Intelligent electric energy meter remote monitoring system and method based on electricity utilization acquisition terminal

Abstract

The invention provides an intelligent electric energy meter remote monitoring system and method based on a power utilization acquisition terminal. The method comprises the following steps: constructing a voltage quality operation base line of an intelligent electric energy meter based on historical data; in the monitoring process, the actually measured voltage data of each intelligent electric energy meter is compared with the base line of the intelligent electric energy meter, and when the actually measured voltage of the target electric energy meter continuously deviates from the base line interval and the deviation direction of the target electric energy meter is inconsistent with the change trend of the head end voltage of the transformer area, the target electric energy meter is marked as a voltage quality suspected abnormal node; for the abnormal node, extracting the split-phase line loss rate of the power supply branch where the abnormal node is located in the same time window, and calculating the correlation degree deviation value of the split-phase line loss rate relative to the historical same period; when the correlation degree deviation value exceeds a threshold value, it is judged that non-load-driven statistical distortion occurs to voltage abnormity corresponding to the suspected abnormal node along with the branch line loss parameter to which the suspected abnormal node belongs, and accordingly the suspected abnormal node is recognized as a user-side hidden fault or an abnormal power consumption behavior. According to the scheme, the hidden abnormity of the user side can be accurately positioned, and the operation and maintenance lean level of the distribution network is improved.

Description

Technical Field

[0001] This application relates to the field of power monitoring technology, and more specifically, to a remote monitoring system and method for smart energy meters based on an electricity consumption acquisition terminal. Background Technology

[0002] With the deepening of smart grid construction, smart meter remote monitoring systems have been widely used. These systems can remotely and automatically collect operating data such as voltage and current from a large number of users, providing a data foundation for power supply companies to achieve line loss analysis, load monitoring, and power quality monitoring. However, existing monitoring functions mainly rely on simple judgments of single data thresholds (such as voltage exceeding limits), and their analytical dimensions and diagnostic depth are still significantly limited.

[0003] Currently, the monitoring of hidden faults on the user side (such as poor service line contact and insulation aging) and abnormal behaviors such as high-tech electricity theft has significant technical shortcomings. Traditional methods mostly rely on manual on-site inspections or isolated voltage over-limit alarms, which have the following pain points: extremely high false alarm rate, unable to effectively distinguish between normal deviations caused by overall voltage fluctuations in the distribution area and genuine local anomalies on the user side; weak location capability, even if an anomaly is detected, it is difficult to accurately locate the specific abnormal user and its power supply phase from hundreds or thousands of users; and insensitivity to hidden anomalies, many initial faults or carefully designed electricity theft behaviors still have voltage values ​​within the acceptable range, only showing a trend change from their own historical normal operation mode. Existing monitoring methods based on fixed thresholds are completely ineffective in this regard, resulting in a large number of anomalies going undetected in a timely manner. Therefore, how to achieve accurate location of hidden anomalies on the user side and improve the level of lean operation and maintenance of distribution networks has become a challenge for the industry. Summary of the Invention

[0004] This application provides a remote monitoring system and method for smart energy meters based on an electricity consumption acquisition terminal, which can accurately locate hidden anomalies on the user side and improve the level of lean operation and maintenance of distribution networks.

[0005] In a first aspect, this application provides a method for remote monitoring of smart energy meters based on an electricity consumption data acquisition terminal, comprising the following steps: Based on the historical operating data of each smart energy meter, a corresponding voltage quality operating baseline is constructed to characterize the normal voltage range and fluctuation characteristics under different load segment conditions. During real-time monitoring, the power consumption acquisition terminal compares the measured voltage data of each smart energy meter with its voltage quality operating baseline. When the measured voltage of the target energy meter continuously deviates from the baseline range and its deviation direction is inconsistent with the voltage change trend at the beginning of the distribution area, it is marked as a suspected abnormal voltage quality node. For nodes suspected of having abnormal voltage quality, the power acquisition terminal extracts the phase line loss rate of the power supply branch within the same time window and calculates the correlation deviation value of the phase line loss rate relative to the historical data of the same period. When the correlation deviation value exceeds the preset threshold, it is determined that the voltage abnormality corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameter of its branch. Based on this, it is identified as a hidden fault or abnormal power consumption behavior on the user side, and a joint alarm information is generated to locate the specific user and branch.

[0006] In this embodiment, constructing the corresponding voltage quality operating baseline based on the historical operating data of each smart energy meter specifically includes: For each smart energy meter, the historical voltage data of the smart energy meter is segmented according to time and load to obtain multiple voltage sample sets for load segments; Statistical analysis was performed on the voltage sample set for each load segment, and the upper and lower quartiles of its voltage value were calculated. Based on the upper and lower quartiles, the normal operating voltage range for each load segment is determined; Establish a mapping relationship between each load segment and its corresponding normal operating voltage range; The voltage quality operating baseline for each smart energy meter is determined based on all the mapping relationships.

[0007] In this embodiment, the electricity data acquisition terminal compares the measured voltage data of each smart meter with its voltage quality operating baseline, specifically including: For each smart meter, obtain the real-time voltage and corresponding real-time load value of the smart meter; Based on the real-time load value, query the voltage quality operating baseline corresponding to the smart energy meter to determine the normal operating voltage range that should be mapped at present; Determine whether the real-time voltage is outside the normal operating range of the voltage for N consecutive sampling periods. If so, it is determined to be outside the baseline range, where N≥3.

[0008] In this embodiment, when the measured voltage of the target electricity meter continuously deviates from the baseline range, and its deviation direction is inconsistent with the voltage change trend at the beginning of the distribution area, it is marked as a suspected voltage quality anomaly node, specifically including: Acquire the voltage data at the beginning of the distribution area that is synchronized with the monitoring time of the target electricity meter; Calculate the voltage change trend coefficient of the target energy meter's measured voltage during the continuous deviation period; Calculate the trend coefficient of the voltage change at the beginning of the transformer area within the same period; If the sign of the voltage change trend coefficient is opposite to that of the first-end voltage change trend coefficient, or if the ratio of their absolute values ​​is greater than a preset proportional threshold, then it is determined that the deviation direction is inconsistent. Target energy meters whose measured voltage continuously deviates from the baseline range and whose deviation direction is inconsistent with the voltage change trend at the beginning of the transformer area are marked as suspected abnormal voltage quality nodes.

[0009] In this embodiment, for nodes suspected of having abnormal voltage quality, the power acquisition terminal extracts the phase line loss rate of the power supply branch within the same time window, and calculates the correlation deviation value of the phase line loss rate relative to historical data for the same period, specifically including: Determine the power supply branch and phase to which the suspected voltage quality anomaly node belongs; The power supply amount of the corresponding phase of the power supply branch within the same time window and the total sales volume of all energy meters under that phase are obtained from the power consumption information collection unit, and the current phase line loss rate is calculated. From the historical database, extract the historical phase line loss rate samples from the M time windows that are closest to the historical load conditions and the current load conditions during the same historical period, where M≥3; Calculate the inverse cosine similarity between the current phase line loss rate and the set of M historical phase line loss rate samples, and use the calculation result as the correlation deviation value of the phase line loss rate relative to the historical data of the same period.

[0010] In this embodiment, when the correlation deviation value exceeds a preset threshold, determining that the voltage anomaly corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameter of its branch specifically includes: Establish a historical regression model of the phase line loss rate of the power supply branch and the total load current; The total load current of the current time window is input into the historical regression model to predict the expected line loss rate under historical normal operating conditions. Calculate the residual between the current measured phase loss rate and the expected line loss rate; When the correlation deviation exceeds the first threshold and the residual exceeds the second threshold, it is determined that a non-load-driven statistical distortion has occurred.

[0011] In this embodiment, the identification of hidden faults or abnormal power consumption behavior on the user side and the generation of joint alarm information located to specific users and branches specifically include: Based on the characteristics of voltage deviation and line loss distortion, a preset fault-abnormal mode is matched to determine the abnormality type; Generate an alarm message containing the anomaly type, anomaly user identifier, power supply branch and phase, voltage deviation, and line loss increment; The alarm message is pushed to the designated interface of the power distribution system or the electricity inspection system.

[0012] Secondly, this application provides a remote monitoring system for smart energy meters based on an electricity consumption data acquisition terminal, comprising: The baseline construction module is used to construct the corresponding voltage quality operating baseline based on the historical operating data of each smart energy meter, which is used to characterize the normal voltage range and fluctuation characteristics under different load segment conditions. The abnormal node marking module is used to compare the measured voltage data of each smart energy meter with its voltage quality operating baseline during real-time monitoring. When the measured voltage of the target energy meter continues to deviate from the baseline range and its deviation direction is inconsistent with the voltage change trend at the beginning of the distribution area, it is marked as a suspected abnormal voltage quality node. The phase line loss analysis module is used to extract the phase line loss rate of the power supply branch within the same time window for nodes with suspected voltage quality abnormalities, and to calculate the correlation deviation value of the phase line loss rate relative to the historical data of the same period. The anomaly determination module is used to determine that when the correlation deviation value exceeds a preset threshold, the voltage anomaly corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameter of its branch. Based on this, it is identified as a hidden fault or abnormal power consumption behavior on the user side, and a joint alarm information is generated to locate the specific user and branch.

[0013] Thirdly, this application provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described remote monitoring method for smart energy meters based on an electricity consumption acquisition terminal.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described remote monitoring method for smart energy meters based on an electricity consumption acquisition terminal.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: In this embodiment, multi-dimensional data correlation analysis enables accurate diagnosis of user-side anomalies in low-voltage distribution networks. First, a voltage quality operating baseline is constructed based on historical operating data from smart meters. This baseline provides a core criterion for accurately identifying early anomalies deviating from the inherent operating patterns of users, laying the data foundation for the entire analysis process. Second, during real-time monitoring, the measured voltage of smart meters is compared with the baseline, and combined with the voltage trend at the head of the distribution area, suspected anomaly nodes are marked. Marking suspected anomaly nodes effectively decouples local anomalies on the user side from global voltage fluctuations in the power grid, completing the initial localization of anomalies from a large number of users to specific branch nodes, significantly improving the directionality of alarm signals. Then, for suspected voltage anomaly nodes, their branch line loss rate is extracted, and the correlation deviation value with the historical same period is calculated. By calculating the correlation deviation value, voltage anomalies can be substantially verified from the perspective of energy loss, thereby forcibly linking surface phenomena (voltage deviation) with physical essence (abnormal line loss), forming a strong chain of evidence for judging hidden faults or electricity theft. Finally, when the correlation deviation value exceeds the threshold, it is determined that the voltage anomaly is accompanied by non-load-driven line loss distortion, thereby identifying hidden faults or abnormal electricity consumption on the user side. This enables a leap from anomaly alarm to cause inference and generates precise instructions that can guide on-site actions and pinpoint specific users and phases. In summary, this application's solution, through a progressive analysis architecture of "dynamic baseline perception - trend collaborative screening - line loss correlation confirmation," achieves precise positioning of hidden anomalies on the user side, thereby promoting the transformation of distribution network operation and maintenance from a reactive post-event handling model to a proactive and precise pre-event control model, and improving the level of lean operation and maintenance. Attached Figure Description

[0016] Figure 1 This is an exemplary flowchart of a remote monitoring method for smart energy meters based on an electricity consumption acquisition terminal, according to some embodiments of this application. Figure 2 This is a flowchart illustrating the process of determining the correlation deviation value according to some embodiments of this application; Figure 3 This is a schematic diagram of the structure of a remote monitoring system for smart energy meters based on an electricity consumption acquisition terminal, according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a computer device that implements a remote monitoring method for smart meters based on an electricity consumption acquisition terminal, according to some embodiments of this application. Detailed Implementation

[0017] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] refer to Figure 1The figure is an exemplary flowchart of a remote monitoring method for smart meters based on an electricity consumption data acquisition terminal, according to some embodiments of this application. This remote monitoring method for smart meters based on an electricity consumption data acquisition terminal mainly includes the following steps: In step 101, based on the historical operating data of each smart energy meter, a corresponding voltage quality operating baseline is constructed to characterize the normal voltage range and fluctuation characteristics under different load segment conditions.

[0019] In this embodiment, the construction of a corresponding voltage quality operating baseline based on the historical operating data of each smart energy meter can be achieved through the following steps: For each smart energy meter, the historical voltage data of the smart energy meter is segmented according to time and load to obtain multiple voltage sample sets for load segments; Statistical analysis was performed on the voltage sample set for each load segment, and the upper and lower quartiles of its voltage value were calculated. Based on the upper and lower quartiles, the normal operating voltage range for each load segment is determined; Establish a mapping relationship between each load segment and its corresponding normal operating voltage range; The voltage quality operating baseline for each smart energy meter is determined based on all the mapping relationships.

[0020] It should be noted that the upper and lower quartiles in this application are statistical quantities used to describe the dispersion and central range of the voltage sample set, respectively referring to the voltage values ​​at the 25th and 75th percentiles after sorting the samples by numerical value; the normal operating voltage range is a benchmark used to judge whether the real-time voltage is abnormal, referring to the numerical range that characterizes normal voltage fluctuations within a specific load segment, determined based on historical statistical patterns; the voltage quality operating baseline is a complete benchmark model used by smart meters to judge voltage anomalies in real-time monitoring, referring to the set composed of the mapping relationships corresponding to all load segments of the energy meter.

[0021] In specific implementation, firstly, historical operating data of the target smart energy meter within a preset historical period is obtained from the electricity information collection system. This historical operating data includes at least a timestamp, voltage measurement value, and corresponding active power value. Load segmentation thresholds and typical time period division rules are set. Typical time periods include one or more combinations of division by season, by weekday and holiday type, and by daily time period. In a preferred embodiment, based on the load current distribution characteristics of the target smart energy meter within the historical period, all historical load current values ​​are statistically sorted and divided into four load intervals according to the equal-frequency segmentation principle. The sum of the sample counts in each interval and the total sample count is strictly guaranteed to be 100%. Each load interval contains 25% of the total sample count, achieving equal-frequency distribution. When the sample count in a load interval is lower than a preset minimum sample threshold (the minimum sample threshold is set to 5% of the total sample count, and not less than 50 valid data points), it is merged with the adjacent interval with the largest sample count. The upper and lower boundaries of each load interval are determined by the active power value at the corresponding quantile point and stored as load segmentation thresholds. When the sample count in a load interval... When the voltage is below a preset minimum sample threshold, adjacent intervals are merged. Furthermore, based on the active power value, historical data is categorized into multiple preset consecutive load intervals. Combined with the typical time period, all historical voltage data points belonging to the same load interval and the same typical time period are aggregated to obtain voltage sample sets for multiple load segments corresponding to the smart meter. Next, for each load segment's voltage sample set, all voltage values ​​within the set are sorted by numerical value. Based on the sorted voltage value sequence, the lower quartile and upper quartile of the sequence are calculated. The voltage range is calculated as follows: the lower quartile refers to the voltage value at the 25th percentile, and the upper quartile refers to the voltage value at the 75th percentile. Then, based on the obtained lower and upper quartiles, the normal operating voltage range corresponding to the load segment is determined. In one specific implementation, the upper quartile Q3 and lower quartile Q1 of the voltage samples for each load segment are first calculated to obtain the interquartile range IQR = Q3 - Q1. Then, the statistical reference range is calculated using the statistical standard box plot outlier judgment rules: lower limit = Q1 - 1.5 × IQR, upper limit = Q3 + 1.5×IQR; The intersection of the above statistical reference interval and the national standard legal voltage qualified interval is taken, and the intersection result is finally used as the normal operating voltage interval under this load segment. This not only conforms to the historical fluctuation characteristics of individual meters but also complies with national mandatory standards. This is only an example and is not intended to limit the specific scope of the invention. Furthermore, a unique load segment identifier is generated for each load segment. The load segment identifier is composed of a combination of the typical time period code and the load interval code corresponding to the segment. Each load segment identifier is then associated with its determined normal operating voltage interval, forming a mapping relationship record between load segments and normal operating voltage intervals. Finally, all mapping relationship records established by the target smart meter are summarized, and these records are organized into a structured data set. This data set is then bound to the unique asset identifier of the target smart meter and persistently stored in the electricity consumption acquisition terminal or remote server. Finally, this data set is used as the voltage quality operating baseline corresponding to the smart meter.

[0022] It should also be noted that the core innovation of this application in constructing a voltage quality operating baseline lies in the use of a method combining "time-load" dual-condition segmentation with robust statistics. Specifically: First, by jointly segmenting historical voltage data according to typical time periods and load levels, the inherent influence of periodic factors and load changes on voltage is effectively removed, enabling the baseline to accurately correspond to specific operating conditions. Second, at the statistical level, the upper and lower quartiles of voltage samples within each segment are used to determine the normal operating voltage range. This method is highly robust to occasional outliers in historical data, thereby ensuring that the baseline can stably and accurately depict the concentrated distribution range of voltage under normal conditions, avoiding the baseline range distortion caused by outlier interference in the traditional mean-standard deviation method. Finally, by establishing a set of mapping relationships from load segments to normal operating voltage ranges, a personalized and structured dynamic baseline model is formed. The beneficial effects of this are as follows: First, because the baseline dynamically matches the historical operating characteristics and real-time operating conditions of individual energy meters, the accuracy and sensitivity of voltage anomaly identification are significantly improved, and subtle deviations relative to their normal mode can be effectively captured; Second, based on a robust statistical and load-correlated baseline model, it is possible to fundamentally distinguish between voltage changes caused by normal load fluctuations and real anomalies, thereby greatly reducing the false alarm rate of the system.

[0023] In step 102, during real-time monitoring, the power acquisition terminal compares the measured voltage data of each smart energy meter with its voltage quality operating baseline. When the measured voltage of the target energy meter continuously deviates from the baseline range and its deviation direction is inconsistent with the voltage change trend at the beginning of the distribution area, it is marked as a suspected abnormal voltage quality node.

[0024] In this embodiment, the comparison between the measured voltage data of each smart energy meter and its voltage quality operating baseline by the power consumption acquisition terminal can be achieved through the following steps: For each smart meter, obtain the real-time voltage and corresponding real-time load value of the smart meter; Based on the real-time load value, query the voltage quality operating baseline corresponding to the smart energy meter to determine the normal operating voltage range that should be mapped at present; Determine whether the real-time voltage is outside the normal operating range of the voltage for N consecutive sampling periods. If so, it is determined to be outside the baseline range, where N≥3.

[0025] It should be noted that the sampling period in this application refers to the time interval between the electricity data acquisition terminal reading voltage and load data from the smart energy meter.

[0026] In practice, firstly, the electricity consumption data acquisition terminal, through its communication module, initiates a data reading request to the target smart meter at a preset fixed collection cycle (e.g., every 15 minutes). The target smart meter responds to the request by replying to the electricity consumption data acquisition terminal via the communication link with its currently measured effective voltage value and current active power value. The electricity consumption data acquisition terminal receives the reply and records the latest received effective voltage value as the real-time voltage and the latest received active power value as the real-time load value, simultaneously adding the same timestamp to both sets of data to form a real-time data record. Secondly, the electricity consumption data acquisition terminal retrieves pre-built and stored data specific to the target smart meter from its local storage or a connected remote database. The voltage quality operating baseline of the smart energy meter is established by comparing the acquired real-time load value with all load segment ranges defined in the voltage quality operating baseline one by one to determine which load segment the real-time load value falls into. After determining the load segment, according to the mapping relationship stored in the voltage quality operating baseline, the upper and lower voltage limits uniquely corresponding to that load segment are directly read. A closed interval is formed with the upper voltage limit as the boundary upper limit and the lower voltage limit as the boundary lower limit, and this closed interval is determined as the normal operating voltage range that should be mapped under the current operating conditions. Then, the power acquisition terminal maintains a first-in-first-out data buffer queue to store data from multiple consecutive acquisition cycles. To determine intermediate results, the terminal compares each real-time voltage reading with the corresponding normal operating voltage range, determining whether the real-time voltage is less than the lower limit or greater than the upper limit of the range. This Boolean logic result ("Yes" or "No") is stored in the queue. The power consumption data acquisition terminal continuously monitors the queue. When the latest consecutive N (N is a preset integer, and N≥3) logic results in the queue are all "Yes", a final determination is generated, classifying the current voltage state of the target smart energy meter as deviating from the baseline range. If at least one of the latest consecutive N logic results in the queue is "No", it is determined that there is no deviation. Finally, the power consumption data acquisition terminal uses the deviation from the baseline range conclusion generated in this determination as... This is a flag signal that triggers subsequent advanced diagnostic procedures. It should be further explained that the value of N is determined by the real-time requirements of voltage anomaly monitoring and the characteristics of signal noise, according to the following quantization rules: when the sampling period T ≤ 5 minutes, N = 5, and the judgment time window is 5 × T, balancing instantaneous disturbance filtering and anomaly response speed; when 5 minutes < T ≤ 10 minutes, N = 4; when T > 10 minutes, N = 3. It should also be noted that before trend fitting, a moving average filter is applied to the voltage sequence. The filter window size is set to twice the sampling period, and the number of data points within the window is odd (e.g., 3-point or 5-point moving averages). This filters only power frequency noise and instantaneous spikes, preserving the true voltage change trend and avoiding distortion of the trend coefficient.

[0027] In this embodiment, when the measured voltage of the target energy meter continuously deviates from the baseline range, and its deviation direction is inconsistent with the voltage change trend at the beginning of the distribution area, it can be marked as a suspected voltage quality abnormality node by the following steps: Acquire the voltage data at the beginning of the distribution area that is synchronized with the monitoring time of the target electricity meter; Calculate the voltage change trend coefficient of the target energy meter's measured voltage during the continuous deviation period; Calculate the trend coefficient of the voltage change at the beginning of the transformer area within the same period; If the sign of the voltage change trend coefficient is opposite to that of the first-end voltage change trend coefficient, or if the ratio of their absolute values ​​is greater than a preset proportional threshold, then it is determined that the deviation direction is inconsistent. Target energy meters whose measured voltage continuously deviates from the baseline range and whose deviation direction is inconsistent with the voltage change trend at the beginning of the transformer area are marked as suspected abnormal voltage quality nodes.

[0028] It should be noted that, in this application, the "first end of the distribution area" refers to the upstream reference measurement node located at the low-voltage side outlet of the distribution transformer, used to characterize the overall voltage level of the entire power supply area; the voltage change trend coefficient is a trend indicator used to quantify the directionality and rate of change of voltage over time within a continuous sampling period; the "first end voltage change trend coefficient" is a trend indicator used to quantify the direction and rate of change of the overall voltage level of the power supply area over time within a continuous sampling period; the "suspected abnormal voltage quality node" refers to a user-side monitoring node whose measured voltage continuously deviates from its normal operating baseline and whose change trend is significantly inconsistent with the overall voltage change characteristics of the distribution area, requiring further diagnosis.

[0029] In specific implementation, firstly, when the power consumption acquisition terminal determines that the measured voltage of a target energy meter continuously deviates from its baseline range, the terminal immediately requests voltage data within N sampling period time windows that are identical to the continuous deviation determination conclusion triggered by the target energy meter from the distribution automation master station system or the intelligent fusion terminal installed at the low-voltage side of the distribution transformer. The voltage data at the beginning of the distribution area is a sequence of effective voltage values ​​collected according to the same sampling period, and the timestamp of each data point is aligned with the sampling timestamp of the target energy meter, thus forming a comparison benchmark sequence that is completely synchronized with the user-side voltage sequence in time. The power consumption acquisition terminal receives this sequence and caches it locally. Secondly, the power consumption acquisition terminal extracts the measured voltage values ​​of the target energy meter within N continuous deviation periods, forming a voltage time series. The terminal performs linear trend analysis on this voltage time series, specifically by fitting a straight line using the least squares method, and then using the slope of this line as... A voltage change trend coefficient is used, where a positive value indicates an upward trend in voltage during the period, and a negative value indicates a downward trend. Its absolute value reflects the rate of change. Then, the power acquisition terminal extracts the acquired voltage data sequence of the distribution area's first-end within the same N sampling period window, performs linear fitting on this sequence, calculates the slope of the fitted line, and uses this slope as the first-end voltage change trend coefficient. This first-end voltage change trend coefficient characterizes the overall direction and intensity of voltage change in the entire power supply area system within the same time period. Further, the power acquisition terminal sets a preset proportional threshold, which can be set based on historical data. The terminal first compares the signs of the voltage change trend coefficient and the first-end voltage change trend coefficient: if their signs are opposite (one positive and one negative), it is directly determined that the deviation directions are inconsistent; if their signs are the same, the ratio of their absolute values ​​is calculated and compared with the proportional threshold. If the ratio of absolute values ​​> 0. If the proportional threshold is exceeded, it is determined that the deviation direction is inconsistent, indicating that the change in voltage on the user side is significantly different from the overall change on the system side. Otherwise, it is determined that the direction is consistent, and its output is a Boolean logic value, namely "inconsistent" or "consistent". Finally, the power acquisition terminal performs a logical "AND" operation: its input is firstly the status flag of the target energy meter being triggered to continuously deviate from the baseline interval, and secondly the determination result of the deviation direction inconsistency (i.e., the Boolean logic output result). If and only if both input conditions are true at the same time, the terminal generates a high-level event marker. The terminal encapsulates the asset number of the target energy meter, the abnormal time window, the calculated voltage change trend coefficient and the head-end voltage change trend coefficient, etc., into a structured abnormal event record, and marks the energy meter as a suspected abnormal voltage quality node in the internal status table.

[0030] It should also be noted that existing technologies for determining user-side voltage anomalies largely rely on single-point threshold exceedances or simple time-series fluctuation analysis. This makes it difficult to distinguish between normal changes caused by overall voltage fluctuations in the distribution area and individual anomalies caused by local line faults, poor contact, or abnormal loads, easily leading to false alarms or missed alarms. This solution introduces a comparative analysis mechanism between the target energy meter's voltage change trend coefficient and the distribution area's head-end voltage change trend coefficient. It jointly identifies anomalies from two dimensions: temporal evolution characteristics and spatial correlation. Anomaly markers are only triggered when the direction or amplitude of user-side voltage changes significantly deviates from the overall system trend. This effectively eliminates the interference of upstream power supply fluctuations on user-side judgment, achieving accurate identification of local anomalies and significantly improving the targeting, reliability, and diagnostic effectiveness of voltage anomaly determination.

[0031] In step 103, for nodes suspected of having abnormal voltage quality, the power acquisition terminal extracts the phase line loss rate of the power supply branch within the same time window and calculates the correlation deviation value of the phase line loss rate relative to the historical data of the same period.

[0032] In this embodiment, reference Figure 2 As shown in the figure, this is a flowchart illustrating the process of determining the correlation deviation value in some embodiments of this application. In this embodiment, for nodes suspected of having abnormal voltage quality, the power acquisition terminal extracts the phase line loss rate of the power supply branch within the same time window. The correlation deviation value of the phase line loss rate relative to historical data of the same period can be calculated using the following steps: In step 1031, the power supply branch and phase to which the suspected voltage quality abnormal node belongs are determined; In step 1032, the power supply of the corresponding phase of the power supply branch within the same time window and the total sales volume of all energy meters under that phase are obtained from the power consumption information acquisition unit, and the current phase line loss rate is calculated. In step 1033, from the historical database, extract the historical phase line loss rate samples of M time windows that are closest to the historical load conditions and the current load conditions during the same historical period, where M≥3; In step 1034, the inverse cosine similarity between the current phase line loss rate and the set of M historical phase line loss rate samples is calculated, and the calculation result is used as the correlation deviation value of the phase line loss rate relative to the historical data of the same period.

[0033] It should be noted that the phase loss rate in this application is an indicator used to reflect the power transmission efficiency and abnormal loss level of a certain phase of the power supply branch during operation; the correlation deviation value is a comparative analysis indicator used to quantify the overall deviation between the current phase loss rate change pattern and the normal line loss characteristics of the same period in history, thereby characterizing the degree of abnormality in the line loss operation state.

[0034] In practice, firstly, the electricity data acquisition terminal queries the local storage or the distribution production management platform's low-voltage network topology database of the target electricity meter marked as a suspected voltage quality anomaly node, based on the asset identifier of the meter. This database records the connection relationship between each smart meter and its upstream power supply point. The terminal determines the physical power supply branch number to which the target meter is connected and the corresponding AC phase of its metering circuit through this query. Then, it associates the determined power supply branch number with the phase code to obtain a unique topology key value for data extraction. Secondly, the electricity data acquisition terminal uses the obtained topology key value to... Two key electricity data points are extracted from the frozen data storage area of ​​the electricity information collection system. The first data point is the total power supply of the corresponding phase of the power supply branch within a time window consisting of N consecutive sampling periods where the voltage anomaly occurs. This power supply comes from the frozen data of the upstream metering point (such as the energy meter on the outgoing side of the branch box) supplying power to this phase. The second data point is the total electricity sales of all user smart energy meters belonging to the same phase under this power supply branch within the same time window. This is obtained by accumulating the frozen electricity sales of these user meters. The terminal calculates the current phase line loss rate using the formula: Current phase line loss rate = (Total power supply - Total electricity sales) The total power supply is multiplied by 100% to obtain a percentage value representing the loss level of the branch phase during the current abnormal period. Then, the power acquisition terminal connects to a historical database storing long-term historical operating data. The terminal uses the calendar characteristics and operating conditions of the current abnormal time window as dual search conditions. The calendar characteristics include season category and weekday type, and the operating conditions are represented by the average total load current value of the power supply branch within the time window. In a preferred embodiment, historical operating condition matching adopts a multi-parameter comprehensive similarity evaluation method. In addition to the average total load current, the average voltage value and the ambient temperature correction coefficient are also introduced as auxiliary parameters. The parameters are normalized and weighted according to preset weights. The M historical windows with the highest comprehensive similarity are selected as matching samples. The terminal further searches the historical database for M windows of the same time period on different dates in the past year that completely match the calendar features of the current window and have the smallest absolute value of the difference between the historical average total load current value and the current value. From the search results, the phase line loss rate data corresponding to each of these M historical windows are extracted to form a historical phase line loss rate sample set containing M data points, which is used to represent the normal line loss level set of this branch phase under similar external conditions.Finally, the power acquisition terminal performs a pattern comparison between the calculated current phase line loss rate value and a historical phase line loss rate sample set containing M values. First, the current phase line loss rate value is copied M times to construct a one-dimensional vector, denoted as vector A, where all elements represent the current phase line loss rate. This vector A is constructed to treat the current line loss rate as a stable state reference vector. By comparing its overall direction with the historical line loss rate fluctuation vector, the structural differences between the current line loss pattern and the historical normal fluctuation pattern are quantified, thus avoiding random error interference from single-point value comparisons. Simultaneously, the M historical sample values ​​are sorted according to their order in the set... The sequence is arranged to construct a historical one-dimensional vector, denoted as vector B. Then, the cosine similarity between vectors A and B is calculated using the following formula: the sum of the products of corresponding elements of the two vectors, divided by the square root of the sum of the squares of the elements in vector A and the square root of the sum of the squares of the elements in vector B. Further, the reciprocal of this cosine similarity value is calculated (i.e., 1 divided by the similarity value). The reciprocal result is used as the final correlation deviation value. The larger this correlation deviation value, the greater the overall difference between the current constant line loss rate pattern and the historically varying normal line loss rate pattern, thus providing a quantitative basis for determining whether there are non-load-driven anomalies.

[0035] In step 104, when the correlation deviation value exceeds the preset threshold, it is determined that the voltage abnormality corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameter of its branch. Based on this, it is identified as a hidden fault or abnormal power consumption behavior on the user side, and a joint alarm information is generated to locate the specific user and branch.

[0036] It should be noted that this application achieves anomaly identification based on the statistical correlation between voltage anomalies and phase line loss rates. The technical principle is as follows: Under normal operating conditions, the phase line loss rate exhibits a stable statistical relationship with the change in total load current, and load fluctuations are the main driving factor. By constructing a load-line loss regression model through historical data, the desired line loss pattern can be obtained. When the voltage of the target energy meter continuously deviates from the baseline, if the correlation deviation between the current phase line loss rate pattern of its branch and the historical normal pattern exceeds a preset threshold, it indicates that the line loss change is no longer driven by the load, but rather an abnormal pattern deviation has occurred. At this time, voltage anomalies and non-load-driven line loss distortion occur simultaneously, indicating that the anomaly originates from a hidden fault on the user side or abnormal electricity consumption behavior.

[0037] In this embodiment, when the correlation deviation value exceeds a preset threshold, determining that the voltage anomaly corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameter of its branch can be achieved by the following steps: Establish a historical regression model of the phase line loss rate of the power supply branch and the total load current; The total load current of the current time window is input into the historical regression model to predict the expected line loss rate under historical normal operating conditions. Calculate the residual between the current measured phase loss rate and the expected line loss rate; When the correlation deviation exceeds the first threshold and the residual exceeds the second threshold, it is determined that a non-load-driven statistical distortion has occurred.

[0038] It should be noted that the historical regression model in this application uses the phase line loss rate of the power supply branch as the dependent variable and the total load current as the independent variable. It is a statistical mapping relationship obtained by fitting historical normal operation data to characterize the typical driving effect of load changes on the line loss rate. Its technical principle is as follows: Under stable operation, line loss is mainly affected by the load level. The phase line loss rate changes in a predictable pattern with the total load. By inputting the total load current of the current time window into the regression model, the expected line loss rate under the same load conditions can be predicted. The expected line loss rate refers to the theoretical line loss level of the power supply branch predicted based on historical normal operation patterns under given load conditions, thus forming a normal reference based on load driving. By comparing the measured line loss rate with the expected value and calculating the residual, the magnitude of the current line loss deviation from the normal pattern can be quantified.

[0039] In practice, firstly, modeling samples are extracted from the normal operation dataset of the historical database. Each sample contains two key parameters within a historical time window: the average phase line loss rate within that window and the average total load current of the corresponding phase of the power supply branch within that window. After collecting a sufficient number of samples, the terminal uses the least squares method, with the average total load current as the independent variable (X) and the average phase line loss rate as the dependent variable (Y), to fit a linear equation Y = a * X + b, where a is the slope and b is the intercept. This linear equation is the historical regression model describing the statistical relationship between the line loss rate and load current of the branch phase under historical normal conditions. Secondly, the average total load current of the corresponding phase of the power supply branch is calculated within the current time window when the voltage anomaly occurs. The terminal calls the historical regression model corresponding to that phase and substitutes the calculated current average total load current value into the linear equation Y = a * X + b. In step b, X performs a calculation, taking the Y value output by the model as the theoretical line loss rate predicted based on historical normal patterns under the current specific load conditions. This value is defined as the expected line loss rate. Then, the absolute difference between the current measured phase line loss rate and the expected line loss rate is calculated, i.e., the residual is calculated as |current measured phase line loss rate - expected line loss rate|. This residual quantifies the absolute magnitude of the deviation of the actual line loss rate from its theoretical prediction. Finally, the terminal presets two independent judgment thresholds: a first threshold for evaluating pattern similarity and a second threshold for evaluating numerical deviation. The terminal compares the previously calculated correlation deviation value with the first threshold and the calculated residual with the second threshold. The terminal performs an AND logic judgment: the terminal makes a final judgment only if the correlation deviation value is greater than the first threshold (indicating an abnormal line loss change pattern) and the residual is greater than the second threshold (indicating an abnormal absolute level of line loss). At this point, it is determined that the voltage anomaly event corresponding to the suspected voltage quality anomaly node is indeed accompanied by a non-load-driven statistical change in the line loss parameters of its branch. Regarding the distortion, it needs further explanation that the first threshold is a quantitative boundary used to determine whether the change pattern of the line loss rate significantly deviates from the historical normal. It is set by statistically analyzing the correlation deviation values ​​of a large number of time windows during the historical normal period and selecting its high percentile (e.g., 95th percentile). The second threshold is a quantitative boundary used to determine whether the line loss rate value significantly deviates from the normal range based on load forecasting. It is set by analyzing the prediction residual distribution of the historical regression model and taking K times (K>2) of the historical residual standard deviation or a fixed proportion of the historical average line loss rate. The first and second thresholds adopt a sliding time window statistical update mechanism. Every preset update cycle, the normal operation data of the most recent 6 months are re-statistically analyzed, and the threshold parameters are automatically adjusted according to the 95th percentile.

[0040] In this embodiment, identifying hidden faults or abnormal power consumption behavior on the user side and generating joint alarm information located to specific users and branches can be achieved through the following steps: Based on the characteristics of voltage deviation and line loss distortion, a preset fault-abnormal mode is matched to determine the abnormality type; Generate an alarm message containing the anomaly type, anomaly user identifier, power supply branch and phase, voltage deviation, and line loss increment; The alarm message is pushed to the designated interface of the power distribution system or the electricity inspection system.

[0041] In practice, firstly, upon confirming statistical distortion in line loss, the system compares the voltage deviation characteristics and line loss increment of the current event with a pre-defined fault-anomaly rule base. This rule base is constructed based on historical fault work order data and data annotated by human experts, and is dynamically optimized and updated through a continuous feedback mechanism. Each rule is assigned a priority identifier. When multiple matching rules exist, the anomaly type with the highest priority is output first. The rule base predefines anomaly categories corresponding to different feature combinations. For example, if the feature is low voltage accompanied by a high line loss increment, it matches as poor contact or aging of the service line; if the feature is normal voltage but a very large line loss increment, it matches as suspected electricity theft. The system outputs a definite anomaly type label through rule matching. Secondly, based on the matching result, the system automatically generates a structured alarm message. This information explicitly includes: the identified anomaly type, the specific user identifier (account number and meter number) of the anomaly, the power supply branch and phase to which it belongs, and the voltage deviation and line loss increment as quantitative evidence. This information is formatted into a standard alarm message. Then, based on the anomaly type, the system automatically selects the target business system for alarm push. Equipment fault alarms are pushed to the power distribution operation and maintenance management system, and abnormal electricity consumption alarms are pushed to the electricity consumption inspection system. The alarm message is automatically delivered through the standard data interface provided by the target system. The data interface adopts a standard communication method based on HTTP or MQTT protocol. The alarm message is encapsulated in JSON or XML format and a failure retransmission mechanism and an authentication mechanism are set to ensure the reliability and security of alarm information transmission, thereby completing a closed-loop process from fault diagnosis to work order triggering.

[0042] On the other hand, in some embodiments, this application provides a remote monitoring system for smart energy meters based on an electricity consumption data acquisition terminal, referencing... Figure 3 The figure is a schematic diagram of the structure of a remote monitoring system for smart energy meters based on an electricity consumption acquisition terminal, according to some embodiments of this application. The system includes: a baseline construction module 301, an abnormal node marking module 302, a phase line loss analysis module 303, and an anomaly determination module 304, which are described below: Baseline construction module 301 is mainly used to construct the corresponding voltage quality operation baseline based on the historical operation data of each smart energy meter, and to characterize the normal voltage range and fluctuation characteristics under different load segment conditions. The abnormal node marking module 302 is mainly used in the real-time monitoring process. The power acquisition terminal compares the measured voltage data of each smart energy meter with its voltage quality operating baseline. When the measured voltage of the target energy meter continues to deviate from the baseline range and its deviation direction is inconsistent with the voltage change trend at the beginning of the distribution area, it is marked as a suspected abnormal node in voltage quality. The phase line loss analysis module 303 is mainly used to extract the phase line loss rate of the power supply branch within the same time window for nodes with suspected abnormal voltage quality, and to calculate the correlation deviation value of the phase line loss rate relative to the historical data of the same period. The anomaly determination module 304 is mainly used to determine that when the correlation deviation value exceeds a preset threshold, the voltage anomaly corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameter of its branch. Based on this, it is identified as a hidden fault or abnormal power consumption behavior on the user side, and a joint alarm information is generated to locate the specific user and branch.

[0043] In addition, this application also provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described remote monitoring method for smart energy meters based on an electricity consumption acquisition terminal.

[0044] In some embodiments, reference Figure 4 This figure is a schematic diagram of the structure of a computer device implementing a remote monitoring method for smart meters based on an electricity consumption data acquisition terminal, according to some embodiments of this application. The remote monitoring method for smart meters based on an electricity consumption data acquisition terminal in the above embodiments can... Figure 4 The computer device shown is used to implement this, and the computer device 400 includes at least one processor 401, a communication bus 402, a memory 403, and at least one communication interface 404.

[0045] Processor 401 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0046] The communication bus 402 can be used to transmit information between the aforementioned components.

[0047] The memory 403 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 403 may exist independently and be connected to the processor 401 via the communication bus 402. The memory 403 may also be integrated with the processor 401.

[0048] The memory 403 stores program code for executing the scheme of this application, and its execution is controlled by the processor 401. The processor 401 executes the program code stored in the memory 403. The program code may include one or more software modules. In the above embodiments, the remote monitoring method of smart energy meters based on electricity consumption acquisition terminals can be implemented by the processor 401 and one or more software modules in the program code in the memory 403.

[0049] Communication interface 404 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0050] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0051] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0052] In addition, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described remote monitoring method for smart energy meters based on an electricity consumption acquisition terminal.

[0053] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0054] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for remote monitoring of smart energy meters based on an electricity consumption data acquisition terminal, characterized in that, include: Based on the historical operating data of each smart energy meter, a corresponding voltage quality operating baseline is constructed to characterize the normal voltage range and fluctuation characteristics under different load segment conditions. During real-time monitoring, the power consumption acquisition terminal compares the measured voltage data of each smart energy meter with its voltage quality operating baseline. When the measured voltage of the target energy meter continues to deviate from the baseline range and its deviation direction is inconsistent with the voltage change trend at the beginning of the distribution area, it is marked as a suspected abnormal voltage quality node. For nodes suspected of having abnormal voltage quality, the power acquisition terminal extracts the phase line loss rate of the power supply branch within the same time window and calculates the correlation deviation value of the phase line loss rate relative to the historical data of the same period. When the correlation deviation value exceeds the preset threshold, it is determined that the voltage abnormality corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameter of its branch. Based on this, it is identified as a hidden fault or abnormal power consumption behavior on the user side, and a joint alarm information is generated to locate the specific user and branch.

2. The method as described in claim 1, characterized in that, Based on the historical operating data of each smart energy meter, the corresponding voltage quality operating baseline is constructed, specifically including: For each smart energy meter, the historical voltage data of the smart energy meter is segmented according to time and load to obtain multiple voltage sample sets for load segments; Statistical analysis was performed on the voltage sample set for each load segment, and the upper and lower quartiles of its voltage values ​​were calculated. Based on the upper and lower quartiles, the normal operating voltage range for each load segment is determined; Establish a mapping relationship between each load segment and its corresponding normal operating voltage range; The voltage quality operating baseline for each smart energy meter is determined based on all the mapping relationships.

3. The method as described in claim 1, characterized in that, The electricity data acquisition terminal compares the measured voltage data of each smart meter with its voltage quality operating baseline, specifically including: For each smart meter, obtain the real-time voltage and corresponding real-time load value of the smart meter; Based on the real-time load value, query the voltage quality operating baseline corresponding to the smart energy meter to determine the normal operating voltage range that should be mapped at present; Determine whether the real-time voltage is outside the normal operating range of the voltage for N consecutive sampling periods. If so, it is determined to be outside the baseline range, where N≥3.

4. The method as described in claim 1, characterized in that, When the measured voltage of the target electricity meter continuously deviates from the baseline range, and the direction of deviation is inconsistent with the voltage change trend at the beginning of the distribution area, it is marked as a suspected voltage quality anomaly node. Specifically, this includes: Acquire the voltage data at the beginning of the distribution area that is synchronized with the monitoring time of the target electricity meter; Calculate the voltage change trend coefficient of the target energy meter's measured voltage during the continuous deviation period; Calculate the trend coefficient of the voltage change at the beginning of the transformer area within the same period; If the sign of the voltage change trend coefficient is opposite to that of the first-end voltage change trend coefficient, or if the ratio of their absolute values ​​is greater than a preset proportional threshold, then it is determined that the deviation direction is inconsistent. Target energy meters whose measured voltage continuously deviates from the baseline range and whose deviation direction is inconsistent with the voltage change trend at the beginning of the transformer area are marked as suspected abnormal voltage quality nodes.

5. The method as described in claim 1, characterized in that, For nodes suspected of having abnormal voltage quality, the power consumption data acquisition terminal extracts the phase line loss rate of the power supply branch within the same time window and calculates the correlation deviation value of the phase line loss rate relative to historical data for the same period. Specifically, this includes: Determine the power supply branch and phase to which the suspected voltage quality anomaly node belongs; The power supply amount of the corresponding phase of the power supply branch within the same time window and the total sales volume of all energy meters under that phase are obtained from the power consumption information collection unit, and the current phase line loss rate is calculated. From the historical database, extract the historical phase line loss rate samples from the M time windows that are closest to the historical load conditions and the current load conditions during the same historical period, where M≥3; Calculate the inverse cosine similarity between the current phase line loss rate and the set of M historical phase line loss rate samples, and use the calculation result as the correlation deviation value of the phase line loss rate relative to the historical data of the same period.

6. The method as described in claim 1, characterized in that, When the correlation deviation value exceeds a preset threshold, determining that the voltage anomaly corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameters of its branch includes: Establish a historical regression model of the phase line loss rate of the power supply branch and the total load current; The total load current of the current time window is input into the historical regression model to predict the expected line loss rate under historical normal operating conditions. Calculate the residual between the current measured phase loss rate and the expected line loss rate; When the correlation deviation exceeds the first threshold and the residual exceeds the second threshold, it is determined that a non-load-driven statistical distortion has occurred.

7. The method as described in claim 1, characterized in that, This identifies hidden faults or abnormal power consumption behavior on the user side, and generates joint alarm information pinpointing the specific user and branch, including: Based on the characteristics of voltage deviation and line loss distortion, a preset fault-abnormal mode is matched to determine the abnormality type; Generate an alarm message containing the anomaly type, anomaly user identifier, power supply branch and phase, voltage deviation, and line loss increment; The alarm message is pushed to the designated interface of the power distribution system or the electricity inspection system.

8. A remote monitoring system for smart energy meters based on an electricity consumption data acquisition terminal, characterized in that, include: The baseline construction module is used to construct the corresponding voltage quality operating baseline based on the historical operating data of each smart energy meter, which is used to characterize the normal voltage range and fluctuation characteristics under different load segment conditions. The abnormal node marking module is used to compare the measured voltage data of each smart energy meter with its voltage quality operating baseline during real-time monitoring. When the measured voltage of the target energy meter continues to deviate from the baseline range and its deviation direction is inconsistent with the voltage change trend at the beginning of the distribution area, it is marked as a suspected abnormal node in voltage quality. The phase line loss analysis module is used to extract the phase line loss rate of the power supply branch within the same time window for nodes with suspected voltage quality abnormalities, and to calculate the correlation deviation value of the phase line loss rate relative to the historical data of the same period. The anomaly determination module is used to determine that when the correlation deviation value exceeds a preset threshold, the voltage anomaly corresponding to the suspected abnormal node is accompanied by a non-load-driven statistical distortion of the line loss parameter of its branch. Based on this, it is identified as a hidden fault or abnormal power consumption behavior on the user side, and a joint alarm information is generated to locate the specific user and branch.

9. A computer device comprising a memory and a processor, the memory storing code, characterized in that, The processor is configured to acquire the code and execute the remote monitoring method for smart energy meters based on an electricity consumption acquisition terminal as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the remote monitoring method for smart energy meters based on an electricity consumption acquisition terminal as described in any one of claims 1 to 7.

Images