invention discloses a user electric quantity abnormity detection analysis method and system, and relates to the technical field of electric quantity monitoring and detection, and the method comprises the steps: obtaining an electric quantity detection value of a target user in a preset period, dividing the period into a plurality of stages, carrying out the stage sampling detection, and obtaining the electric quantity of a first stage and the electric quantity of a second stage. A target reference electric quantity is set according to the two electric quantities. When the absolute value of the deviation between the target detection electric quantity and the reference electric quantity does not meet the preset threshold value, the system sends out an abnormal early warning signal, and a preset electric quantity checking strategy is called according to the signal to check the electricity consumption of the user. The technical problem that single reason troubleshooting is insufficient in existing electric quantity anomaly detection is solved, and the technical effect of improving the electric quantity anomaly detection accuracy and efficiency is achieved by recognizing and solving electric quantity anomaly caused by user behaviors, hardware faults or software problems.

Description

Technical Field

[0001] The present application relates to the technical field of power monitoring and detection, and in particular to a method and system for detecting and analyzing abnormal power consumption of a user. Background Art

[0002] With the rapid development of smart grids and renewable energy, users' electricity consumption patterns have become more complex. Traditional power monitoring methods have difficulty accurately identifying the specific causes of power anomalies, which not only affects users' power consumption experience, but also threatens the stability of power supply. Power anomalies may be caused by changes in users' power consumption behavior, hardware equipment failures, or errors in power metering software. Therefore, there is an urgent need for an efficient power anomaly detection and analysis method that comprehensively considers multiple factors, accurately locates the causes of anomalies, and provides corresponding verification and processing strategies. This method can improve the level of intelligence in power management and ensure the safe and stable operation of the power system.

[0003] In the current related technologies, there is a technical problem of insufficient single cause investigation in abnormal power detection. Summary of the invention

[0004] The present application provides a method and system for detecting and analyzing user power anomalies. By obtaining the power consumption detection value of the target user within a predetermined period, the period is divided into multiple stages, and stage sampling detection is performed to obtain the power consumption of the first stage and the second stage respectively. The target reference power is set according to these two power quantities. When the absolute value of the deviation between the target detection power and the reference power does not meet the predetermined threshold, the system sends an abnormal warning signal, and calls the predetermined power verification strategy based on the signal to verify the user's power consumption, thereby achieving the technical effect of improving the accuracy and efficiency of power anomaly detection by identifying and resolving power anomalies caused by user behavior, hardware failures or software problems.

[0005] This application provides a method for detecting and analyzing abnormal power consumption of users, including:

[0006] Obtain a target detection power, where the target detection power refers to a detection value of the power consumption of a target user in a predetermined period; divide the predetermined period into multiple stages, where the multiple stages include a first stage and a second stage; introduce a stage detection logic to perform stage sampling detection on the power consumption of the target user in the first stage and the second stage, and obtain a first power consumption in the first stage and a second power consumption in the second stage respectively; set a target baseline power consumption of the target user according to the first power consumption and the second power consumption; issue an abnormal warning signal when the absolute value of the power consumption deviation between the target detection power consumption and the target baseline power consumption does not meet a predetermined deviation threshold; and call a predetermined power consumption verification strategy according to the abnormal warning signal to verify the power consumption of the target user.

[0007] This application also provides a user power anomaly detection and analysis system, including:

[0008] A target detection power acquisition module, the target detection power acquisition module is used to obtain the target detection power, the target detection power refers to the detection value of the power consumption of the target user in a predetermined period; a predetermined period division module, the predetermined period division module is used to divide the predetermined period into multiple stages, the multiple stages include a first stage and a second stage; a stage sampling detection module, the stage sampling detection module is used to introduce a stage detection logic to perform stage sampling detection on the power consumption of the target user in the first stage and the second stage, and obtain the first power of the first stage and the second power of the second stage respectively; a target reference power setting module, the target reference power setting module is used to set the target reference power of the target user according to the first power and the second power; an abnormal warning signal issuing module, the abnormal warning signal issuing module is used to issue an abnormal warning signal when the absolute value of the power deviation between the target detection power and the target reference power does not meet the predetermined deviation threshold; a power consumption verification module, the power consumption verification module is used to call a predetermined power verification strategy according to the abnormal warning signal to verify the power consumption of the target user.

[0009] The user power anomaly detection and analysis method and system proposed in this application first obtains the target user's power consumption detection value within a predetermined period, divides the period into multiple stages, and performs stage sampling detection to obtain the power consumption of the first stage and the second stage respectively. The target reference power is set based on these two power quantities. When the absolute value of the deviation between the target detection power and the reference power does not meet the predetermined threshold, the system sends an abnormal warning signal, and calls the predetermined power verification strategy based on the signal to verify the user's power consumption. By identifying and resolving power anomalies caused by user behavior, hardware failures or software problems, the technical effect of improving the accuracy and efficiency of power anomaly detection is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0011] Figure 1 A flowchart of a method for detecting and analyzing abnormal power consumption of a user provided in an embodiment of the present application;

[0012] Figure 2A schematic diagram of the structure of a user power anomaly detection and analysis system provided in an embodiment of the present application.

[0013] Explanation of the reference numerals: target detection power acquisition module 10 , predetermined period division module 20 , stage sampling detection module 30 , target reference power setting module 40 , abnormal warning signal issuing module 50 , power consumption verification module 60 . DETAILED DESCRIPTION

[0014] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0015] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0016] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0017] The present application embodiment provides a method for detecting and analyzing abnormal power consumption of a user, such as Figure 1 As shown, the method includes:

[0018] Step S100, obtaining the target detection power, which refers to the detection value of the target user's power consumption in a predetermined period. Specifically, in the user power anomaly detection and analysis scheme, the determination of the predetermined period takes into account a variety of factors, such as residential users often use a month as a cycle to match electricity bill settlement and reflect seasonal changes in electricity consumption, commercial users choose a week or a month according to business characteristics, industrial users determine it based on production plans and equipment maintenance cycles, and the power supply department also considers management and statistical needs; after the cycle is determined, the power management system is set and managed, such as creating a power consumption monitoring record for the user at the beginning of the cycle, organizing and storing data at the end, and setting a reminder function. The acquisition of target power detection relies on metering equipment such as smart meters. Smart meters sample current and voltage through mutual inductors and convert them into digital signal processing. They can also be equipped with microprocessors and memory for data calculation, storage and summary recording at time intervals. The intelligent power monitoring system with new Internet of Things technology can realize refined monitoring of equipment. The frequency of data collection is determined based on the user type. Residential users generally collect data every 15 minutes or 30 minutes, and commercial and industrial users may collect data more frequently. At the same time, the data accuracy is guaranteed to reach a high level, and verification and error correction technology is used to ensure data accuracy. The collected data is transmitted to the power management system through power line carrier communication, wireless communication or Ethernet, and stored in the database. A power consumption record sheet is established for each user and relevant information is recorded. Backup and recovery measures are also taken to ensure data security and reliability, thereby accurately obtaining the power consumption of the target user within the predetermined period, providing a data basis for subsequent analysis.

[0019] Step S200, dividing the predetermined period into a plurality of stages, the plurality of stages including a first stage and a second stage. Specifically, in the user power anomaly detection and analysis scheme, the predetermined period is divided into a plurality of stages in order to analyze the power consumption in more detail and meet the needs of data statistical analysis. From the perspective of power consumption patterns, residential users' power consumption is affected by daily cycles, seasons, and daily activities, commercial users are related to business hours and business types, and industrial users depend on production processes, etc. Based on this, the stages can be divided according to time, power consumption behavior characteristics, and external factors. The time dimension can be divided by day, week, month or finer time periods, such as dividing a day into the morning rush hour, morning and other stages; in terms of electricity consumption behavior characteristics, it can be divided according to residents' specific habits, commercial business activities and industrial production processes, such as residents' weekends and weekdays, restaurant lunch and dinner times, different stages of industrial production, etc.; when combined with external factors, seasons, weather, holidays, etc. are considered, such as summer and winter, high temperature and normal temperature, holidays and non-holidays, and weather-affected industries such as agriculture are divided according to rainfall and temperature, so as to reasonably divide the first stage and the second stage into multiple stages, providing a basis and framework for power analysis and anomaly detection, so as to accurately understand the power consumption situation and discover problems.

[0020] Step S300, introduce the stage detection logic to perform stage sampling detection on the target user's power consumption in the first stage and the second stage, and obtain the first power consumption in the first stage and the second power consumption in the second stage respectively. Specifically, in the user power anomaly detection and analysis scheme, the design of the stage detection logic is based on clarifying the purpose and focus of each stage detection, such as focusing on high-load power consumption during peak power consumption, focusing on the stability of basic power consumption during the valley period, etc. The detection methods include real-time monitoring and sampling combined with equal time intervals or adaptive sampling and event-driven sampling, etc., which are selected according to the characteristics of the power system, the user's power consumption behavior pattern and cost efficiency. At the same time, there are algorithms and rules, such as judging abnormal fluctuations by calculating the historical average and standard deviation, data filtering and preprocessing, and correlation analysis of abnormal situations. In the first stage, according to the characteristics of its peak power consumption period, typical time periods such as morning, noon, and afternoon or key time points of industrial production are selected for sampling. The collected data is validated and aggregated and averaged to obtain the first power, and the reasons for the difference are compared with historical data for analysis. In the second stage, the detection strategy is adjusted according to the power consumption situation. For example, the frequency is reduced during non-peak hours at night but attention is paid to standby energy consumption. Sampling is performed at corresponding time points. The data is processed in the same way to obtain the second power consumption and compared with the historical data to analyze the cause of the abnormal change. The first and second power consumption are obtained through the stage detection logic to provide basic data for subsequent analysis to understand power consumption behavior and discover potential problems.

[0021] Step S400, the target baseline electricity quantity of the target user is set according to the first electricity quantity and the second electricity quantity. Specifically, when setting the target baseline electricity quantity in the user electricity anomaly detection and analysis scheme, the first electricity quantity and the second electricity quantity must first be preprocessed and analyzed. Verify data accuracy, including checking equipment operation, data transmission and format, etc., to ensure data reliability; analyze data characteristics, calculate statistical indicators such as mean, median, standard deviation, draw electricity change curves, and analyze the relationship between two-stage electricity quantities and change trends. The setting methods include weighted average method, which assigns weights to the first and second electricity quantities according to the importance of the stage or the historical electricity consumption ratio, etc., to calculate the target baseline electricity quantity; trend analysis method based on historical data, which uses time series analysis method to model and predict historical data, and adjusts the baseline electricity quantity considering the electricity consumption trend; empirical method combining user type and electricity consumption scenario, which is set according to the characteristics of different user types and industry experience standards. Adjustments and optimizations are then carried out, with real-time feedback adjustments, such as timely updates based on the deviation between electricity consumption and benchmark electricity consumption and the reasons, and the establishment of a dynamic adjustment mechanism; periodic evaluation and optimization, quarterly or annual review and analysis, using the accuracy of electricity anomaly detection as indicators, and considering factors such as market changes and changes in electricity consumption behavior to optimize the target benchmark electricity consumption, thereby providing a reliable reference standard for accurate detection of electricity anomalies.

[0022] In a possible implementation, the target baseline power of the target user is set according to the first power and the second power, and step S400 further includes step S410, counting the number of the multiple stages and recording it as the total number of stages. Specifically, the predetermined period is divided into multiple stages in the previous step. The stages include the first stage, the second stage and possibly other more stages, each of which represents a different time interval or power consumption characteristic interval within the predetermined period. For example, the predetermined period is one month, and it may be divided into a stage every ten days, so there are three stages. The number of stages is counted and recorded as the total number of stages. By counting the divided stages one by one, for example, if the predetermined period is divided by day, there are 30 days in a month, and each day is a stage, then the total number of stages is 30; if it is divided by week, there are about 4 weeks in a month (there may be discrepancies, determined according to the specific number of days in the month and the start date), and the total number of stages is 4. After the statistics are completed, the total number of stages will be used as an important parameter for subsequent calculations, which reflects the degree of detail of the division of the predetermined period.

[0023] Step S420, according to the preset detection ratio set in the stage detection logic, the number of stage detections is obtained in combination with the total number of stages. Specifically, a preset detection ratio is set in the stage detection logic, such as a preset detection ratio of 5%. The setting of the ratio is based on a comprehensive consideration of the accuracy and efficiency of power detection. A lower detection ratio may reduce the workload of data collection and processing, but may lead to an incomplete understanding of the power consumption situation; a higher detection ratio can grasp the power consumption situation in more detail, but it will increase the burden and cost of the system. A 5% preset detection ratio means that in all stages, 5% of the stages will be selected for more detailed detection and analysis, and the number of stage detections will be calculated based on the total number of stages and the preset detection ratio. The calculation method is to multiply the total number of stages by the preset detection ratio. For example, if the total number of stages is 30 and the predetermined detection ratio is 5% (i.e. 0.05), then the number of stage detections = 30 × 0.05 = 1.5. Since the number of stage detections usually needs to be an integer, it can be rounded up according to actual conditions. For example, it can be rounded up to 2, which means that 2 stages will be selected from the 30 stages for more detailed detection. The number of stage detections will determine the number of data samples used in the subsequent calculation of the average stage power.

[0024] Step S430, combining the first power, the second power and the stage detection number, calculate the average stage power. Specifically, the first power and the second power have been obtained, which represent the power consumption of the first stage and the second stage respectively. At the same time, according to the stage detection number determined previously, the number of stages that need to participate in the average calculation is known, and the first power, the second power and the power of other related stages (if there are more stages) are averaged together. Assuming that in addition to the first and second stages, there are other stages with power of the third power, the fourth power, etc. (until the number of stages determined by the stage detection number is reached), the average stage power is calculated by adding these power and then dividing it by the stage detection number. For example, if the stage detection number is 3 (including the first, second and another stage), the power is the first power X1, the second power X2, and the third power X3, then the average stage power = (X1 + X2 + X3) / 3. Through this calculation, an indicator reflecting the average power consumption of each detection stage is obtained, which can be used as an important reference value for evaluating the overall power consumption of users.

[0025] Step S440, the product of the total number of stages and the average stage power is adjusted for tolerance to obtain the target reference power. Specifically, the product of the total number of stages and the average stage power is adjusted for tolerance. The purpose of the tolerance adjustment is to take into account various uncertain factors in the actual power consumption process, such as occasional fluctuations of power-consuming equipment, measurement errors, slight changes in user power consumption habits, etc. By setting a certain tolerance range, the target reference power can be more in line with the actual situation and avoid overly strict or loose judgment standards. For example, the tolerance range can be set to ±5%, which means that on the calculated basic value, it can be floated up and down by 5% to determine the final target reference power range, and the product of the total number of stages and the average stage power is calculated to obtain a preliminary power value. Then adjust according to the set tolerance range. If the tolerance range is ±5%, the lower limit of the target reference power is the product value multiplied by (1-0.05), and the upper limit is the product value multiplied by (1+0.05). For example, the total number of stages is 30, the average stage power is 10 kWh, and the product is 30×10=300 kWh. Then the range of the target benchmark power is [300×(1-0.05),300×(1+0.05)]=[285,315] kWh. This range is the final target benchmark power. When the user's actual power consumption is within this range, it can be considered relatively normal; if it exceeds this range, further power anomaly detection and analysis may be required. By counting the number of stages, determining the number of stage detections, calculating the average stage power, and adjusting the tolerance, the target benchmark power is reasonably obtained, which provides a clear reference standard for user power anomaly detection and analysis, and helps to accurately determine whether the user's power consumption is abnormal.

[0026] Step S500, when the absolute value of the power deviation between the target detection power and the target reference power does not meet the predetermined deviation threshold, an abnormal warning signal is issued. Specifically, in the user power abnormality detection and analysis scheme, the target detection power and the target reference power are first obtained. The target detection power is the actual power consumption recorded in real-time monitoring, and the target reference power is the expected normal power consumption range obtained by integrating multiple factors. Then calculate their power deviation absolute value, that is, | target detection power - target reference power |. The setting of the predetermined deviation threshold comprehensively considers factors such as the user's power consumption type (residential, commercial, and industrial power consumption rules are different), historical data statistical analysis results (power consumption fluctuation size), and power system stability and power supply cost. For example, ordinary residential users may set it to ±8%. Compare the absolute value of the power deviation with the predetermined deviation threshold. If it does not meet (that is, | power deviation | > predetermined deviation threshold), it is determined that the power is abnormal. At this time, the system sends an abnormal warning signal through a monitoring interface warning message, a text message or an email to relevant personnel, sets an audible and visual alarm device, and records the abnormal information in the database, providing a trigger mechanism for subsequent verification and problem solving, and ensuring the normal operation of the power system and the reasonable power consumption of users.

[0027] Step S600, according to the abnormal warning signal, the predetermined power verification strategy is called to verify the power consumption of the target user. Specifically, in the user power anomaly detection and analysis scheme, the system receives the abnormal warning signal through a special signal monitoring module. The module is connected to the power anomaly detection module, and after receiving the signal through the communication interface and other means, the relevant information is recorded in the log file and a preliminary analysis is performed, such as judging the severity of the anomaly according to the power deviation. Then, according to the analysis results, a suitable strategy is selected from the predetermined power verification strategy library for calling. The strategy library is divided into user behavior, hardware and software verification strategies according to the verification object. If the preliminary analysis shows that the user behavior may be abnormal, the user behavior verification strategy is called, the historical power consumption behavior log is obtained to establish a baseline, and the actual power consumption behavior in the abnormal warning time period is compared to calculate the behavior anomaly index. If the threshold is reached, a confirmation signal is sent to the user; if the user behavior is normal or suspected of hardware problems, the hardware verification strategy is called to collect the operating characteristics of distributed power equipment, analyze the operating signals of the center and edge devices to determine whether there is a hardware failure; if the hardware is normal, the software verification strategy is called to check the power metering software version, operating status and database records, and if there is a problem, update or data repair operations are performed. Through this process, abnormal power consumption can be investigated and analyzed to find out the cause and ensure the stable operation of the power system.

[0028] In a possible implementation, the predetermined power verification strategy is called according to the abnormal warning signal to verify the power consumption of the target user. Step S600 further includes step S610, extracting the user behavior verification strategy in the predetermined power verification strategy. Specifically, the predetermined power verification strategy library contains multiple strategies, which are stored and managed in a structured manner. The user behavior verification strategy has a clear identification and position in the library. The library organizes the strategy in the form of a directory structure or a database table. Each strategy has a corresponding index or keyword for quick search and extraction. When it is necessary to extract the user behavior verification strategy, the system performs the extraction operation through the corresponding interface or function, which involves the access permission verification of the strategy library to ensure that only authorized modules or programs can extract the strategy. Once the verification is passed, the system uses a suitable reading method to extract the user behavior verification strategy according to the storage location and format of the strategy. For example, if the strategy is stored in the form of a text file, the system will use a file reading function to open the file and read the content therein; if it is stored in a database, the corresponding database query statement will be executed to obtain the strategy data. The extracted strategy is usually loaded in the form of a data structure or object in the memory for subsequent processing and application.

[0029] Step S620, analyze the historical electricity usage behavior log of the target user to generate a target electricity usage behavior baseline. Specifically, the historical electricity usage behavior log of the target user is stored in the database of the power management system or a special log storage system. The log contains rich information, such as electricity consumption in different time periods of the day, the start and stop time of electrical equipment, electricity consumption on special dates, etc. First, it is necessary to obtain relevant historical electricity usage behavior log data from the storage system. This may involve database query operations, filtering out the required log records according to the target user's identification and time range, and organizing and preprocessing the acquired log data, such as removing duplicate records, correcting erroneous data (such as obvious electricity consumption anomalies may be caused by data collection errors and need to be corrected or eliminated), and sorting the data in chronological order for subsequent analysis, using data analysis and statistical methods to generate a target electricity usage behavior baseline. A common method is to calculate the mean and standard deviation of historical electricity consumption data. For example, for each time period (such as morning, afternoon, and evening of each day), the mean value of electricity consumption in the time period is calculated as the central value of the baseline, and the standard deviation is calculated to reflect the fluctuation range of electricity consumption. The frequency of use and time distribution of electrical equipment are analyzed to determine the electrical equipment and power consumption characteristics that users usually use in different time periods. For example, through analysis, it is found that users usually turn on the TV and air conditioner at night, and the power consumption of these two devices accounts for a certain proportion of the total power consumption at night. These characteristics will also be included in the target electricity consumption behavior baseline, considering the impact of seasonal factors on electricity consumption behavior, and generating electricity consumption behavior baselines for different seasons. For example, in summer, due to the frequent use of air conditioners, the electricity consumption baseline will be adjusted accordingly; in winter, there may be different baseline characteristics due to the use of heating equipment.

[0030] Step S630, according to the user behavior verification strategy, the actual power consumption behavior of the target user in the predetermined period is compared with the target power consumption behavior baseline to obtain the behavior abnormality index. Specifically, within the predetermined period, the actual power consumption behavior data of the target user is collected in real time. The process of obtaining the target detection power is similar. The power consumption is monitored in real time through devices such as smart meters, and information such as the opening and closing time of the power-consuming equipment is recorded. The collected actual power consumption behavior data needs to be processed and analyzed in a timely manner. First, the data is sorted in the same format and time granularity as the historical power consumption behavior log so as to be compared with the baseline. For example, if the baseline is to count the power consumption in hours, then the actual power consumption behavior data also needs to be summarized and analyzed by hours. Then, the actual power consumption data is preliminarily checked for rationality to ensure the accuracy and reliability of the data. For example, if it is found that the power consumption in a certain hour is negative or exceeds the power consumption range that can be generated by the rated power of the device, the data needs to be further verified or corrected. According to the user behavior verification strategy, the sorted actual power consumption behavior data is compared with the target power consumption behavior baseline. The comparison includes the amount of electricity consumption, the distribution of electricity consumption time, and the use of electrical equipment. For the comparison of electricity consumption, the difference or ratio between the actual electricity consumption and the baseline electricity consumption can be calculated. For example, for a certain time period, the actual electricity consumption is X and the baseline electricity consumption is Y, and the difference |XY| or ratio can be calculated. For the comparison of power consumption time distribution, the degree of deviation between the actual power consumption equipment on and off time and the baseline time can be analyzed. For example, if the TV is usually turned on between 7 and 10 p.m. in the baseline, but it is actually turned on at 6 p.m. on a certain day, record this time deviation. Taking these factors into consideration, calculate the behavior anomaly index through a certain algorithm. This index can be a comprehensive value that reflects the degree of deviation of the actual power consumption behavior from the baseline. For example, the power consumption difference, time deviation and other factors can be weighted and summed to obtain a value between 0 and 1. The larger the value, the higher the degree of anomaly.

[0031] Step S640: If the behavior abnormality index reaches a predetermined abnormality threshold, a power usage behavior confirmation signal is sent to the target user to obtain a target behavior verification result. Specifically, a predetermined abnormal threshold of the behavior abnormality index is set in advance. The threshold is determined based on experience and analysis of the user's electricity usage behavior. Generally speaking, if the behavior abnormality index exceeds the threshold, it is considered that there may be abnormal electricity usage behavior that needs further confirmation. The setting of the threshold needs to balance the false alarm rate and the missed alarm rate. If the threshold is set too high, some real abnormal behaviors may be ignored (missed alarm); if the threshold is set too low, confirmation signals may be sent to users frequently, causing unnecessary interference (false alarm). For example, after analysis and practical verification of a large number of user electricity usage data, it is determined that when the behavior abnormality index exceeds 0.3, it is the predetermined abnormal threshold. When the calculated behavior abnormality index reaches or exceeds this predetermined abnormal threshold, the system will trigger the operation of sending an electricity usage behavior confirmation signal to the target user. Sending an electricity usage behavior confirmation signal to the target user can be achieved in a variety of ways, such as text messages, mobile phone application push notifications or emails, etc. The signal will contain some basic information, such as the time period when the abnormality occurred, the initial detection time, and the time of the initial detection. After receiving the signal, the user can reply according to his actual power consumption. For example, the user can confirm whether there are new power consumption equipment, whether special power consumption activities have been carried out, or whether there are other situations that may cause abnormal power consumption. The system obtains the target behavior verification result according to the user's reply. If the user confirms that there is abnormal power consumption behavior, the specific situation can be further understood to provide more accurate information for subsequent analysis and processing; if the user denies the existence of abnormal behavior, other reasons may need to be further investigated, such as hardware equipment failure or software problems. By extracting user behavior verification strategies, analyzing historical power consumption behavior logs to generate baselines, comparing actual behaviors with baselines to calculate abnormal indexes, and sending confirmation signals to obtain verification results, the power consumption behavior of target users can be effectively verified, possible abnormal situations can be discovered and confirmed in time, and important user behavior basis can be provided for power anomaly detection and analysis.

[0032] In one possible implementation, if the behavior abnormality index reaches a predetermined abnormality threshold, a power usage behavior confirmation signal is sent to the target user to obtain a target behavior verification result. Step S640 further includes step S641. Based on the target behavior verification result, if the power usage behavior of the target user is normal, the hardware verification strategy in the predetermined power verification strategy is extracted. Specifically, the target behavior verification results are analyzed. If both user feedback and system analysis indicate that the electricity usage behavior is consistent with the behavioral baseline and there is no abnormal change, then it is determined that the target user's electricity usage behavior is normal. For example, the user clearly states that no new electricity-consuming equipment has been added and that daily electricity usage habits have not changed. At the same time, the system compares its actual electricity usage behavior with the baseline and all indicators are within the normal range. The system turns its attention to the hardware and prepares to extract the hardware verification strategy. After excluding the electricity usage behavior factor, the power anomaly may be caused by hardware equipment problems. The hardware verification strategy has a specific storage location and identification in the predetermined power verification strategy library. The system accurately extracts the strategy through the corresponding interface or program based on the structure and index of the library. For example, the system finds the hardware verification strategy-related files or data records from the classification directory of the strategy library and loads them into the memory to prepare for the subsequent verification of distributed power equipment.

[0033] Step S642: collecting operation characteristics of the distributed power equipment corresponding to the target user according to the hardware verification strategy to obtain target equipment operation information. Specifically, based on the target user information, the distributed power equipment related to it is determined, such as electricity meters, wires, transformers and various types of power-consuming equipment at the user end, and the system establishes connections with these devices through communication interfaces or sensors. For example, for smart meters, data reading is realized by using their communication protocols; for other equipment, corresponding sensors are installed to collect operation characteristics. Taking electricity meters as an example, power consumption, current, voltage and other data are obtained through power line carrier communication and other technologies; for transformers, relevant information is collected with the help of temperature, oil level sensors and voltage and current monitoring equipment. The hardware verification strategy clarifies the scope of operation characteristic collection, including electrical parameters, temperature, humidity, operating time, equipment status, etc. The collection method varies from device to device. For example, in addition to reading electricity data, the electricity meter also checks the clock and communication module; the transformer monitors temperature, oil level, output voltage and current, etc. The collected data is sorted and stored, and can be stored in real time in the memory or recorded in the database for subsequent analysis and processing. For example, the real-time collected equipment operation data is stored in chronological order to form an equipment operation data sequence, which provides a basis for subsequent analysis.

[0034] Step S643, extracting the central device operation information in the target device operation information, wherein the central device operation information includes the first operation signal of the first central device. Specifically, in distributed power equipment, central devices such as transformers play a key role, and their operation status affects the stability of the power system. The system identifies the central device from the target device operation information according to the preset identifier, and extracts its operation information, including the first operation signal of the first central device. For example, the main transformer in a certain area is determined as the first central device, and its voltage output, current input and output, temperature and equipment operation status and other signals are obtained. These signals reflect the core operation parameters of the transformer. If the first central device is an electric meter, an oscilloscope can be used to check its output signal waveform. The oscilloscope can intuitively display the waveform characteristics, which is used to determine whether the electric meter is working normally. For example, by observing whether the waveform is distorted, whether the frequency is stable, and whether the amplitude is within the normal range, the performance of the electric meter is evaluated. The system collects and preliminarily processes other types of first operation signals, such as real-time monitoring of temperature signals. If the temperature is abnormal, it indicates that there may be a problem, which provides a basis for determining whether the device meets the first predetermined operation constraint.

[0035] Step S644: when the first operation signal meets the first predetermined operation constraint of the first central device, extract the edge device operation information in the target device operation information, the edge device operation information includes the second operation signal of the first edge device. Specifically, the first predetermined operating constraint is a standard range set based on the technical specifications of the central equipment and historical experience. After obtaining the first operating signal of the first central equipment, the system compares it with the first predetermined operating constraint. For example, for a transformer, if the output voltage fluctuation is within the allowable range, the oil temperature is normal, and the load rate is reasonable, it is determined that the operation is normal. Only when the first operating signal meets the first predetermined operating constraint, the system will further extract the edge device operating information to ensure that under the premise that the central equipment is normal, possible hardware problems can be more accurately detected. Edge devices include user-end electrical equipment and small equipment on branch lines. The system extracts the operating information of edge devices from the target device operating information, such as the second operating signal of the first edge device, including current consumption, voltage value, working status, power factor and other parameters. For different edge devices, corresponding methods are used to obtain operating signals, such as obtaining the current and operating status information of household air conditioners through smart sockets, and installing transformers in branch line distribution boxes to obtain line current and voltage signals, providing data support for subsequent analysis of the operating status of edge devices.

[0036] Step S645: Analyze the second operation signal to obtain a target hardware verification result. Specifically, the second operation signal of the first edge device is analyzed and compared with the normal operation parameters. For example, the normal operation current range is calculated according to the rated power of the air conditioner. If the actual monitored current deviates from the range, it may indicate that the air conditioner is faulty. The operation time and frequency of the equipment are analyzed. For example, if a certain equipment is frequently started or runs for too long during an abnormal period of time, it may be a fault in the equipment control circuit or affected by external factors. A trend analysis algorithm is used to observe the change trend of the operation signal to predict whether the equipment may fail. For example, if the branch line current is gradually increased and exceeds the normal fluctuation range, it may indicate problems such as line aging, leakage or overload. The target hardware verification result is determined based on the analysis result. If the edge device operation signal is found to be abnormal, the abnormal type, degree and related equipment information are recorded to provide a basis for subsequent fault diagnosis and maintenance. If all edge device operation signals are normal, other factors still need to be considered comprehensively, such as the long-term operation trend of the central equipment and the stability of the overall power system, and a comprehensive assessment is conducted to ensure the reliable operation of the power system. When there is no abnormality in the power consumption behavior, the hardware verification strategy is fully implemented, from the collection of equipment operation characteristics to the analysis of central and edge device information, hardware problems are accurately checked, and accurate hardware judgment is provided for power anomaly detection to ensure the normal operation of the power system.

[0037] In a possible implementation, the second operation signal is analyzed to obtain a target hardware verification result, and step S645 further includes step S6451, according to the target hardware verification result, if the distributed power device is normal, then the software verification strategy in the predetermined power verification strategy is extracted. Specifically, the target hardware verification result is analyzed, and if no abnormality is found in the hardware device through the analysis of the operation information of the central device and the edge device, it is determined that the distributed power device is normal, for example, the various operation parameters of the central device meet the predetermined operation constraints, and the operation signal of the edge device is also within the normal range, and no signs of hardware failure or abnormality are shown. At this point, since problems with electricity usage behavior and hardware equipment have been ruled out, the possible cause of the abnormal electricity consumption turns to the software level. The system prepares to extract the software verification strategy in the predetermined electricity consumption verification strategy. Based on a systematic troubleshooting idea, when both electricity usage behavior and hardware equipment are judged to be normal, software problems become the focus of attention, because the software may have vulnerabilities, errors or outdated versions, which affect the accuracy of electricity metering and the correctness of data processing. The software verification strategy has its specific storage location and identification in the predetermined electricity consumption verification strategy library. The system accurately extracts the software verification strategy through the corresponding program interface or module according to the organization method and indexing mechanism of the strategy library. For example, the system finds the corresponding software verification strategy file or data record from the software policy classification of the strategy library and loads it into the memory to prepare for the subsequent verification and processing of the electricity metering software. It is necessary to ensure the accuracy and completeness of the extraction to ensure that the software verification strategy can be correctly applied to the electricity metering software corresponding to the target user.

[0038] Step S6452: remotely update and upgrade the power metering software corresponding to the target user according to the software verification strategy. Specifically, according to the information of the target user, the system determines the electricity metering software related to the target user, which involves the query of the associated records of the user's metering equipment and software in the power management system. For example, the corresponding electricity metering software identification and communication address are found through the user's meter number or account information. The system establishes a remote connection with the software through network communication technology. If it is a smart meter system based on the Internet of Things, it may communicate with the electricity metering software through a wireless network (such as GPRS, Wi-Fi, etc.) or power line carrier communication; if it is in the power management system within the enterprise, it may be connected through a local area network or a dedicated communication line. According to the requirements of the software verification strategy, the electricity metering software is remotely updated and upgraded. First, the system checks the version information of the current software, obtains the latest version list by communicating with the software version management server, or directly obtains the version number from the electricity metering software itself, and compares the current version with the latest version. If it is found that the current version is not the latest version, the system will download the latest software installation package or update file. The download process may need to consider network bandwidth and stability, and use breakpoint resume and other technologies to ensure the complete download of the file. After the download is completed, the system starts the software update and upgrade program through a remote control command, and the new The software version is installed on the target device. During the installation process, some compatibility checks and data backup operations may be performed to prevent data loss or software failure during the update process. For example, the power metering data and software configuration files are backed up before the update, and data recovery and software configuration are reset after the update is completed. At the same time, the system will monitor the status of the entire update and upgrade process and provide timely feedback on whether there are errors or abnormalities. If there are problems, the system will handle them accordingly according to the preset error handling mechanism, such as rolling back to the last stable version or retrying the update. After the update and upgrade is completed, the system will perform some basic functional tests on the power metering software to ensure that the software can operate normally and accurately measure power. For example, some simulated power data is sent for testing to check whether the software's metering results are correct, and whether the software's communication function, data storage function, etc. are normal. Through testing, the effect of the software update and upgrade is verified to ensure that the accuracy and reliability of power metering are improved. When there is no abnormality in the distributed power equipment, the system will conduct a comprehensive check and remote update and upgrade of the power metering software corresponding to the target user to eliminate the impact of software problems on power detection and ensure the normal operation and data accuracy of the entire power detection and analysis system.

[0039] In a possible implementation, when the first operation signal meets the first predetermined operation constraint of the first central device, the edge device operation information in the target device operation information is extracted, and the edge device operation information includes the second operation signal of the first edge device. Step S644 further includes step S6441, analyzing the first operation signal to generate a first operation spectrum. Specifically, firstly, the first operation signal of the first central device is obtained, and the voltage, current or other related electrical signals collected in real time by the sensor are preprocessed, including noise removal, filtering and other operations to improve the quality and reliability of the signal. For example, a digital filter is used to filter out high-frequency noise in the signal to make the signal smoother. Then, the preprocessed signal is subjected to fast Fourier transform (FFT) or other spectrum analysis methods to convert the first operation signal in the time domain into the frequency domain to obtain a first operation spectrum. The spectrum can show the distribution of the signal on different frequency components and help analyze the operation status of the device. For example, the spectrum can be used to find out whether there is interference or harmonic components of a specific frequency, which may be related to the failure or abnormal operation of the device.

[0040] Step S6442, spline fitting is performed on the first running spectrum based on the principle of spline curves to obtain a first spline curve. Specifically, the spline curve has good local controllability and smoothness, and is very suitable for fitting complex data. Based on the principle of spline curves, a suitable spline function type (such as a cubic spline function) is selected, and the node distribution of the spline curve is determined. The selection of nodes can be determined according to the characteristics of the spectrum and the analysis requirements. Generally, nodes are set at key frequency points of the spectrum or places where data changes greatly. Through the least squares method or other fitting algorithms, the spline curve is made as close as possible to the data points of the first running spectrum. During the fitting process, the parameters of the spline curve are continuously adjusted to achieve the best fitting effect. For example, by optimizing the control point coordinates of the spline curve, the error between the fitting curve and the spectrum data is minimized. The first spline curve obtained by fitting can more concisely describe the main features of the spectrum while retaining the local details of the data.

[0041] Step S6443, obtain the first fitting polynomial of the first spline curve, and predict the target fitting operation value at the target time through the first fitting polynomial. Specifically, extract the corresponding first fitting polynomial expression from the first spline curve. The polynomial can accurately describe the mathematical relationship of the spline curve. For example, for a cubic spline curve, a fitting expression in the form of a cubic polynomial may be obtained. The first fitting polynomial is used to predict the target fitting operation value at the target time according to the current time and the operation trend of the equipment. The target time needs to be substituted into the fitting polynomial for calculation. For example, if the equipment operation value is to be predicted one hour in the future, the corresponding predicted value is calculated by the fitting polynomial according to the current time and the equipment operation status data. Some other factors can also be considered in the prediction process, such as the historical operation trend of the equipment, the current load condition, etc., and the prediction results can be appropriately corrected and optimized to improve the accuracy of the prediction.

[0042] Step S6444, when the target fitting operation value exceeds the first predetermined maximum fitting operation value in the first predetermined operation constraint, an early warning instruction is issued. Specifically, the first predetermined operation constraint of the first central device is pre-set, including the first predetermined maximum fitting operation value, which is determined according to the design specifications, historical operation experience and safety requirements of the device, and represents the maximum allowable operating value range of the device under normal operation. The predicted target fitting operation value is compared with the first predetermined maximum fitting operation value. If the target fitting operation value exceeds the predetermined maximum fitting operation value, it means that the device may have abnormal operation in the future, and the system immediately issues an early warning instruction. The early warning instruction can be issued in various forms, such as displaying a red warning message on the monitoring interface, sending text messages or emails to relevant managers, triggering sound and light alarm devices, etc., so as to attract the attention of relevant personnel in time and take corresponding measures for prevention and treatment.

[0043] Step S6445, based on the early warning instruction, the system performs a predictive check on the first central device. Specifically, once the early warning instruction is received, the system starts to perform a predictive check on the first central device. First, more detailed information about the current operating status of the device is collected, including real-time operating parameters, device temperature, vibration and other physical quantities. This information can be obtained through sensors or monitoring systems connected to the device. Then, the collected information is comprehensively analyzed, and combined with the previous prediction results, the type and location of the problem that may occur in the device are determined. For example, if it is predicted that the current value may exceed the normal range, further check the circuit connection, load condition and related electrical components of the device for abnormalities. Some professional detection tools and methods, such as insulation resistance testing and current harmonic analysis, can be used to conduct in-depth detection and diagnosis of the equipment. According to the predicted time window, maintenance personnel are arranged in advance to inspect and maintain the equipment to ensure that problems can be discovered and solved in time before the equipment becomes abnormal, so as to avoid the impact of equipment failure on the normal operation of the power system, realize predictive maintenance and management of the equipment, and improve the reliability and stability of the power system.

[0044] In a possible implementation, spline fitting is performed on the first operating spectrum based on the principle of spline curves to obtain a first spline curve, and step S6442 further includes step S64421, randomly picking up the first peak point in the first operating spectrum and forming a first sample set. Specifically, the first operating spectrum is analyzed to identify the peak points therein. These peak points represent the frequency positions where the energy in the spectrum is relatively concentrated, which may be related to the key operating characteristics or potential problems of the device. A plurality of first peak points are randomly selected in the spectrum through a certain algorithm or method. For example, a random number generator can be set to randomly select a certain number of peak point positions according to the data length and distribution of the spectrum. The selected peak points should be representative to a certain extent, covering different frequency regions and amplitude ranges of the spectrum. These selected first peak points are combined together to form a first sample set to provide basic data for subsequent analysis and fitting.

[0045] Step S64422, a polynomial fitting analysis is performed on the first sample set to obtain a first sample curve. Specifically, a suitable polynomial fitting method is selected to analyze the first sample set. Common polynomial fitting methods such as the least squares method determine the coefficients of the polynomial by minimizing the sum of squares of the errors between the actual data points and the fitting polynomial. According to the characteristics of the sample set and the analysis requirements, a suitable polynomial degree is selected. Generally speaking, the higher the polynomial degree, the more accurately the fitting curve can pass through the sample points, but it may also lead to overfitting, that is, it is too sensitive to small fluctuations in the data and loses the accurate description of the overall trend of the data. Therefore, it is necessary to balance the fitting accuracy and model complexity. For example, for simpler spectral features, a quadratic polynomial may be selected for fitting; for complex spectra, a higher degree polynomial may be required. In the fitting process, the coefficients of the polynomial are calculated so that the fitting curve can be as close as possible to the peak points in the first sample set. The first sample curve obtained is a mathematical description of the local characteristics of the spectrum represented by these peak points, which can help us further analyze the laws and trends of the spectrum.

[0046] Step S64423, randomly pick up the first check peak point in the first operating spectrum, and determine whether the first check peak point belongs to the first predetermined neighborhood of the first sample curve. Specifically, randomly select some points in the first operating spectrum again as the first check peak points. For each first check peak point, determine its frequency and amplitude position in the spectrum. Then, based on the characteristics of the first sample curve and the pre-set neighborhood definition method, determine whether the check peak point belongs to the first predetermined neighborhood of the first sample curve. The definition of the neighborhood can be based on frequency distance, amplitude difference or other related spectrum parameters. For example, a frequency range and amplitude deviation range can be set as the boundary conditions of the neighborhood. If the frequency and amplitude of the check peak point are within this range, it is considered to belong to the neighborhood of the first sample curve and marked as the first neighborhood point; if it is not within this range, it is marked as the first non-neighborhood point. This judgment process helps to evaluate the representativeness and coverage of the first sample curve for the overall characteristics of the spectrum.

[0047] Step S64424: If it is, the first check peak point is used as the first neighborhood point; if it is not, the first check peak point is used as the first non-neighborhood point. Specifically, if it is not within this range, it is marked as the first non-neighborhood point. This judgment process helps to evaluate the representativeness and coverage of the first sample curve for the overall characteristics of the spectrum.

[0048] Step S64425, when the first ratio of the first neighborhood point to the first non-neighborhood point reaches the predetermined ratio constraint, the first sample curve is used as the first spline curve. Specifically, the number of the first neighborhood points and the first non-neighborhood points is counted, and the first ratio between them is calculated. This ratio reflects the fitting quality and adaptability of the first sample curve to the spectrum data. When the first ratio reaches the predetermined ratio constraint, it indicates that the first sample curve can better describe the characteristics of the spectrum to a certain extent, and it is used as the first spline curve. The predetermined ratio constraint is pre-set based on experience and the requirements for the accuracy of spectrum analysis. If the ratio does not meet the constraint condition, it may be necessary to readjust the selection of the sample set, the parameters of the polynomial fitting or the definition of the neighborhood, etc., and analyze and judge again until the condition is met. The first spline curve determined in this way can more accurately represent the main characteristics of the first operating spectrum, providing a more reliable basis for subsequent equipment operation analysis, prediction and anomaly detection.

[0049] The embodiment of the present application obtains the power consumption detection value of the target user within a predetermined period, divides the period into multiple stages, performs stage sampling detection, and obtains the power consumption of the first stage and the second stage respectively. The target reference power is set according to these two power consumptions. When the absolute value of the deviation between the target detection power and the reference power does not meet the predetermined threshold, the system sends an abnormal warning signal, and calls the predetermined power verification strategy to verify the user's power consumption based on the signal, thereby achieving the technical effect of improving the accuracy and efficiency of power anomaly detection by identifying and resolving power anomalies caused by user behavior, hardware failures or software problems.

[0050] In the above, refer to Figure 1 The user power anomaly detection and analysis method according to an embodiment of the present invention is described in detail. Figure 2 A user power anomaly detection and analysis system according to an embodiment of the present invention is described.

[0051] The user power anomaly detection and analysis system according to the embodiment of the present invention is used to solve the technical problem of insufficient single cause investigation in the existing power anomaly detection, and achieves the technical effect of improving the accuracy and efficiency of power anomaly detection by identifying and solving power anomalies caused by user behavior, hardware failure or software problems. The user power anomaly detection and analysis system includes: a target detection power acquisition module 10, a predetermined period division module 20, a stage sampling detection module 30, a target reference power setting module 40, an abnormal warning signal issuing module 50, and a power consumption verification module 60.

[0052] The target detection power acquisition module 10 is used to acquire the target detection power, where the target detection power refers to the detection value of the power consumption of the target user in a predetermined period.

[0053] The predetermined period division module 20 is used to divide the predetermined period into a plurality of stages, wherein the plurality of stages include a first stage and a second stage.

[0054] The stage sampling detection module 30 is used to introduce stage detection logic to perform stage sampling detection on the power consumption of the target user in the first stage and the second stage, and obtain the first power consumption in the first stage and the second power consumption in the second stage respectively.

[0055] The target reference power setting module 40 is used to set the target reference power of the target user according to the first power and the second power.

[0056] The abnormal warning signal issuing module 50 is used to issue an abnormal warning signal when the absolute value of the power deviation between the target detection power and the target reference power does not meet a predetermined deviation threshold.

[0057] The power consumption verification module 60 is used to call a predetermined power consumption verification strategy to verify the power consumption of the target user according to the abnormal warning signal.

[0058] The specific configuration of the target reference power setting module 40 will be described in detail below. As described above, the target reference power of the target user is set according to the first power and the second power, and the target reference power setting module 40 further includes: a total stage number counting unit, the total stage number counting unit is used to count the number of the multiple stages, recorded as the total stage number; a stage detection number acquisition unit, the stage detection number acquisition unit is used to obtain the stage detection number according to the predetermined detection ratio set in the stage detection logic, combined with the total stage number; an average stage power calculation unit, the average stage power calculation unit is used to combine the first power, the second power and the stage detection number to calculate the average stage power; a tolerance adjustment unit, the tolerance adjustment unit is used to perform tolerance adjustment on the product of the total stage number and the average stage power to obtain the target reference power.

[0059] The specific configuration of the power consumption verification module 60 will be described in detail below. As described above, the power consumption of the target user is verified by calling a predetermined power consumption verification strategy according to the abnormal warning signal. The power consumption verification module 60 further includes: a user behavior verification strategy extraction unit, the user behavior verification strategy extraction unit is used to extract the user behavior verification strategy in the predetermined power consumption verification strategy; a target power consumption behavior baseline generation unit, the target power consumption behavior baseline generation unit is used to analyze the historical power consumption behavior log of the target user to generate a target power consumption behavior baseline; a behavior abnormality index acquisition unit, the behavior abnormality index acquisition unit is used to compare the actual power consumption behavior of the target user in the predetermined period with the target power consumption behavior baseline according to the user behavior verification strategy, and obtain a behavior abnormality index; a target behavior verification result acquisition unit, the target behavior verification result acquisition unit is used to send a power consumption behavior confirmation signal to the target user if the behavior abnormality index reaches a predetermined abnormal threshold, and obtain a target behavior verification result.

[0060] Among them, if the behavior abnormality index reaches a predetermined abnormal threshold, a power consumption behavior confirmation signal is sent to the target user to obtain a target behavior verification result. The target behavior verification result acquisition unit further includes: a hardware verification strategy extraction subunit, the hardware verification strategy extraction subunit is used to extract the hardware verification strategy in the predetermined power verification strategy according to the target behavior verification result, if the power consumption behavior of the target user is normal; an operation feature collection subunit, the operation feature collection subunit is used to collect the operation features of the distributed power equipment corresponding to the target user according to the hardware verification strategy to obtain the target equipment operation information; a central equipment operation information extraction subunit Element, the central device operation information extraction subunit is used to extract the central device operation information in the target device operation information, and the central device operation information includes a first operation signal of a first central device; an edge device operation information extraction subunit, the edge device operation information extraction subunit is used to extract the edge device operation information in the target device operation information when the first operation signal meets the first predetermined operation constraint of the first central device, and the edge device operation information includes a second operation signal of the first edge device; a target hardware verification result analysis subunit, the target hardware verification result analysis subunit is used to analyze the second operation signal to obtain a target hardware verification result.

[0061] Among them, the target hardware verification result is obtained by analyzing the second operating signal, and the target hardware verification result analysis subunit further includes: a software verification strategy extraction microunit, the software verification strategy extraction microunit is used to extract the software verification strategy in the predetermined power verification strategy based on the target hardware verification result if the distributed power equipment has no abnormality; a remote update and upgrade microunit, the remote update and upgrade microunit is used to remotely update and upgrade the power metering software corresponding to the target user according to the software verification strategy.

[0062] Among them, when the first operation signal meets the first predetermined operation constraint of the first central device, the edge device operation information in the target device operation information is extracted, and the edge device operation information includes the second operation signal of the first edge device, and the edge device operation information extraction subunit further includes: a first operation spectrum analysis microunit, the first operation spectrum analysis microunit is used to analyze the first operation signal to generate a first operation spectrum; a first spline curve acquisition microunit, the first spline curve acquisition microunit is used to perform spline fitting on the first operation spectrum based on the spline curve principle to obtain a first spline curve; a target fitting operation value prediction microunit, the target fitting operation value prediction microunit is used to obtain a first fitting polynomial of the first spline curve, and predict the target fitting operation value at the target time through the first fitting polynomial; an early warning instruction issuing microunit, the early warning instruction issuing microunit is used to issue an early warning instruction when the target fitting operation value exceeds the first predetermined maximum fitting operation value in the first predetermined operation constraint; a predictive verification microunit, the predictive verification microunit is used to perform a predictive verification on the first central device based on the early warning instruction.

[0063] Among them, spline fitting is performed on the first operating spectrum based on the spline curve principle to obtain a first spline curve, and the first spline curve acquisition micro unit further includes: a first sample set composition unit, the first sample set composition unit is used to randomly pick up the first peak point in the first operating spectrum and form a first sample set; a first sample curve acquisition unit, the first sample curve acquisition unit is used to perform polynomial fitting analysis on the first sample set to obtain a first sample curve; a first predetermined neighborhood judgment unit, the first predetermined neighborhood judgment unit is used to randomly pick up a first verification peak point in the first operating spectrum, and judge whether the first verification peak point belongs to the first predetermined neighborhood of the first sample curve; a first non-neighborhood point judgment unit, the first non-neighborhood point judgment unit is used to use the first verification peak point as the first neighborhood point if it belongs, and use the first verification peak point as the first non-neighborhood point if it does not belong; a predetermined ratio constraint judgment unit, the predetermined ratio constraint judgment unit is used to use the first sample curve as the first spline curve when the first ratio of the first neighborhood point to the first non-neighborhood point reaches a predetermined ratio constraint.

[0064] The user power anomaly detection and analysis system provided in the embodiment of the present invention can execute the user power anomaly detection and analysis method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0065] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0066] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for detecting and analyzing abnormal power consumption of users, characterized in that: include: Obtaining a target detection power, where the target detection power refers to a detection value of power consumption of a target user in a predetermined period; Dividing the predetermined period into a plurality of stages, wherein the plurality of stages include a first stage and a second stage; Introducing a stage detection logic to perform stage sampling detection on the target user's power consumption in the first stage and the second stage, and obtaining a first power consumption in the first stage and a second power consumption in the second stage respectively; Setting a target baseline power of the target user according to the first power and the second power; When the absolute value of the power deviation between the target detection power and the target reference power does not meet the predetermined deviation threshold, an abnormal warning signal is issued; The predetermined power consumption verification strategy is called according to the abnormal warning signal to verify the power consumption of the target user.

2. The method for detecting and analyzing abnormal power consumption of users according to claim 1, characterized in that: include: Counting the number of the multiple stages and recording it as the total number of stages; According to the predetermined detection ratio set in the stage detection logic, the stage detection number is obtained in combination with the total number of stages; Combining the first electric quantity, the second electric quantity and the number of stage detections, calculating an average stage electric quantity; The product of the total number of stages and the average stage power is adjusted for tolerance to obtain the target reference power.

3. The method for detecting and analyzing abnormal power consumption of users according to claim 1, characterized in that: include: Extracting the user behavior verification strategy from the predetermined power verification strategy; Analyze the historical electricity usage behavior log of the target user to generate a target electricity usage behavior baseline; According to the user behavior verification strategy, the actual power usage behavior of the target user in the predetermined period is compared with the target power usage behavior baseline to obtain a behavior abnormality index; If the behavior abnormality index reaches a predetermined abnormality threshold, a power usage behavior confirmation signal is sent to the target user to obtain a target behavior verification result.

4. The method for detecting and analyzing abnormal power consumption of users according to claim 3 is characterized in that: Also includes: According to the target behavior verification result, if the target user's power consumption behavior is normal, extracting the hardware verification strategy in the predetermined power verification strategy; Collecting the operation characteristics of the distributed power equipment corresponding to the target user according to the hardware verification strategy to obtain the target equipment operation information; Extracting central device operation information from the target device operation information, wherein the central device operation information includes a first operation signal of a first central device; When the first operation signal meets the first predetermined operation constraint of the first central device, extracting edge device operation information from the target device operation information, the edge device operation information including the second operation signal of the first edge device; The second operation signal is analyzed to obtain a target hardware verification result.

5. The method for detecting and analyzing abnormal power consumption of users according to claim 4, characterized in that: Also includes: According to the target hardware verification result, if the distributed power equipment has no abnormality, extracting the software verification strategy in the predetermined power verification strategy; The electric quantity metering software corresponding to the target user is remotely updated and upgraded according to the software verification strategy.

6. The method for detecting and analyzing abnormal power consumption of users according to claim 4, characterized in that: Also includes: Analyzing the first operating signal to generate a first operating spectrum; Performing spline fitting on the first operating spectrum based on the spline curve principle to obtain a first spline curve; Obtaining a first fitting polynomial of the first spline curve, and predicting a target fitting running value at a target time by using the first fitting polynomial; When the target fitting operation value exceeds a first predetermined maximum fitting operation value in the first predetermined operation constraint, issuing a warning instruction; A predictive check is performed on the first central device based on the early warning instruction.

7. The method for detecting and analyzing abnormal user power consumption according to claim 6, characterized in that: include: Randomly picking up the first peak point in the first running spectrum and forming a first sample set; Performing a polynomial fitting analysis on the first sample set to obtain a first sample curve; Randomly picking up a first verification peak point in the first operating spectrum, and determining whether the first verification peak point belongs to a first predetermined neighborhood of the first sample curve; If it does, take the first check peak point as the first neighborhood point; if it does not, take the first check peak point as the first non-neighborhood point; When a first ratio of the first neighborhood point to the first non-neighborhood point reaches a predetermined ratio constraint, the first sample curve is used as the first spline curve.

8. User power anomaly detection and analysis system, characterized by: The system is used to implement the user power anomaly detection and analysis method according to any one of claims 1 to 7, and the system includes: A target detection power acquisition module, the target detection power acquisition module is used to acquire the target detection power, the target detection power refers to the detection value of the power consumption of the target user in a predetermined period; A predetermined period division module, the predetermined period division module is used to divide the predetermined period into a plurality of stages, the plurality of stages including a first stage and a second stage; a stage sampling detection module, the stage sampling detection module is used to introduce a stage detection logic to perform stage sampling detection on the power consumption of the target user in the first stage and the second stage, and obtain a first power consumption in the first stage and a second power consumption in the second stage respectively; a target reference power setting module, the target reference power setting module being used to set a target reference power of the target user according to the first power and the second power; An abnormal warning signal issuing module, the abnormal warning signal issuing module is used to issue an abnormal warning signal when the absolute value of the power deviation between the target detection power and the target reference power does not meet a predetermined deviation threshold; The power consumption verification module is used to call a predetermined power consumption verification strategy according to the abnormal warning signal to verify the power consumption of the target user.

Images