Health evaluation method of UHV oil chromatography monitoring device based on in-service data

By constructing a health assessment model for data in transit, the completeness, stability and accuracy of the ultra-high pressure oil chromatography monitoring device is evaluated in real time, and the problem of the state of the device cannot be evaluated in real time in the prior art is solved, and the ultra-high pressure oil chromatography is realized Efficient fault identification and performance improvement of monitoring devices.

CN114354783BActive Publication Date: 2024-12-03WUHAN NARI LIABILITY OF STATE GRID ELECTRIC POWER RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111550880.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-17
Publication Date
2024-12-03
Estimated Expiration
2041-12-17

AI Technical Summary

Technical Problem

The existing ultra-high voltage oil chromatography monitoring device cannot effectively evaluate its status in real time during operation, resulting in uneven monitoring quality and insufficient control of operation and inspection personnel, and failure to detect faults in a timely manner.

Method used

By constructing a health assessment model based on the data in the process, obtain and pretreat the dissolution concentration value in the specified component gas oil of the oil chromatography monitoring device, calculate the completeness, stability and accuracy of the device, evaluate its health in real time, and Output fault information based on the evaluation results.

Benefits of technology

Real-time health assessment of UHV oil chromatography monitoring devices is realized, fault identification, guidance performance improvement and operation and maintenance decision-making is guided, and the status maintenance level is improved, avoiding the cost and power outage time caused by return to the factory for maintenance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114354783B_ABST
    Figure CN114354783B_ABST
Patent Text Reader

Abstract

The present invention provides a method for evaluating the health of a UHV oil chromatogram monitoring device based on in-service data, comprising the following steps: obtaining the dissolved concentration value of the specified component gas collected by the UHV oil chromatogram monitoring device in the evaluation period as a target data set; preprocessing and cleaning the target data set; performing integrity evaluation of the UHV oil chromatogram monitoring device; performing stability evaluation of the UHV oil chromatogram monitoring device; performing accuracy evaluation of the UHV oil chromatogram monitoring device; and calculating the health evaluation value of the UHV oil chromatogram monitoring device according to the integrity evaluation, stability evaluation and accuracy evaluation results of the UHV oil chromatogram monitoring device. The present invention can effectively evaluate the operating status of the UHV oil chromatogram monitoring device in real time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid safety control and protection, and in particular relates to a health assessment method for an ultra-high voltage oil chromatography monitoring device based on in-service data. Background Art

[0002] Currently, in the field of status monitoring of UHV oil-containing equipment (transformers, high-voltage reactors, etc.), oil chromatography online monitoring technology is the main means of obtaining equipment status information. However, several serious accidents involving large transformers in recent years have exposed the uneven quality of current monitoring devices and the lack of control over the operating status of the devices by operation and maintenance personnel. Currently, monitoring devices are only returned to the factory for inspection after a failure occurs, and their status cannot be effectively evaluated during operation. Therefore, conducting a real-time, scientific and reliable evaluation of the currently operating UHV oil chromatography monitoring devices, and guiding the performance improvement of the monitoring devices based on the evaluation results and better serving on-site operation and maintenance are the key to solving the above pain points. Summary of the invention

[0003] The purpose of the present invention is to address the shortcomings of the above-mentioned background technology and to provide a method for evaluating the health of a UHV oil chromatography monitoring device based on in-service data. Through a large amount of oil chromatography online monitoring data accumulated by the UHV equipment monitoring and early warning platform currently deployed in the entire network, a set of complete monitoring device level evaluation models is constructed to fully mine the status data information of the main equipment and effectively evaluate the operating status of the monitoring device in real time.

[0004] The technical solution adopted by the present invention is: a method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data, characterized in that it comprises the following steps:

[0005] S1, obtaining the dissolved concentration value of the specified component gas in the oil collected by the UHV oil chromatography monitoring device during the evaluation period as the target data set;

[0006] S2, preprocessing and cleaning the target data set;

[0007] S3, judging whether the target data set after preprocessing and cleaning contains the dissolved concentration value data of all the specified component gases in each acquisition cycle of the UHV oil chromatography monitoring device, and calculating the integrity rate of the UHV oil chromatography monitoring device according to the judgment result;

[0008] S4, based on the target data set after pretreatment and cleaning, calculating the average value and standard deviation of the dissolved concentration value of each specified component gas in oil in each set time period and comparing them, and calculating the stability rate of the ultra-high voltage oil chromatography monitoring device according to the comparison results;

[0009] S5, based on the target data set after preprocessing and cleaning, judging the consistency level between the dissolved concentration value in oil of the specified component gas measured by the UHV oil chromatography monitoring device and the offline test measurement result; calculating the accuracy of the UHV oil chromatography monitoring device according to the judgment result;

[0010] S6, calculating the health evaluation value of the UHV oil chromatography monitoring device and the fault information of the UHV oil chromatography monitoring device according to the integrity rate, stability rate and accuracy rate of the UHV oil chromatography monitoring device.

[0011] In the above technical scheme, the specified component gases include H2, CH4, C2H6, C2H4, C2H2, CO, and CO2; the target data set includes data groups collected at each monitoring time point; the data groups include the dissolved concentration values ​​of 7 component gases H2, CO, CO2, CH4, C2H6, C2H4, and C2H2 in oil.

[0012] In the above technical solution, in step S1, an evaluation period is selected on the monitoring platform of the ultra-high voltage oil chromatography monitoring device, and the historical collected data corresponding to the evaluation period in the monitoring platform is called as the target data set; after step S2 is executed, if there are less than 6 groups of valid data in the evaluation period, it is necessary to return to step S1 and reselect the evaluation period.

[0013] In the above technical solution, step S2 specifically includes the following steps: query all the vacant time periods in the original data of the target data set through SQL SELECT statements, and the on-site operation and maintenance personnel verify whether the vacancy in the time period is caused by the network protection action organized by the State Grid or the regular equipment maintenance. If so, the time period is directly eliminated and not included in the statistical evaluation scope. If not, the vacancy is treated as data missing and included in the statistical evaluation scope; then, the Java program is used to find out the repeated data with exactly the same monitoring values ​​of all components of the gas in the original data of the target data set, that is, the 7 characteristic gas concentration values ​​of each data are exactly the same, and verify whether the cause of the repeated data is manual debugging. If so, the data of the time period is directly eliminated and not included in the statistical evaluation scope. If not, the initial set of data is taken as the valid value, and the subsequent repeated data are all treated as null values ​​and included in the statistical evaluation scope; finally, the Java program is used to determine whether there are garbled characters, null values ​​and abnormal data in the original data of the target data set. If so, these types of data are uniformly classified as invalid null values ​​and included in the statistical evaluation scope, and the data in the statistical evaluation scope is used as the target data set after preprocessing and cleaning. This data preprocessing and cleaning technology is closely integrated with the actual operation and maintenance needs, comprehensively cleaning all types of abnormal and invalid data involved in the monitoring platform ledger, greatly increasing the proportion of valuable data on the monitoring platform, and laying the foundation for subsequent device performance evaluation.

[0014] In the above technical solution, in step S3, according to the target data set after pretreatment and cleaning, it is judged that: in each acquisition cycle of the ultra-high voltage oil chromatographic monitoring device, if there is at least one data group in the target data set and the data group includes the dissolved concentration values ​​of 7 component gases in oil, then the data is considered to be complete and has feedback data, otherwise it is considered that the data is missing; the total number of times the ultra-high voltage oil chromatographic monitoring device in the target data set should be fed back within the evaluation time period is defined as N, and the actual number of times there is feedback data is n, then the data completeness calculation result is C = n / N × 100%. This completeness calculation method is closely combined with the requirements of the State Grid Corporation for the online rate of the monitoring device, standardizes the continuity and acquisition density of the device monitoring operation, and effectively reflects the integrity of the device acquisition.

[0015] In the above technical solution, in step S4, the standard deviation σ of each component gas is calculated as follows:

[0016]

[0017] Where M represents the total number of samples of a certain gas component, u represents the average value of the dissolved concentration of a certain gas component in oil, and x i It represents the dissolved concentration value of the i-th oil of a certain component gas in the sample.

[0018] If the calculated standard deviation of a certain gas in a certain set time period is σ≤30%μ, the measurement result of the gas in the ultra-high voltage oil chromatography monitoring device in the set time period is considered to be stable, otherwise it is unstable. This stability rate calculation method is closely combined with the actual production and operation. In the case of insufficient offline data samples, only the existing online monitoring data is used to accurately obtain the stability rate data of the device, getting rid of the dependence on offline data.

[0019] In the above technical solution, in step S4, the evaluation period is defined to include K set time periods, and the set time period during which the measurement result of a certain component gas is stable is k, then the measurement stability rate of this component gas is S=k / K×100%; the measurement stability rate of each component gas is calculated respectively; based on the measurement stability rate of each component gas and the calculation weight of each component gas, where the weight is set by the operation and maintenance personnel according to the actual operation and maintenance needs of each substation, the stability rate of the ultra-high voltage oil chromatography monitoring device is calculated. Based on this stability rate calculation method, faults exposed by the operating device due to insufficient stability can be effectively discovered, proving the effectiveness of this method.

[0020] In the above technical scheme, in step S5, the deviation rate of each component gas in each data group in the target data set is calculated respectively, and the deviation rate = |online value-offline value| / offline value; wherein the online value refers to the oil solubility concentration value of a certain component gas recorded in a certain data group in the target data set, and the offline value refers to the oil solubility concentration value of the component gas obtained through regular offline testing; the average value of the deviation rate of the 7 component gases in the set time period is calculated respectively, and according to the accuracy range of the calculated average value, the accuracy of each component gas in the set time period is obtained, wherein different numerical ranges of the average value of the deviation rate of each component gas are respectively associated with an accuracy rate; according to the accuracy of each component gas in the set time period and the calculation weight of each component gas, wherein the weight is set by the operation and maintenance personnel according to the actual operation and maintenance requirements of each substation, the accuracy of the ultra-high voltage oil chromatography monitoring device in the set time period is calculated; the accuracy of the ultra-high voltage oil chromatography monitoring device in all set time periods within the evaluation period is calculated and the average value is taken as the accuracy of the ultra-high voltage oil chromatography monitoring device. This accuracy calculation method combines and taps into the value of offline and online data, and achieves real-time acquisition of device measurement accuracy data without power outages or operational interruptions.

[0021] In the above technical solution, in step S6, the calculation formula of the health evaluation value Value is:

[0022] Value=a*C+b*S+c*V

[0023] Among them, C represents the completeness rate, a represents the weight of the completeness rate; S represents the stability rate, b represents the weight of the stability rate; V represents the accuracy rate, c represents the weight of the accuracy rate; each weight value is set by the operation and maintenance personnel according to the actual operation and maintenance needs of each substation;

[0024] And set the qualified score according to actual needs. If the device score is qualified, it is judged that the device is operating normally; if the device score is unqualified, it is judged whether the single parameter meets the standard: generally, if the completeness rate is less than 90%, the stability rate and accuracy rate are less than 80%, the single parameter does not meet the standard, and the associated UHV oil chromatography monitoring device fault information is output according to the judgment result;

[0025] If the completeness rate does not meet the standard, it is determined that there is a fault in the communication between the acquisition control system of the UHV oil chromatography monitoring device and the server;

[0026] If the stability rate does not meet the standard, it is determined that the accuracy of the sensor unit of the UHV oil chromatography monitoring device is affected by the operating environment and there is a fault;

[0027] If the accuracy rate does not meet the standard, it is determined that there is a fault in the sensitivity of the sensor unit of the ultra-high voltage oil chromatography monitoring device.

[0028] The monitoring device evaluation technology is closely integrated with the needs of on-site operation and maintenance, and evaluates the performance of the device operation in real time and objectively. Not only can it intuitively understand the health of the device operation as a whole, but it can also discover the direction of device problems from specific performance indicators such as the integrity, stability, and accuracy of the device. Generally speaking, low integrity is related to communication problems between the device's acquisition control system and the server, low stability is related to the low accuracy of the sensor unit and the influence of the operating environment, and low accuracy is directly related to the sensitivity of the sensor unit. The evaluation results effectively support the device operation and maintenance decisions, improve the level of condition maintenance, and output real-time fault information of the device, realizing efficient fault identification of the device.

[0029] The present invention provides a health evaluation system for an ultra-high voltage oil chromatography monitoring device based on in-service data, comprising a target data acquisition module, a data processing module, a completeness calculation module, a stability calculation module, an accuracy calculation module and a health evaluation module; wherein:

[0030] The target data acquisition module is used to obtain the oil-dissolved concentration value of the specified component gas collected by the UHV oil chromatography monitoring device during the evaluation period as the target data set;

[0031] The data processing module is used to preprocess and clean the target data set;

[0032] The completeness calculation module is used to determine whether the target data set after preprocessing and cleaning contains the dissolved concentration value data of all specified component gases in oil in each acquisition cycle of the UHV oil chromatography monitoring device, calculate the completeness of the UHV oil chromatography monitoring device according to the judgment result, and output the obtained completeness to the health assessment module;

[0033] The stability rate calculation module is used to calculate the average value and standard deviation of the dissolved concentration value of each specified component gas in oil in each set time period based on the pre-processed and cleaned target data set, and compare them, calculate the stability rate of the UHV oil chromatography monitoring device according to the comparison result, and output the obtained stability rate to the health assessment module;

[0034] The accuracy calculation module is used to judge the consistency level between the oil dissolved concentration value of the specified component gas measured by the UHV oil chromatographic monitoring device and the offline test measurement result based on the pre-processed and cleaned target data set; calculate the accuracy of the UHV oil chromatographic monitoring device according to the judgment result, and output the obtained accuracy to the health assessment module;

[0035] The health evaluation module is used to calculate the health evaluation value of the UHV oil chromatogram monitoring device and the fault information of the UHV oil chromatogram monitoring device according to the integrity rate, stability rate and accuracy rate of the UHV oil chromatogram monitoring device.

[0036] The beneficial effects of the present invention are: in view of the current situation that the operating quality of the oil chromatography monitoring device in the UHV substation in operation is uneven and there is no reliable and effective means to evaluate the health level of the device in real time, the present invention proposes a health evaluation method for the UHV oil chromatography monitoring device based on a large amount of historical monitoring data accumulated by the device, which can effectively and in real time identify the faults of the UHV oil chromatography monitoring device. The method constructs a set of complete health evaluation models to evaluate the integrity of the acquisition strategy of the monitoring device, the stability of the measurement level, and the accuracy of the measurement results. It comprehensively analyzes from multiple dimensions to obtain the overall operating status of the oil chromatography monitoring device in the UHV substation, and can identify the source of the device's fault from the analysis results, thereby guiding the direction of device performance improvement, that is, if the completeness rate is low, it is necessary to adjust the setting of the acquisition strategy and check the operating status of the communication system, if the stability rate is low, it is necessary to check the accuracy of the sensor unit, and if the accuracy rate is low, it is necessary to check and calibrate the sensitivity of the sensor unit, etc., thereby supporting the device operation and maintenance decision-making and improving the level of status maintenance.

[0037] Compared with the method of returning the device to the factory for test and maintenance, this method can not only comprehensively evaluate the performance of the device in combination with objective factors such as the on-site operating environment, but also promptly discover device problems, indicate the causes of the problems, and provide decisions for device operation and maintenance, effectively improving the level of condition-based maintenance. It has strong real-time and reliability, and can also avoid the labor and time costs caused by other methods that require power outages and disassembly. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of the health evaluation method of the UHV oil chromatography monitoring device based on in-service data of the present invention;

[0039] Figure 2 This is a flow chart of raw data preprocessing and cleaning of the present invention;

[0040] Figure 3 It is the classification of the accuracy of the measurement results of seven gases in the standard "Q / GDW 10536-2017 Technical Specifications for Online Monitoring Devices for Dissolved Gases in Transformer Oil". DETAILED DESCRIPTION

[0041] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments to facilitate a clear understanding of the present invention, but they do not constitute a limitation on the present invention.

[0042] The present invention provides a health evaluation system for an ultra-high voltage oil chromatography monitoring device based on in-service data, comprising a target data acquisition module, a data processing module, a completeness calculation module, a stability calculation module, an accuracy calculation module and a health evaluation module.

[0043] The target data acquisition module is used to obtain the dissolved oil concentration values ​​of the specified component gases collected by the UHV oil chromatography monitoring device during the evaluation period as the target data set. The operator selects the historical monitoring data of the UHV oil chromatography monitoring device in a certain period of time as the target data set for status evaluation according to the purpose. The target data set includes the data groups collected at each monitoring time point; each group of data is the dissolved oil concentration value (μL / L) of the seven gases H2, CH4, C2H6, C2H4, C2H2, CO, and CO2 at a certain monitoring time point. If the selected time period is long (for example, nearly one year or half a year), the comprehensiveness of the device evaluation results will be strong, and the objective operation level of the device can be better reflected. If the selected time period is short (for example, the same day or the same week), the real-time nature of the device evaluation results will be strong, and the real-time operation level can be truthfully reflected.

[0044] The data processing module is used to preprocess and clean the target data set; it mainly includes eliminating a large amount of duplicate data caused by equipment debugging, and the entire period of time when data is missing due to network protection operations or main equipment maintenance. Data anomalies and missing data caused by such objective factors should not be included in the data statistics. At the same time, subjective factors such as garbled data, gaps and data anomalies caused by device operation should be classified as data missing and included in the data statistics.

[0045] The completeness calculation module is used to determine whether the target data set after preprocessing and cleaning contains the dissolved concentration value data of all specified component gases in oil in each collection cycle of the UHV oil chromatography monitoring device. The completeness of the UHV oil chromatography monitoring device is calculated based on the judgment result, and the obtained completeness is output to the health assessment module. The completeness calculation module is used to evaluate the data collection of the seven component gases in the collection cycle. If there is more than one feedback (data collection of all seven component gases) in the collection cycle, the data collection in this cycle is considered complete. The completeness of the device data collection is calculated by counting the collection status of the unit cycle during the entire period of time. The completeness of data collection reflects the continuous reliability level of the device in the time dimension.

[0046] The stability rate calculation module is used to calculate the average value and standard deviation of the dissolved concentration value of each specified component gas in each set time period based on the target data set after preprocessing and cleaning, and compare them. According to the comparison results, the stability rate of the ultra-high voltage oil chromatography monitoring device is calculated, and the obtained stability rate is output to the health assessment module. The stability rate calculation module fits and determines the concentration measurement baseline of each of the seven component gases (i.e., the average value in each set time period) based on the historical data of the dissolved concentration value of the seven component gases collected by the device. Based on this baseline, the mean square error is calculated for each group of online monitoring data (the dissolved concentration value of the seven component gases in oil). For time periods with a mean square error value greater than 30%, it is unstable, otherwise it is stable. The stability ratio of each time period is calculated to obtain the stability rate of each single component gas, and the weight is determined according to the importance of each component gas to obtain the stability rate of the device measurement level. The stability rate of the device measurement level reflects the degree of drift of the gas concentration value measured by online monitoring in a continuous time, and reflects the accuracy level of the device measurement.

[0047] The accuracy calculation module is used to determine the level of agreement between the oil-dissolved concentration value of the specified component gas measured by the UHV oil chromatographic monitoring device and the offline test measurement result based on the preprocessed and cleaned target data set; calculate the accuracy of the UHV oil chromatographic monitoring device based on the judgment result, and output the obtained accuracy to the health assessment module. The accuracy calculation module uses the results of regular offline test data as a benchmark to calculate the deviation rate of each set of online monitoring data (the oil-dissolved concentration values ​​of the seven component gases), such as Figure 3 The accuracy range of the reference standard DL / T1432 is divided into different accuracy levels of A, B, and C. The accuracy is calculated according to the accuracy level of the deviation rate and the deviation level of the dissolved concentration value of each component gas in oil, and the weight is determined according to the importance of each component gas to obtain the accuracy of the device. The accuracy of the device measurement results reflects the level of agreement between the gas concentration value measured by online monitoring and the offline test measurement results, and reflects the sensitivity level of the device measurement.

[0048] The health assessment module is used to calculate the health assessment value of the UHV oil chromatography monitoring device based on the completeness rate, stability rate and accuracy rate of the UHV oil chromatography monitoring device. The health assessment module sets the weights of each indicator according to the external input instructions used to reflect actual needs, calculates the health assessment value of the device, and sets the qualified score according to the user's actual operation and maintenance needs. If the device evaluation fails, it determines whether the single parameter is low. Generally, it is stipulated that the completeness rate is less than 90%, and the stability rate and accuracy rate are less than 80% as the single parameter is low, and the associated UHV oil chromatography monitoring device fault information is output according to the judgment result;

[0049] If the completeness rate does not meet the standard, it is determined that there is a fault in the communication between the acquisition control system of the UHV oil chromatography monitoring device and the server;

[0050] If the stability rate does not meet the standard, it is determined that the accuracy of the sensor unit of the UHV oil chromatography monitoring device is affected by the operating environment and there is a fault;

[0051] If the accuracy rate does not meet the standard, it is determined that there is a fault in the sensitivity of the sensor unit of the ultra-high voltage oil chromatography monitoring device.

[0052] like Figure 1 As shown, the present invention provides a method for evaluating the health of a UHV oil chromatography monitoring device based on in-service data, which mainly includes the steps of selecting monitoring data, preprocessing and cleaning data, evaluating the integrity, stability and accuracy of the device, evaluating the health of the device, etc., wherein each calculation weight is set by the operation and maintenance personnel in various specific application environments and device models, specifically:

[0053] Step 1: Select the target period for device evaluation

[0054] Select the target data set, each set of data is the oil dissolved concentration value (μL / L) of the seven gases H2, CH4, C2H6, C2H4, C2H2, CO, and CO2 at a certain monitoring time point, and select the historical measurement data of the target period to be evaluated in the original ledger of the monitoring platform of the UHV oil chromatography monitoring device as the original data of the target data set. The evaluation results generated by calling the historical measurement data of different target periods represent the performance of the device during that period.

[0055] After the next step of data cleaning is completed, the effectiveness of the selected target period must be evaluated first. If there is no valid data or too little valid data in the period, the target period must be reselected. It is not recommended to select a target period that is too short to ensure the reliability of the evaluation results and reduce random errors.

[0056] Step 2: Raw data preprocessing and cleaning

[0057] like Figure 2The flowchart of the raw data preprocessing and cleaning of the present invention is shown. First, all the vacant time periods in the raw data of the target data set are queried through the SQL SELECT statement to verify whether the vacancy in the time period is caused by objective factors such as the network protection action organized by the State Grid or regular equipment maintenance. If so, the time period is directly eliminated and not included in the statistical evaluation scope. If not, the vacancy is treated as data missing (null value) and included in the statistical evaluation scope; then, all repeated data groups (7 characteristic gas component monitoring values ​​are exactly the same) are found out by the Java program, and it is verified whether the cause of the repeated data group is manual debugging. If so, the data of the time period is directly eliminated and not included in the statistical evaluation scope. If not, the initial set of data is taken as the valid value, and the subsequent repeated data are all treated as null values ​​and included in the statistical evaluation scope; finally, the Java program is used to determine whether there are garbled codes, null values ​​and abnormal data. If so, these types of data are uniformly classified as invalid null values ​​and included in the statistical evaluation scope, so that the raw data preprocessing and cleaning are completed. All data within the statistical evaluation scope are taken as the target data set.

[0058] Step 3.1: Completeness Assessment of the Installation’s Data Collection Strategy

[0059] According to the UHV station operation and maintenance management requirements, the standard measurement cycle of the UHV oil chromatography monitoring device is set to sample once every 4 hours, and the data is measured 6 times a day, that is, 0:00-3:59, 4:00-7:59, 8:00-11:59, 12:00-15:59, 16:00-19:59, and 20:00-23:59. Data collection is required in 6 time periods, namely 0:00-3:59, 4:00-7:59, 8:00-11:59, 12:00-15:59, 16:00-19:59, and 20:00-23:59. The UHV oil chromatography monitoring device measures the dissolved concentration values ​​of the 7 component gases H2, CO, CO2, CH4, C2H6, C2H4, and C2H2 in oil as a set of data feedback each time. If there are more than or equal to 1 set of data feedback records in each time period of the target data set, the data is considered to be complete and has feedback, otherwise it is considered that the data is missing. Define the total number of times the device in the target data set should feedback in the target time period as N, and the actual number of times there is feedback data is n, then the data completeness calculation result is C=n / N×100%.

[0060] Step 3.2: Measurement level stability assessment of the device

[0061] For the seven gas components of a single device, the average value μ and standard deviation σ of the dissolved concentration of each component gas in oil are calculated on a daily basis. The standard deviation formula is:

[0062]

[0063] Where M represents the total number of samples of a certain component gas, u represents the average value of the dissolved concentration of a certain component gas in oil, and Xi represents the i-th dissolved concentration of a certain component gas in oil in the sample. If the standard deviation of a certain component gas in a single day is calculated to be σ≤30%μ;

[0064] The gas is considered stable on that day, otherwise it is unstable. Define the total number of days for measuring a certain component gas in the target data set as K, and the number of days with stable measurement results as k, then the stability rate of the gas is S = k / K × 100%. According to this method, the stability rates of the seven characteristic gases are calculated respectively. The respective weights are determined according to the importance of each gas and the actual operation and maintenance needs of each substation. Among them, H2 and C2H2 are key gases. It is generally recommended to calculate the stability rate of the device with a weight of 25% for H2 and C2H2, and 10% for the remaining five gases, that is:

[0065] S=[25%S(H2)+25%S(C2H2)+10%S(CH4)+10%S(C2H4)+10%S(C2H6)+10%S(CO)+10%S(CO2)]

[0066] Step 3.3: Evaluate the accuracy of the device's measurement results

[0067] Accuracy refers to the level of agreement between the dissolved gas concentration value in oil measured by the monitoring device and the offline test measurement results. Based on the results of regular offline test data, a single device is measured on a daily basis, and the standard "Q / GDW 10536-2017 Technical Specifications for Dissolved Gas Monitoring Devices in Transformer Oil" (such as Figure 3 The accuracy range is divided into A, B, and C different accuracy levels. The deviation rate is calculated for each set of online monitoring data (7 component gases), and the deviation rate = |online value-offline value| / offline value. The average deviation rate of each gas component on the day is calculated and rated according to the accuracy range. The standards (A: 100, B: 60, C: 30, D: 0) are selected for graded assessment and scoring to obtain the accuracy rating of each component on the day. The weight is determined according to the importance of each gas and the actual operation and maintenance needs of each substation. Among them, H2 and C2H2 are key gases. It is recommended to calculate the accuracy assessment of the device on the day with a weight of 25% for H2 and C2H2 respectively, and 10% for the other 5 gases each:

[0068] V=[25%V(H2)+25%V(C2H2)+10%V(CH4)+10%V(C2H4)+10%V(C2H6)+10%V(CO)+10%V(CO2)]

[0069] Calculate the accuracy of each valid day within the target period and take the average to get the average daily accuracy.

[0070] Step 4: According to the above-obtained device integrity rate, stability rate and accuracy rate, set the weight of each indicator according to the user's actual operation and maintenance needs, and calculate the device health evaluation value. For example, you can refer to the setting of integrity accounting for 20%, stability and accuracy each accounting for 40% to calculate the comprehensive evaluation of the device, that is:

[0071] Value=20%C+40%S+40%V

[0072] And set the qualified score according to actual needs. If the device score is unqualified, judge whether the single parameter is low. Generally, it is stipulated that the completeness rate is less than 90%, the stability rate and accuracy rate are less than 80% to be considered as low single parameter. And output the fault information of the associated UHV oil chromatography monitoring device according to the judgment result;

[0073] If the completeness rate does not meet the standard, it is determined that there is a fault in the communication between the acquisition control system of the UHV oil chromatography monitoring device and the server;

[0074] If the stability rate does not meet the standard, it is determined that the accuracy of the sensor unit of the UHV oil chromatography monitoring device is affected by the operating environment and there is a fault;

[0075] If the accuracy rate does not meet the standard, it is determined that there is a fault in the sensitivity of the sensor unit of the ultra-high voltage oil chromatography monitoring device.

[0076] Generally speaking, low integrity is related to communication problems between the device's acquisition control system and the server, low stability is related to low precision of the sensor unit and the influence of the operating environment, and low accuracy is directly related to the sensitivity of the sensor unit.

[0077] The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.

Claims

1. A method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data, characterized in that: The following steps are involved: S1, obtaining the dissolved concentration value of the specified component gas in the oil collected by the UHV oil chromatography monitoring device during the evaluation period as the target data set; S2, preprocessing and cleaning the target data set; S3, judging whether the target data set after preprocessing and cleaning contains the dissolved concentration value data of all designated component gases in each acquisition cycle of the UHV oil chromatography monitoring device, and calculating the integrity rate of the UHV oil chromatography monitoring device according to the judgment result; S4, based on the target data set after pretreatment and cleaning, calculating the average value and standard deviation of the dissolved concentration value of each specified component gas in oil in each set time period and comparing them, and calculating the stability rate of the ultra-high voltage oil chromatography monitoring device according to the comparison results; S5, based on the target data set after preprocessing and cleaning, judging the consistency level between the dissolved concentration value in oil of the specified component gas measured by the UHV oil chromatography monitoring device and the offline test measurement result; calculating the accuracy of the UHV oil chromatography monitoring device according to the judgment result; S6, calculating a health evaluation value of the UHV oil chromatogram monitoring device according to the integrity rate, stability rate, and accuracy rate of the UHV oil chromatogram monitoring device, and obtaining fault information of the UHV oil chromatogram monitoring device; In step S6, the calculation formula of the health evaluation value is: Value=a*C+b*S+c*V Among them, C represents the completeness rate, a represents the weight of the completeness rate; S represents the stability rate, b represents the weight of the stability rate; V represents the accuracy rate, c represents the weight of the accuracy rate; The qualified score Value (T) is set according to actual needs. If the device score is unqualified, it is judged whether the single parameter meets the standard; and the fault information of the associated UHV oil chromatography monitoring device is output according to the judgment result; If the completeness rate does not meet the standard, it is determined that there is a fault in the communication between the acquisition control system of the UHV oil chromatography monitoring device and the server; If the stability rate does not meet the standard, it is determined that the accuracy of the sensor unit of the UHV oil chromatography monitoring device is affected by the operating environment and there is a fault; If the accuracy rate does not meet the standard, it is determined that there is a fault in the sensitivity of the sensor unit of the ultra-high voltage oil chromatography monitoring device.

2. The method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data according to claim 1 is characterized in that: The specified component gases include H2, CH4, C2H6, C2H4, C2H2, CO, and CO2; the target data set includes data groups collected at each monitoring time point; the data groups include the dissolved concentration values ​​of the seven component gases H2, CO, CO2, CH4, C2H6, C2H4, and C2H2 in oil.

3. The method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data according to claim 1 is characterized in that: In step S1, an evaluation period is selected on the monitoring platform of the ultra-high voltage oil chromatography monitoring device, and the historical collected data corresponding to the evaluation period in the monitoring platform is called as the target data set; after step S2 is executed, if there are less than 6 groups of valid data in the evaluation period, it is necessary to return to step S1 and reselect the evaluation period.

4. The method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data according to claim 1 is characterized in that: Step S2 specifically includes the following steps: query all the vacant time periods in the original data of the target data set through the SQL SELECT statement, verify whether the vacancy in the time period is caused by the network protection action organized by the State Grid or the regular equipment maintenance, if so, directly eliminate the time period and do not include it in the statistical evaluation scope, if not, treat the vacancy as data missing and include it in the statistical evaluation scope; then, use the Java program to find out the repeated data groups with exactly the same gas monitoring values ​​of all components in the original data of the target data set, verify whether the cause of the repeated data group is manual debugging, if so, directly eliminate the data of the time period and do not include it in the statistical evaluation scope, if not, take the initial set of data as the valid value, and treat the subsequent repeated data as null values ​​and include it in the statistical evaluation scope; finally, use the Java program to determine whether there are garbled characters, null values ​​and abnormal data in the original data of the target data set, if so, these types of data are uniformly classified as invalid null values ​​and included in the statistical evaluation scope, and the data in the statistical evaluation scope is used as the target data set after preprocessing and cleaning.

5. The method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data according to claim 2 is characterized in that: In step S3, it is judged based on the target data set after preprocessing and cleaning: in each acquisition cycle of the ultra-high voltage oil chromatography monitoring device, if there is at least one data group in the target data set and the data group includes the dissolved concentration values ​​of 7 component gases in oil, then the data is considered to be complete and has feedback data, otherwise it is considered that the data is missing; the total number of times that the ultra-high voltage oil chromatography monitoring device in the target data set should feedback within the evaluation time period is defined as N, and the actual number of times there is feedback data is n, then the data completeness calculation result is C=n / N×100%.

6. The method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data according to claim 2 is characterized in that: In step S4, the standard deviation σ of each component gas is calculated as follows: Where M represents the total number of samples of a certain gas component, μ represents the average value of the dissolved concentration of a certain gas component in oil, and x i It represents the dissolved concentration value of the i-th oil of a certain component gas in the sample. If the calculated standard deviation of a certain gas in a certain set time period is σ≤30%μ It is considered that the measurement result of the gas in the set time period by the ultra-high pressure oil chromatography monitoring device is stable, otherwise it is unstable.

7. The method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data according to claim 6 is characterized in that: In step S4, the evaluation period is defined to include K set time periods, and the set time periods during which the measurement results of a certain component gas are stable are k, then the measurement stability rate of the component gas is S=k / K×100%; the measurement stability rate of each component gas is calculated respectively; based on the measurement stability rate of each component gas and the calculation weight of each component gas, the stability rate of the ultra-high voltage oil chromatography monitoring device is calculated.

8. The method for evaluating the health of an ultra-high voltage oil chromatography monitoring device based on in-service data according to claim 7 is characterized in that: In step S5, the deviation rate of each component gas in each data group in the target data set is calculated respectively, and the deviation rate = |online value - offline value| / offline value; wherein the online value refers to the oil solubility concentration value of a certain component gas recorded in a certain data group in the target data set, and the offline value refers to the oil solubility concentration value of the component gas obtained through regular offline tests; the average value of the deviation rate of the seven component gases in the set time period is calculated respectively, and the accuracy level of each component gas in the set time period is obtained according to the accuracy range of the calculated average value, wherein different numerical ranges of the average value of the deviation rate of each component gas are respectively associated with accurate values; according to the accurate value of each component gas in the set time period and the calculation weight of each component gas, the accuracy rate of the ultra-high voltage oil chromatography monitoring device in the set time period is calculated; The accuracy of the UHV oil chromatographic monitoring device in all set time periods within the evaluation period is calculated and the average value is taken as the accuracy of the UHV oil chromatographic monitoring device.

9. A health evaluation system for ultra-high voltage oil chromatography monitoring device based on in-service data, used to implement the method according to any one of claims 1 to 8, characterized in that: It includes target data acquisition module, data processing module, completeness rate calculation module, stability rate calculation module, accuracy rate calculation module and health assessment module; among which, The target data acquisition module is used to obtain the oil-dissolved concentration value of the specified component gas collected by the UHV oil chromatography monitoring device during the evaluation period as the target data set; The data processing module is used to preprocess and clean the target data set; The completeness calculation module is used to determine whether the target data set after preprocessing and cleaning contains the dissolved concentration value data of all specified component gases in oil in each acquisition cycle of the UHV oil chromatography monitoring device, calculate the completeness of the UHV oil chromatography monitoring device according to the judgment result, and output the obtained completeness to the health assessment module; The stability rate calculation module is used to calculate the average value and standard deviation of the dissolved concentration value of each specified component gas in oil in each set time period based on the pre-processed and cleaned target data set, and compare them, calculate the stability rate of the UHV oil chromatography monitoring device according to the comparison result, and output the obtained stability rate to the health assessment module; The accuracy calculation module is used to judge the consistency level between the oil dissolved concentration value of the specified component gas measured by the UHV oil chromatographic monitoring device and the offline test measurement result based on the pre-processed and cleaned target data set; calculate the accuracy of the UHV oil chromatographic monitoring device according to the judgment result, and output the obtained accuracy to the health assessment module; The health evaluation module is used to calculate the health evaluation value of the UHV oil chromatogram monitoring device and the fault information of the UHV oil chromatogram monitoring device according to the integrity rate, stability rate and accuracy rate of the UHV oil chromatogram monitoring device.