Monitoring operation and maintenance system for booster station

Abstract

The invention discloses a booster station monitoring operation and maintenance system, and relates to the technical field of data management. A plurality of heterogeneous intelligent sensors are deployed in a booster station to monitor the operation state of equipment in real time, an electromagnetic field sensor is utilized to sense the electromagnetic interference intensity, and when abnormal high-frequency pulses are detected, a data anomaly detection mode is triggered to detect the abnormal high-frequency pulses; calculating an error compensation value according to the abnormal degree, correcting abnormal data, ensuring the accuracy of the data, performing equipment state evaluation according to the recovered data, and if potential safety hazards are found, sending intelligent early warning to operation and maintenance personnel and providing suggestions for adjusting monitoring strategies or optimizing shielding measures; according to the invention, the anti-interference capability and accuracy of booster station monitoring data are effectively improved, the risk of operation and maintenance misjudgment caused by electromagnetic interference is reduced, the reliability of intelligent monitoring in an ultrahigh-voltage power transmission scene is enhanced, and a powerful guarantee is provided for safe and stable operation of a power system.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a monitoring and operation and maintenance system for a booster station. Background Art

[0002] In modern power systems, booster stations are important hubs for power transmission and conversion, and their operating status directly affects the safety and stability of the power grid. Traditional monitoring and operation of booster stations mainly rely on manual inspections and regular maintenance, which have problems such as poor real-time performance, delayed response, and complex troubleshooting. In recent years, with the development of the Internet of Things, big data, and artificial intelligence technologies, intelligent monitoring and operation systems have been gradually applied to booster station management, realizing real-time monitoring and predictive maintenance of equipment status through data collection, remote monitoring, and intelligent analysis.

[0003] The prior art has the following deficiencies: In a booster station, strong electromagnetic interference (such as electromagnetic pulses caused by lightning strikes and high-voltage equipment failures) may cause abnormal drift in the data collection of smart sensors, thereby affecting the accuracy of the monitoring system. Traditional filtering and error correction algorithms are unable to cope with this high-intensity, instantaneous and sudden interference, resulting in distorted equipment status assessment and even triggering erroneous operation and maintenance decisions. This problem is particularly prominent in ultra-high voltage transmission scenarios, but because its occurrence probability is low and difficult to reproduce, existing systems often lack targeted optimization measures, affecting the reliability and accuracy of the intelligent monitoring and operation of booster stations. Summary of the invention

[0004] The object of the present invention is to provide a booster station monitoring and operation and maintenance system to address the deficiencies in the background technology.

[0005] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a booster station monitoring and operation and maintenance system, including a data acquisition module, a data anomaly detection module, a multi-source data correction module and an early warning adjustment module; Data acquisition module: multiple heterogeneous intelligent sensors are set up in the booster station to collect equipment operation status data in real time; among them, the electromagnetic interference intensity is monitored in real time through the electromagnetic field sensor, and when an abnormal high-frequency pulse is detected, the system is triggered to enter the data anomaly detection mode; Data anomaly detection module: Analyzes the correlation between data collected by different types of sensors, determines the contingency of abnormal drift of multiple sensors at the same time, and analyzes the fluctuation of electromagnetic pulse interference in combination with the readings of electromagnetic field sensors; Multi-source data correction module: Combined with the data anomaly detection results, the module evaluates the degree of data drift anomaly of smart sensors in extreme electromagnetic pulse interference environments, calculates the error compensation value based on the degree of drift anomaly, and corrects the smart sensor data; Early warning and adjustment module: After completing data correction, the device status is evaluated based on the restored data. If extreme electromagnetic pulse interference causes a safety hazard to the equipment, an early warning is sent to the operation and maintenance personnel, and adjustment suggestions are provided, including adjusting the monitoring strategy or strengthening shielding measures.

[0006] Preferably, the data acquisition module includes multiple heterogeneous intelligent sensors, including temperature sensors, vibration sensors, optical sensors, current / voltage sensors and electromagnetic field sensors, which are used to respectively collect the temperature, vibration, discharge phenomenon, current, voltage and electromagnetic interference intensity of the equipment to improve the accuracy of data acquisition.

[0007] Preferably, after analyzing the correlation between the data collected by different types of sensors, the frequency of data drift of different sensors in the same time window is determined, and compared with the historical data under normal working conditions to generate a data drift frequency anomaly index. The method for obtaining the data drift frequency anomaly index is as follows: within a fixed time window, collect time series data of different sensors: suppose the data collected by sensors A and B are: ; ;in: is sensor A at time The collected values, is sensor B at time The collected values, n is the number of data points in the time window; Calculate the Pearson correlation coefficient of sensors A and B under normal operating conditions : ;in: and are the means of historical data, is sensor A at time The collected values, is sensor B at time Collected values, calculate the Pearson correlation coefficient of the current time window : ;in: and are the means of the current time window respectively; Calculate the correlation deviation between the correlation coefficient of the current time window and the historical correlation coefficient : ; Define the data drift frequency as , the proportion of windows whose correlation deviation Δr exceeds the threshold θ: ;in: is an indicator function, if If , it takes 1, otherwise it takes 0, m is the total number of historical time windows; the calculated correlation deviation and data drift frequency The data drift frequency anomaly index is obtained after weighted average summation calculation.

[0008] Preferably, after analyzing the fluctuation of electromagnetic pulse interference, an electromagnetic pulse interference fluctuation index is generated, and the method for obtaining the electromagnetic pulse interference fluctuation index is: In the time window Q, the electromagnetic field strength signal x(t) is collected, where x(t) represents the electromagnetic field strength at time t, and the sampling frequency is assumed to be fs; the short-time Fourier transform is applied to the collected electromagnetic field strength signal, and the calculation formula is: ;in: represents the time-frequency spectrum at time t and frequency f, x(c) represents the electromagnetic field intensity value at the moment when the sampling point index is c, and is a discrete time signal sequence obtained by discretizing the electromagnetic field intensity signal x(t) at the sampling frequency fs. is the window function, f is the frequency component, is the kernel function of short-time Fourier transform; Set the window length to W and calculate the time-varying power spectral density. The expression is: ;in, Represents the power spectrum at time t and frequency f; calculates the power spectrum energy per unit time, the expression is: ;in, represents the electromagnetic field power at time t, and the maximum electromagnetic field power of the electromagnetic pulse is taken as the peak power , and calculate the average power in the entire time window , calculate the electromagnetic pulse interference fluctuation index, the expression is: ; Where QSZ is the electromagnetic pulse interference fluctuation index.

[0009] Preferably, a multi-source data correction module: combines the data anomaly detection results to evaluate the degree of data drift anomaly of the intelligent sensor in the extreme electromagnetic pulse interference environment, normalizes the data drift frequency anomaly index and the electromagnetic pulse interference fluctuation index so that they are both between [0,1], and calculates the data drift anomaly factor of the intelligent sensor based on the normalized data drift frequency anomaly index and the electromagnetic pulse interference fluctuation index.

[0010] Preferably, the acquired smart sensor data drift anomaly factor is compared with a threshold value pre-set according to historical data. If the smart sensor data drift anomaly factor is greater than or equal to the pre-set threshold value, it indicates that the smart sensor data drift anomaly degree in the extreme electromagnetic pulse interference environment is high, and the corresponding smart sensor data is classified as abnormal data, requiring data correction; if the smart sensor data drift anomaly factor is less than the pre-set threshold value, it indicates that the smart sensor data drift anomaly degree in the extreme electromagnetic pulse interference environment is low, and the corresponding smart sensor data is classified as normal data.

[0011] Preferably, the error compensation value is calculated based on the degree of drift anomaly, and the smart sensor data is corrected, specifically including: calculating the data drift amount based on historical data comparison to quantify the error offset, calculating the sensor data drift amount : ;in: is the current sensor measurement value, is the reference value; the error compensation value is used to adjust the abnormal data. Assuming the error compensation value is C, the calculation formula is: ; Where: C is the error compensation value, λ is the drift correction coefficient, which is used to adjust the compensation amplitude: The calculation method of the drift correction coefficient is: ; is the data drift anomaly factor of the smart sensor, is the maximum smart sensor data drift anomaly factor recorded in historical data; Use the compensation value C to correct the original sensor data: ; For the corrected sensor data, perform the correction data deviation verification: calculate the deviation between the corrected data and the historical data , the expression is: ; is the reference value of the sensor data under normal conditions. Less than 5%: This indicates that the calibration data is reliable and can be used continuously. Greater than or equal to 5%: This indicates that the calibration has failed and the data should be discarded or recalibrated.

[0012] Preferably, after completing the data correction, the early warning adjustment module analyzes the equipment status. If it is found that the equipment operating status exceeds the safe range, different levels of early warning information are sent to the operation and maintenance personnel according to the abnormality level, and corresponding adjustment suggestions are provided, including increasing the data collection frequency, optimizing the filtering strategy, strengthening electromagnetic shielding, optimizing the grounding system or adjusting the sensor layout.

[0013] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The present invention realizes accurate monitoring, anomaly analysis, correction and operation and maintenance optimization of intelligent sensor data in electromagnetic interference environment through four major modules: data collection, data anomaly detection, multi-source data correction and early warning adjustment. In view of the problem that traditional technology is difficult to effectively deal with extreme electromagnetic pulse interference, the present invention uses electromagnetic field sensors to monitor the interference intensity in real time, combines multi-sensor data fusion analysis to identify the cause of abnormal data drift, and calculates the sensor data drift anomaly factor through the data drift frequency anomaly index and electromagnetic pulse interference fluctuation index to ensure that data correction is accurate and reliable. In addition, through the error compensation strategy, the interfered data is adaptively corrected to improve the accuracy of data recovery, thereby avoiding distortion of equipment status assessment or false triggering of protection mechanisms due to erroneous data, and improving the operational reliability of the booster station.

[0014] 2. The implementation of the present invention effectively solves the problems of high-intensity transient interference that is difficult to eliminate, low error correction accuracy, and high misjudgment rate in traditional booster station monitoring and operation and maintenance, and improves the anti-interference ability and intelligence level of the system. By dynamically adjusting the data acquisition frequency, optimizing the filtering strategy, strengthening electromagnetic shielding, etc., the stability of the booster station monitoring system in complex electromagnetic environments is improved, accurate assessment of the equipment status is ensured, and the operation and maintenance personnel are notified in time through an intelligent early warning mechanism to prevent equipment anomalies from evolving into serious failures. The present invention is suitable for ultra-high voltage power transmission scenarios, and can be widely used in intelligent operation and maintenance of power systems, enhancing power grid security, and realizing accurate, intelligent, and efficient booster station management. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] 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.

[0016] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0017] 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.

[0018] For examples, see Figure 1As shown, the booster station monitoring and operation and maintenance system described in this embodiment includes a data acquisition module, a data anomaly detection module, a multi-source data correction module and an early warning adjustment module; Data acquisition module: multiple heterogeneous intelligent sensors are set up in the booster station to collect equipment operation status data in real time; among them, the electromagnetic interference intensity is monitored in real time through the electromagnetic field sensor, and when an abnormal high-frequency pulse is detected, the system is triggered to enter the data anomaly detection mode; Data anomaly detection module: Analyzes the correlation between data collected by different types of sensors, determines the contingency of abnormal drift of multiple sensors at the same time, and analyzes the fluctuation of electromagnetic pulse interference in combination with the readings of electromagnetic field sensors; Multi-source data correction module: Combined with the data anomaly detection results, the module evaluates the degree of data drift anomaly of smart sensors in extreme electromagnetic pulse interference environments, calculates the error compensation value based on the degree of drift anomaly, and corrects the smart sensor data; Early warning and adjustment module: After completing data correction, the device status is evaluated based on the restored data. If extreme electromagnetic pulse interference causes a safety hazard to the equipment, an early warning is sent to the operation and maintenance personnel, and adjustment suggestions are provided, including adjusting the monitoring strategy or strengthening shielding measures.

[0019] In order to ensure the comprehensiveness and reliability of the data, the data acquisition module uses the following types of heterogeneous intelligent sensors to collect information on different physical quantities in the booster station: Temperature sensor: installed on key equipment such as main transformers, circuit breakers, busbars, etc., to collect temperature data on the equipment surface and surrounding environment in real time to monitor equipment heat dissipation and abnormal heating.

[0020] Vibration sensor: Deployed on the transformer casing, cooling system and circuit breaker mechanism components to detect mechanical vibration characteristics and identify possible looseness, wear or abnormal operating conditions inside the equipment.

[0021] Optical sensors (visible light + infrared): installed at key equipment and connection points, using imaging technology to monitor abnormal discharge, insulation damage and local overheating on the surface of the equipment.

[0022] Current / voltage sensor: used to monitor the current and voltage status of busbars, cables and transformers in real time, and to determine the stability of power transmission and whether there are abnormal power fluctuations.

[0023] Electromagnetic field sensor: This sensor is one of the key innovations of the present invention. It is mainly used for real-time monitoring of the electromagnetic environment inside the substation, especially for rapid detection and early warning of instantaneous high-intensity electromagnetic pulses (such as lightning strikes, arc discharges, switch operations, etc.).

[0024] The electromagnetic field sensor uses a high-sensitivity Hall effect probe or a wide-band electromagnetic wave receiving module, which can detect electromagnetic wave signals ranging from low frequency (50Hz-60Hz industrial frequency electromagnetic field) to high frequency (MHz-GHz level transient pulses). The specific working process is as follows: The sensor continuously monitors the strength of the ambient electromagnetic field and records the baseline value of background electromagnetic noise; dynamic thresholds are set so that the system can distinguish between normal electromagnetic field fluctuations (such as normal electromagnetic radiation when the equipment is running) and abnormal interference signals.

[0025] Analyze real-time electromagnetic waveforms through fast Fourier transform, extract spectrum features, and identify abnormal high-frequency electromagnetic pulses; combine short-time energy analysis to calculate the instantaneous intensity and duration of electromagnetic pulses to determine whether the interference level has been reached; If a large-amplitude electric or magnetic field transition is detected in a short period of time (for example, less than 1 millisecond), and the frequency falls within a specific range (such as 1MHz-100MHz caused by lightning strikes, kHz-MHz caused by equipment switching operations), it is preliminarily judged as a high-frequency interference event.

[0026] To prevent misjudgment, the electromagnetic field sensor cross-compares signal anomalies with other sensor data such as temperature, vibration, and current; for example, if electromagnetic interference occurs at the same time as the optical sensor detects arc discharge, or the current sensor detects instantaneous current fluctuations, the source of the interference can be further confirmed and marked.

[0027] When the electromagnetic field sensor detects an abnormal high-frequency pulse and after signal verification and confirmation, the system automatically enters the data anomaly detection mode and performs the following operations: The raw data of all sensors at the current moment are stored and marked as an abnormal moment to prevent subsequent calculation errors based on abnormal data.

[0028] Increase the frequency of data collection and improve the data update rate to more accurately capture abnormal trends.

[0029] Send electromagnetic interference warnings to the monitoring center or operation and maintenance personnel, providing information such as interference type, intensity, and impact range.

[0030] Provide basic data for subsequent abnormal data correction and equipment status assessment, ensuring that the system can automatically or manually handle data drift problems that may be caused by electromagnetic interference.

[0031] Data anomaly detection module: When the booster station is in normal operation, the data of various sensors should conform to certain physical laws. For example, temperature changes should be related to load conditions, current and voltage fluctuations should conform to power transmission characteristics, and vibration sensor signals should correspond to the working status of mechanical equipment. This module compares the data acquisition timestamps of different sensors to ensure that the data is in the same time window during analysis to avoid incorrect association judgments caused by data acquisition delays.

[0032] Sliding window statistical analysis is used to calculate the trend and amplitude of data changes of each sensor within a period of time (such as 1 second, 5 seconds, 1 minute), and establish a time series curve; for example, if the values ​​of the temperature sensor and the vibration sensor rise synchronously and the trend is consistent with the historical operating mode, the data is considered normal; but if the temperature data suddenly increases and the vibration data does not change significantly, there may be a sensor false alarm or a sudden abnormal event.

[0033] Calculate the correlation coefficient between different types of sensor data, for example: temperature sensor vs. load current sensor → temperature changes are usually closely related to load conditions; vibration sensor vs. equipment switch status sensor → when the equipment is turned on or off, the vibration should change accordingly; optical sensor (infrared temperature) vs. physical temperature sensor → monitor whether the temperature rise is reflected at the same time. If the abnormal change of a sensor cannot find a reasonable corresponding relationship in the data of other sensors, it may be due to the sensor itself, rather than external interference.

[0034] When multiple sensors are determined to be abnormal, it is necessary to further determine whether the probability of their simultaneous occurrence is accidental or is affected by common external factors (such as electromagnetic interference). This module uses the following methods to evaluate the randomness of abnormal drift: Calculate the frequency of data drift in different sensors within the same time window and compare it with historical data under normal working conditions; if multiple sensors drift significantly at the same time within a short period of time (milliseconds or seconds), and the amplitude is far beyond the historical fluctuation range, it indicates that the abnormal drift may not be a random phenomenon, but is affected by external interference.

[0035] If the abnormal sensors are distributed in multiple areas of the entire booster station (for example, different parts of the main transformer, busbar, circuit breaker and other equipment), it may be a systemic problem, such as external interference; if the abnormality is only concentrated in multiple sensors of a single device, it may be a fault of the device itself, rather than a station-wide interference.

[0036] If the data drift is sudden (such as an instantaneous data jump), it may be caused by a short-term strong electromagnetic pulse; if the data drift occurs gradually (such as a slow deviation over a few minutes or longer), it is more likely to be caused by sensor aging or environmental factors (such as temperature and humidity).

[0037] After analyzing the correlation between the data collected by different types of sensors, the frequency of data drift of different sensors in the same time window is determined, and compared with the historical data under normal working conditions to generate the data drift frequency anomaly index. The method for obtaining the data drift frequency anomaly index is as follows: In a fixed time window (such as T = 10 seconds), collect time series data from different sensors: Assume that the data collected by sensors A and B are: ; ;in: is sensor A at time The collected values, is sensor B at time The collected value, n is the number of data points in the time window (such as collecting 10 points per second, then n=100).

[0038] Calculate the Pearson correlation coefficient of sensors A and B under normal operating conditions : ;in: and are the means of historical data, is sensor A at time The collected values, is sensor B at time Calculate the Pearson correlation coefficient for the current time window : ;in: and are the means of the current time window respectively; Calculate the correlation deviation between the correlation coefficient of the current time window and the historical correlation coefficient : If Δr is small (e.g. <0.1), it means that the correlation of the sensor data remains basically stable and the drift is not obvious; if Δr is large (e.g. >0.5), it means that the correlation of the sensor data changes significantly and abnormal drift may occur.

[0039] Define the data drift frequency as , which indicates the proportion of windows whose correlation deviation Δr exceeds a certain threshold θ (such as 0.3) in the last m time windows (such as 60 windows in the last 10 minutes): ;in: is an indicator function, if If yes, it takes 1, otherwise it takes 0, and m is the total number of historical time windows (e.g., if the sampling is 60 times in 10 minutes, then m=60); The calculated correlation bias and data drift frequency The data drift frequency anomaly index DDFAI is obtained by weighted average summation calculation.

[0040] If DDFAI<0.2, it means that the data is stable and there is no obvious drift; if 0.2≤DDFAI<0.5, it means that there may be slight drift and it needs to be observed; if DDFAI≥0.5, it means that there is serious data drift, which may be caused by electromagnetic interference and measures need to be taken.

[0041] After discovering abnormal data from multiple sensors, this module will combine the readings from the electromagnetic field sensor to analyze whether the electromagnetic pulse is the main cause of the data drift.

[0042] Obtain the time series data of the electromagnetic field sensor and compare it with the abnormal occurrence time of other sensors; if the electromagnetic field intensity fluctuates violently (such as suddenly increasing to the kV / m level) instantly before the data drift of multiple sensors, it indicates that the data anomaly is related to the electromagnetic pulse.

[0043] Through the spectrum analysis function of the electromagnetic field sensor, the frequency range of the interference signal is extracted and the characteristics of common interference sources are compared: Lightning interference → more high-frequency components (MHz-GHz), short duration (<1ms), and steep waveform; Switching operation interference → mainly sudden changes in waveforms in the kHz-MHz range, which may last for tens of milliseconds; Arc discharge → longer duration, wider signal frequency, and accompanied by abnormal data from infrared optical sensors. If the interference spectrum characteristics match the known electromagnetic pulse type, it can be confirmed that the data drift is caused by electromagnetic interference.

[0044] The relationship between the electromagnetic pulse intensity and the sensor data drift amplitude is statistically analyzed to analyze whether the degree of data anomaly shows a linear or nonlinear correlation with the electromagnetic field change; for example: if the electromagnetic field fluctuation is small (such as <100V / m), but the data drifts significantly, it may be a sensor failure rather than interference; if the electromagnetic field instantly jumps to the kV / m level, and the data of multiple sensors suddenly change at the same time, it can be determined that the data anomaly is caused by the electromagnetic pulse.

[0045] After analyzing the fluctuation of electromagnetic pulse interference, an electromagnetic pulse interference fluctuation index is generated. The method for obtaining the electromagnetic pulse interference fluctuation index is as follows: In the time window Q, collect the electromagnetic field strength signal x(t), where x(t) represents the electromagnetic field strength at time t (unit: V / m). Assume the sampling frequency is fs (e.g. 10kHz).

[0046] Apply short-time Fourier transform to the collected electromagnetic field strength signal, and the calculation formula is: ;in: represents the time-frequency spectrum at time t and frequency f, x(c) represents the electromagnetic field intensity value at the sampling point index c, which is a discrete time signal sequence obtained by discretizing the electromagnetic field intensity signal x(t) at the sampling frequency fs. That is, ,in, is the sampling period, is the window function (such as Hanning window, Hamming window), f is the frequency component, is the kernel function of the short-time Fourier transform.

[0047] Set the window length to W (e.g. 10ms, corresponding to W=0.01×fs sampling points). Set the window overlap rate (e.g. 50%). Calculate the number of windows M: ; Wherein, the window step size = W × (1-overlap rate), N is the total number of data points of the signal. Calculate the time-varying power spectral density, the expression is: ;in, Represents the power spectrum at time t and frequency f; calculates the power spectrum energy per unit time, the expression is: ;in, represents the electromagnetic field power at time t, and the maximum electromagnetic field power of the electromagnetic pulse is taken as the peak power , and calculate the average power in the entire time window , calculate the electromagnetic pulse interference fluctuation index, the expression is: ; Where QSZ is the electromagnetic pulse interference fluctuation index.

[0048] If QSZ<0.3: The electromagnetic interference fluctuation is small and has little impact on the system. If 0.3≤QSZ<0.7: The electromagnetic interference fluctuation is obvious and the impact range needs to be monitored. If QSZ≥0.7: The electromagnetic interference fluctuation is serious and may affect the sensor data or system stability. It is recommended to take protective measures (such as shielding and filtering).

[0049] Multi-source data correction module: Combined with the data anomaly detection results, the data drift anomaly degree of the intelligent sensor in the extreme electromagnetic pulse interference environment is evaluated, and the data drift frequency anomaly index and the electromagnetic pulse interference fluctuation index are normalized so that they are both between [0,1]. The data drift anomaly factor of the intelligent sensor is calculated based on the normalized data drift frequency anomaly index and the electromagnetic pulse interference fluctuation index.

[0050] For example, the present invention can use the following formula to calculate the data drift anomaly factor of the intelligent sensor, and the calculation expression is: ; In the formula, is the data drift anomaly factor of the smart sensor, is the data drift frequency anomaly index, QSZ is the electromagnetic pulse interference fluctuation index, is the weight coefficient of the data drift frequency anomaly index and the electromagnetic pulse interference fluctuation index (which can be optimized based on experimental experience or machine learning), and Both are greater than 0.

[0051] The acquired smart sensor data drift anomaly factor is compared with a threshold value pre-set according to historical data. If the smart sensor data drift anomaly factor is greater than or equal to the pre-set threshold value, it indicates that the smart sensor data drift anomaly degree in the extreme electromagnetic pulse interference environment is high, and the corresponding smart sensor data is classified as abnormal data, requiring data correction; if the smart sensor data drift anomaly factor is less than the pre-set threshold value, it indicates that the smart sensor data drift anomaly degree in the extreme electromagnetic pulse interference environment is low, and the corresponding smart sensor data is classified as normal data.

[0052] Combined with the degree of drift anomaly, the error compensation value is calculated and the smart sensor data is corrected, including: Calculate the data drift based on historical data comparison to quantify the error offset and calculate the sensor data drift : ;in: is the current sensor measurement value, is a reference value (obtained from historical data, other redundant sensors, or a prediction model).

[0053] The error compensation value is used to adjust the abnormal data to make it close to the true value. Assuming the error compensation value is C, the calculation formula is: ; Where: C is the error compensation value, λ is the drift correction coefficient (0<λ≤1), which is used to adjust the compensation amplitude: The calculation method of the drift correction coefficient is: ; is the data drift anomaly factor of the smart sensor, It is the maximum smart sensor data drift anomaly factor recorded in the historical data.

[0054] Use the compensation value C to correct the original sensor data: ; is the corrected sensor data, if If the data drift exceeds the normal working range, it means that the data drift is too serious and may have failed. In this case, the data will be discarded or manual intervention will be requested. If the C calculation is unstable (such as multiple sensors are abnormal), the historical average value is used as the final correction value.

[0055] In order to ensure the reliability of the corrected data, it is necessary to verify the deviation of the corrected data: calculate the deviation between the corrected data and the historical data , the expression is: ; is the reference value of the sensor data under normal conditions. Less than 5%: This indicates that the calibration data is reliable and can be used continuously. Greater than or equal to 5%: This indicates that the calibration has failed and the data should be discarded or recalibrated.

[0056] For the corrected data, if its credibility is high, it can be used to update the historical database to improve the accuracy of subsequent corrections.

[0057] Early warning and adjustment module: After completing data correction, the device status is evaluated based on the restored data. If extreme electromagnetic pulse interference causes a safety hazard to the equipment, an early warning is sent to the operation and maintenance personnel, and adjustment suggestions are provided, including adjusting the monitoring strategy or strengthening shielding measures.

[0058] Check whether the corrected data covers all key monitoring points (such as main transformers, circuit breakers, busbars, cables, etc.). Ensure that the data is smooth and without missing. If there are still abnormal data, request manual review or use redundant data to supplement.

[0059] Compare the corrected data with the historical operating data to determine whether the fluctuations in the values ​​are reasonable. If the data still shows abnormal deviations (such as abnormally large changes in equipment temperature in a short period of time), re-evaluate the correction strategy to prevent erroneous data from affecting the status assessment results.

[0060] Based on the restored data, check the following key equipment operating parameters: Main transformer: Are the winding temperature, oil temperature, current, and voltage normal? Is there any abnormal vibration or electromagnetic noise? Circuit breaker: Is the opening and closing state abnormal? Is there any abnormal heating of the contacts? Busbar and cable: Is there any voltage fluctuation, current overload, or overheating of the cable joint? Electromagnetic environment: Is there still high-intensity interference that may cause protection misoperation or measurement error?

[0061] Combine historical data with equipment rated parameters to analyze whether the restored data exceeds the normal range. Use trend analysis to determine whether the anomaly continues to increase and whether it may cause equipment failure or safety hazards. If the data fluctuates violently, it may be that the electromagnetic pulse interference has not been completely eliminated, or the equipment is damaged, resulting in abnormal measurement values.

[0062] Calculate the impact level of abnormality: Mild abnormality (such as short-term interference, equipment operation is still within the normal range): You can continue to observe and take no measures for the time being. Moderate abnormality (such as voltage and current exceeding the standard but not exceeding the protection threshold): It is recommended to adjust the monitoring strategy, increase the frequency of data collection, and pay close attention to the trend of changes. Severe abnormality (such as overheating of the main transformer, abnormal discharge of the cable, and severe current fluctuations): Immediately send an early warning to the operation and maintenance personnel and take protective measures.

[0063] If it is found that the equipment parameters exceed the safe operating range or have a continuous abnormal trend, the early warning mechanism will be triggered. The early warning information includes: abnormal equipment, abnormal parameters, possible risks, and recommended measures.

[0064] Warning level classification: Low-level warning (yellow): Minor abnormalities are detected, but the equipment is still running stably. It is recommended to strengthen data monitoring. Intermediate warning (orange): The equipment status is abnormal and may affect operation. It is recommended that operation and maintenance personnel check as soon as possible. High-level warning (red): There are safety hazards, such as equipment overheating, arc discharge, etc. It is recommended to check and take measures immediately.

[0065] If it is found that the monitoring data is still affected by electromagnetic interference, the following measures can be taken: Improve data collection frequency: Increase the sampling frequency of key equipment to capture more detailed changes. Optimize data filtering strategy: Use more advanced signal processing algorithms to reduce the impact of electromagnetic interference. Enable redundant sensors: Enable backup sensors at key measurement points to ensure data reliability. Increase data comparison: Cross-compare current data with historical data and adjacent equipment data to improve monitoring accuracy.

[0066] Strengthen shielding and protection measures: If it is confirmed that the electromagnetic pulse interference is still strong and affects the operation of the equipment, it is necessary to further optimize the protection strategy: Enhance physical shielding: Add electromagnetic shielding measures for key equipment, such as metal shielding covers, Faraday cages, etc. Optimize the grounding system: Check and improve the grounding of the equipment to reduce the coupling path of electromagnetic interference. Add filtering devices: Install EMI filters on the sensor or cable ports to reduce high-frequency interference. Adjust the sensor layout: Keep the sensors that are susceptible to interference away from high electromagnetic radiation areas, such as transformers, high-voltage switches, etc.

[0067] The above is only a specific implementation 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 covered by the protection scope of the present application.

Claims

1. The booster station monitoring and operation system is characterized by: It includes data acquisition module, data anomaly detection module, multi-source data correction module and early warning adjustment module; Data acquisition module: multiple heterogeneous intelligent sensors are set up in the booster station to collect equipment operation status data in real time; among them, the electromagnetic interference intensity is monitored in real time through the electromagnetic field sensor, and when an abnormal high-frequency pulse is detected, the system is triggered to enter the data anomaly detection mode; Data anomaly detection module: Analyzes the correlation between data collected by different types of sensors, determines the contingency of abnormal drift of multiple sensors at the same time, and analyzes the fluctuation of electromagnetic pulse interference in combination with the readings of electromagnetic field sensors; Multi-source data correction module: Combined with the data anomaly detection results, it evaluates the degree of data drift anomaly of smart sensors in extreme electromagnetic pulse interference environments, and calculates the error compensation value based on the degree of drift anomaly to correct the smart sensor data; Early warning and adjustment module: After completing data correction, the device status is evaluated based on the restored data. If extreme electromagnetic pulse interference causes a safety hazard to the equipment, an early warning is sent to the operation and maintenance personnel, and adjustment suggestions are provided, including adjusting the monitoring strategy or strengthening shielding measures.

2. The booster station monitoring and operation and maintenance system according to claim 1 is characterized in that: The data acquisition module includes multiple heterogeneous intelligent sensors, including temperature sensors, vibration sensors, optical sensors, current / voltage sensors and electromagnetic field sensors, which are used to respectively collect the temperature, vibration, discharge phenomenon, current, voltage and electromagnetic interference intensity of the equipment to improve the accuracy of data acquisition.

3. The booster station monitoring and operation and maintenance system according to claim 1 is characterized in that: After analyzing the correlation between the data collected by different types of sensors, the frequency of data drift of different sensors in the same time window is determined, and compared with the historical data under normal working conditions to generate the data drift frequency anomaly index. The method for obtaining the data drift frequency anomaly index is as follows: within a fixed time window, collect time series data of different sensors: Assume that the data collected by sensors A and B are: ; ;in: is sensor A at time The collected values, is sensor B at time The collected values, n is the number of data points in the time window; Calculate the Pearson correlation coefficient of sensors A and B under normal operating conditions : ;in: and are the means of historical data, is sensor A at time The collected values, is sensor B at time Collected values, calculate the Pearson correlation coefficient of the current time window : ;in: and are the means of the current time window respectively; Calculate the correlation deviation between the correlation coefficient of the current time window and the historical correlation coefficient : ; Define the data drift frequency as , the proportion of windows whose correlation deviation Δr exceeds the threshold θ: ;in: is an indicator function, if If , it takes 1, otherwise it takes 0, m is the total number of historical time windows; the calculated correlation deviation and data drift frequency The data drift frequency anomaly index is obtained after weighted average summation calculation.

4. The booster station monitoring and operation and maintenance system according to claim 3 is characterized in that: After analyzing the fluctuation of electromagnetic pulse interference, an electromagnetic pulse interference fluctuation index is generated. The method for obtaining the electromagnetic pulse interference fluctuation index is as follows: In the time window Q, the electromagnetic field strength signal x(t) is collected, where x(t) represents the electromagnetic field strength at time t, and the sampling frequency is assumed to be fs; the short-time Fourier transform is applied to the collected electromagnetic field strength signal, and the calculation formula is: ;in: represents the time-frequency spectrum at time t and frequency f, x(c) represents the electromagnetic field intensity value at the moment when the sampling point index is c, and is a discrete time signal sequence obtained by discretizing the electromagnetic field intensity signal x(t) at the sampling frequency fs. is the window function, f is the frequency component, is the kernel function of short-time Fourier transform; Set the window length to W and calculate the time-varying power spectral density. The expression is: ;in, Represents the power spectrum at time t and frequency f; calculates the power spectrum energy per unit time, the expression is: ;in, represents the electromagnetic field power at time t, and the maximum electromagnetic field power of the electromagnetic pulse is taken as the peak power , and calculate the average power in the entire time window , calculate the electromagnetic pulse interference fluctuation index, the expression is: ; Where QSZ is the electromagnetic pulse interference fluctuation index.

5. The booster station monitoring and operation and maintenance system according to claim 4 is characterized in that: Multi-source data correction module: Combined with the data anomaly detection results, the data drift anomaly degree of the intelligent sensor in the extreme electromagnetic pulse interference environment is evaluated, and the data drift frequency anomaly index and the electromagnetic pulse interference fluctuation index are normalized so that they are both between [0,1]. The data drift anomaly factor of the intelligent sensor is calculated based on the normalized data drift frequency anomaly index and the electromagnetic pulse interference fluctuation index.

6. The booster station monitoring and operation and maintenance system according to claim 5 is characterized in that: The acquired smart sensor data drift anomaly factor is compared with a threshold value pre-set according to historical data. If the smart sensor data drift anomaly factor is greater than or equal to the pre-set threshold value, it indicates that the smart sensor data drift anomaly degree in the extreme electromagnetic pulse interference environment is high, and the corresponding smart sensor data is classified as abnormal data, requiring data correction; if the smart sensor data drift anomaly factor is less than the pre-set threshold value, it indicates that the smart sensor data drift anomaly degree in the extreme electromagnetic pulse interference environment is low, and the corresponding smart sensor data is classified as normal data.

7. The booster station monitoring and operation and maintenance system according to claim 6, characterized in that: Combined with the degree of drift anomaly, the error compensation value is calculated to correct the smart sensor data, including: calculating the data drift amount based on historical data comparison to quantify the error offset, and calculating the sensor data drift amount : ;in: is the current sensor measurement value, is the reference value; the error compensation value is used to adjust the abnormal data. Assuming the error compensation value is C, the calculation formula is: ; Where: C is the error compensation value, λ is the drift correction coefficient, which is used to adjust the compensation amplitude: The calculation method of the drift correction coefficient is: ; is the data drift anomaly factor of the smart sensor, is the maximum smart sensor data drift anomaly factor recorded in historical data; Use the compensation value C to correct the original sensor data: ; For the corrected sensor data, perform the correction data deviation verification: calculate the deviation between the corrected data and the historical data , the expression is: ; is the reference value of the sensor data under normal conditions. Less than 5%: This indicates that the calibration data is reliable and can be used continuously. Greater than or equal to 5%: This indicates that the calibration has failed and the data should be discarded or recalibrated.

8. The booster station monitoring and operation and maintenance system according to claim 7 is characterized in that: After completing the data correction, the early warning adjustment module analyzes the equipment status. If it is found that the equipment operating status exceeds the safe range, different levels of early warning information are sent to the operation and maintenance personnel according to the abnormality level, and corresponding adjustment suggestions are provided, including increasing the data collection frequency, optimizing the filtering strategy, strengthening electromagnetic shielding, optimizing the grounding system or adjusting the sensor layout.

Images