A method and system for monitoring electric energy

Abstract

The invention discloses an electric energy monitoring method and system, and specifically relates to the technical field of energy monitoring. The method and system are used to monitor the target parameters of electric power equipment in real time, extract the load fluctuation rate characteristics and the sampling interval deviation characteristics, calculate the synchronization accuracy value of the equipment data in each time period, and divide the synchronization accuracy into different levels by comparing and analyzing with the gradient standard threshold. Corresponding processing measures are implemented for different situations. For accurate synchronization, the system maintains normal operation. For inaccurate synchronization, the system triggers an early warning and starts a self-diagnosis and calibration program. For incompletely accurate synchronization, the severity of the synchronization error is predicted, the alarm threshold is adjusted in advance, and corresponding maintenance measures are taken, so as to effectively avoid fault diagnosis errors caused by data asynchrony, improve the safety of electric power equipment and the reliability of the monitoring system, and reduce the risk of accidents.

Description

Technical Field

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

[0002] Power energy monitoring refers to the real-time collection, analysis and management of various data in the power system through advanced technical means to ensure that the operating status, energy consumption, equipment health, etc. of power equipment can be monitored and controlled in a timely and accurate manner. This process usually involves collecting data such as current, voltage, power, temperature, etc. of power equipment such as transformers, distribution boards, generators, etc., and using smart sensors and Internet of Things technology to achieve remote monitoring and early warning functions.

[0003] Through power energy monitoring, enterprises and institutions can optimize the efficiency of their power resources, reduce energy waste, reduce failures, and improve the accuracy of energy management. This not only helps to reduce operating costs, but also improves the safety and reliability of the power system to meet the needs of sustainable development. The widespread application of power energy monitoring systems, especially in smart grids, industrial automation and green buildings, is becoming an important part of modern energy management.

[0004] The prior art has the following deficiencies:

[0005] In large-scale power monitoring systems, hundreds or even thousands of sensors and devices are usually involved, and the data transmission between these devices usually needs to rely on different network nodes for synchronization. In some cases, data synchronization errors may occur due to system configuration or transmission delays. For example, the data collected by some devices are not synchronized to the central monitoring platform in a timely manner, or the data timestamps between different devices are inconsistent. In addition, data asynchrony may cause the monitoring system to be unable to present the comprehensive status of the equipment in real time, resulting in errors in fault diagnosis, especially under high load or emergency conditions, which may lead to the inability to accurately assess the safety status of power equipment and increase the risk of accidents. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for monitoring electric power to solve the deficiencies in the background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for monitoring electric energy, comprising the following steps:

[0008] S1: Determine the target parameters to be monitored according to the characteristics and usage environment of different power equipment, including the current, voltage, temperature and oil level of the transformer, and the current, voltage and load of the switchboard;

[0009] S2: preprocessing the target parameters obtained in several time periods, and extracting features of the preprocessed target parameters, respectively extracting the load fluctuation rate features of the power equipment and the sampling interval deviation features of the sensor data in the target parameters;

[0010] S3: According to the extracted load fluctuation rate characteristics and sampling interval deviation characteristics of the power equipment, the accuracy weight assignment of the target parameter in each time period is determined, and the synchronization accuracy value of the data between the power equipment is obtained by weighted average calculation of the accuracy weight assignment of the target parameter in each time period;

[0011] S4: Compare and analyze the calculated synchronization accuracy value of the data between the power devices with the gradient standard threshold, and divide the synchronization accuracy of the data between the power devices into accuracy synchronization, incomplete accuracy synchronization and inaccuracy synchronization according to the analysis result;

[0012] S5: For accurate synchronization, the power monitoring system operates normally and no intervention is required. For inaccurate synchronization, an early warning signal should be triggered immediately, and the equipment self-diagnosis program should be started to calibrate and maintain the power equipment;

[0013] S6: For incomplete accuracy synchronization, the severity of the synchronization error of the data between the power equipment within a fixed time period is predicted. If the error severity is high, the alarm threshold is lowered in advance and corresponding maintenance measures are taken.

[0014] Preferably, in S2, the extracted load fluctuation rate characteristics of the power equipment are analyzed to generate an abnormal load fluctuation rate index of the power equipment, and the method for obtaining the abnormal load fluctuation rate index of the power equipment is:

[0015] Obtain the current, voltage, temperature and load fluctuation rate of the power equipment from the power equipment sensor, perform normalization, and calculate the normalized data matrix The covariance matrix Σ of is: ; where n is the sample size, is the transpose of the data matrix, and the covariance matrix Σ is decomposed to obtain the eigenvalues ​​and eigenvectors: ;in is the eigenvector, is the corresponding eigenvalue, select the matrix composed of the first k eigenvectors , then the new data is expressed as: ; Where Z is the data after dimensionality reduction, the data after dimensionality reduction Z is inversely transformed to reconstruct the original data ,Right now: ; Reconstruction error is the difference between the original data and the reconstructed data, calculated as: ;in, is the original data of the i-th sample. According to the distribution of the reconstruction error, a threshold is set : ; Where μ(e) is the mean of all reconstruction errors, σ(e) is the standard deviation of all reconstruction errors, α is a constant, and the load fluctuation anomaly index is defined as the ratio of the sample reconstruction error to the threshold, expressed as: ; Among them, QSA is the abnormal index of load fluctuation rate of power equipment.

[0016] Preferably, in S2, the sampling interval deviation characteristics of the extracted sensor data are analyzed to generate a sampling interval deviation index, and the sampling interval deviation index is obtained by:

[0017] Collect the sampling interval data of the power equipment, set the sampling interval data to x(t), where t is the time series and x(t) represents the sampling interval. Select Daubechies wavelet and perform multi-layer wavelet decomposition on the sampling interval data to obtain low-frequency components and high-frequency components. The expression is: ;in, and are the low-frequency and high-frequency components, and They are low-frequency and high-frequency wavelet basis functions, respectively, and the high-frequency components of each scale are calculated , use inverse wavelet transform to reconstruct the coefficients of the low-frequency part and the high-frequency part back to the sampling interval signal, calculate the error between the reconstructed signal and the original signal, and use it as the reconstructed signal of the sampling interval deviation , is the reconstructed signal obtained by inverse wavelet transform; calculate the original signal x(t) and the reconstructed signal The difference between , the expression is: ; Sum the errors at each time point to get the sampling interval deviation index, which is expressed as: ; where N is the total number of data points, and are the i-th sampling points of the original signal and the reconstructed signal respectively, and BVB is the sampling interval deviation index.

[0018] Preferably, in S3, the accuracy weight assignment of the target parameter in each time period is determined according to the extracted load fluctuation rate characteristics and sampling interval deviation characteristics of the power equipment, specifically:

[0019] The power equipment load fluctuation rate anomaly index and sampling interval deviation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as inputs of the machine learning model. The machine learning model predicts the accuracy weighted labels of the target parameters in each time period for each group of comprehensive feature vectors as the prediction target, and minimizes the sum of the prediction errors of the accuracy weighted labels of the target parameters in all time periods as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy weighted values ​​of the target parameters in each time period are determined according to the output results of the model, wherein the machine learning model is a polynomial regression model, and the synchronization accuracy value of the data between the power equipment is obtained by weighted average calculation of the accuracy weighted values ​​of the target parameters in each time period.

[0020] Preferably, in S4, the calculated synchronization accuracy value of the data between the power devices is compared and analyzed with the gradient standard threshold, specifically:

[0021] Compare the acquired synchronization accuracy value of the data between the power devices with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the synchronization accuracy value of the data between the power devices with the first standard threshold and the second standard threshold respectively;

[0022] If the synchronization accuracy value of the data between the power devices is greater than the second standard threshold, it means that the synchronization accuracy of the data between the power devices is high, and a high-accuracy synchronization signal is generated at this time, and the synchronization accuracy of the data between the power devices is divided into accuracy synchronization;

[0023] If the synchronization accuracy value of the data between the power devices is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the synchronization accuracy of the data between the power devices is general, and a medium accuracy synchronization signal is generated at this time, and the synchronization accuracy of the data between the power devices is divided into incomplete accuracy synchronization;

[0024] If the synchronization accuracy value of the data between the power devices is less than the first standard threshold, it means that the synchronization accuracy of the data between the power devices is low. At this time, a low-accuracy synchronization signal is generated, and the synchronization accuracy of the data between the power devices is divided into inaccurate synchronization.

[0025] Preferably, in S6, for incomplete accuracy synchronization, the severity of synchronization error of data between power devices within a fixed time period is predicted, specifically:

[0026] For incomplete accuracy synchronization, that is, the synchronization accuracy value of the data between the power equipment generated within a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the synchronization accuracy values ​​generated in the subsequent fixed time period that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold are collected, and a corresponding data set is established, the mean and standard deviation of the data set are calculated, and after analyzing them, the severity of the synchronization error of the data between the power equipment within the fixed time period is predicted based on the analysis results.

[0027] Preferably, if the mean value of the synchronization accuracy values ​​in the data set is greater than or equal to the reference threshold value of the mean value of the synchronization accuracy values, and the standard deviation of the synchronization accuracy values ​​is less than the reference threshold value of the standard deviation of the synchronization accuracy values, it indicates that the synchronization error between different power devices is slight and the error severity is low, and there is no need to change the alarm threshold value and continue monitoring;

[0028] If the mean value of the synchronization accuracy is greater than or equal to the reference threshold of the mean value of the synchronization accuracy, and the standard deviation of the synchronization accuracy is greater than or equal to the reference threshold of the standard deviation of the synchronization accuracy, it means that the synchronization error between different power equipment fluctuates greatly and the error severity is high. At this time, it is necessary to lower the alarm threshold and take warning;

[0029] If the mean value of the synchronization accuracy is less than the reference threshold value of the synchronization accuracy, and the standard deviation of the synchronization accuracy is greater than or equal to the reference threshold value of the synchronization accuracy, it means that the synchronization error between different power equipment is serious and volatile, and the error severity is high. The alarm threshold should be lowered, early warning should be strengthened, and equipment maintenance calibration should be initiated;

[0030] If the mean of the synchronization accuracy values ​​is less than the reference threshold of the mean of the synchronization accuracy values, and the standard deviation of the synchronization accuracy values ​​is less than the reference threshold of the standard deviation of the synchronization accuracy values, it means that the predicted synchronization error between different power equipment is stable but the error is large, and the error severity is medium. At this time, lower the alarm threshold, strengthen monitoring and take early warning measures.

[0031] The present invention also provides an electric energy monitoring system, including a data acquisition module, a feature extraction module, a synchronization accuracy value calculation module, a synchronization accuracy division module, an early warning and fault diagnosis module and a dynamic adjustment module;

[0032] Data acquisition module: Determine the target parameters to be monitored according to the characteristics and usage environment of different power equipment, including the current, voltage, temperature and oil level of the transformer, and the current, voltage and load of the switchboard;

[0033] Feature extraction module: pre-processes the target parameters obtained in several time periods, and extracts features from the pre-processed target parameters, respectively extracting the load fluctuation rate features of the power equipment and the sampling interval deviation features of the sensor data in the target parameters;

[0034] Synchronization accuracy value calculation module: Determine the accuracy weight assignment of the target parameter in each time period according to the extracted load fluctuation rate characteristics and sampling interval deviation characteristics of the power equipment, and obtain the synchronization accuracy value of the data between the power equipment after weighted average calculation of the accuracy weight assignment of the target parameter in each time period;

[0035] Synchronization accuracy division module: compares and analyzes the calculated synchronization accuracy value of the data between the power devices with the gradient standard threshold, and divides the synchronization accuracy of the data between the power devices according to the analysis result, and divides it into accuracy synchronization, incomplete accuracy synchronization and inaccuracy synchronization;

[0036] Early warning and fault diagnosis module: For accurate synchronization, the power monitoring system operates normally without intervention. For inaccurate synchronization, an early warning signal should be triggered immediately, and the equipment self-diagnosis program should be started to calibrate and maintain the power equipment;

[0037] Dynamic adjustment module: For incomplete accuracy synchronization, the severity of the synchronization error of the data between power equipment within a fixed time period is predicted. If the error severity is high, the alarm threshold is lowered in advance and corresponding maintenance measures are taken.

[0038] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0039] 1. The present invention combines the characteristics of power equipment and the target parameters of the operating environment monitoring, and extracts and analyzes the characteristics of equipment load fluctuation rate and sampling interval deviation, so as to effectively solve the problem of equipment data asynchrony in the power monitoring system. The system can acquire and process the data of multiple devices in real time, obtain the synchronization accuracy value through weighted average calculation, and judge the accuracy of data synchronization based on the threshold, so as to realize real-time classification and response of different synchronization accuracies. For accurate synchronization, the system maintains normal operation; for inaccurate synchronization, the early warning is triggered in time and the equipment self-diagnosis program is started; and for incomplete accuracy synchronization, the severity of the synchronization error is predicted in advance, the alarm threshold is flexibly adjusted and preventive measures are taken.

[0040] 2. The present invention significantly improves the real-time performance and accuracy of the power monitoring system in a large-scale equipment environment, and avoids fault diagnosis errors caused by data synchronization errors. By predicting the synchronization error and adjusting the alarm threshold in real time, the missed and false alarms of faults are reduced, and the maintenance efficiency and safety of power equipment are improved. Especially under high load or emergency conditions, the system can accurately reflect the operating status of the equipment and provide timely and effective maintenance and correction measures, thereby reducing the probability of power system accidents and ensuring the stability and safety of power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced 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.

[0042] Figure 1 The present invention is a flow chart of the method.

[0043] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Example 1, please refer to Figure 1 and Figure 2 As shown, the electric power energy monitoring method described in this embodiment includes the following steps:

[0046] S1: Determine the target parameters to be monitored according to the characteristics and usage environment of different power equipment, including the current, voltage, temperature and oil level of the transformer, and the current, voltage and load of the switchboard;

[0047] S2: preprocessing the target parameters obtained in several time periods, and extracting features of the preprocessed target parameters, respectively extracting the load fluctuation rate features of the power equipment and the sampling interval deviation features of the sensor data in the target parameters;

[0048] S3: According to the extracted load fluctuation rate characteristics and sampling interval deviation characteristics of the power equipment, the accuracy weight assignment of the target parameter in each time period is determined, and the synchronization accuracy value of the data between the power equipment is obtained by weighted average calculation of the accuracy weight assignment of the target parameter in each time period;

[0049] S4: Compare and analyze the calculated synchronization accuracy value of the data between the power devices with the gradient standard threshold, and divide the synchronization accuracy of the data between the power devices into accuracy synchronization, incomplete accuracy synchronization and inaccuracy synchronization according to the analysis result;

[0050] S5: For accurate synchronization, the power monitoring system operates normally and no intervention is required. For inaccurate synchronization, an early warning signal should be triggered immediately, and the equipment self-diagnosis program should be started to calibrate and maintain the power equipment;

[0051] S6: For incomplete accuracy synchronization, the severity of the synchronization error of the data between the power equipment within a fixed time period is predicted. If the error severity is high, the alarm threshold is lowered in advance and corresponding maintenance measures are taken.

[0052] In S1, according to the characteristics and use environment of different power equipment, the target parameters to be monitored are determined, including the current, voltage, temperature and oil level of the transformer, and the current, voltage and load of the switchboard, specifically:

[0053] Transformers are key equipment in power systems, and their operating status directly affects the stability of the entire power grid. The target parameters for transformer monitoring include: current monitoring, using current transformers (CTs) to monitor the input and output currents of transformers. CTs can accurately measure currents and convert high current signals into lower measurable signals. Current transformers (CTs) and smart sensors CTs are used to measure high currents in transformers, which can convert high current signals into low voltage signals for easy data collection and monitoring. Equipped with digital smart current sensors, connected to data acquisition equipment, the current signals are transmitted to the central monitoring platform in real time.

[0054] Voltage monitoring uses a voltage transformer (VT) to measure the high voltage output of the transformer. The voltage transformer is able to reduce the high voltage signal to a low voltage signal suitable for measurement. Voltage transformer (VT) and smart sensor The voltage transformer (VT) converts the voltage signal into a low voltage signal for processing and collection. The smart voltage sensor cooperates with the data acquisition module to collect the voltage data of the transformer in real time and send it to the remote monitoring platform through the Internet of Things (IoT) technology.

[0055] Temperature monitoring, the temperature monitoring of transformers is very critical, usually by installing temperature sensors to monitor the temperature of the oil tank or the core components of the transformer. Temperature sensors (such as thermocouples, thermistors) are installed on the transformer oil level, the surface of the oil tank or the windings of the transformer to monitor the temperature and transmit data through wireless communication modules.

[0056] Oil level monitoring: The oil level of the transformer directly affects the cooling effect. Usually, a liquid level sensor or a float sensor is used to monitor the oil level. The liquid level sensor or float sensor detects the transformer oil level in real time through a float sensor (located inside the oil tank). If the oil level is too low, an alarm will be automatically triggered to prevent the transformer from overheating or damage. Through intelligent liquid level sensors, the oil level information is integrated with other equipment data to achieve multi-parameter synchronous monitoring.

[0057] The distribution board is a key device for power distribution, and its target parameters include current, voltage and load. Monitoring of the distribution board can help operation and maintenance personnel determine the operation of the distribution system and detect faults in a timely manner. Current monitoring uses current transformers (CTs) to monitor the current of each branch line in the distribution board to ensure that the system load does not exceed the designed carrying capacity. Current transformers (CTs) and smart current sensors Current transformers (CTs) are installed on each output line of the distribution board to monitor the current of each line in real time and transmit data to the central monitoring system. Smart current sensors are used to detect the load current of the distribution board in real time, and can transmit data to the cloud through wireless communication technologies such as Wi-Fi or Zigbee.

[0058] Voltage monitoring, monitor the voltage of the distribution board through voltage transformer (VT) or intelligent voltage sensor to ensure that the system voltage is maintained within a safe range. Voltage transformer (VT) and intelligent voltage sensor Voltage transformer is installed at the input end of the distribution board to monitor the voltage level and convert it into a standardized electrical signal. Equipped with intelligent voltage sensor, it collects voltage data in real time and detects voltage fluctuations or instability.

[0059] Load monitoring, load monitoring of the distribution board is usually estimated by collecting data such as current, voltage and power factor to understand the load status of the distribution board in real time. Power sensor, power factor sensor, smart meter power sensor: monitor real-time power data, including active power and reactive power. Detect power factor to determine whether the system is operating within the normal load range to avoid overload operation. Used to monitor the load status of the distribution board as a whole, provide accurate power, current and voltage data, and support remote transmission and real-time alarm.

[0060] All monitoring devices (such as current transformers, temperature sensors, liquid level sensors, power sensors, etc.) transmit the collected real-time data to the central monitoring platform through wireless communication modules (such as LoRa, Zigbee, Wi-Fi or 4G / 5G communication). The monitoring platform processes and analyzes the data to display the working status of the transformer and distribution board in real time, and generate alarms, warnings and maintenance reports.

[0061] Through the Internet of Things (IoT) technology, real-time data is transmitted to the cloud or central server for centralized management and data analysis. This ensures fast information transmission between monitoring devices, reduces data loss and delay, and improves system response speed.

[0062] S2: Preprocess the target parameters obtained in several time periods, and perform feature extraction on the preprocessed target parameters, respectively extracting the load fluctuation rate characteristics of the power equipment and the sampling interval deviation characteristics of the sensor data in the target parameters.

[0063] Preprocess the target parameters obtained in several time periods using statistical methods, such as the 3σ principle (3 times the standard deviation method) to identify and remove outliers that are far from the normal range. For missing or lost sample points, linear interpolation, spline interpolation or other interpolation algorithms can be used to fill the gaps to ensure the continuity and integrity of the data. For the noise in the sensor data, filters (such as Kalman filtering or low-pass filtering) can be applied to remove the noise to ensure more accurate data.

[0064] Since the target parameters of different power equipment have different dimensions (such as current, voltage, temperature, and oil level), it is necessary to standardize the data for subsequent feature analysis. Common standardization methods include: Z-score normalization: By calculating the mean and standard deviation of each target parameter, the data is converted into a standard normal distribution with a mean of 0 and a standard deviation of 1. Min-Max normalization: All data are scaled to the [0,1] interval for easy comparison and analysis.

[0065] In a multi-device system, the sampling frequencies and timestamps of different devices may deviate, so it is necessary to ensure data time synchronization: align the data collected by different devices according to the timestamps to ensure that the data of each device is compared in the same time period. In the case of asynchronous timestamps, interpolation methods can be used to fill in the missing data in the time interval to ensure the continuity of the time window.

[0066] The load fluctuation rate feature refers to the fluctuation amplitude of the load of power equipment, reflecting the stability of the equipment load. Equipment with large fluctuations may mean that the equipment operation status is unstable or the network load is uneven, which may affect the accuracy of data synchronization. Calculate the standard deviation or coefficient of variation of the equipment load: reflect the amplitude of load fluctuation. Calculate the standard deviation (or root mean square value) of the equipment load in each time period, that is, calculate the degree to which the load data deviates from the average value. The ratio of the standard deviation to the average value can be used to measure the relative volatility of the equipment load. If the transformer load fluctuates greatly, it may mean that the equipment load is unbalanced, and system synchronization may be delayed or inaccurate. Calculate the speed at which the load changes over time to reflect the instantaneous change of the load. A common method is to calculate the load change rate by calculating the load change per unit time. If the transformer load changes at a large rate, it may indicate that the equipment is in an unstable state and may affect data synchronization.

[0067] The sampling interval deviation feature refers to the inconsistent intervals at which sensors collect data at different times, which may cause synchronization errors. The deviation of the sampling frequency and timestamp of sensor data will affect the synchronization of data, resulting in disordered data timing or inability to match data from different devices. By calculating the deviation between the time interval of each sampling and the expected sampling interval, the time synchronization error of the device is reflected. The standard deviation of the sampling interval deviation of different devices is calculated to reflect the degree of sampling time inconsistency. If the sampling interval deviation of a device is too large, it may indicate that there is a problem with the clock synchronization of the device, affecting the real-time and accuracy of data collection.

[0068] After analyzing the extracted power equipment load fluctuation rate characteristics, the power equipment load fluctuation rate abnormality index is generated. The method for obtaining the power equipment load fluctuation rate abnormality index is as follows:

[0069] The current (I), voltage (V), temperature (T) and load fluctuation rate (L) of the power equipment are obtained from the power equipment sensor, and these data are standardized to ensure that the mean of each variable is 0 and the variance is 1. Calculate the standardized data matrix The covariance matrix Σ is used to measure the linear relationship between each feature: ; where n is the sample size, is the transpose of the data matrix. Perform eigenvalue decomposition on the covariance matrix Σ to obtain eigenvalues ​​and eigenvectors: ;in is the eigenvector, is the corresponding eigenvalue. According to the size of the eigenvalue, the first k principal components (corresponding to the largest eigenvalue) are selected, and the original data is projected into a low-dimensional space using the selected principal components. Select the matrix composed of the first k eigenvectors , then the new data is expressed as: ; Where Z is the data after dimensionality reduction. Perform an inverse transformation on the reduced-dimensional data Z (using the inverse transformation of the principal component) to reconstruct the original data ,Right now: ; Reconstruction error is the difference between the original data and the reconstructed data, calculated as: ;in, is the original data of the i-th sample. According to the distribution of the reconstruction error, a threshold is set , used to determine whether a sample is abnormal. Usually, the mean reconstruction error plus a multiple of the standard deviation can be selected as the threshold: ; where μ(e) is the mean of all reconstruction errors, σ(e) is the standard deviation of all reconstruction errors, and α is a constant, usually 2-3, which is used to control the sensitivity of anomaly determination. Greater than threshold , then the sample is considered abnormal. The load fluctuation rate abnormality index is defined as the ratio of the sample reconstruction error to the threshold, and the expression is: ; Among them, QSA is the abnormal index of load fluctuation rate of power equipment.

[0070] The larger the abnormal index of load fluctuation rate of power equipment, the greater the abnormality of load fluctuation of equipment, which may reflect the problem of data synchronization. This is because a larger abnormal index indicates that there is a significant deviation between the load fluctuation of equipment and the expected pattern or historical data, which may be due to data collection delay, sensor failure or network delay, resulting in the failure of data synchronization between different devices, thus affecting the real-time and accuracy of the monitoring system. A larger abnormal index usually indicates that the accuracy of data synchronization between devices is poor, and equipment calibration or system optimization may be required immediately.

[0071] On the contrary, if the load fluctuation rate abnormality index of the power equipment is small, it means that the load fluctuation of the equipment is close to the expected value and the synchronization is good. This usually means that the data synchronization between the devices is more accurate, the transmission and acquisition timestamps are consistent, and the system can reflect the status of the equipment in real time. A smaller abnormality index means that there is no obvious deviation in the data, and the system can provide more accurate equipment status information, thereby reducing the error of fault diagnosis and improving the safety and stability of the power system.

[0072] After analyzing the sampling interval deviation characteristics of the extracted sensor data, a sampling interval deviation index is generated. The method for obtaining the sampling interval deviation index is as follows:

[0073] Collect the sampling interval data of the power equipment, and set the sampling interval data to x(t), where t is the time series, and x(t) represents the sampling interval (the unit can be seconds, milliseconds, etc.). Select a suitable wavelet function (such as Daubechies wavelet, Haar wavelet, Symlet wavelet, etc.) for wavelet transform. Wavelet transform can decompose the signal into multiple frequency components so as to analyze the local characteristics of the data at different scales. Common wavelet functions: Haar wavelet: simple and fast, but more sensitive to irregular changes. Daubechies wavelet: suitable for situations where the signal changes are relatively smooth. Symlet wavelet: relatively smooth for the signal.

[0074] Select a wavelet basis function (such as Daubechies wavelet). Perform multi-layer wavelet decomposition on the sampling interval data to obtain low-frequency components and high-frequency components. The expression is: ;in, and are the low-frequency (approximation) and high-frequency (detail) components, and are the wavelet basis functions for low frequency and high frequency, respectively. Calculate the high frequency component of each scale ,if If the amplitude exceeds a certain threshold, it means that the fluctuation of the sampling interval is abnormal. ;in, is the maximum value of the high frequency component.

[0075] Use inverse wavelet transform to reconstruct the coefficients of the low-frequency part and the high-frequency part back to the sampling interval signal, and calculate the error between the reconstructed signal and the original signal as the reconstructed signal of the sampling interval deviation. , is the reconstructed signal obtained by inverse wavelet transform; calculate the original signal x(t) and the reconstructed signal The difference between , the expression is: ; Sum the errors at each time point to get the sampling interval deviation index, which is expressed as: ; where N is the total number of data points, and are the i-th sampling points of the original signal and the reconstructed signal respectively, and BVB is the sampling interval deviation index.

[0076] The larger the sampling interval deviation index, the less accurate the data synchronization between power equipment is. This is because a larger deviation index means that the sampling interval fluctuates significantly in time, which may cause the collected data to fail to accurately and in real time reflect the status of the equipment due to equipment failure, network delay or data loss. Therefore, the inconsistency of the sampling interval will affect the timeliness and consistency of the data, thereby reducing the monitoring system's accurate assessment of the equipment status and increasing errors and risks.

[0077] On the contrary, the smaller the sampling interval deviation index, the higher the accuracy of data synchronization between power equipment. A smaller deviation index means that the sampling interval is basically stable, the data collection and transmission of each device can be kept consistent, and the synchronization is more accurate. This stable synchronization helps to improve the accuracy of the power monitoring system, so that real-time data can better reflect the actual operating status of the equipment, help to promptly discover potential problems, and reduce the risk of failure.

[0078] S3: According to the extracted load fluctuation rate characteristics and sampling interval deviation characteristics of the power equipment, the accuracy weight assignment of the target parameters in each time period is determined, and the synchronization accuracy value of the data between the power equipment is obtained by weighted average calculation of the accuracy weight assignment of the target parameters in each time period.

[0079] The power equipment load fluctuation rate anomaly index and sampling interval deviation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as inputs of the machine learning model. The machine learning model predicts the accuracy weighted labels of the target parameters in each time period for each group of comprehensive feature vectors as the prediction target, and minimizes the sum of the prediction errors of the accuracy weighted labels of the target parameters in all time periods as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy weighted values ​​of the target parameters in each time period are determined according to the output results of the model, wherein the machine learning model is a polynomial regression model, and the synchronization accuracy value of the data between the power equipment is obtained by weighted average calculation of the accuracy weighted values ​​of the target parameters in each time period.

[0080] The method for obtaining the accuracy weight assignment of the target parameter in each time period is as follows: from the comprehensive feature vector training data of the trained machine learning model, the corresponding function expression is obtained: ; In the formula, is the output function of the model, QSA is the abnormal index of load fluctuation rate of power equipment, BVB is the sampling interval deviation index, Assign accuracy weights to the target parameters in each time period.

[0081] S4: Compare and analyze the calculated synchronization accuracy value of the data between the power devices with the gradient standard threshold, and divide the synchronization accuracy of the data between the power devices into accuracy synchronization, incomplete accuracy synchronization and inaccuracy synchronization according to the analysis result.

[0082] Compare the acquired synchronization accuracy value of the data between the power devices with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the synchronization accuracy value of the data between the power devices with the first standard threshold and the second standard threshold respectively;

[0083] If the synchronization accuracy value of the data between the power devices is greater than the second standard threshold, it means that the synchronization accuracy of the data between the power devices is high, and a high-accuracy synchronization signal is generated at this time, and the synchronization accuracy of the data between the power devices is divided into accuracy synchronization;

[0084] If the synchronization accuracy value of the data between the power devices is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the synchronization accuracy of the data between the power devices is general, and a medium accuracy synchronization signal is generated at this time, and the synchronization accuracy of the data between the power devices is divided into incomplete accuracy synchronization;

[0085] If the synchronization accuracy value of the data between the power devices is less than the first standard threshold, it means that the synchronization accuracy of the data between the power devices is low. At this time, a low-accuracy synchronization signal is generated, and the synchronization accuracy of the data between the power devices is divided into inaccurate synchronization.

[0086] S5: For accurate synchronization, the power monitoring system operates normally and no intervention is required. For inaccurate synchronization, an early warning signal should be triggered immediately, and the equipment self-diagnosis program should be started to calibrate and maintain the power equipment.

[0087] When the data synchronization accuracy value of the power equipment is higher than the second standard threshold, the system determines that the accuracy is synchronized. At this time, it means that the data synchronization between the power equipment is stable, and the monitoring system can accurately reflect the equipment status in real time. The data acquisition and transmission of the equipment and sensors are in normal state, and the information consistency is good. The system does not need to perform any operation or adjustment. The monitoring platform can continue to display the operating status of the power equipment in real time, and conduct continuous monitoring and data analysis. At this time, no equipment maintenance or calibration is required because the accuracy of the data has been verified. Since the synchronization accuracy has reached the ideal state, no abnormal warning will be triggered.

[0088] When the data synchronization accuracy value of the power equipment is lower than the first standard threshold, the system determines it as inaccurate synchronization. When the synchronization accuracy value is lower than the first standard threshold, the system should immediately trigger a high-priority warning signal to remind the operator that there is a serious problem with the data synchronization between the devices. The warning signal can be visual, auditory or message push to ensure that relevant personnel receive warnings in time. Through the warning signal, the system should provide a list of devices with low synchronization accuracy and point out possible risks or abnormal equipment status so that operators can evaluate the potential impact.

[0089] The system should immediately initiate the self-diagnosis procedure of the power equipment to check and identify the root cause of the inaccurate data synchronization. The self-diagnosis procedure may include but is not limited to: Check the working status of the sensor to confirm whether there is a hardware failure (such as damaged sensor, loose connection, etc.). Check the network communication status to evaluate whether there is any delay or loss in data transmission. Check the time synchronization of the data acquisition equipment to verify whether there is clock offset or timestamp confusion. Check whether the load fluctuations and other external factors of the power equipment affect the performance of the sensor. The self-diagnosis program can not only provide real-time feedback, but also automatically generate fault reports to help technicians quickly locate the source of the problem.

[0090] Based on the self-diagnosis results, the system can start an automatic calibration procedure to fix problems such as sampling intervals and time synchronization. For example, if the sampling frequency of the sensor is found to be inaccurate or the device clock is out of sync, the system should automatically adjust these parameters to restore the synchronization of the device. If the system detects a hardware failure (such as a damaged sensor, circuit problem, etc.), the maintenance process should be started immediately to replace or repair the device. The maintenance process may include: Replace the sensor: If the sensor is found to be faulty or aging, replace it in time. If network delay or packet loss problems are found, start network optimization or repair measures. Schedule regular maintenance and calibration based on the device status and maintenance logs to prevent synchronization problems from recurring.

[0091] During the calibration and repair process, the system can dynamically monitor the performance of power equipment and sensors and make optimization adjustments. For example, by adjusting the frequency of data collection, increasing network bandwidth or improving algorithms, ensure that future synchronization accuracy is no longer disturbed. During the repair process, the system should strengthen the monitoring frequency of related equipment and sensors to ensure that synchronization problems do not recur. Additional redundant monitoring systems can be introduced to further improve the stability of the system. After the problem is solved, the system should generate a detailed fault analysis report, recording the time of occurrence, handling process and results of the problem. This will help with future fault tracking, prevention and maintenance. Through the analysis of failure modes, the system can retrain existing machine learning models, optimize the model's predictive capabilities, and improve the accuracy of synchronization accuracy prediction.

[0092] S6: For incomplete accuracy synchronization, the severity of the synchronization error of the data between the power equipment within a fixed time period is predicted. If the error severity is high, the alarm threshold is lowered in advance and corresponding maintenance measures are taken.

[0093] For incomplete accuracy synchronization, that is, the synchronization accuracy value of the data between the power equipment generated within a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the synchronization accuracy values ​​generated in the subsequent fixed time period that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold are collected, and a corresponding data set is established, the mean and standard deviation of the data set are calculated, and after analyzing them, the severity of the synchronization error of the data between the power equipment within the fixed time period is predicted based on the analysis results.

[0094] If the mean value of the synchronization accuracy value in the data set is greater than or equal to the reference threshold value of the synchronization accuracy value mean, and the standard deviation of the synchronization accuracy value is less than the reference threshold value of the synchronization accuracy value standard deviation, it means that the synchronization error between different power devices is slight and the error severity is low. No need to change the alarm threshold, continue monitoring.

[0095] If the synchronization accuracy value mean is greater than or equal to the reference threshold of the synchronization accuracy value mean, and the synchronization accuracy value standard deviation is greater than or equal to the reference threshold of the synchronization accuracy value standard deviation, it means that the synchronization error between different power equipment fluctuates greatly and the error severity is high. At this time, the alarm threshold needs to be lowered to detect potential problems in advance and take warnings.

[0096] If the mean value of synchronization accuracy is less than the reference threshold value of synchronization accuracy, and the standard deviation of synchronization accuracy is greater than or equal to the reference threshold value of synchronization accuracy, it means that the synchronization error between different power equipment is serious and volatile, and the error severity is high. At this time, the system should lower the alarm threshold, strengthen early warning, and start equipment maintenance or calibration.

[0097] If the synchronization accuracy value mean is less than the reference threshold of the synchronization accuracy value mean, and the synchronization accuracy value standard deviation is less than the reference threshold of the synchronization accuracy value standard deviation, it means that the predicted synchronization error between different power equipment is stable but the error is large, and the error severity is medium. At this time, it is also necessary to consider lowering the alarm threshold in order to strengthen monitoring and take early warning measures.

[0098] When the system predicts that the severity of the synchronization error is high, the system should lower the alarm threshold in advance to ensure that it can respond before the error increases further. The specific implementation steps are as follows:

[0099] Dynamically adjust the alarm threshold: according to the predicted severity of the error, lower the alarm threshold. This means that even if the synchronization accuracy value is not lower than the original standard threshold, the system will trigger the warning signal at the stage where the error is relatively light. By lowering the alarm threshold, potential synchronization problems can be detected earlier to avoid more serious problems caused by error accumulation. For example: original alarm threshold: between the first standard threshold and the second standard threshold. Adjusted alarm threshold: for example, set to mean -1 standard deviation to make the system more sensitive and able to warn of potential synchronization problems in advance. When the system detects a high degree of error severity, take corresponding maintenance measures in time to ensure device synchronization and system stability: the system will start the automatic calibration program to adjust parameters such as sampling frequency and clock synchronization to ensure the accuracy of data collection. If the system detects problems such as network delay or packet loss, it will immediately start network optimization to ensure the timeliness and integrity of data transmission.

[0100] If the device hardware fails (such as a damaged sensor), the system will automatically initiate a maintenance request and arrange for staff to replace the device. In the case of large errors, the system will increase the frequency of regular inspections to ensure long-term stable operation of the equipment. The system can retrain the algorithm model based on the synchronization error data to optimize future synchronization accuracy prediction and error detection capabilities.

[0101] In this embodiment, the target parameters to be monitored are determined according to the characteristics and usage environment of different power equipment, such as the current, voltage, temperature and oil level of the transformer, and the current, voltage and load of the distribution board, and the collected data is preprocessed and feature extracted, with a focus on extracting the load fluctuation rate feature and the sampling interval deviation feature. Then, based on these features, the accuracy weight in each time period is calculated and weighted averaged to obtain the synchronization accuracy value of the data between the devices. By comparing with the standard threshold, the system divides the synchronization accuracy into accuracy synchronization, incomplete accuracy synchronization and inaccuracy synchronization, and takes corresponding measures accordingly. For accuracy synchronization, the system operates normally; for inaccuracy synchronization, an early warning is triggered and the equipment self-diagnosis and calibration are started; for incomplete accuracy synchronization, the system predicts the severity of the error, reduces the alarm threshold in advance and takes maintenance measures to ensure the stable and safe operation of the power equipment.

[0102] Embodiment 2, an electric energy monitoring system described in this embodiment includes a data acquisition module, a feature extraction module, a synchronization accuracy value calculation module, a synchronization accuracy division module, an early warning and fault diagnosis module and a dynamic adjustment module;

[0103] Data acquisition module: Determine the target parameters to be monitored according to the characteristics and usage environment of different power equipment, including the current, voltage, temperature and oil level of the transformer, and the current, voltage and load of the switchboard;

[0104] Feature extraction module: pre-processes the target parameters obtained in several time periods, and extracts features from the pre-processed target parameters, respectively extracting the load fluctuation rate features of the power equipment and the sampling interval deviation features of the sensor data in the target parameters;

[0105] Synchronization accuracy value calculation module: Determine the accuracy weight assignment of the target parameter in each time period according to the extracted load fluctuation rate characteristics and sampling interval deviation characteristics of the power equipment, and obtain the synchronization accuracy value of the data between the power equipment after weighted average calculation of the accuracy weight assignment of the target parameter in each time period;

[0106] Synchronization accuracy division module: compares and analyzes the calculated synchronization accuracy value of the data between the power devices with the gradient standard threshold, and divides the synchronization accuracy of the data between the power devices according to the analysis result, and divides it into accuracy synchronization, incomplete accuracy synchronization and inaccuracy synchronization;

[0107] Early warning and fault diagnosis module: For accurate synchronization, the power monitoring system operates normally without intervention. For inaccurate synchronization, an early warning signal should be triggered immediately, and the equipment self-diagnosis program should be started to calibrate and maintain the power equipment;

[0108] Dynamic adjustment module: For incomplete accuracy synchronization, the severity of the synchronization error of the data between power equipment within a fixed time period is predicted. If the error severity is high, the alarm threshold is lowered in advance and corresponding maintenance measures are taken.

[0109] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0110] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0111] Those of ordinary skill 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 to be beyond the scope of this application.

[0112] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for monitoring electric power, characterized in that: The following steps are involved: S1: Determine the target parameters to be monitored according to the characteristics and usage environment of different power equipment, including the current, voltage, temperature and oil level of the transformer, and the current, voltage and load of the switchboard; S2: preprocessing the target parameters obtained in several time periods, and extracting features of the preprocessed target parameters, respectively extracting the load fluctuation rate features of the power equipment and the sampling interval deviation features of the sensor data in the target parameters; After analyzing the sampling interval deviation characteristics of the extracted sensor data, a sampling interval deviation index is generated. The method for obtaining the sampling interval deviation index is as follows: Collect the sampling interval data of the power equipment, set the sampling interval data to x(t), where t is the time series and x(t) represents the sampling interval. Select Daubechies wavelet and perform multi-layer wavelet decomposition on the sampling interval data to obtain low-frequency components and high-frequency components. The expression is: ;in, and are the low-frequency and high-frequency components, and They are low-frequency and high-frequency wavelet basis functions, respectively, and the high-frequency components of each scale are calculated , use inverse wavelet transform to reconstruct the coefficients of the low-frequency part and the high-frequency part back to the sampling interval signal, calculate the error between the reconstructed signal and the original signal, and use it as the reconstructed signal of the sampling interval deviation , is the reconstructed signal obtained by inverse wavelet transform; calculate the original signal x(t) and the reconstructed signal The difference between , the expression is: ; Sum the errors at each time point to get the sampling interval deviation index, which is expressed as: ; where N is the total number of data points, and are the i-th sampling points of the original signal and the reconstructed signal, respectively, and BVB is the sampling interval deviation index; S3: According to the extracted load fluctuation rate characteristics and sampling interval deviation characteristics of the power equipment, the accuracy weight assignment of the target parameter in each time period is determined, and the synchronization accuracy value of the data between the power equipment is obtained by weighted average calculation of the accuracy weight assignment of the target parameter in each time period; S4: Compare and analyze the calculated synchronization accuracy value of the data between the power devices with the gradient standard threshold, and divide the synchronization accuracy of the data between the power devices into accuracy synchronization, incomplete accuracy synchronization and inaccuracy synchronization according to the analysis result; S5: For accurate synchronization, the power monitoring system operates normally and no intervention is required. For inaccurate synchronization, an early warning signal should be triggered immediately, and the equipment self-diagnosis program should be started to calibrate and maintain the power equipment; S6: For incomplete accuracy synchronization, the severity of the synchronization error of the data between the power equipment within a fixed time period is predicted. If the error severity is high, the alarm threshold is lowered in advance and corresponding maintenance measures are taken.

2. The electric power energy monitoring method according to claim 1, characterized in that: In S2, the extracted load fluctuation rate characteristics of the power equipment are analyzed to generate the power equipment load fluctuation rate abnormality index. The method for obtaining the power equipment load fluctuation rate abnormality index is: Obtain the current, voltage, temperature and load fluctuation rate of the power equipment from the power equipment sensor, perform normalization, and calculate the normalized data matrix The covariance matrix Σ is: ; Where n is the number of samples, is the transpose of the data matrix, and the covariance matrix Σ is decomposed to obtain the eigenvalues ​​and eigenvectors: ;in is the eigenvector, is the corresponding eigenvalue, select the matrix composed of the first k eigenvectors , then the new data is expressed as: ; Where Z is the data after dimensionality reduction, the data after dimensionality reduction Z is inversely transformed to reconstruct the original data ,Right now: ; Reconstruction error is the difference between the original data and the reconstructed data, calculated as: ;in, is the original data of the i-th sample. According to the distribution of the reconstruction error, a threshold ϵ is set: ; Where μ(e) is the mean of all reconstruction errors, σ(e) is the standard deviation of all reconstruction errors, α is a constant, and the load fluctuation anomaly index is defined as the ratio of the sample reconstruction error to the threshold, expressed as: ; Among them, QSA is the abnormal index of load fluctuation rate of power equipment.

3. The electric power energy monitoring method according to claim 2, characterized in that: In S3, the accuracy weight assignment of the target parameters in each time period is determined according to the extracted load fluctuation rate characteristics of the power equipment and the sampling interval deviation characteristics, which is specifically: The power equipment load fluctuation rate anomaly index and sampling interval deviation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as inputs of the machine learning model. The machine learning model predicts the accuracy weighted labels of the target parameters in each time period for each group of comprehensive feature vectors as the prediction target, and minimizes the sum of the prediction errors of the accuracy weighted labels of the target parameters in all time periods as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy weighted values ​​of the target parameters in each time period are determined according to the output results of the model, wherein the machine learning model is a polynomial regression model, and the synchronization accuracy value of the data between the power equipment is obtained by weighted average calculation of the accuracy weighted values ​​of the target parameters in each time period.

4. The electric power energy monitoring method according to claim 3 is characterized in that: In S4, the calculated synchronization accuracy value of the data between the power devices is compared and analyzed with the gradient standard threshold, specifically: Compare the acquired synchronization accuracy value of the data between the power devices with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the synchronization accuracy value of the data between the power devices with the first standard threshold and the second standard threshold respectively; If the synchronization accuracy value of the data between the power devices is greater than the second standard threshold, it means that the synchronization accuracy of the data between the power devices is high, and a high-accuracy synchronization signal is generated at this time, and the synchronization accuracy of the data between the power devices is divided into accuracy synchronization; If the synchronization accuracy value of the data between the power devices is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the synchronization accuracy of the data between the power devices is general, and a medium accuracy synchronization signal is generated at this time, and the synchronization accuracy of the data between the power devices is divided into incomplete accuracy synchronization; If the synchronization accuracy value of the data between the power devices is less than the first standard threshold, it means that the synchronization accuracy of the data between the power devices is low. At this time, a low-accuracy synchronization signal is generated, and the synchronization accuracy of the data between the power devices is divided into inaccurate synchronization.

5. The electric power energy monitoring method according to claim 1, characterized in that: In S6, for incomplete accuracy synchronization, the severity of synchronization error of data between power equipment within a fixed time period is predicted, specifically: For incomplete accuracy synchronization, that is, the synchronization accuracy value of the data between the power equipment generated within a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the synchronization accuracy values ​​generated in the subsequent fixed time period that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold are collected, and a corresponding data set is established, the mean and standard deviation of the data set are calculated, and after analyzing them, the severity of the synchronization error of the data between the power equipment within the fixed time period is predicted based on the analysis results.

6. The electric power energy monitoring method according to claim 5, characterized in that: If the mean value of the synchronization accuracy values ​​in the data set is greater than or equal to the reference threshold value of the synchronization accuracy value mean value, and the standard deviation of the synchronization accuracy value is less than the reference threshold value of the synchronization accuracy value standard deviation, it means that the synchronization error between different power equipment is slight and the error severity is low, and there is no need to change the alarm threshold value and continue monitoring; If the mean value of the synchronization accuracy is greater than or equal to the reference threshold of the mean value of the synchronization accuracy, and the standard deviation of the synchronization accuracy is greater than or equal to the reference threshold of the standard deviation of the synchronization accuracy, it means that the synchronization error between different power equipment fluctuates greatly and the error severity is high. At this time, it is necessary to lower the alarm threshold and take warning; If the mean value of the synchronization accuracy is less than the reference threshold value of the synchronization accuracy, and the standard deviation of the synchronization accuracy is greater than or equal to the reference threshold value of the synchronization accuracy, it means that the synchronization error between different power equipment is serious and volatile, and the error severity is high. The alarm threshold should be lowered, early warning should be strengthened, and equipment maintenance calibration should be initiated; If the mean of the synchronization accuracy values ​​is less than the reference threshold of the mean of the synchronization accuracy values, and the standard deviation of the synchronization accuracy values ​​is less than the reference threshold of the standard deviation of the synchronization accuracy values, it means that the predicted synchronization error between different power equipment is stable but the error is large, and the error severity is medium. At this time, lower the alarm threshold, strengthen monitoring and take early warning measures.

7. An electric power monitoring system, used to implement an electric power monitoring method according to any one of claims 1 to 6, characterized in that: It includes data acquisition module, feature extraction module, synchronization accuracy value calculation module, synchronization accuracy division module, early warning and fault diagnosis module and dynamic adjustment module; Data acquisition module: Determine the target parameters to be monitored according to the characteristics and usage environment of different power equipment, including the current, voltage, temperature and oil level of the transformer, and the current, voltage and load of the switchboard; Feature extraction module: pre-processes the target parameters obtained in several time periods, and extracts features from the pre-processed target parameters, respectively extracting the load fluctuation rate features of the power equipment and the sampling interval deviation features of the sensor data in the target parameters; Synchronization accuracy value calculation module: Determine the accuracy weight assignment of the target parameter in each time period according to the extracted load fluctuation rate characteristics and sampling interval deviation characteristics of the power equipment, and obtain the synchronization accuracy value of the data between the power equipment after weighted average calculation of the accuracy weight assignment of the target parameter in each time period; Synchronization accuracy division module: compares and analyzes the calculated synchronization accuracy value of the data between the power devices with the gradient standard threshold, and divides the synchronization accuracy of the data between the power devices according to the analysis result, and divides it into accuracy synchronization, incomplete accuracy synchronization and inaccuracy synchronization; Early warning and fault diagnosis module: For accurate synchronization, the power monitoring system operates normally without intervention. For inaccurate synchronization, an early warning signal should be triggered immediately, and the equipment self-diagnosis program should be started to calibrate and maintain the power equipment; Dynamic adjustment module: For incomplete accuracy synchronization, the severity of the synchronization error of the data between power equipment within a fixed time period is predicted. If the error severity is high, the alarm threshold is lowered in advance and corresponding maintenance measures are taken.

Images