CN118211837A - Performance evaluation method and device for intelligent ammeter - Google Patents

Abstract

The invention relates to the field of intelligent electric meters, and discloses a performance evaluation method and device of an intelligent electric meter, wherein the method comprises the following steps: taking the ammeter environment data, ammeter operation data and ammeter maintenance data as key evaluation indexes of performance evaluation, and acquiring time sequence data of the key evaluation indexes; establishing a performance attenuation model, inputting the time sequence data into the performance attenuation model, and obtaining a natural attenuation mode of the ammeter performance attenuation; and identifying abnormal fluctuation in the fluctuation trend of the intelligent ammeter, carrying out risk assessment on the abnormal fluctuation, judging an abnormal reason, and judging whether the abnormal reason is a tampering behavior. According to the invention, the electric meter environment data, the electric meter operation data and the electric meter maintenance data are used as key evaluation indexes for performance evaluation, so that the natural attenuation mode of the electric meter performance is obtained in consideration of the influence of environment and aging on the performance of the electric meter, and further the tamper behavior of hacker drip is identified.

Description

Performance evaluation method and device for intelligent ammeter

Technical Field

The invention relates to the field of intelligent electric meters, in particular to a performance evaluation method and device of an intelligent electric meter.

Background

The intelligent ammeter is a novel ammeter integrating functions of energy metering, data acquisition, remote communication, control and the like, and adopts an advanced communication technology, so that the intelligent ammeter can be remotely communicated with a management system of an electric company or a user, and functions of real-time data transmission, remote meter reading, remote control and the like can be realized.

The intelligent ammeter performance evaluation can be remotely and real-timely carried out through the communication technology of the intelligent ammeter, most performance evaluation methods only evaluate ammeter operation data, influence of other factors on ammeter performance is ignored, so that the ammeter performance evaluation is inaccurate, whether the performance of the intelligent ammeter is reduced from a normal aging process or influenced by malicious tampering behaviors is difficult to judge through performance evaluation results, and the recognition accuracy of the ammeter tampering behaviors is low.

Disclosure of Invention

The invention provides a performance evaluation method of a smart electric meter, which solves the problem that the smart electric meter has low accuracy in recognizing the tampering behavior of the electric meter.

According to an aspect of the present invention, there is provided a performance evaluation method of a smart meter, including: taking the ammeter environment data, ammeter operation data and ammeter maintenance data as key evaluation indexes of performance evaluation, and acquiring time sequence data of the key evaluation indexes; establishing a performance attenuation model, wherein the performance attenuation model applies a time sequence analysis technology to correlate the working time length of an ammeter with the performance attenuation of the ammeter, and inputs the time sequence data into the performance attenuation model to acquire a natural attenuation mode of the performance attenuation of the ammeter; and obtaining the fluctuation trend of the performance attenuation of the ammeter according to the natural attenuation mode, identifying abnormal fluctuation in the fluctuation trend, carrying out risk assessment on the abnormal fluctuation and judging an abnormal reason, and judging whether the abnormal reason is a tampering behavior.

Further, the inputting the time series data into the performance attenuation model, and obtaining a natural attenuation mode of the ammeter performance attenuation specifically includes: and respectively inputting the ammeter environment data, the ammeter operation data and the time sequence data of the ammeter maintenance data into the performance attenuation model, determining a coefficient relation between the ammeter environment data and the ammeter performance attenuation through simple regression analysis, using the time sequence analysis to correlate the ammeter operation data with the ammeter performance attenuation, and using a statistical analysis method to correlate the ammeter maintenance data with the ammeter performance attenuation so as to obtain a natural attenuation mode of the ammeter performance attenuation.

Still further, the performance decay model applies cross-validation techniques to improve evaluation performance, in particular, by partitioning the time series data into multiple test sets to detect stability and accuracy of the performance decay model.

Furthermore, the performance attenuation model simulates the environmental condition of the electric meter, predicts the performance attenuation change of the electric meter based on the environmental data of the electric meter, and identifies the performance degradation or the failure of the electric meter caused by the change of environmental factors according to the prediction result, so as to formulate a preventive maintenance strategy.

Further, the fluctuation trend of the performance attenuation of the electric meter is obtained according to the natural attenuation mode, specifically: and performing anomaly detection on the time series data, performing data cleaning on the detected anomaly data, analyzing the time series data by using a time series analysis method by using a performance attenuation model, and continuously adjusting parameters of the performance attenuation model according to seasonal and periodic fluctuation of the time series data so as to improve the prediction effect of the fluctuation trend obtained by the natural attenuation mode, and then obtaining the fluctuation trend of the performance attenuation of the ammeter.

Further, the identifying the abnormal fluctuation in the fluctuation trend, performing risk assessment on the abnormal fluctuation, and judging an abnormal reason, and judging whether the abnormal reason is a tampering behavior specifically includes: determining a threshold value of normal fluctuation of current ammeter performance attenuation, wherein the threshold value is obtained after statistical analysis is carried out based on the time sequence data and the fluctuation trend, if fluctuation that the current ammeter performance attenuation exceeds the threshold value is detected, the abnormal fluctuation is judged, the risk assessment is carried out, the abnormality cause is judged, and if the abnormal fluctuation or the risk assessment exceeds a preset value, the ammeter performance early warning is triggered; and feeding back the risk assessment and the abnormality reasons to a smart meter maintenance system, prompting the smart meter maintenance system to prompt the meter with reduced meter performance and feed back the meter to be maintained, and displaying the trend of reduced meter performance to maintenance personnel and operators through a data visualization technology.

Furthermore, after the intelligent ammeter maintenance system processes ammeter maintenance matters, maintenance results are collected and fed back to the performance attenuation model, and further adjustment and optimization are carried out on the performance attenuation model, so that the evaluation capability of the performance attenuation model is ensured to be more consistent with the actual situation.

Further, if the risk assessment does not meet a preset risk threshold trend result, performing adaptive correction on the risk threshold according to the deviation of the risk assessment, performing data restoration and parameter calibration operations, and correcting abnormal key assessment indexes by using a moving average method of time sequence analysis to ensure long-term accuracy of the key assessment indexes.

Further, the judging whether the abnormal cause is a tampering behavior specifically includes: judging the abnormal reason of the abnormal fluctuation, judging whether the abnormal reason is the natural attenuation of the performance of the ammeter or the tampering behavior of the ammeter, and correcting and remotely modifying the key evaluation index if the abnormal reason is the natural attenuation and the abnormal fluctuation exceeds an error range; if the tampering behavior is the tampering behavior, isolating the tampered ammeter from the network, isolating a system part affected by the tampering, recovering data of the key evaluation index, performing data audit and check, confirming the integrity and accuracy of the key evaluation index, acquiring a detailed report of a processing result, and notifying a manager of a tampering event and the detailed report.

Further, the method for judging the cause of the abnormality specifically includes: and exploring the mapping relation between the abnormal fluctuation and the abnormal cause of the known ammeter by using a correlation rule mining technology, and confirming the causal relation between the abnormal fluctuation and the abnormal cause after verifying the mapping relation by applying causal relation analysis.

Further, the performance decay model is one of an autoregressive integral moving average model, a seasonal autoregressive integral moving average model and an exponential smoothing model.

Still further, the meter environment data includes temperature, humidity, PM2.5, and noise, the meter operation data includes current, voltage, active power, reactive power, and meter load, and the meter maintenance data includes calibration times, fault times, and maintenance times.

According to another aspect of the present invention, there is provided a performance evaluation apparatus of a smart meter, including: the evaluation module is used for taking the ammeter environment data, ammeter operation data and ammeter maintenance data as key evaluation indexes of performance evaluation and obtaining time sequence data of the key evaluation indexes; the simulation module is used for establishing a performance attenuation model, the performance attenuation model applies a time sequence analysis technology to correlate the working time length of the ammeter with the performance attenuation of the ammeter, and the time sequence data is input into the performance attenuation model to acquire a natural attenuation mode of the performance attenuation of the ammeter; the identification module is used for obtaining the fluctuation trend of the ammeter performance attenuation according to the natural attenuation mode, identifying abnormal fluctuation in the fluctuation trend, carrying out risk assessment on the abnormal fluctuation and judging an abnormal reason, judging whether the abnormal reason is a tampering behavior, and feeding back the risk assessment and the abnormal reason to the intelligent ammeter maintenance system for reference of subsequent maintenance actions.

According to another aspect of the present invention, there is provided an electronic apparatus including: at least one processor, and a memory communicatively coupled to the at least one processor;

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the performance evaluation method of any one of the smart meters according to the embodiments of the present invention.

According to another aspect of the present invention, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the performance evaluation method of any one of the smart meters according to the embodiments of the present invention.

The performance evaluation method and device for the intelligent ammeter provided by the embodiment of the invention have the beneficial effects that: according to the method, the electric meter environment data, the electric meter operation data and the electric meter maintenance data are used as key evaluation indexes for performance evaluation, so that the performance of the electric meter is considered to be influenced by environment and aging on the basis of evaluating the internal operation data of the electric meter, the electric meter performance evaluation is more comprehensive, in the remote performance evaluation of the intelligent electric meter, the influence of environment and aging on the electric meter performance can be evaluated by adding time sequence data of the electric meter environment data and the electric meter maintenance data into a performance attenuation model, the natural attenuation trend of the electric meter performance is obtained, and the tampering behavior can be identified in the natural attenuation trend of the electric meter performance.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.

Drawings

The drawings are included to provide a better understanding of the present invention and are not to be construed as limiting the invention. Wherein:

FIG. 1 is a flowchart of a method for evaluating performance of a smart meter according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a performance evaluation apparatus for a smart meter for implementing an embodiment of the present invention;

fig. 3 is a schematic diagram of an electronic device according to an embodiment of the invention.

In the figure, 100, a performance evaluation device of a smart meter; 11. an evaluation module; 12. a simulation module; 13. an identification module; 200. an electronic device; 201. a calculation unit; 202. a ROM; 203. a RAM; 204. a bus; 205. an I/O interface; 206. an input unit; 207. an output unit; 208. a storage unit; 209. and a communication unit.

Detailed Description

Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding, and are to be considered merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

As shown in fig. 1, an embodiment of the present invention discloses a performance evaluation method for a smart meter, including: s1, taking ammeter environment data, ammeter operation data and ammeter maintenance data as key evaluation indexes of performance evaluation, and acquiring time sequence data of the key evaluation indexes; s2, a performance attenuation model is established, the performance attenuation model applies a time sequence analysis technology to correlate the working time length of the ammeter with the performance attenuation of the ammeter, and time sequence data are input into the performance attenuation model to obtain a natural attenuation mode of the performance attenuation of the ammeter; and S3, obtaining a fluctuation trend of the ammeter performance attenuation according to the natural attenuation mode, identifying abnormal fluctuation in the fluctuation trend, performing risk assessment on the abnormal fluctuation, judging the abnormal reason, and judging whether the abnormal reason is a tampering behavior.

According to the performance evaluation method of the intelligent electric meter, the electric meter environment data, the electric meter operation data and the electric meter maintenance data are used as key evaluation indexes for performance evaluation, so that the performance of the electric meter is considered to be influenced by environment and aging on the basis of evaluating the electric meter internal operation data, and further the performance evaluation of the electric meter is more comprehensive. The time sequence analysis technology is applied to the performance attenuation model to correlate the working time length of the electric meter with the performance attenuation of the electric meter, so that the relation between the performance attenuation of the electric meter and the working time length of the electric meter is obtained, the time sequence analysis is carried out on the time sequence of the key evaluation index added to the performance attenuation model on the basis, the trend evaluation accuracy of the performance attenuation of the electric meter can be improved, and an accurate natural attenuation mode of the performance attenuation of the electric meter is obtained. The natural attenuation mode can be used for accurately predicting the performance attenuation fluctuation trend of the electric meter, so that the current performance attenuation fluctuation trend of the electric meter is compared with the predicted fluctuation trend, abnormal fluctuation of the fluctuation trend is found, risk assessment and abnormal cause judgment are carried out on the abnormal fluctuation, and if the abnormal fluctuation and the risk assessment exceed a preset threshold, intelligent electric meter early warning is triggered. Whether the abnormal fluctuation causes are tamper behaviors or not can be identified through the performance attenuation model, and finally risk assessment and the abnormal causes are fed back to the intelligent ammeter maintenance system for reference of maintenance actions of the follow-up intelligent ammeter.

Therefore, in the remote performance evaluation of the intelligent electric meter, the influence of environment and aging on the performance of the electric meter can be evaluated by adding the time sequence data of the environment data of the electric meter and the maintenance data of the electric meter into the performance attenuation model, the natural attenuation trend of the performance of the electric meter is obtained, the tampering behavior can be identified in the natural attenuation trend of the performance of the electric meter, the electric meter subjected to the tampering behavior is maintained, and the potential risks and threats are reduced.

Specifically, in evaluating a performance decay model of a smart meter, a relationship between a working time length and performance decay is analyzed by first recording the working time length of a group of meters and using a linear regression model. The data set of the working time length T and the performance parameter P of the ammeter is obtained, and the following performance attenuation model can be constructed: p=alpha+beta+t+epsilon, where alpha is the intercept, beta represents the effect of the on-time on the performance decay, and epsilon is the error term. The estimated value of the parameter can be obtained through least square solution, and the predictability of the estimated value is further evaluated through the R side of the linear regression model, so that the natural attenuation mode of the ammeter performance is obtained.

It should be noted that the performance decay model may contain a plurality of sub-models, such as a regression model, a mixture gaussian model, and a time series model.

In an alternative embodiment of the present invention, step S2 is specifically: the method comprises the steps of respectively inputting the time sequence data of the ammeter environment data, the ammeter operation data and the ammeter maintenance data into a performance attenuation model, determining the coefficient relation between the ammeter environment data and the ammeter performance attenuation through simple regression analysis, analyzing the correlation between the ammeter operation data and the ammeter performance attenuation through the time sequence, and obtaining the natural attenuation mode of the ammeter performance attenuation through the correlation between the ammeter maintenance data and the ammeter performance attenuation through a statistical analysis method.

Specifically, when analyzing the temperature data through the performance decay model, it is found that the temperature Tm is related to the performance parameter P of the electricity meter, and a simple linear regression model p=gamma+delta+tm+epsilon can be used to help understand and predict the effect of the temperature rise on the performance. For example, delta represents the magnitude of the effect of temperature on performance decay. In terms of real-time monitoring of current and voltage, an anomaly detection algorithm may be employed, such as based on a Gaussian mixture model (GaussianMixtureModel, GMM), which can learn the distribution of normal operating conditions from historical current and voltage data and is used to identify potential anomaly patterns in real time. Then, a time series prediction model, such as an autoregressive moving average (ARMA) model, is used to analyze the load data L and the performance parameter P, construct a model p=f (L), and predict the effect of high load on the performance of the electricity meter.

In addition, in analyzing the relationship between the number of calibrations and the performance by the performance decay model, a poisson regression model may be used, taking into account that the number of calibrations is count data. For example, if the number of calibrations is denoted by C, the model may be expressed as p=exp (eta+ thetaC) +epsilon, where the parameters eta and theta are estimated by maximum likelihood. Analysis between power down frequency and performance decay may also employ a two-term regression statistical model, where whether power down is indicated by power down event B being a 0-1 variable, then the two-term regression statistical model is p=iotata+kappa ab+epsilon, thereby enabling an estimation of the potential impact of power down on performance. The maintenance records of the electricity meter can be statistically analyzed by a survival analysis method, and a Kaplan-Meier estimator is used for evaluating the relationship between the probability of unattenuated performance and the maintenance times in a specific time. By grouping the number of repairs, the probability of performance decay at different repair frequencies can be obtained. The relationship between age and performance decay of an electricity meter can be studied by using a Cox proportional hazards model, and correlating the risk of electricity meter age a with the risk of performance degradation by constructing a risk function h (t) =h_0 (t) exp (phiA), where h_0 (t) is a baseline risk function and phi is the coefficient of influence of electricity meter age on risk. The data obtained by the environmental humidity sensor and the performance parameters of the ammeter can be analyzed by adopting a linear regression model, and the model is set to be P=lambda+ muH +epsilon, wherein H represents humidity reading, and the influence amplitude of humidity change on the performance of the ammeter can be disclosed. Finally, a degradation model, such as a locally weighted scatter smoothing technique (LOESS), is used to fit the degradation trend of meter metering accuracy over time and identify abnormal degradation, which helps predict potential faults.

At the same time, manufacturer information is incorporated into the multiple linear regression model through the performance decay model. If M represents a different manufacturer code, the model can be extended to:

P=omega+xim+ zetaT + sigmaTm + cdots +epsilon, taking into account the overall effect of the differences between different manufacturers on the decay of the meter performance. Through the gradual data collection, model construction and analysis processes, a comprehensive performance attenuation model can be established, and the model provides a natural attenuation mode of ammeter performance attenuation and provides support for maintenance decisions.

In an alternative embodiment of the invention, the performance decay model employs a cross-validation technique to improve the evaluation performance, in particular by partitioning the time series data into multiple test sets to detect the stability and accuracy of the performance decay model.

In particular, in carrying out the analysis of the relationship between environmental condition indicators and electrical energy losses, it is first necessary to deploy temperature, humidity and air quality sensors that are capable of monitoring environmental data in real time, while at the same time extracting operating data, including electrical energy losses, load fluctuations and power factors, from the electricity meters of the electrical power system. For example, the temperature sensor records data once an hour, and if three consecutive recordings have a temperature anomaly exceeding 40 ℃ (exceeding the normal operating range), the corresponding data point will be marked as suspicious and processed during the data cleansing process. Data cleansing involves the program automatically identifying and removing those outliers that result from sensor errors or data transmission interruptions, which may be inaccurate due to sensor failure or environmental abrupt changes, for example, if the humidity sensor readings suddenly jump from 60% to 20% throughout the day. These data points are corrected by linear interpolation or other data smoothing techniques to ensure continuity and reliability of the data set. After data preprocessing, a correlation coefficient calculation formula, such as a pearson correlation coefficient, is applied to evaluate the strength of the relationship between the environmental index and the electric energy loss. For example, if the calculation result shows that the correlation coefficient between the temperature and the power loss is 0.8, it indicates that there is a strong positive correlation between the two. Such analysis helps to select the environmental indicators that are most likely to affect meter performance. Based on the selected metrics, a predictive model is constructed by enhancing a decision tree algorithm, such as a random forest or gradient boosting tree, that is capable of handling a large number of features and preventing overfitting. Training of the model involves multiple iterations, using different subsets of data in each round, for example, using K-fold cross-validation, ensuring that the model has sufficient generalization capability. Using this enhanced decision tree model to simulate the effects of different environmental conditions, such as a temperature increase from 25 ℃ to 30 ℃ may result in a 3% increase in the energy consumption of the electricity meter. The model helps to understand how the meter performance will change under certain environmental conditions by identifying these patterns. Further, the model may predict that under certain future environmental conditions, such as a predicted average temperature of 35 ℃, the power consumption of the meter is predicted to rise by 5%, while load fluctuations may increase by 2%. Such predictions may help the operation and maintenance team identify weak electricity meters that are susceptible to environmental changes, thereby developing targeted preventive maintenance strategies, such as pre-replacement of heat-dissipating components, to reduce potential performance degradation risks. To ensure accuracy of the model, the data set is periodically updated and the model retrained. As environmental conditions change and new data is added, the model parameters need to be adjusted to accommodate the new data characteristics.

In an alternative embodiment of the present invention, the performance degradation model simulates the environmental conditions of the electric meter, predicts the performance degradation change of the electric meter based on the environmental data of the electric meter, and identifies the performance degradation or the failure of the electric meter caused by the change of environmental factors according to the prediction result, thereby making a preventive maintenance strategy.

Specifically, in a smart meter monitoring system, environmental monitoring sensors are arranged around to capture temperature and humidity data in real time. If during a high temperature period of the day, the sensor detects that the ambient temperature exceeds a predetermined threshold, such as 35 ℃, while the relative humidity is below 30%, an abnormal condition is identified. The values are collected by special sensors, such as NTC thermistors, and the relation between the resistance and the temperature can be described by Steinhart-Hart equation, and the humidity sensor adopts a capacitive or resistive element, and the output has a certain function relation with the ambient humidity. At this point, the system will automatically trigger an alarm and the anomaly information will be transmitted to the central monitoring system in real time over a wireless network, such as using LoRaWAN technology. The data summary analyzer then works, for example, using a multiple linear regression algorithm, to combine the environmental parameters with known impact models to extract core evaluation metrics.

For example, calculating an index of environmental comfort may be achieved by a weighting algorithm, such as environmental index=w1×temperature+w2×humidity+w3×pm2.5+w4×noise, where w1, w2, w3, and w4 are weighting coefficients of the parameters. Based on the index, the performance impact assessment tool further works to predict the performance change of the electricity meter by using a decision tree method, so as to know the specific impact of the electricity meter environment data on the electricity meter function. If the environmental index is below a certain threshold, it is determined that the meter performance is negatively affected. According to the analysis result, the adaptive adjustment of the temperature and the humidity is triggered, and the optimal temperature and humidity adjusting strategy is determined by utilizing a PID control algorithm, namely, the P (proportion), I (integral) and D (derivative) parameters of the controller are adjusted to obtain the optimal response. If the analysis determines that the environmental conditions will significantly affect the performance of the meter, the meter performance optimization program will automatically start, which may be implemented by a simple feedback control loop, such as automatically adjusting some regulatory parameters inside the meter, such as calibrating the temperature compensation coefficient inside the meter, based on the difference between the actual and target values of the temperature and humidity. Finally, a general environmental conditioning command will be issued, the command including specific values, such as lowering the room temperature to 26 ℃, and raising the relative humidity to 50%. The regulating and controlling device such as air conditioner and humidifier starts working according to the instructions, and adopts a strategy similar to PID control to finely regulate until the set environmental parameters are reached, so that the stable operation of the whole system and the maximization of the performance of the ammeter are ensured.

In an alternative embodiment of the present invention, the fluctuation trend of the power meter performance attenuation is obtained according to the natural attenuation mode, specifically: performing anomaly detection on the time series data, performing data cleaning on the detected anomaly data, analyzing the time series data by a performance attenuation model by using a time series analysis method, continuously adjusting parameters of the performance attenuation model according to seasonal and periodic fluctuation of the time series data so as to improve the prediction effect of fluctuation trend obtained by a natural attenuation mode, and then obtaining the fluctuation trend of ammeter performance attenuation.

Specifically, in the initial stage of ammeter data analysis, normalized cleaning operation is carried out on the collected raw data, and wrong readings and irrelevant noise are removed. For example, a simple threshold filter is set to reject abnormally high readings or abnormally low readings that exceed the set power range of the meter, ensuring the accuracy of subsequent analysis. And then, carrying out normalization processing on the time sequence data, and normalizing the electric meter reading into a [0,1] interval so as to eliminate data deviation caused by different electric meters and acquisition environments. When the statistical analysis is carried out, a moving average algorithm is applied, the daily average power consumption of the last week is calculated, and the hidden consumption trend is revealed through smoothing the periodic fluctuation. In order to construct a prediction model of the performance attenuation model, an autoregressive integral moving average model is selected, parameters of a plurality of models are selected and compared by utilizing a red pool information quantity criterion, and finally a parameter set for obtaining optimal performance by fitting historical data is determined. In the model optimization link, a cross-validation technology is adopted to evaluate the generalization capability of the model, and the training process is repeated and the result is compared by dividing the data set into a plurality of training and testing combinations so as to avoid the over-fitting phenomenon. And the model parameter adjustment adopts a gradient descent method, and the parameter of the autoregressive integral moving average model is gradually adjusted until the natural attenuation mode prediction error of the ammeter performance attenuation is minimum on the verification set.

In an alternative embodiment of the present invention, step S3 is specifically: determining a threshold value of normal fluctuation of the current ammeter performance attenuation, wherein the threshold value is obtained after statistical analysis is performed based on time sequence data and fluctuation trend, if the current ammeter performance attenuation exceeds the fluctuation of the threshold value, judging the ammeter performance attenuation as abnormal fluctuation, carrying out risk assessment and judging an abnormal reason, and if the abnormal fluctuation or the risk assessment exceeds a preset value, triggering ammeter performance early warning; and feeding back the risk assessment and the abnormality reasons to the intelligent ammeter maintenance system, prompting the ammeter with reduced ammeter performance and feeding back the ammeter to be maintained to the intelligent ammeter maintenance system, and displaying the trend of reduced ammeter performance to maintenance personnel and operators through a data visualization technology.

Specifically, to set a reasonable abnormal fluctuation threshold, statistical analysis is performed on the fluctuation trend, and a 95% confidence interval of the performance evaluation time series data is calculated, and the abnormal fluctuation is identified by taking the 95% confidence interval as the threshold. If the actual observed electricity consumption exceeds the threshold, potential electricity meter performance problems or future faults can be found in time. An anomaly detection algorithm is designed by using an early warning mechanism, when predicted data exceeds an anomaly threshold value at several continuous time points, the system automatically triggers an alarm, for example, a threshold value is set, for example, 3 standard deviations of the data points exceeding the average value are regarded as anomaly values, and once the data at several continuous time points is detected to exceed the threshold value, the anomaly is regarded as occurring, and early warning is triggered. The mechanism ensures that the alarm information is targeted and reliable based on a time series prediction model and ammeter historical performance fluctuation data. The data visualization link comprises the steps of displaying consumption trend and abnormal conditions of the ammeter under different time scales, and intuitively presenting the performance degradation process of the ammeter by adopting a heat map and a trend line diagram. The high consumption interval can be easily identified through the change of the color, and by means of the trend graph, an operation and maintenance person can intuitively see how the current electric quantity consumption fluctuates compared with the historical data.

In an alternative embodiment of the invention, after the smart meter maintenance system processes meter maintenance items, the smart meter maintenance system collects maintenance results and feeds the maintenance results back to the performance attenuation model, and further adjusts and optimizes the performance attenuation model to ensure that the evaluation capability of the performance attenuation model is more consistent with the actual situation.

The feedback and maintenance results of the maintenance team are recorded in the database, and the new data points are fed back into the performance evaluation performance attenuation model for retraining, so that the model can learn the latest equipment performance change condition, the prediction accuracy is further improved, and the model evaluation capability is ensured to be more consistent with the actual condition.

In an optional embodiment of the present invention, if the risk assessment does not conform to the preset risk threshold trend result, performing adaptive correction on the risk threshold according to the deviation of the risk assessment, performing data restoration and parameter calibration operations, and correcting the abnormal key assessment index by using a moving average method of time series analysis, so as to ensure long-term accuracy of the key assessment index.

Specifically, a standard consumption pattern is established by analyzing the time-series data of the smart meter in detail. For example, electricity meters from one plant have shown an average daily consumption of 1000kWh from monday to friday in the past year record, while weekend consumption drops to 800kWh. Through statistical analysis, the standard consumption pattern is determined and the daily fluctuation range is calculated, assuming 950kWh to 1050kWh. Next, using an anomaly detection algorithm, such as an isolated forest (IsolationForest), data points that differ significantly from the normal mode can be detected, which points represent anomalies in the readings of the meter. For example, if a day records 1200kWh usage, the algorithm will flag it as a potential anomaly.

After the abnormality detection result is obtained, a risk scoring system is established, and the detected abnormality data is compared with a risk threshold value by using a scoring criterion. It is assumed that the scoring system classifies the detected anomalies into low risk, scores 1-3, medium risk, scores 4-6, high risk, scores 7-9. If the outlier score exceeds 6, the system will trigger an adaptive adjustment mechanism. Taking threshold adjustment as an example, if the consumption exceeds 1100kWh for two consecutive days, the scoring system may temporarily raise the risk threshold, considering higher than 1100kWh as abnormal, the adaptive adjustment data excludes some sporadic peak effects. The automatic correction flow initiated therewith is intended to take a data repair strategy,

Smoothing techniques such as moving average, which are commonly used in time series analysis, are applied to smooth abnormal data, thereby alleviating the influence of abnormal points. Assuming that an outlier of 1200kWh occurs in wednesday and both wednesday use 1000kWh, the outlier was corrected to (1000+1200+1000)/3= 1066.7kWh using the three-day moving average, and the data was adjusted closer to the normal mode. Next, a comparison calibration strategy is used to adjust key parameters such as power factor or maximum demand peak. The strategy applies a simple design assuming a normal value of 0.95 for the power factor, and if the corrected data shows an actual value of 0.90, it can be adjusted using the formula (p_real/p_ideal) x 100, where p_real is the actual power factor and p_ideal is the ideal power factor. The corrected data set provides a correction reference, and after parameter adjustment, the scoring system can review the data again to ensure the accuracy of the performance decay model.

Finally, continuously verifying the validity of the calibration process through parameter adjustment feedback and data comparison analysis. For example, the amounts of electricity used for the same period of time before and after correction are compared, and whether the deviation is significantly improved is verified by the statistical analysis of the summary. If the difference is within the statistically allowable error range, the calibration process is proved to be effective. Quantitative analysis of these calibration effects is fed directly into the system, further optimizing the algorithm. If the correction data on a specific day is found to be frequently in error, the fault-tolerant parameters of the isolated forest algorithm are adjusted.

In an optional embodiment of the present invention, determining whether the cause of the anomaly is a tampering behavior specifically includes: judging the abnormal reason of the abnormal fluctuation, judging whether the abnormal reason is the natural attenuation of the performance of the ammeter or the tampering behavior of the ammeter, and correcting and remotely modifying the key evaluation index if the abnormal fluctuation exceeds an error range and the natural attenuation is carried out; if the electric meter is tampered, isolating the tampered electric meter from the network, isolating a system part affected by tampering, recovering data of the key evaluation index, performing data audit and check, confirming the integrity and accuracy of the key evaluation index, acquiring a detailed report of a processing result, and notifying a tampering event and the detailed report to a manager.

Specifically, in the process of continuously monitoring the voltage, a real-time data acquisition system is used for reading the current voltage value through an installed sensor and comparing the current voltage value with a natural attenuation mode of a performance attenuation model to judge whether an unnatural attenuation mode occurs. If the value is found to deviate from the normal range, the future voltage trend is predicted through time series analysis and a pattern recognition algorithm, such as an autoregressive integral moving average model, so that the potential health problem of the equipment is early recognized. When the current fluctuation analysis reveals anomalies, these current readings and their time stamps are transmitted to the data processing center. Applied here is an anomaly detection algorithm, such as a cluster-based anomaly detection technique, that maps data points into a feature space to identify outliers that deviate from most of the data points. Precisely locating the point in time when the anomaly occurred helps to trace back any tamper activity associated therewith in the maintenance log.

Thereafter, the electric energy accumulated error is calculated by correcting the readings of the electric meter. For example, the error analysis may use the formula ee= (Er-Et)/Et 100%, where Er is the reading of the electricity meter and Et is the theoretical correct reading by standard equipment. If the error exceeds industry standards, typically + -2% is considered to be a normal acceptable range, the calibration parameters of the meter can be fine-tuned using remote calibrated modem technology without going to the field. On the physical security, a vibration sensor and a switch state monitor are deployed to automatically check the integrity of the tamper-proof device periodically. Any unusual physical activity will immediately trigger an alarm through the security information and event management System (SIEM) while correlating the operational records of the device to determine its legitimacy. In the field of network security, encryption algorithms are used to encrypt communication data, for example, advanced Encryption Standard (AES), and in combination with a key management protocol, ensure the integrity and confidentiality of the transmission data. By monitoring the communication mode in real time, an intrusion detection system is used to detect abnormal behavior in network traffic and identify unauthorized access or attack attempts. Meanwhile, seal integrity checking is implemented, such as automatic detection technology based on machine vision is introduced to identify seal damage, the method compares seal picture changes, and an image identification algorithm is utilized to identify abnormal modes, so that potential physical tampering behaviors are accurately pulled.

And predicting the performance degradation curve of the equipment according to the accumulated historical data by using a regression analysis tool for the equipment power consumption related abnormal fluctuation. For the performance decay model, linear regression or polynomial regression is employed to fit the power consumption characteristics of the device to predict its decay period. Finally, high-precision load characteristic monitoring of the electricity meter is performed regularly, which involves the use of high-speed sampling digitizing equipment to obtain fine load curves. These data will be used to support a more accurate statistical model, such as gaussian process regression, to analyze and predict the performance changes of the electricity meter, and the results will be directly related to maintenance planning and life extension strategies.

The method for judging the cause of the abnormality in an alternative embodiment of the present invention specifically includes: and exploring the mapping relation between the abnormal fluctuation and the abnormal cause of the known ammeter by using a correlation rule mining technology, and confirming the causal relation between the abnormal fluctuation and the abnormal cause after verifying the mapping relation by applying causal relation analysis.

Specifically, once abnormal fluctuations of the electricity meter are identified, association rule mining techniques are used to explore potential links between these fluctuations and failure modes of known electricity meters, as well as mapping relationships that may exist between them; if the specific fluctuation mode of the ammeter is found to be related to a certain fault mode, next, causal relation analysis is applied to verify the direct connection between the fluctuation and the fault, and whether the fluctuation is the direct cause of the fault of the ammeter is confirmed; after confirming that the fluctuation is the cause of the fault, constructing the discovery as a maintenance problem to be processed; deep mining the fault cause and searching for other potential influencing factors related to the fault cause; by means of fault pattern recognition, these factors are further analyzed to identify paths that may lead to fault spread; introducing a maintenance decision system which recommends an optimal maintenance strategy based on various evaluation criteria in order to select the action which is most effective in preventing the fault distribution; planning and executing a subsequent concrete maintenance action plan, and continuously monitoring the effect of these actions to ensure that the problem is resolved.

In an alternative embodiment of the invention, the performance decay model is one of an autoregressive integral moving average model, a seasonal autoregressive integral moving average model, and an exponential smoothing model.

In an alternative embodiment of the invention, the meter environmental data includes temperature, humidity, PM2.5, and noise, the meter operational data includes current, voltage, active power, reactive power, and meter load, and the meter maintenance data includes calibration times, fault times, and maintenance times.

As shown in fig. 2, the performance evaluation apparatus 100 of the smart meter may include:

The evaluation module 11 is configured to obtain time series data of key evaluation indexes by using the electric meter environment data, the electric meter operation data and the electric meter maintenance data as key evaluation indexes for performance evaluation;

The simulation module 12 is used for establishing a performance attenuation model, the performance attenuation model applies a time sequence analysis technology to correlate the working time length of the ammeter with the performance attenuation of the ammeter, and time sequence data is input into the performance attenuation model to obtain a natural attenuation mode of the performance attenuation of the ammeter;

the identification module 13 is configured to obtain a fluctuation trend of the power meter performance attenuation according to the natural attenuation mode, identify abnormal fluctuation in the fluctuation trend, perform risk assessment on the abnormal fluctuation, judge an abnormal reason, judge whether the abnormal reason is a tampering behavior, and feed back the risk assessment and the abnormal reason to the smart power meter maintenance system for reference of subsequent maintenance actions.

The specific functions and examples of the modules and sub-modules of the apparatus in the embodiments of the present invention may be described with reference to the relevant descriptions of the corresponding steps in the foregoing method embodiments, which are not repeated herein.

In the technical scheme of the invention, the acquisition, storage, application and the like of the related user personal information all conform to the regulations of related laws and regulations, and the public sequence is not violated.

According to embodiments of the present invention, the present invention also provides an electronic device, a readable storage medium and a computer program product.

FIG. 3 shows a schematic block diagram of an example electronic device 200 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable electronic devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 3, the electronic device 200 includes a computing unit 201 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 202 or a computer program loaded from a storage unit 208 into a Random Access Memory (RAM) 203. In the RAM 203, various programs and data required for the operation of the device 200 can also be stored. The computing unit 201, ROM 202, and RAM 203 are connected to each other through a bus 204. An input/output (I/O) interface 205 is also connected to bus 204.

Various components in the electronic device 200 are connected to the I/O interface 205, including: an input unit 206 such as a keyboard, a mouse, etc.; an output unit 207 such as various types of displays, speakers, and the like; a storage unit 208 such as a magnetic disk, an optical disk, or the like; and a communication unit 209 such as a network card, modem, wireless communication transceiver, etc. The communication unit 209 allows the electronic device 200 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.

The computing unit 201 may be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of computing unit 201 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 201 performs the respective methods and processes described above, for example, a performance evaluation method of a smart meter. For example, in some embodiments, a method of performance evaluation of a smart meter may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 208. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 200 via the ROM 202 and/or the communication unit 209. When the computer program is loaded into the RAM 203 and executed by the computing unit 201, one or more steps of a performance evaluation method of the smart meter described above may be performed. Alternatively, in other embodiments, the computing unit 201 may be configured to perform a method of performance evaluation of a smart meter by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present invention may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present invention, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution disclosed in the present invention can be achieved, and are not limited herein.

The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the principle of the present invention should be included in the protection scope of the present invention.

Claims (15)

1. A performance evaluation method of a smart meter, comprising:

taking the ammeter environment data, ammeter operation data and ammeter maintenance data as key evaluation indexes of performance evaluation, and acquiring time sequence data of the key evaluation indexes;

Establishing a performance attenuation model, wherein the performance attenuation model applies a time sequence analysis technology to correlate the working time length of an ammeter with the performance attenuation of the ammeter, and inputs the time sequence data into the performance attenuation model to acquire a natural attenuation mode of the performance attenuation of the ammeter; and obtaining the fluctuation trend of the performance attenuation of the ammeter according to the natural attenuation mode, identifying abnormal fluctuation in the fluctuation trend, carrying out risk assessment on the abnormal fluctuation and judging an abnormal reason, and judging whether the abnormal reason is a tampering behavior.

2. The method according to claim 1, wherein the inputting the time series data into the performance decay model obtains a natural decay pattern of the meter performance decay, in particular:

And respectively inputting the ammeter environment data, the ammeter operation data and the time sequence data of the ammeter maintenance data into the performance attenuation model, determining a coefficient relation between the ammeter environment data and the ammeter performance attenuation through simple regression analysis, using the time sequence analysis to correlate the ammeter operation data with the ammeter performance attenuation, and using a statistical analysis method to correlate the ammeter maintenance data with the ammeter performance attenuation so as to obtain a natural attenuation mode of the ammeter performance attenuation.

3. The method according to claim 2, wherein the performance decay model improves evaluation performance by applying a cross-validation technique, in particular by partitioning the time series data into a plurality of test sets to detect stability and accuracy of the performance decay model.

4. The method of claim 2, wherein the performance degradation model simulates an environmental condition of the electricity meter, predicts the change in performance degradation of the electricity meter based on the environmental data of the electricity meter, and identifies a decrease in performance of the electricity meter or occurrence of an electricity meter failure due to a change in environmental factors based on the prediction results, thereby developing a preventive maintenance strategy.

5. The method according to claim 1, wherein the fluctuation trend of the electric meter performance attenuation is obtained according to the natural attenuation mode, specifically:

And performing anomaly detection on the time series data, performing data cleaning on the detected anomaly data, analyzing the time series data by using a time series analysis method by using a performance attenuation model, and continuously adjusting parameters of the performance attenuation model according to seasonal and periodic fluctuation of the time series data so as to improve the prediction effect of the fluctuation trend obtained by the natural attenuation mode, and then obtaining the fluctuation trend of the performance attenuation of the ammeter.

6. The method according to claim 1, wherein the identifying of the abnormal fluctuation in the fluctuation trend performs risk assessment and abnormality cause judgment on the abnormal fluctuation, specifically:

Determining a threshold value of normal fluctuation of current ammeter performance attenuation, wherein the threshold value is obtained after statistical analysis is carried out based on the time sequence data and the fluctuation trend, if fluctuation that the current ammeter performance attenuation exceeds the threshold value is detected, the abnormal fluctuation is judged, the risk assessment is carried out, the abnormality cause is judged, and if the abnormal fluctuation or the risk assessment exceeds a preset value, the ammeter performance early warning is triggered;

And feeding back the risk assessment result and the abnormality cause result to a smart meter maintenance system, prompting the smart meter maintenance system of the meter with reduced meter performance and feeding back the meter to be maintained, and displaying the trend of reduced meter performance to maintenance personnel and operators through a data visualization technology.

7. The method of claim 6, wherein the smart meter maintenance system collects maintenance results after processing meter maintenance events and feeds the maintenance results back to the performance decay model, and further adjusts and optimizes the performance decay model to ensure that the evaluation capability of the performance decay model is more consistent with the reality.

8. The method of claim 6, wherein if the risk assessment does not meet a preset risk threshold trend result, performing adaptive correction on the risk threshold according to the deviation of the risk assessment, performing data restoration and parameter calibration operations, and correcting abnormal key assessment indexes by using a moving average method of time series analysis to ensure long-term accuracy of the key assessment indexes.

9. The method according to claim 1, wherein the determining whether the cause of the abnormality is a tampering action is specifically:

Judging the abnormal reason of the abnormal fluctuation, judging whether the abnormal reason is the natural attenuation of the performance of the ammeter or the tampering behavior of the ammeter, and correcting and remotely modifying the key evaluation index if the abnormal reason is the natural attenuation and the abnormal fluctuation exceeds an error range;

If the tampering behavior is the tampering behavior, isolating the tampered ammeter from the network, isolating a system part affected by the tampering, recovering data of the key evaluation index, performing data audit and check, confirming the integrity and accuracy of the key evaluation index, acquiring a detailed report of a processing result, and notifying a manager of a tampering event and the detailed report.

10. The method according to claim 9, wherein the abnormality cause judging method is:

And exploring the mapping relation between the abnormal fluctuation and the abnormal cause of the known ammeter by using a correlation rule mining technology, and confirming the causal relation between the abnormal fluctuation and the abnormal cause after verifying the mapping relation by applying causal relation analysis.

11. The method of claim 1, wherein the performance decay model is one of an autoregressive integral moving average model, a seasonal autoregressive integral moving average model, and an exponential smoothing model.

12. The method of claim 1, wherein the meter environmental data includes temperature, humidity, PM2.5, and noise, the meter operational data includes current, voltage, active power, reactive power, and meter load, and the meter maintenance data includes calibration times, fault times, and maintenance times.

13. A performance evaluation device of a smart meter, comprising:

the evaluation module is used for taking the ammeter environment data, ammeter operation data and ammeter maintenance data as key evaluation indexes of performance evaluation and obtaining time sequence data of the key evaluation indexes;

The simulation module is used for establishing a performance attenuation model, the performance attenuation model applies a time sequence analysis technology to correlate the working time length of the ammeter with the performance attenuation of the ammeter, and the time sequence data is input into the performance attenuation model to acquire a natural attenuation mode of the performance attenuation of the ammeter;

The identification module is used for obtaining the fluctuation trend of the ammeter performance attenuation according to the natural attenuation mode, identifying abnormal fluctuation in the fluctuation trend, carrying out risk assessment on the abnormal fluctuation and judging an abnormal reason, and judging whether the abnormal reason is a tampering behavior.

14. An electronic device, comprising: at least one processor, and a memory communicatively coupled to the at least one processor;

wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-12.

15. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-12.