Intelligent detection method for line loss of intelligent power grid based on big data of Internet of Things

Abstract

The invention discloses a smart power grid line loss intelligent detection method based on Internet of Things big data, and relates to the technical field of smart power grid electric energy metering, and the method comprises the following steps: S1, extracting a voltage signal and a current signal of each electric energy metering node in a set time window, carrying out the frequency domain transformation to obtain a frequency spectrum, calculating the energy concentration index of a high-frequency band, and calculating the energy concentration index of the high-frequency band; and identifying a high-frequency harmonic interference time section according to the change of the energy concentration degree. According to the method, a closed-loop mechanism of high-frequency harmonic interference identification, error traceability, cause discrimination and dynamic correction is constructed, fragment-level and node-level error compensation is realized, and the accuracy and the stability of line loss anomaly identification of the smart grid are improved by combining electric quantity consistency verification and trend model updating; and the reliability and the intelligent level of the system in a complex interference environment are obviously enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grid electric energy metering, and in particular to a smart grid line loss intelligent detection method based on Internet of Things big data. Background Art

[0002] "Smart Grid Line Loss Intelligent Detection Based on IoT Big Data" involves deploying IoT sensor terminals at key grid nodes to collect multi-dimensional data, including current, voltage, power factor, and electricity usage, in real time. Leveraging a big data platform, this system aggregates, cleans, and deeply mines vast amounts of spatially and temporally distributed grid operation information. Using intelligent algorithms like machine learning and pattern recognition, it dynamically identifies the location, type (technical or managerial), and changing trends of power losses during power transmission. This approach transcends the limitations of traditional line loss calculations, which rely on manual meter reading and static statistics, enabling real-time perception, accurate assessment, and dynamic early warning of grid line losses. This provides technical support for energy efficiency optimization and abnormal power usage monitoring in smart grids.

[0003] The existing technology has the following deficiencies: In the smart grid's line loss monitoring and energy consumption assessment processes, smart meters serve as key data collection terminals. Their energy measurement accuracy directly impacts the system's accuracy in identifying line loss data and the reliability of subsequent decision-making. However, in actual operating environments, the energy metering chips within some smart meters are susceptible to reduced interference immunity or misjudgment of electrical signal characteristics due to long-term exposure to high-frequency harmonic interference. This is particularly true in areas where high-power nonlinear loads (such as welding equipment, electric vehicle fast charging devices, and inverters) frequently operate. The metering chips may be affected by non-power frequency current components, leading to misjudgment of effective power. This type of interference is sudden, localized, and intermittent, causing random measurement offsets within the metering chips over short periods of time. Even if the individual error amplitude is small, the accumulated error over time can cause systematic deviations in the line loss calculation results, disrupting the error compensation mechanism of the line loss model and even causing false alarms and misidentification of power theft.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent detection method for smart grid line losses based on IoT big data, construct a closed-loop mechanism for high-frequency harmonic interference identification, error tracing, cause determination and dynamic correction, realize segment-level and node-level error compensation, combine power consistency verification and trend model update, improve the accuracy and stability of smart grid line loss anomaly identification, significantly enhance the reliability and intelligence level of the system in complex interference environments, and solve the problems in the above-mentioned background technology.

[0006] In order to achieve the above objectives, the present invention provides the following technical solution: a smart grid line loss intelligent detection method based on IoT big data, comprising the following steps: S1 extracts the voltage and current signals of each energy metering node within a set time window, performs frequency domain transformation to obtain the spectrum, calculates the energy concentration index of the high-frequency band, and identifies the high-frequency harmonic interference time segment based on the energy concentration change; S2, based on the identified high-frequency harmonic interference time segment, calls the historical standard active power curve template of the corresponding electric energy metering node, performs error fitting between the current active power curve and the template, and extracts the active power deviation vector; S3: Input the active power deviation vector, terminal device type, instantaneous load change characteristics, and high-frequency harmonic spectrum characteristics into the abnormal cause classification model to determine whether the output power deviation is caused by the high-frequency harmonic interference of the power metering chip, and mark the abnormal power error segment; S4, for abnormal power error segments, generates high-frequency harmonic interference correction factors based on the time duration, numerical offset direction and amplitude, and constructs a correction weight sequence for short-period data at the power metering node level; S5, based on the correction weight sequence, the original electric energy metering data is subjected to segment weighted correction, the electric energy consistency check matrix between multiple electric energy metering nodes is calculated, and the systematic error is corrected as a whole; S6, input the corrected data into the line loss analysis module, update the warning judgment threshold and error tolerance boundary in the line loss trend model, and realize closed-loop correction of electricity metering interference data and real-time and stable identification and response of line loss anomalies.

[0007] Preferably, step S1 includes: Within the set time window, the voltage and current signals of the energy metering node are continuously collected and weighted by the window function to eliminate the influence of spectrum leakage; Perform fast Fourier transform on the weighted signal to obtain the voltage spectrum and current spectrum, and calculate the frequency domain energy accumulation value of the high frequency band; Calculate the frequency domain energy concentration index to form time series data; Based on dynamic thresholds and disturbance duration judgment, the time segment of high-frequency harmonic interference is identified, and an interference event log is generated for subsequent error compensation.

[0008] Preferably, step S2 includes: Based on the identified high-frequency harmonic interference time segment, the active power measurement values ​​continuously recorded by the corresponding energy metering node between the start and end time of the high-frequency harmonic interference are extracted, the current active power curve is constructed, and the active power curve is normalized; Call the standard active power curve template under the historical operating conditions of the corresponding node; Perform time-series fitting on the current active power curve and the standard template, use dynamic time warping algorithm or window alignment strategy to align the shapes, and calculate the point-by-point power difference. The statistical characteristics of the power difference sequence are extracted to construct the active power deviation vector, which is used to quantify the impact of high-frequency harmonic interference on the metering node.

[0009] Preferably, step S3 includes: Extract active power deviation vector, terminal equipment type, instantaneous load change characteristics and high-frequency harmonic spectrum characteristics; The acquired features are uniformly input into the trained power anomaly cause classification model, which is built based on a supervised learning algorithm. The classification result of whether the output power deviation is caused by the energy metering chip being interfered with by high-frequency harmonics; The time segment corresponding to the classification result is marked as an abnormal power error segment and serves as the basis for generating subsequent correction parameters.

[0010] Preferably, step S4 includes: Extract the time duration, numerical offset direction and numerical offset amplitude of the abnormal power energy error fragment; Generate a corresponding correction factor based on the offset direction and offset amplitude, and set the correction gradient according to the time distribution within the segment; Mapping correction factors to abnormal segments and dividing them into short-period time periods; Correction weights are assigned in each time period to construct a short-period data correction weight sequence at the energy metering node level.

[0011] Preferably, step S5 includes: Based on the short-cycle data correction weight sequence, the abnormal power error fragments in the original power metering data are corrected at the fragment level to generate a corrected power sequence; Convert the corrected power data of each electric energy metering node into equal-period electric quantity data, and construct an electric quantity consistency check matrix between multiple metering nodes; Based on the degree of deviation of the power difference between each node in the power consistency check matrix, the correction weight of the corresponding node is adjusted to generate new corrected power data; Repeat the verification and correction process until the overall consistency error of the system converges to within the tolerance threshold.

[0012] Preferably, step S6 includes: The corrected electric energy metering data is aggregated according to the grid topology, and the basic line loss data sequence of each power supply unit is recalculated; Input the corrected basic line loss data series into the line loss trend model, replace the original data of the corresponding time period, and update the model parameters; Adjust the line loss abnormality warning threshold and error tolerance based on the updated line loss trend model; The updated judgment results are integrated with the line loss analysis process to form a closed-loop correction and real-time response control mechanism for electricity metering data.

[0013] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention can realize a closed-loop processing mechanism from time period identification of high-frequency harmonic interference, error tracing, cause determination to refined correction and consistency verification of data across the entire network, effectively solving the problems of power misjudgment, line loss misidentification, and false alarms caused by interference in traditional electric energy metering systems in complex electromagnetic environments. This method not only has the interference perception capability of multi-dimensional feature fusion, but also introduces dynamic correction factors and weight distribution strategies to achieve error compensation at the fragment and node levels. At the same time, through the power consistency matrix and trend model update mechanism, the accuracy and stability of the corrected data in the entire network analysis are fully guaranteed. Ultimately, the smart grid can maintain high sensitivity, high accuracy, and sustainable response capabilities to line loss anomalies in the face of complex operating scenarios such as high-frequency interference and nonlinear load concentration, thereby improving the reliability, accuracy, and intelligence level of the line loss analysis system. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0015] Figure 1 This is a flow chart of the method for intelligent detection of smart grid line losses based on IoT big data of the present invention. DETAILED DESCRIPTION

[0016] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0017] The present invention provides Figure 1 The smart grid line loss intelligent detection method based on IoT big data includes the following steps: S1 extracts the voltage and current signals of each energy metering node within a set time window, performs frequency domain transformation to obtain the voltage and current spectra, calculates the frequency domain energy concentration index within the high-frequency band, and identifies the time segments with high-frequency harmonic interference based on the changes in the frequency domain energy concentration index; Aiming at the adverse effects of high-frequency harmonic interference on the accuracy of electric energy metering in smart grids, a high-frequency harmonic disturbance identification method based on frequency domain energy concentration analysis is proposed to achieve time period positioning and dynamic marking of high-frequency electromagnetic disturbances. The specific steps include: For each electric energy metering node, the original waveform data of the voltage signal and current signal within the corresponding time period is continuously extracted within a set time window (for example, 1 second, 5 seconds, or 10 seconds). The above waveform data should be obtained from the electric energy metering terminal through synchronous sampling to ensure the alignment accuracy of the voltage signal and the current signal in the time dimension. All data must meet the minimum sampling frequency of not less than 10kHz to ensure sufficient high-frequency resolution in frequency domain analysis. To avoid spectrum leakage problems caused by edge effects, the original waveform data must be preprocessed through a window function (such as Hamming window, Blackman window, etc.) before entering the frequency domain transformation processing to enhance the stability of the spectrum calculation.

[0018] To avoid spectral leakage caused by signal truncation during frequency domain transformation of raw waveform data—the diffusion of spectral energy at non-real frequencies, which affects the accurate identification of high-frequency interference components—window function weighting should be performed on the raw waveform data before frequency domain transformation. Specifically, the raw voltage and current signals within a selected time window are sampled separately to obtain data segments of fixed length. Subsequently, a window function with edge attenuation characteristics (such as a Hamming window or a Blackman window) is selected and multiplied point-by-point with each sampled data segment in an equal-length manner. This operation can be viewed as smoothing the abrupt transitions at the waveform's start and end boundaries while preserving the bulk of the signal energy. This, in turn, suppresses the spectral spreading effect introduced by non-periodic truncation during subsequent frequency domain transformation. The introduction of the window function effectively improves the energy localization accuracy of each frequency component in the spectrum, enhancing the stability and reliability of frequency domain energy concentration analysis.

[0019] A Fast Fourier Transform (FFT) is performed on the processed voltage and current signals to obtain their corresponding voltage and current spectra, extracting the amplitude information corresponding to each frequency component. Within the spectral data, a specific high-frequency band is designated as the harmonic interference assessment interval, for example, the 150Hz to 2kHz frequency range, which covers the high-order harmonics generated by typical power electronic equipment. Based on this high-frequency range, the voltage and current spectra are integrated to calculate the accumulated frequency domain energy within this band. To further enhance the discriminative power of interference localization, the Normalized High-Frequency Energy Ratio (N-HFER) is introduced as a frequency domain energy concentration metric. This metric is defined as the ratio of the integrated energy within the high-frequency range to the total energy across the entire frequency band. This metric objectively reflects the concentration of high-frequency energy in the overall signal, enabling the identification of sudden high-frequency harmonic disturbances.

[0020] The calculated frequency domain energy concentration index is constructed into a time series data, and the time window boundaries are combined to identify periods where the frequency domain energy concentration is significantly higher than the conventional benchmark. To enhance the robustness of recognition, a dynamic mean and standard deviation reference curve of the energy concentration is constructed based on the sliding window method. It is determined whether the concentration index of a certain time window exceeds a dynamic threshold (for example, the mean plus two standard deviations) as the trigger condition for the presence of high-frequency harmonic interference. In addition, to avoid misjudging occasional disturbances, a minimum disturbance duration threshold is set (for example, two consecutive time windows are triggered) to ensure that the identified interference time period has sufficient temporal continuity, thereby forming a stable interference segment marker.

[0021] Based on the above trigger criteria, the time segments where high-frequency harmonic disturbances occur are marked and archived to form a time series interference event log. Each interference event log should contain its start and end time, frequency domain energy concentration peak, interference intensity level, and the number and sampling identifier of the corresponding electric energy metering node. This interference event log can be used for subsequent electric energy metering deviation analysis and error correction model calls, and can serve as an important basis for power deviation judgment and error attribution in interference environments. In addition, Through the above-mentioned frequency domain energy concentration analysis and interference period calibration mechanism, the automatic identification and period positioning of high-frequency harmonic disturbances in complex electromagnetic environments are realized, which significantly improves the accuracy of identifying abnormal fragments in electricity metering data and provides stable and reliable basic data support for the subsequent smart grid line loss error compensation mechanism.

[0022] This step establishes a highly sensitive, high-resolution high-frequency interference detection mechanism for subsequent energy metering error identification and correction. Its core objective is to accurately identify high-frequency harmonic disturbances in the environment of smart grid energy metering nodes through a frequency domain perspective. High-frequency harmonic interference is often caused by nonlinear loads (such as inverters and electric vehicle fast-charging devices). Its disturbance characteristics are sudden, localized, and non-periodic, making it difficult to directly identify using time-domain signals. Therefore, voltage and current signals are acquired within a set time window and their spectral characteristics are extracted through frequency domain transformation, focusing on the energy distribution in the high-frequency band. By introducing a frequency-domain energy concentration metric, the proportion of high-frequency energy in the overall signal is quantified, effectively measuring the interference intensity and its changing trend. If the high-frequency energy concentration within a certain period is significantly higher than the normal operating level, it can be preliminarily determined that external harmonic interference was present during that period. This step not only provides a clear time-based basis for subsequent power deviation calculation and error attribution, but also significantly enhances the system's ability to detect abnormal interference events in complex electromagnetic environments. It is a fundamental step and prerequisite for dynamic error correction and accurate line loss analysis.

[0023] S2, based on the identified time period with high-frequency harmonic interference, calls a standard active power curve template under the historical operating conditions corresponding to the current energy metering node, performs error fitting between the current active power curve and the standard active power curve, and extracts the active power deviation vector of the current energy metering node under the influence of the interference; For the time periods identified as having high-frequency harmonic interference, an error fitting method based on a standard active power curve template was constructed to quantify and locate the deviation of electric energy metering data under the influence of interference. By extracting the active power deviation vector, the degree of disturbance caused by high-frequency interference on metering accuracy can be accurately reflected. The specific steps include: During a time period identified as containing high-frequency harmonic interference, the active power measurements continuously recorded by the corresponding energy metering node are extracted to construct the current active power curve. This curve uses time as the horizontal axis and the active power output of each cycle by the energy metering chip as the vertical axis, forming a power change curve with time series characteristics. To improve fitting accuracy, the extracted data should be linearly interpolated to ensure curve consistency at different sampling frequencies. In addition, to facilitate alignment with the template data, the current curve is normalized to make its dimension uniform and more comparable.

[0024] From the operational database of the platform to which the energy metering node belongs, a historical active power curve template corresponding to the node under normal operating conditions is retrieved. The selection of the template curve requires comprehensive consideration of the terminal device type corresponding to the energy metering node, the operating time period (e.g., weekdays vs. non-weekdays), the load level, and historical load characteristics. The template curve is then matched to the standard active power variation pattern that best matches the current operating scenario. The template curve is generated through historical big data clustering analysis and a typical curve extraction mechanism. It features strong representativeness, stable time series, and clear morphology, making it a key reference for assessing the degree of deviation from the current curve.

[0025] The template curve can be generated by using cluster analysis and typical curve extraction mechanism based on the large amount of historical active power data accumulated in the electric energy metering platform. The specific steps are as follows: The historically collected active power curves are preliminarily screened and classified according to the terminal device type, operating period, load level and other attributes of the electric energy metering node to ensure the similarity of the input data.

[0026] Using time series clustering algorithms (such as K-means dynamic time warping variant, DBSCAN, etc.), curves with similar power change trends are clustered into several sub-cluster groups. In each sub-cluster, one or more most representative typical curves are extracted as standard templates for this type of operating state through central curve extraction, mean curve extraction, or morphological matching optimization.

[0027] This template curve has obvious timing stability, clear load change patterns and small error interference. Therefore, it can be used as a reliable benchmark for fitting comparison in practical applications. Its function is to provide a reference form for subsequent error identification of the current operating power curve, thereby effectively determining whether there is abnormal deviation and assisting in judging whether high-frequency interference has a substantial impact on measurement accuracy. It is the key basis in the entire error tracing analysis chain.

[0028] Using a time series fitting error-based approach, the current active power curve is compared with the standard active power curve template for error fitting analysis. In this solution, a dynamic time warping algorithm (Dynamic Time Warping) or a window alignment fitting strategy is preferably used to align the shapes of the two power curves on the time axis, and the deviation between the power values ​​at each moment is calculated point by point to construct a power deviation sequence. To enhance the ability to characterize the impact of disturbances, multiple statistical features such as peak offset, average offset, and deviation volatility are further extracted from the deviation sequence to comprehensively construct an active power deviation vector. This vector can be regarded as the deviation signature of the power curve of the metering node under the influence of high-frequency harmonic interference, quantitatively reflecting the degree and form of the interference impact.

[0029] The implementation process of fitting analysis includes the following key steps: The current active power curve and the selected standard active power curve template are normalized separately to ensure the consistency of the two curves in numerical dimension and time scale, so as to eliminate the deviation influence caused by different operating scenarios or measurement equipment.

[0030] The Dynamic Time Warping (DTW) algorithm is used to nonlinearly align the two curves along the time axis. This involves constructing a distance matrix, comparing the distances between each sampling point on the curves, and searching for the alignment path with the minimum cumulative distance among all possible paths. This allows for morphological matching of data with unequal lengths or asynchronous fluctuations. For scenarios where the curves are of the same length or synchronization is more critical, a window alignment strategy can be used. By setting a fixed-length sliding window, the amplitude difference between the two curves at the same time position is calculated point by point. After alignment is complete, the power difference between the current curve and the template curve is extracted at each time point to construct a time-varying power deviation sequence.

[0031] This sequence truly reflects the degree of deviation between the energy metering curve and the normal reference curve within the high-frequency interference section, and serves as the fundamental data source for subsequently constructing the power deviation vector, determining the cause of abnormal energy errors, and generating correction parameters. This refined fitting method enables fine-grained dynamic deviation characterization of energy metering data under the influence of high-frequency disturbances, enhancing the accuracy of error location and the intelligence of disturbance response.

[0032] A one-to-one correspondence is established between the extracted active power deviation vector and the interference time segment, and the vector is stored in a structured manner in the deviation tracing data set for use by the subsequent abnormal cause identification module. This deviation vector not only contains the overall deviation characteristics of the disturbance segment, but also retains the dynamic response pattern of the power curve under interference, which can provide a key quantitative basis for determining whether the interference originates from the abnormal response behavior of the electric energy metering chip. In addition, this step bypasses the drawbacks of traditional absolute threshold-based judgment through template matching and fitting, avoiding misjudgment due to the complexity of grid load changes. It has good adaptability and non-obvious judgment strategies, significantly enhancing the electric energy metering system's ability to trace and analyze complex interference environments.

[0033] The core function of this step is to quantitatively analyze and extract features of the active power deviations generated by energy metering nodes under high-frequency harmonic interference, providing an accurate and traceable data foundation for subsequent abnormal error identification and dynamic correction. In the actual operation of smart grids, due to the complex power consumption environment, energy metering data may experience short-term deviations during interference periods. However, it is often difficult to determine whether these deviations are caused by high-frequency harmonic interference or whether they have systemic impacts simply through absolute numerical determination. Through this step, the active power variation curve of the current energy metering node is first extracted during the identified interference time period and compared with the standard active power curve template generated under historical operating conditions. This effectively constructs a reference baseline that conforms to the characteristics of the node and represents normal operation. Based on this, a fitting analysis is used to identify the deviations between the current power curve and the standard template on the time and amplitude axes. Multiple statistical features that constitute the power deviation vector, such as peak error, mean error, and fluctuation frequency, are extracted, thus achieving a structured characterization of abnormal energy metering behavior under interference. This deviation vector not only quantifies the degree of anomaly in the current data but also serves as an input feature for subsequent anomaly classification models, helping to determine whether the error is caused by metering chip distortion. In short, this step plays a key role in the overall methodology, transforming interference perception into a quantitative expression. It serves as a bridge from "detecting interference" to "identifying the nature of the error," significantly improving the accuracy, interpretability, and practicality of energy metering error analysis.

[0034] S3: The active power deviation vector, the terminal device type corresponding to the current energy metering node, the instantaneous load change characteristics, and the high-frequency harmonic spectrum characteristics are input into the power anomaly cause classification model. The output classifies whether the power deviation is caused by the energy metering chip being interfered with by high-frequency harmonics and marks it as an abnormal energy error segment. To further determine whether the extracted active power deviation vector is caused by the abnormal response behavior of the energy metering chip in a high-frequency harmonic interference environment, a multi-feature fusion-based power anomaly cause classification model is proposed. By constructing a discrimination mechanism based on device type, load dynamic characteristics, and spectral characteristics as input, it can achieve intelligent identification of power deviation properties and accurate labeling of abnormal error fragments. This specific implementation method includes the following steps: The terminal device type information corresponding to the current node is extracted from the energy metering system and used as a static prior attribute to input the classification model, which influences the characteristics of power fluctuations. Different types of terminal devices (such as welding equipment, industrial air conditioners, and electric vehicle charging stations) have significantly different load characteristics and anti-interference capabilities during operation, which may cause different forms of spectrum perturbations and power curve fluctuations. Therefore, setting device type as an independent input feature in the model helps improve the model's adaptability and recognition accuracy in multiple scenarios.

[0035] Extract the instantaneous load variation characteristics within the current interference time segment, primarily including indicators such as the load current change rate, voltage fluctuation amplitude, and active power rise / fall gradient, as time-sensitive features reflecting the load's dynamic response behavior. These characteristics are obtained by calculating the first-order or second-order derivatives of the power parameters within continuous time segments. Combined with the suddenness of the disturbance and nonlinear load characteristics, they help determine whether power deviations are caused by abnormal load operation or whether they exhibit obvious external interference characteristics.

[0036] Based on the voltage and current spectra obtained through the aforementioned frequency domain transformation, high-frequency harmonic spectrum features are extracted, including frequency-domain statistics such as the energy proportion of high-order harmonic components, the location of harmonic peak frequencies, and the rate of spectrum change. These features can be used to characterize the spectral shape and intensity variations of the interference signal, serving as an important basis for identifying the interference source in the model. By combining spectral features with the power deviation vector as input, the model's ability to identify the specific causal path of "misjudgment of the metering chip due to harmonic interference" can be enhanced.

[0037] The active power deviation vector, terminal device type, instantaneous load variation characteristics, and high-frequency harmonic spectrum characteristics are fed into a trained power anomaly cause classification model. The model, which can be built using a supervised learning algorithm (such as a gradient boosting tree, random forest, or lightweight neural network), outputs a classification result of the power deviation cause. If the model determines that the power deviation is highly correlated with high-frequency harmonic interference and inconsistent with the load's inherent operating logic, the corresponding time segment is marked as an "abnormal power error segment under harmonic interference." This segment labeling serves as a prerequisite for the subsequent generation of error correction parameters, establishing a complete closed-loop path from interference identification and error quantification to cause attribution at the data level.

[0038] To build a classification model for the causes of power anomalies based on a supervised learning algorithm, it is first necessary to construct a high-quality training sample set. This sample set consists of annotated active power deviation events in historical power metering data. Each sample contains four types of input features: active power deviation vector, terminal device type, instantaneous load change characteristics, and high-frequency harmonic spectrum characteristics. The label is the cause type of the corresponding deviation event (such as "normal load fluctuation", "high-frequency harmonic interference", "metering chip failure", etc.). Next, the structured sample data is subjected to feature normalization, category encoding (such as unique hot encoding of device type), and data cleaning to ensure the uniformity and validity of the input features. Subsequently, a suitable supervised learning algorithm is selected. For example, the gradient boosting tree can be used to process nonlinear, heterogeneous feature data and has strong interpretability, the random forest has strong robustness and resistance to overfitting, and the lightweight neural network is suitable for large-scale and diverse data scenarios. During model training, cross-validation and hyperparameter tuning are used to improve model generalization capabilities. After training, model performance is evaluated using metrics such as precision, recall, and F1 score. The causal classification results output by the model undergo logical verification and expert review to ensure their stability and reliability in practical applications. Ultimately, the model can be integrated into an electric energy data processing platform to automatically determine the cause, dynamically label, and respond to real-time power deviation data.

[0039] The core function of this step is to use an intelligent discrimination mechanism to causally classify active power deviations that occur during energy metering. This allows for clear identification of whether the deviations are caused by abnormal response behavior of the energy metering chip in an environment with high-frequency harmonic interference, thereby accurately labeling abnormal energy error fragments. In actual power grid operations, active power deviations can be caused by a variety of factors, such as normal load fluctuations, equipment startup and shutdown, grid voltage disturbances, and external high-frequency harmonic interference. Numerical deviations in power data alone cannot accurately determine their causes, potentially leading to false alarms, miscalibration, and even misidentification of power theft. Therefore, by constructing a classification model for the causes of energy anomalies that integrates multidimensional features, it is possible to effectively distinguish abnormal behaviors from different sources at the data level. Specifically, the model uses the active power deviation vector as the core behavioral feature, the terminal device type as the static attribute of the operating environment, the instantaneous load change characteristics as the dynamic response indicator, and the high-frequency harmonic spectrum characteristics as the external interference label to form a complete feature input vector. The deviation causes are then classified and identified using supervised learning models (such as gradient boosting trees, random forests, or lightweight neural networks). When the model output indicates a deviation from metering error caused by high-frequency interference, the current period is marked as an "abnormal power error segment," providing a precise target range for subsequent correction mechanisms. This step establishes an intelligent decision-making path from power offset perception to cause analysis, opening up the logical chain of interference identification, error diagnosis, and data correction. This significantly improves the power metering system's ability to understand and respond accurately to complex interference environments, and serves as a key link in connecting perception and processing within the entire intelligent line loss detection system.

[0040] S4, for abnormal power error segments, generates corresponding high-frequency harmonic interference correction factors based on their time duration, numerical offset direction, and numerical offset amplitude characteristics, and constructs a short-period data correction weight sequence at the power metering node level; To quantitatively correct abnormal energy error fragments that occur in energy metering nodes under the influence of high-frequency harmonic interference, a correction factor generation method based on the characteristics of abnormal fragments is proposed. A short-period data correction weight sequence at the energy metering node level is constructed for subsequent accurate repair of energy data. This implementation method includes the following steps: For identified abnormal power error segments, key characteristic parameters are extracted, including duration, direction, and amplitude. Duration measures the duration of the interference and is crucial for controlling the distribution of correction intensity along the time axis. The direction of the deviation indicates whether the error is biased high or low, crucial for determining the direction of the correction. The amplitude reflects the severity of the error and is the core parameter for determining the magnitude of the correction. These characteristic parameters are obtained through statistical calculation of power deviation data, ensuring targeted and accurate correction processing.

[0041] Based on the above characteristics, a correction factor corresponding to each abnormal segment is constructed. Specifically, if the offset direction is positive, the correction factor is negative, and vice versa, so that the correction effect exhibits symmetrical compensation characteristics. The absolute value of the correction factor is set according to the magnitude of the offset amplitude; the larger the offset amplitude, the stronger the correction factor. Taking into account the influence of the interference duration, the effect of the correction factor can be distributed within the segment according to the time gradient, so that the correction weight at the beginning and end of the segment changes gradually, thereby avoiding abrupt corrections that would cause distortion of the power curve structure.

[0042] Each correction factor is precisely mapped to the corresponding abnormal segment interval in a time series, and then divided into several short-period time periods, such as one interval per second or half a second. Within each interval, different weights are assigned based on its relative position within the entire time period, forming a set of continuous, time-sensitive correction weights. This process ensures that each correction weight corresponds to a specific power measurement point, enabling precise adjustment at the sampling point level. This approach constructs a correction weight sequence that is both segment-continuous and time-flexible, ensuring the smoothness and structural stability of data correction.

[0043] This step aims to quantitatively correct identified abnormal energy error fragments, establishing a correction mechanism with time-series awareness and interference sensitivity, thereby accurately restoring the true state of energy metering data under the influence of high-frequency harmonic interference. In actual smart grid environments, high-frequency harmonic interference is characterized by sudden, intermittent, and localized characteristics, which can cause energy metering chips to experience systematic offsets within a short period of time. Simply marking abnormal fragments and shielding or eliminating related data not only fails to retain critical energy consumption information, but also may undermine the continuity and analytical integrity of the overall power curve.

[0044] Therefore, this step constructs a targeted correction factor by extracting the core features of each abnormal segment, such as the time duration, offset direction, and offset amplitude, and further decomposes it into a continuous correction weight sequence at the sampling granularity to achieve flexible adjustment of the original data. This correction method not only takes into account the numerical size of the error, but also integrates the interference persistence and trend information. It can apply dynamic compensation to the sampling points affected by the interference point by point without destroying the original data structure, significantly improving the smoothness and credibility of data repair. The final correction result will be used in subsequent key links such as power grid line loss calculation and early warning analysis, effectively avoiding global assessment deviations caused by the diffusion of local abnormal errors, and ensuring the continuity and accuracy of the smart grid data processing process. Therefore, this step plays a key role in the transition from "anomaly identification" to "error correction" in the entire method system, and is an important technical link to support the dynamic error compensation mechanism.

[0045] S5, based on the short-cycle data correction weight sequence, performs segment-level weighted correction on the original electric energy metering data, calculates the power consistency check matrix between multiple electric energy metering nodes, and comprehensively corrects the systematic errors caused by high-frequency harmonic interference in the electric energy metering data of each node; To address the systematic offset problem caused by local energy metering errors in multi-node power grid data due to high-frequency harmonic interference, a segment-level weighted correction method based on a short-period data correction weight sequence is proposed. Furthermore, an energy consistency verification mechanism is used to achieve overall correction and balance of metering data across multiple energy metering nodes. This includes the following steps: Based on the short-cycle data correction weight sequence constructed above, it is applied to the original power data of the corresponding power metering node, and the measurement values ​​of the abnormal power error segments are adjusted at the segment level. The specific approach is: for each sampling point identified as affected by interference, the corrected power value is calculated according to its correction weight in the time series to obtain a new power time series. This correction operation is performed in a point-to-point manner to ensure the temporal continuity and physical rationality of the data. Outside the segment, that is, the data part that is not affected by the interference, the original measurement value remains unchanged, thereby constructing a complete power curve that integrates the original data and the corrected segment, realizing the retention of the full curve based on local interference compensation.

[0046] The corrected power data from each energy metering node is uniformly converted into energy data with equal periods (e.g., per minute, per 5 minutes, etc.). The energy data between different nodes is aligned along the time axis to facilitate cross-node comparison. In actual power grid structures, multiple energy metering nodes are often located within the same power supply branch or substation. Their main incoming lines and branch outgoing lines should adhere to the energy conservation relationship. Based on this relationship, the energy logical association structure is extracted and energy consistency relationship pairs are constructed between nodes. This forms multiple pairs of "parent-child" or "adjacent node" combinations for subsequent consistency verification.

[0047] Based on the above-mentioned node combination relationship, an electricity consistency check matrix is ​​constructed. Each element of the matrix is ​​used to describe the difference in electricity between two related electricity metering nodes within a specific time period. For example, in a typical bus-branch architecture, the electricity of the bus node is used as a reference and compared with the sum of the electricity of all the downstream branch nodes to calculate the difference in electricity per unit time. Under normal operating conditions, the difference should be kept within an acceptable error tolerance range; if the difference deviates for a long time, it means that the metering data of one or some nodes still has residual errors or has not been fully corrected. Through this check matrix, the global metering consistency level can be quantitatively evaluated and possible error concentration nodes can be identified.

[0048] The process of constructing an energy consistency check matrix is ​​based on the physical connections between energy metering nodes in the power grid topology. The energy data between nodes is systematically compared according to the principle of energy conservation. First, the grid operation model determines to which parent node each node logically belongs or whether a parallel connection exists. This results in several pairs of "energy-related node pairs," such as an incoming node and its subordinate branch nodes. Then, within a unified time period (e.g., every 5 or 15 minutes), the energy data corresponding to each node is aggregated. For each group of related nodes, the difference between the energy of the reference node (e.g., the bus or parent node) and the sum of the energy of the target nodes is calculated and entered into the corresponding position in the check matrix. Each row of the matrix represents a time period, and each column represents a set of energy comparison relationships. The values ​​in the matrix reflect the magnitude and direction of the energy differences. Analysis of this matrix can quickly identify whether there is unreasonable energy distribution between energy metering nodes and, furthermore, determine whether any nodes have uncorrected deviations in their correction values. This is a crucial foundation for global data consistency assessment and correction adjustments.

[0049] According to the degree of deviation in the power consistency check matrix, the short-cycle correction weights of the nodes with prominent deviations are further adjusted. This adjustment is based on the principle of minimum deviation. By fine-tuning the correction amplitude without destroying the shape of the single-node data curve, the overall error is minimized. After the adjustment is completed, a new corrected power sequence is regenerated, and the check matrix operation is performed again until the overall consistency error of the system converges to within the tolerance threshold. Through the above process, the transition from single-node fragment-level correction to multi-node global consistency correction is achieved, which effectively improves the physical accuracy and structural rationality of the data and suppresses the diffusion of secondary measurement errors caused by interference correction imbalance.

[0050] The core purpose of this step is to further improve the overall data consistency and logical closure of the entire power grid metering system, building on the local error correction achieved at individual energy metering nodes. This effectively eliminates systematic errors caused by high-frequency harmonic interference. In complex distribution networks, multiple energy metering nodes have clear physical power supply relationships, such as between trunk lines and branches, between substation buses and outgoing lines, and between distribution transformers and end users. The power at these nodes should statistically satisfy the principle of energy conservation. However, due to the regional and asymmetric nature of high-frequency harmonic interference, its impact on different nodes varies. Even if local segment correction is achieved at the node level, without coordinating inter-node relationships, system-wide energy distribution imbalances may still occur, compromising the accuracy of line loss analysis and load assessment. Therefore, by combining each node's raw energy metering data with its corresponding short-period correction weight sequence for segment-level weighting, fine-grained error compensation can be achieved within the interference segment. Furthermore, by constructing an energy consistency check matrix, the energy levels of multiple nodes are time-series compared, quantifying their differences, and thus revealing the distribution characteristics of residual errors across nodes. This process not only helps identify nodes with under- or overcorrection but also enables secondary balancing of errors across the entire network. Ultimately, this step transitions from "local correction" to "system consistency optimization," preserving the accuracy of repairs at each metering point while ensuring the consistency and physical rationality of data across the entire network. This is a key step in ensuring the accuracy of line loss detection, power tracing, and energy efficiency assessment in smart grids.

[0051] S6: Input the corrected energy metering data into the power grid line loss analysis module, update the warning threshold and error tolerance boundary in the line loss trend model, and realize closed-loop correction of energy metering interference data and real-time stable identification and response control of smart grid line loss anomalies; To achieve closed-loop correction of energy metering data affected by high-frequency harmonic interference and ensure the accuracy and responsiveness of line loss analysis in interference environments, a process is proposed to dynamically update the line loss trend model based on the corrected data. This process uses the corrected energy metering data as input to reconstruct the basis for line loss determination and update the warning threshold and tolerance boundaries, thereby achieving stable identification and real-time control of line loss anomalies in smart grids. Specifically, it includes the following steps: After short-term weight correction and node consistency leveling, the energy metering data is reorganized according to the grid topology. The corresponding line loss value for each power supply unit or distribution branch is recalculated, including the difference between the incoming power and each outgoing power within a unit time period. This results in a corrected basic line loss data sequence. To ensure data continuity and model input consistency, this sequence must be time-aligned and dimensionally unified with historical line loss data to ensure that no abnormal mutations or scale mismatches are introduced at the model input layer.

[0052] Based on the corrected line-loss data, a structured update is performed on the existing line-loss trend model. A line-loss trend model is a mathematical or statistical model used to describe the variation in line losses in a specific power grid region over different time periods. It often includes time series models, regression prediction models, or multi-factor regression models. During the update process, the time window corresponding to the corrected data segment is preferentially introduced. The original noisy data is replaced with the real data within this window, and model parameters, such as the sliding mean, volatility, and gradient of change, are recalculated. This ensures that the model reflects the actual operating status of the current network more realistically, stably, and with predictive accuracy.

[0053] Based on the updated line loss trend model, the reference threshold and error tolerance for line loss anomaly warnings are dynamically adjusted. Traditional warning mechanisms often use static thresholds or empirical settings, which are prone to false alarms or missed alarms in complex electromagnetic interference environments. However, by introducing the latest correction data and combining it with the dynamic output of the trend model, adaptive adjustment of the warning threshold can be achieved. For example, based on indicators such as the confidence interval and residual distribution of the model output, a multi-level threshold response system is established for different operating conditions. The tolerance margin is relaxed when the load fluctuates significantly, while the error tolerance range is compressed under stable conditions, thereby improving the system's ability to identify true anomalies and its interference tolerance.

[0054] This dynamically updated judgment system is integrated with the existing line-loss analysis process to form a closed-loop mechanism capable of self-sensing, self-correcting, and self-adjusting operations. When new data flows in, the system first completes error identification and correction operations, then embeds the correction results into the line-loss analysis process, triggering trend model retraining and threshold refreshing. Ultimately, the real-time judgment results are used to drive alarm mechanisms, anomaly flagging, or automated operation and maintenance decisions. This mechanism ensures the continuity and credibility of line-loss assessment results even when energy metering data is subject to periodic or intermittent high-frequency interference, significantly enhancing the operational stability and anomaly response capabilities of smart grids in complex environments.

[0055] This step aims to integrate the previous results of identifying and correcting energy metering errors for high-frequency harmonic interference into the power line loss analysis system, establishing a closed-loop response mechanism based on improved data quality. This will comprehensively enhance the stability, accuracy, and adaptability of smart grid line loss anomaly identification. Smart grid line loss analysis typically relies on power and energy data from each energy metering node for calculation and trend modeling. However, in operating environments prone to high-frequency electromagnetic interference, raw metering data may exhibit periodic offsets, localized mutations, or systematic distortion. Directly using this data for line loss modeling and warning threshold setting can lead to false alarms, missed alarms, and even affect energy scheduling and operation and maintenance strategy decisions. This step first inputs high-quality energy metering data, which has undergone short-term weight adjustment and consistency verification, into the line loss analysis process. This eliminates interference-induced error fragments at the data source level, enhancing the stability of trend analysis. Subsequently, by updating the historical data window and prediction structure of the line loss trend model, the model is more closely aligned with current operating conditions, avoiding the accumulation of bias caused by relying on distorted historical data. On this basis, the early warning threshold and error tolerance boundaries are dynamically adjusted to ensure a sensitive but not overly reactive response to different load conditions, equipment operating cycles, or grid topology changes, thus transforming the early warning mechanism from static judgment to dynamic perception. Ultimately, this step forms a full-process closed-loop control chain of "interference identification - error correction - model update - threshold adjustment - stable identification." This enables the power grid to maintain stable, accurate, and real-time line loss monitoring capabilities even in complex electromagnetic environments. This significantly improves the smart grid's ability to perceive operational anomalies and its emergency control capabilities, and is a key link in supporting its high reliability and highly intelligent operation.

[0056] The aforementioned smart grid line loss intelligent detection method based on IoT big data enables a closed-loop processing mechanism, encompassing everything from identifying the time period of high-frequency harmonic interference, tracing the source of errors, determining the cause, to refined corrections and network-wide data consistency verification. This effectively addresses the problems of traditional electricity metering systems in complex electromagnetic environments, such as power misjudgment, line loss misidentification, and false alarms caused by interference. This method not only possesses interference perception capabilities through multi-dimensional feature fusion but also introduces dynamic correction factors and weight distribution strategies to achieve segment- and node-level error compensation. Furthermore, through the power consistency matrix and trend model update mechanism, the accuracy and stability of the corrected data in network-wide analysis are fully guaranteed. Ultimately, this enables the smart grid to maintain high sensitivity, high accuracy, and sustainable response capabilities to line loss anomalies in complex operating scenarios such as high-frequency interference and concentrated nonlinear loads, thereby enhancing the reliability, precision, and intelligence of the line loss analysis system.

[0057] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0058] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0059] It should be noted that, in this document, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0060] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0061] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0062] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0063] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0064] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0065] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0066] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

Claims

1. An intelligent detection method for smart grid line loss based on IoT big data, characterized in that: The following steps are involved: S1 extracts the voltage and current signals of each energy metering node within a set time window, performs frequency domain transformation to obtain the spectrum, calculates the energy concentration index of the high-frequency band, and identifies the high-frequency harmonic interference time segment based on the energy concentration change; S2, based on the identified high-frequency harmonic interference time segment, calls the historical standard active power curve template of the corresponding electric energy metering node, performs error fitting between the current active power curve and the template, and extracts the active power deviation vector; S3: Input the active power deviation vector, terminal device type, instantaneous load change characteristics, and high-frequency harmonic spectrum characteristics into the abnormal cause classification model to determine whether the output power deviation is caused by the high-frequency harmonic interference of the power metering chip, and mark the abnormal power error segment; S4, for abnormal power error segments, generates high-frequency harmonic interference correction factors based on the time duration, numerical offset direction and amplitude, and constructs a correction weight sequence for short-period data at the power metering node level; S5, based on the correction weight sequence, the original electric energy metering data is subjected to segment weighted correction, the electric energy consistency check matrix between multiple electric energy metering nodes is calculated, and the systematic error is corrected as a whole; S6, input the corrected data into the line loss analysis module, update the warning judgment threshold and error tolerance boundary in the line loss trend model, and realize closed-loop correction of electricity metering interference data and real-time and stable identification and response of line loss anomalies.

2. The intelligent detection method for smart grid line loss based on Internet of Things big data according to claim 1 is characterized in that: Step S1 includes: Within the set time window, the voltage and current signals of the energy metering node are continuously collected and weighted by the window function to eliminate the influence of spectrum leakage; Perform fast Fourier transform on the weighted signal to obtain the voltage spectrum and current spectrum, and calculate the frequency domain energy accumulation value of the high frequency band; Calculate the frequency domain energy concentration index to form time series data; Based on dynamic thresholds and disturbance duration judgment, the time segment of high-frequency harmonic interference is identified, and an interference event log is generated for subsequent error compensation.

3. The intelligent detection method for smart grid line loss based on Internet of Things big data according to claim 1 is characterized in that: Step S2 includes: Based on the identified high-frequency harmonic interference time segment, the active power measurement values ​​continuously recorded by the corresponding energy metering node between the start and end time of the high-frequency harmonic interference are extracted, the current active power curve is constructed, and the active power curve is normalized; Call the standard active power curve template under the historical operating conditions of the corresponding node; Perform time-series fitting on the current active power curve and the standard template, use dynamic time warping algorithm or window alignment strategy to align the shapes, and calculate the point-by-point power difference. The statistical characteristics of the power difference sequence are extracted to construct the active power deviation vector, which is used to quantify the impact of high-frequency harmonic interference on the metering node.

4. The intelligent detection method for smart grid line loss based on Internet of Things big data according to claim 1 is characterized in that: Step S3 includes: Extract active power deviation vector, terminal equipment type, instantaneous load change characteristics and high-frequency harmonic spectrum characteristics; The acquired features are uniformly input into the trained power anomaly cause classification model, which is built based on a supervised learning algorithm. The classification result of whether the output power deviation is caused by the energy metering chip being interfered with by high-frequency harmonics; The time segment corresponding to the classification result is marked as an abnormal power error segment and serves as the basis for generating subsequent correction parameters.

5. The intelligent detection method for smart grid line loss based on Internet of Things big data according to claim 1 is characterized in that: Step S4 includes: Extract the time duration, numerical offset direction and numerical offset amplitude of the abnormal power energy error fragment; Generate a corresponding correction factor based on the offset direction and offset amplitude, and set the correction gradient according to the time distribution within the segment; Mapping correction factors to abnormal segments and dividing them into short-period time periods; Correction weights are assigned in each time period to construct a short-period data correction weight sequence at the energy metering node level.

6. The intelligent detection method for smart grid line loss based on Internet of Things big data according to claim 1 is characterized in that: Step S5 includes: Based on the short-cycle data correction weight sequence, the abnormal power error fragments in the original power metering data are corrected at the fragment level to generate a corrected power sequence; Convert the corrected power data of each electric energy metering node into equal-period electric quantity data, and construct an electric quantity consistency check matrix between multiple metering nodes; Based on the degree of deviation of the power difference between each node in the power consistency check matrix, the correction weight of the corresponding node is adjusted to generate new corrected power data; Repeat the verification and correction process until the overall consistency error of the system converges to within the tolerance threshold.

7. The intelligent detection method for smart grid line loss based on Internet of Things big data according to claim 1 is characterized in that: Step S6 includes: The corrected electric energy metering data is aggregated according to the grid topology, and the basic line loss data sequence of each power supply unit is recalculated; Input the corrected basic line loss data series into the line loss trend model, replace the original data of the corresponding time period, and update the model parameters; Adjust the line loss abnormality warning threshold and error tolerance based on the updated line loss trend model; The updated judgment results are integrated with the line loss analysis process to form a closed-loop correction and real-time response control mechanism for electricity metering data.

Images