Intelligent online monitoring system for intelligent electric energy meter

Abstract

The invention relates to the technical field of intelligent electric energy meter on-line monitoring, and discloses an intelligent on-line monitoring system for an intelligent electric energy meter, and the system comprises the steps: collecting the power utilization power of a special power line of a tenant in each time period of D days, respectively constructing the power utilization behaviors of workdays and holidays, and judging the stability of the power utilization behaviors; acquiring a plurality of electric appliances, and identifying the power utilization power of the tenant to obtain the on-off state of each electric appliance; using the credibility score to quantify the power consumption credibility of each tenant: obtaining all tenants, evaluating the similarity between the power consumption behavior of each tenant and the power consumption behavior of the power line in the public area, and calculating the credibility score of each tenant according to the abnormal behavior of each tenant; the credibility score of each tenant is updated in real time, whether the power consumption behavior of the power line in the public area is abnormal or not is judged, and when the abnormal power consumption behavior occurs in the public area, an alarm is given according to the tenant portrait; and intelligent and refined power utilization supervision is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of online monitoring of smart electric energy meters, and in particular to an intelligent online monitoring system for smart electric energy meters. Background Art

[0002] In multi-tenant scenarios, separate management of electricity usage in public areas and tenant-specific electricity can improve transparency and effectively prevent disputes over electricity bills and electricity theft. This helps landlords gain a detailed understanding of power consumption characteristics in different areas, facilitates energy-saving management and fault location, and improves overall management efficiency.

[0003] While most current smart energy meter systems offer remote data collection capabilities, they still face significant deficiencies in tenant behavior modeling and electricity theft detection. First, existing systems typically rely solely on total power consumption changes for anomaly detection, lacking in-depth analysis of individual tenant usage patterns, resulting in low detection accuracy. Second, most systems fail to model factors such as tenant active hours and differences in holiday and weekday behavior, making it difficult to identify abnormal fluctuations in behavior patterns. Furthermore, existing systems often identify electricity theft based solely on set thresholds, lacking a comprehensive assessment of multiple factors such as historical behavior, behavioral consistency, and public electricity similarity, making them prone to false positives and missed detections. Furthermore, existing technologies generally lack tenant interaction interfaces, hindering tenant self-inspection and transparent system operation. Furthermore, system parameters are not flexibly adjustable, resulting in poor adaptability to different scenarios. Finally, the lack of real-time monitoring of system operational status makes it difficult to promptly detect and address any data collection module failures.

[0004] The present invention proposes an intelligent online monitoring system for smart electricity meters, which realizes more intelligent and refined electricity consumption supervision by constructing tenant portraits, refined behavior modeling, credibility scoring mechanism and multi-dimensional anomaly detection strategy. Summary of the Invention

[0005] The present invention provides an intelligent online monitoring system for an intelligent electric energy meter, which is used to promote the solution of the problems mentioned in the above background technology.

[0006] The present invention provides the following technical solution: an intelligent online monitoring system for a smart electric energy meter, comprising: The data collection module is used to collect the power consumption of the public area power lines and the tenant-specific power lines, and upload them to the database for storage; The tenant profile building module is used to build an electricity usage behavior model for each tenant and output a tenant profile, specifically including: Divide a day into 24 time periods and collect the electricity consumption of tenants' dedicated power lines in each time period on day D. Build electricity consumption behaviors on weekdays and holidays, and determine the stability of electricity consumption behaviors. Obtain multiple electrical appliances, identify the tenant's power consumption, and obtain the on / off status of each appliance; The electricity theft credibility scoring module is used to quantify the credibility of each tenant's electricity use. This includes: obtaining all tenants, evaluating the similarity between each tenant's electricity use behavior and the electricity use behavior of the public area power lines, and calculating each tenant's credibility score based on each tenant's abnormal behavior; Anomaly detection and alarm module, used to determine whether the power usage behavior of power lines in public areas is abnormal and issue alarms based on tenant profiles; The visualization display module is used to generate visualization content based on tenant profiles and electricity usage behaviors, and then display the visualization content to tenants; The system management and configuration module is used to associate the collected power consumption with tenants, adjust the calculation standard of the credibility score, and adjust system parameters.

[0007] Optionally, constructing tenants' electricity usage behaviors on weekdays and holidays respectively and determining the stability of the electricity usage behaviors includes: The electricity usage behavior of the tenant Divided into electricity consumption on weekdays and electricity consumption on holidays; For period t, collect the electricity consumption of tenants on the dedicated power line during period t on each working day, calculate the average of all electricity consumption, and record the result as the working day electricity consumption behavior during period t ; For period t, collect the electricity consumption of the tenants on the dedicated power line during each holiday period t, calculate the average of all electricity consumption, and record the result as the holiday electricity consumption behavior during period t ; The stability of electricity consumption on weekdays and holidays is calculated separately, specifically: Get the power consumption on day c in time period t , calculate the time period variance of power consumption , where when a=1, is the time period variance of the working day, when a=2, is the period variance of holidays, The number of days that are working days or holidays; Power fluctuation coefficient during set period , used to reflect the degree of diurnal fluctuation of tenants’ electricity consumption; , a=1,2.

[0008] Optionally, acquiring multiple electrical appliances, identifying the electricity consumption of the tenant, and obtaining the on / off status of each electrical appliance includes: Obtain all electrical appliances used by tenants, connect each appliance to a separate circuit, and measure the actual power of each appliance , is the actual power of the kth appliance; Set the switch status of the kth appliance in time period t ,when When the appliance k is turned off, When , the appliance k is turned on; Get the power consumption in time period t ; The switch states of b electrical appliances are combined into a state vector ; Use the exhaustive method to calculate the switch status of each appliance, specifically: , get the switch state of the kth appliance in time period t .

[0009] Optionally, the electricity theft credibility scoring module is used to quantify the electricity usage credibility of each tenant, including: Obtain all power consumption of the power lines in the public area during period t, calculate the mean of all power consumption, and obtain the power consumption behavior of the public area ; Get the electricity usage behavior of the u-th tenant ; Use the Pearson correlation coefficient to calculate the correlation between the electricity usage behavior in the public area and the electricity usage behavior of the u-th tenant. ; The abnormal behavior of tenant u triggers alarm and steals electricity; Get the total number of abnormal behaviors of tenant u within D days ; Calculate the history score of tenant u .

[0010] Optionally, the electricity theft credibility scoring module is used to quantify the electricity usage credibility of each tenant and further includes: Set power thresholds to determine whether tenants are active during a time period; Calculate the power threshold as follows: Obtain the power consumption of all tenants in each time period of the day, calculate the mean and standard deviation, and then sum the mean and standard deviation to use as the power threshold. Compare the power consumption of tenant u in time period t with the power threshold. If the power consumption is greater than or equal to the power threshold, tenant u is active in time period t. If the power consumption is less than the power threshold, tenant u is inactive during time period t; Get the total number of periods during which tenant u has abnormal behavior ; Get the total number of periods during which tenant u has abnormal behavior during active periods ; Calculate the consistency score of tenant u , ,in, is a constant; Get the daytime fluctuation degree of tenant u on weekdays and holidays, calculate the average, and record the result as the average fluctuation degree ; Calculate the credibility score of tenant u for stealing electricity ,in, , , and is the weight coefficient.

[0011] Optionally, the anomaly detection and alarm module is used to determine whether the power usage behavior of the power lines in the public area is abnormal and issue an alarm based on the tenant profile, including: Get the electricity consumption behavior of the public area in the current period , calculate the power difference of electricity consumption behavior in adjacent time periods; Set the abnormal difference threshold; Compare the difference with the abnormal difference threshold. If the difference is greater than the abnormal difference threshold, determine whether the current period is an active period for any tenant. If the current time period is not an active time period for any tenant, it is considered that the electricity usage in the public area is abnormal; Setting a score difference threshold, wherein the score difference threshold is less than 0; Obtain each tenant's credibility score for the current period, obtain the credibility score for the previous period, calculate the difference, and compare the difference with the score difference threshold; If the difference is less than the score difference threshold, the tenant corresponding to the difference is recorded as a candidate tenant; Calculate the similarity between each candidate tenant's electricity usage behavior and the electricity usage behavior in the public area; Set similarity threshold; The tenants corresponding to the electricity usage behaviors whose similarity is greater than the similarity threshold are recorded as marked tenants; The personal information of the marked tenant is pushed to the landlord as an alert.

[0012] Optionally, the visualization display module is used to generate visualization content based on tenant portraits and electricity usage behaviors and display the visualization content to tenants, including: The visualization content includes a tenant portrait radar chart, a line chart reflecting tenant electricity usage behavior, a heat map comparison of public area power lines and tenant-specific power lines, and a ranking display interface for all tenants' trust scores.

[0013] Optionally, the system management and configuration module is used to associate the collected power consumption with tenants, adjust the calculation standard of the credibility score, and adjust system parameters, including: Adjust the calculation criteria for credibility scores, specifically: The weight coefficient is dynamically adjusted by the landlord , , and size; The weight coefficient satisfies ; Adjusting system parameters, including similarity threshold, anomaly difference threshold, and score difference threshold, which are dynamically adjusted by the landlord; The data upload status of each power line is detected in real time. When data upload interruption is detected, it is determined to be offline and the landlord is notified.

[0014] The present invention has the following beneficial effects: 1. This smart energy meter uses an intelligent online monitoring system that differentiates electricity usage between weekdays and holidays to more accurately characterize tenants' living habits and electricity usage patterns over different time periods. The calculation of time-period mean and variance enables the model to not only capture tenants' average power levels but also measure their electricity usage volatility, establishing a quantitative foundation for behavioral stability and avoiding misjudgment of natural electricity fluctuations.

[0015] 2. The smart electricity meter uses an intelligent online monitoring system, which connects each appliance to the circuit individually and measures the actual power, providing high-precision annotated data. By constructing a device state vector and combining it with an exhaustive method to infer the current appliance status, it maps the total power consumption to the specific device operating status. This enhances the reliability of the model in actual application scenarios, allows for further analysis of active periods, and lays a solid foundation for energy efficiency analysis and load optimization.

[0016] 3. This intelligent online monitoring system for smart energy meters compares tenants' own electricity usage with that of public areas. Combined with historical anomaly records, it assesses the credibility of tenant behavior from multiple perspectives, effectively avoiding the limitations of single-feature judgments and making the PTS score more comprehensive and discriminatory. The Pearson correlation coefficient allows for rapid identification of suspicious tenants in highly similar scenarios. Historical behavior records serve as negative indicators, allowing repeated anomalies to be tracked continuously within the system, improving the overall system's long-term security and control capabilities.

[0017] 4. This smart energy meter uses an intelligent online monitoring system that incorporates dynamic power thresholds, using a data-driven approach to determine tenant activity, eliminating the subjective nature of manually set thresholds. By intersecting tenants' abnormal behavior with their active periods, it can determine whether the anomaly is motivated and further confirm its credibility. This effectively avoids false positives and missed negatives, improving the accuracy of anomaly identification. Furthermore, by combining the diurnal fluctuation coefficients for weekdays and holidays, it can more comprehensively assess the stability and habitual deviations of tenants' electricity usage. Adjustable weighting coefficients are introduced to achieve dynamic adaptation in different scenarios.

[0018] 5. The intelligent online monitoring system for this smart energy meter integrates multiple triggering mechanisms, including sudden changes in public power, sudden drops in PTS scores, and the matching of tenant behavior with public loads. This system forms a systematic, hierarchical early warning logic. It uses power difference and score difference thresholds to control the frequency of alarms and avoid false triggers. A secondary comparison is performed between candidate tenants and public behavior to enhance the credibility and interpretability of alarm results. This enables real-time detection and immediate response, ensuring the system's high sensitivity and responsiveness to electricity theft.

[0019] 6. This smart meter's intelligent online monitoring system provides landlords with a high degree of control, including PTS score weight adjustment, anomaly threshold configuration, and device status monitoring. By dynamically adjusting score weights, the system optimizes risk assessment models based on actual operational needs, achieving behavioral assessments that are more responsive to scenario characteristics. Real-time device status monitoring promptly identifies anomalies in the data acquisition module, reducing the risk of system blind spots and data gaps, and ensuring the platform's long-term operational stability and data integrity. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Schematic diagram of the process of the present invention.

[0021] Figure 2 Schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0023] Example 1, refer to Figure 2 , an intelligent online monitoring system for a smart electric energy meter, comprising: The data collection module is used to collect the power consumption of the public area power lines and the tenant-specific power lines, and upload them to the database for storage; The tenant profile building module is used to build an electricity usage behavior model for each tenant and output a tenant profile, specifically including: Divide a day into 24 time periods and collect the electricity consumption of tenants' dedicated power lines in each time period on day D. Build electricity consumption behaviors on weekdays and holidays, and determine the stability of electricity consumption behaviors. Obtain multiple electrical appliances, identify the tenant's power consumption, and obtain the on / off status of each appliance; The steps of constructing tenants' electricity usage behaviors on weekdays and holidays and determining the stability of electricity usage behaviors include: The electricity usage behavior of the tenant Divided into electricity consumption on weekdays and electricity consumption on holidays; For period t, collect the electricity consumption of tenants on the dedicated power line during period t on each working day, calculate the average of all electricity consumption, and record the result as the working day electricity consumption behavior during period t ; For period t, collect the electricity consumption of the tenants on the dedicated power line during each holiday period t, calculate the average of all electricity consumption, and record the result as the holiday electricity consumption behavior during period t ; The stability of electricity consumption on weekdays and holidays is calculated separately, specifically: Get the power consumption on day c in time period t , calculate the time period variance of power consumption , where when a=1, is the time period variance of the working day, when a=2, is the period variance of holidays, The number of days that are working days or holidays; Power fluctuation coefficient during set period , used to reflect the degree of diurnal fluctuation of tenants’ electricity consumption; , a=1,2.

[0024] By categorizing tenants' electricity usage into two categories, weekdays and holidays, and analyzing power data across different time periods, this module can establish a tenant behavior model that more closely reflects real-life patterns. This behavioral model not only improves the accuracy of the profile but also effectively explores the characteristics of tenants' electricity usage habits on different day types, providing a more solid data foundation for subsequent modules such as credibility scoring and anomaly detection. Furthermore, by introducing the time period fluctuation coefficient as a stability evaluation indicator, it is possible to assess the regularity and stability of tenant behavior from both horizontal time periods and vertical days, providing greater judgment in identifying potential anomalies or deviations.

[0025] The acquiring of multiple electrical appliances, identifying the electricity consumption of the tenant, and obtaining the on / off status of each electrical appliance includes: Obtain all electrical appliances used by tenants, connect each appliance to a separate circuit, and measure the actual power of each appliance , is the actual power of the kth appliance; Set the switch status of the kth appliance in time period t ,when When the appliance k is turned off, When , the appliance k is turned on; Get the power consumption in time period t ; The switch states of b electrical appliances are combined into a state vector ; Use the exhaustive method to calculate the switch status of each appliance, specifically: , get the switch state of the kth appliance in time period t .

[0026] By connecting each tenant's electrical appliance to the measurement circuit one by one and obtaining the actual power parameters of each appliance, a direct mapping relationship between each appliance and power is established, which is more interpretable and practical, significantly reducing identification errors. By combining an exhaustive method to traverse the appliance combination state, the state vector mapping is matched with the observed power to accurately reconstruct the appliance's on / off state, providing precise underlying data support for subsequent power usage pattern analysis, energy-saving recommendations, and behavioral anomaly detection.

[0027] The electricity theft credibility scoring module is used to quantify the credibility of each tenant's electricity use. This includes: obtaining all tenants, evaluating the similarity between each tenant's electricity use behavior and the electricity use behavior of the public area power lines, and calculating each tenant's credibility score based on each tenant's abnormal behavior; The electricity theft credibility scoring module is used to quantify the electricity usage credibility of each tenant, including: Obtain all power consumption of the power lines in the public area during period t, calculate the mean of all power consumption, and obtain the power consumption behavior of the public area ; Get the electricity usage behavior of the u-th tenant ; Use the Pearson correlation coefficient to calculate the correlation between the electricity usage behavior in the public area and the electricity usage behavior of the u-th tenant. ; The abnormal behavior of tenant u triggers alarm and steals electricity; Get the total number of abnormal behaviors of tenant u within D days ; Calculate the history score of tenant u .

[0028] By calculating similarity indicators between tenants' electricity usage and that of public areas, we can effectively identify anomalous behavior suspected of connecting public electricity to private use, helping to pinpoint potential electricity theft. Furthermore, by recording and quantifying the number of historical anomalous behaviors, we build a behavioral profile of tenants, providing a dynamic and quantifiable input for credibility scoring, significantly improving the stability and rationality of the scoring results. By integrating comparative analysis of individual behavioral characteristics with overall system behavior, we achieve greater intelligence and judgment.

[0029] The electricity theft credibility scoring module is used to quantify the electricity usage credibility of each tenant and also includes: Set power thresholds to determine whether tenants are active during a time period; Calculate the power threshold as follows: Obtain the power consumption of all tenants in each time period of the day, calculate the mean and standard deviation, and then sum the mean and standard deviation to use as the power threshold. Compare the power consumption of tenant u in time period t with the power threshold. If the power consumption is greater than or equal to the power threshold, tenant u is active in time period t. If the power consumption is less than the power threshold, tenant u is inactive during time period t; Get the total number of periods during which tenant u has abnormal behavior ; Get the total number of periods during which tenant u has abnormal behavior during active periods ; Calculate the consistency score of tenant u , ,in, is a constant; Get the daytime fluctuation degree of tenant u on weekdays and holidays, calculate the average, and record the result as the average fluctuation degree ; Calculate the credibility score of tenant u for stealing electricity ,in, , , and is the weight coefficient.

[0030] By introducing a dynamic power threshold model, the system correlates tenant activity status with actual electricity usage data, effectively addressing the bias caused by differences in electricity usage between tenants and improving the accuracy of abnormal behavior detection. By analyzing the consistency between abnormal behavior periods and active periods, the system identifies abnormal events that occur during unexpected time periods, improving the system's sensitivity and discernment of electricity theft. A multi-factor fusion scoring mechanism is introduced, calculating a weighted credibility score based on behavioral stability, similarity, consistency, and historical records. The scoring results are more objective and comprehensive, dynamically reflecting changes in tenant behavior risks.

[0031] Anomaly detection and alarm module, used to determine whether the power usage behavior of power lines in public areas is abnormal and issue alarms based on tenant profiles; The anomaly detection and alarm module is used to determine whether the power usage behavior of the power lines in the public area is abnormal and issue an alarm based on the tenant profile, including: Get the electricity consumption behavior of the public area in the current period , calculate the power difference of electricity consumption behavior in adjacent time periods; Set the abnormal difference threshold; Compare the difference with the abnormal difference threshold. If the difference is greater than the abnormal difference threshold, determine whether the current period is an active period for any tenant. If the current time period is not an active time period for any tenant, it is considered that the electricity usage in the public area is abnormal; Setting a score difference threshold, wherein the score difference threshold is less than 0; Obtain each tenant's credibility score for the current period, obtain the credibility score for the previous period, calculate the difference, and compare the difference with the score difference threshold; If the difference is less than the score difference threshold, the tenant corresponding to the difference is recorded as a candidate tenant; Calculate the similarity between each candidate tenant's electricity usage behavior and the electricity usage behavior in the public area; Set similarity threshold; The tenants corresponding to the electricity usage behaviors whose similarity is greater than the similarity threshold are recorded as marked tenants; The personal information of the marked tenant is pushed to the landlord as an alert.

[0032] The anomaly detection and alarm module achieves multi-dimensional judgment and precise identification of power theft by monitoring sudden changes in public line power in real time, combining tenant activity status and credibility score changes. This module integrates behavioral graph similarity analysis with dynamic score fluctuation analysis to significantly improve the accuracy and effectiveness of alarms. By setting a dual threshold mechanism for difference and similarity, it can promptly respond to potential anomalies while effectively reducing false alarm rates. The phased processing logic for candidate tenant screening and final tag push improves operability and controllability in large-scale tenant scenarios, significantly enhancing the system's practicality and intelligence.

[0033] The visualization display module is used to generate visualization content based on tenant profiles and electricity usage behaviors, and then display the visualization content to tenants; The visualization display module is used to generate visualization content based on tenant profiles and electricity usage behaviors and display the visualization content to tenants, including: The visualization content includes a tenant portrait radar chart, a line chart reflecting tenant electricity usage behavior, a heat map comparison of public area power lines and tenant-specific power lines, and a ranking display interface for all tenants' trust scores.

[0034] The system management and configuration module is used to associate the collected power consumption with tenants, adjust the calculation standard of the credibility score, and adjust system parameters.

[0035] The system management and configuration module is used to associate the collected power consumption with tenants, adjust the calculation standard of the credibility score, and adjust system parameters, including: Adjust the calculation criteria for credibility scores, specifically: The weight coefficient is dynamically adjusted by the landlord , , and size; The weight coefficient satisfies ; Adjusting system parameters, including similarity threshold, anomaly difference threshold, and score difference threshold, which are dynamically adjusted by the landlord; The data upload status of each power line is detected in real time. When data upload interruption is detected, it is determined to be offline and the landlord is notified.

[0036] This system provides platform managers with a flexible and controllable system adjustment mechanism, supporting dynamic adjustment of the weights of various factors in the credibility score. This allows for flexible adaptation based on actual application scenarios and behavioral patterns, thereby fulfilling personalized and differentiated management needs. By setting reasonable constraints, the overall stability and interpretability of the scoring system during adjustments are ensured. The system's parameter configuration capabilities allow landlords to optimize their strategies based on actual data distribution and anomaly risk levels, improving the system's responsiveness and accuracy.

[0037] In this embodiment, refer to Figure 1 , which is a method flow chart of the functional modules of the system, specifically: the method of each module in the data acquisition module, tenant portrait construction module, electricity theft credibility scoring module, anomaly detection and alarm module, visualization display module and system management and configuration module; It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0038] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. An intelligent online monitoring system for smart electric energy meters, characterized in that: include: The data collection module is used to collect the power consumption of the public area power lines and the tenant-specific power lines, and upload them to the database for storage; The tenant profile building module is used to build an electricity usage behavior model for each tenant and output a tenant profile, specifically including: Divide the day into 24 periods; Construct tenants' electricity usage behaviors on weekdays and holidays respectively, and determine the stability of electricity usage behaviors; Obtain multiple electrical appliances, identify the tenant's power consumption, and obtain the on / off status of each appliance; The electricity theft credibility scoring module is used to quantify the credibility of each tenant's electricity use, including: evaluating the similarity between each tenant's electricity use behavior and the electricity use behavior of the public area power lines, and calculating each tenant's credibility score based on each tenant's abnormal behavior; Anomaly detection and alarm module, used to determine whether the power usage behavior of power lines in public areas is abnormal and issue alarms based on tenant profiles; The visualization display module is used to generate visualization content based on tenant profiles and electricity usage behaviors, and then display the visualization content to tenants; The system management and configuration module is used to associate the collected power consumption with tenants, adjust the calculation standard of the credibility score, and adjust system parameters.

2. The intelligent online monitoring system for smart electric energy meters according to claim 1, characterized in that: The steps of constructing tenants' electricity usage behaviors on weekdays and holidays and determining the stability of electricity usage behaviors include: The electricity usage behavior of the tenant Divided into electricity consumption on weekdays and electricity consumption on holidays; For period t, collect the electricity consumption of tenants on the dedicated power line during period t on each working day, calculate the average of all electricity consumption, and record the result as the working day electricity consumption behavior during period t ; For period t, collect the electricity consumption of the tenants on the dedicated power line during each holiday period t, calculate the average of all electricity consumption, and record the result as the holiday electricity consumption behavior during period t ; The stability of electricity consumption on weekdays and holidays is calculated separately, specifically: Get the power consumption on day c in time period t , calculate the time period variance of power consumption , where when a=1, is the time period variance of the working day, when a=2, is the period variance of holidays, The number of days that are working days or holidays; Power fluctuation coefficient during set period , used to reflect the degree of diurnal fluctuation of tenants’ electricity consumption; ，a=1，2。 3. The intelligent online monitoring system for smart electric energy meters according to claim 2, characterized in that: The acquiring of multiple electrical appliances, identifying the electricity consumption of the tenant, and obtaining the on / off status of each electrical appliance includes: Obtain all electrical appliances used by tenants, connect each appliance to a separate circuit, and measure the actual power of each appliance , is the actual power of the kth appliance; Set the switch status of the kth appliance in time period t ,when When the appliance k is turned off, When , the appliance k is turned on; Get the power consumption in time period t ; The switch states of b electrical appliances are combined into a state vector ; Use the exhaustive method to calculate the switch status of each appliance, specifically: , get the switch state of the kth appliance in time period t .

4. The intelligent online monitoring system for smart electric energy meters according to claim 3, characterized in that: The electricity theft credibility scoring module is used to quantify the electricity usage credibility of each tenant, including: Obtain all power consumption of the power lines in the public area during period t, calculate the mean of all power consumption, and obtain the power consumption behavior of the public area ; Get the electricity usage behavior of the u-th tenant ; Use the Pearson correlation coefficient to calculate the correlation between the electricity usage behavior in the public area and the electricity usage behavior of the u-th tenant. ; The abnormal behavior of tenant u triggers alarm and steals electricity; Get the total number of abnormal behaviors of tenant u within D days ; Calculate the history score of tenant u .

5. The intelligent online monitoring system for smart electric energy meters according to claim 4, characterized in that: The electricity theft credibility scoring module is used to quantify the electricity usage credibility of each tenant and also includes: Set power thresholds to determine whether tenants are active during a time period; Calculate the power threshold as follows: Obtain the power consumption of all tenants in each time period of the day, calculate the mean and standard deviation, and then sum the mean and standard deviation to use as the power threshold. Compare the power consumption of tenant u in time period t with the power threshold. If the power consumption is greater than or equal to the power threshold, tenant u is active in time period t. If the power consumption is less than the power threshold, tenant u is inactive during time period t; Get the total number of periods during which tenant u has abnormal behavior ; Get the total number of periods during which tenant u has abnormal behavior during active periods ; Calculate the consistency score of tenant u , ,in, is a constant; Get the daytime fluctuation degree of tenant u on weekdays and holidays, calculate the average, and record the result as the average fluctuation degree ; Calculate the credibility score of tenant u for stealing electricity ,in, , , and is the weight coefficient.

6. The intelligent online monitoring system for smart electric energy meters according to claim 5, characterized in that: The anomaly detection and alarm module is used to determine whether the power usage behavior of the power lines in the public area is abnormal and issue an alarm based on the tenant profile, including: Get the electricity consumption behavior of the public area in the current period , calculate the power difference of electricity consumption behavior in adjacent time periods; Set the abnormal difference threshold; Compare the difference with the abnormal difference threshold. If the difference is greater than the abnormal difference threshold, determine whether the current period is an active period for any tenant. If the current time period is not an active time period for any tenant, it is considered that the electricity usage in the public area is abnormal; Setting a score difference threshold, wherein the score difference threshold is less than 0; Obtain each tenant's credibility score for the current period, obtain the credibility score for the previous period, calculate the difference, and compare the difference with the score difference threshold; If the difference is less than the score difference threshold, the tenant corresponding to the difference is recorded as a candidate tenant; Calculate the similarity between each candidate tenant's electricity usage behavior and the electricity usage behavior in the public area; Set similarity threshold; The tenants corresponding to the electricity usage behaviors whose similarity is greater than the similarity threshold are recorded as marked tenants; Calculate the difference between the power difference and the actual power of each appliance, and obtain the appliance with the smallest difference, which is recorded as the target appliance; Obtain the tenants who have used the target appliance among the marked tenants and record them as target tenants; Push the target tenant’s personal information to the landlord to alert him.

7. The intelligent online monitoring system for smart electric energy meters according to claim 1, characterized in that: The visualization display module is used to generate visualization content based on tenant profiles and electricity usage behaviors and display the visualization content to tenants, including: The visualization content includes a tenant portrait radar chart, a line chart reflecting tenant electricity usage behavior, a heat map comparison of public area power lines and tenant-specific power lines, and a ranking display interface for all tenants' trust scores.

8. The intelligent online monitoring system for smart electric energy meters according to claim 6, characterized in that: The system management and configuration module is used to associate the collected power consumption with tenants, adjust the calculation standard of the credibility score, and adjust system parameters, including: Adjust the calculation criteria for credibility scores, specifically: The weight coefficient is dynamically adjusted by the landlord , , and size; The weight coefficient satisfies ; Adjusting system parameters, including similarity threshold, anomaly difference threshold, and score difference threshold, which are dynamically adjusted by the landlord; The data upload status of each power line is detected in real time. When data upload interruption is detected, it is determined to be offline and the landlord is notified.

Images