Measurement anomaly identification and anomaly association analysis method based on rule base

Abstract

The invention discloses a rule base-based metering anomaly recognition and anomaly association analysis method, which comprises the following steps of: S1, establishing a metering anomaly recognition rule base by taking a corresponding anomaly event caused by a metering device fault as an anomaly recognition target; and S2, carrying out association analysis on the abnormal measurement events, analyzing the possibility of abnormal occurrence in similar or same time, and indicating that the events caused by the abnormity possibly have concurrency and relevance. The efficiency and the accuracy of health management of the metering equipment are improved, and a new technical means is provided for construction of a novel electric power system.

Description

Technical Field

[0001] The present invention relates to the field of metering device anomaly recognition, and in particular to a metering anomaly recognition and anomaly association analysis method based on a rule base. Background Art

[0002] With the rapid development of new power systems, the stable operation and management of metering equipment have become the key to ensuring the high-quality development of the power grid. At present, there is no automatic identification and alarm function for abnormal metering devices. Traditional methods of identifying abnormal metering equipment often rely on regular manual inspections and empirical judgments, which is not only inefficient, but also difficult to reflect the operating status of metering equipment in real time and accurately. With the accuracy of power system data acquisition and collection becoming more and more accurate and reliable, by analyzing the historical data of each sub-indicator collected by the collection equipment and real-time monitoring of the operating status of the substation, timely analyzing the reliability and rationality of each indicator data, timely developing abnormal data, and comprehensively analyzing possible related abnormalities, the early prediction of abnormalities can be achieved, thereby improving the operational management level of metering equipment.

[0003] In the daily operation and maintenance of power supply services, the healthy and stable operation of metering equipment is crucial to ensure the stability and reliability of power supply. However, the existing technology has the following problems in the identification of metering equipment anomalies: First, there are many types of existing metering anomalies and many reasons for the anomalies. The causes and failures of metering anomalies have not been comprehensively analyzed, which makes it difficult to meet the needs of comprehensive analysis and mastery of metering anomalies; Second, most of the existing anomaly identification methods require manual participation and lack digital and intelligent processing capabilities, which not only reduces work efficiency but also increases the risk of human misjudgment; Third, the existing metering anomaly identification mainly conducts post-event source analysis of problems or failures, and lacks correlation analysis to explore possible anomalies in other metering devices; Fourth, due to the limitations of the identification method, the application scope of the metering anomaly identification results is limited, making it difficult to provide effective data support for the optimization decision of the power system. Summary of the invention

[0004] The purpose of the present invention is to provide a method for metering anomaly identification and anomaly association analysis based on a rule base, which not only improves the efficiency and accuracy of metering equipment health management, but also provides a new technical means for the construction of new power systems.

[0005] In order to achieve the above objectives, the present invention is implemented by the following technical solutions:

[0006] A method for measuring anomaly identification and anomaly correlation analysis based on a rule base is characterized in that it includes the following steps:

[0007] Step S1, taking the corresponding abnormal events caused by the failure of the metering device as the abnormality identification target, and establishing a metering abnormality identification rule base;

[0008] Step S2, conduct correlation analysis of measurement abnormal events to analyze the possibility of abnormalities occurring at similar or the same time, indicating that the events caused by the abnormalities may be concurrent and related.

[0009] The metering device failure includes one or more of: functional failure of the metering device, abnormal wiring of the metering device, abnormal clock of the metering device or abnormal electricity consumption behavior of the user.

[0010] The functional failure of the metering device is determined based on the following factors: uneven electric energy indication value, electric energy meter flying away, electric energy meter running backward, electric energy meter stopping, electric energy meter phase failure, voltage imbalance, current loss and current imbalance;

[0011] The uneven electric energy indication value refers to the inequality between the total electric energy indication value of the electric energy meter and the sum of the electric energy indication values ​​of each rate. The judgment rule is as follows:

[0012] If | Positive / reverse active energy indication -∑ 各费率电能示值 Positive / reverse active energy indication|>

[0013] The rate number × K indicates that the electric energy value is uneven; where K is the index judgment threshold value, which is 0.01, and the rate number is specified by the multi-function electric energy meter communication protocol specification, with a maximum value of 63;

[0014] The electric energy meter flying away specifically means that the daily power consumption of the electric energy meter significantly exceeds the normal value, and the judgment rule is:

[0015] like This indicates that the energy meter has flown away. In the formula, K is the index judgment threshold value, which is 60%. The calculation method for the possible maximum daily power consumption is the maximum daily current × voltage value × 24;

[0016] The electric energy meter running backwards specifically means that the current meter reading data is smaller than the previous data, and the judgment rule is:

[0017] If the current electric energy reading minus the previous day's electric energy reading < 0, it indicates that the electric energy meter is running backwards;

[0018] The electric energy meter stops running specifically when the electric energy meter stops running under actual power consumption. The judgment rule is: if the difference between the daily positive / reverse active total electric energy indications of the electric energy meter within 2 days is equal to 0, and 3 points of any phase of the three-phase current monitored during this period are greater than 0.1A, it indicates that the electric energy meter stops running;

[0019] The voltage phase failure is specifically: one or more phases of the metering circuit voltage are disconnected, and the judgment rules are as follows:

[0020] Three-phase three-wire phase loss: the voltage of any phase in the AC phase is less than the K* reference voltage, and the voltage of the other phase is not less than the K* reference voltage; three-phase four-wire phase loss: the voltage of any phase is less than the K* reference voltage, and the voltage of any phase in the other two phases is not less than the K* reference voltage; at least 3 voltage points must be monitored within a day, and a voltage phase loss event is detected on the energy meter or terminal, and a voltage phase loss anomaly is generated;

[0021] The voltage imbalance is specifically the voltage imbalance exceeding the limit, and the judgment rule is:

[0022]

[0023] To ensure the phase voltage (U a ,U b ,U c ) are all greater than 0, the voltage unbalance rate is calculated according to the maximum value - minimum value / maximum value of the three-phase voltage to determine whether it exceeds the limit. The value of K is 0.01. At least 10 points need to be monitored in one day, and the data diagnosis requirements are met for 3 consecutive days, then a voltage unbalance exception is generated;

[0024] The current loss is specifically when any one or two phases of the three-phase current is less than the starting current, and the load current of other phases is greater than 5% of the rated current. The judgment rules are as follows:

[0025] Three-phase three-wire current loss: the current of any phase in the AC phase is less than 0.5% of the rated current, and the current of the other phase is not less than 5% of the rated current;

[0026] Three-phase four-wire current loss: the current of any phase is less than 0.5% of the rated current, and the current of at least one of the other two phases is not less than 10% of the rated current; at least three current points must be monitored within a day, and if a current loss event is detected, a voltage loss abnormality is generated;

[0027] The current imbalance is specifically the current imbalance exceeding the limit, and the judgment rules are as follows:

[0028]

[0029] The current imbalance rate is calculated according to the three-phase current maximum value - minimum value / maximum value to determine whether it exceeds the limit. The value of K is 0.01. At least 10 points need to be monitored in one day, and if the data diagnosis requirements are met for 3 consecutive days, a voltage imbalance abnormality will be generated.

[0030] The abnormality of the wiring of the metering device is determined by the following factors: abnormal reverse power, abnormal phase sequence and reverse flow;

[0031] The reverse power anomaly specifically refers to the reverse power occurrence in normal electricity users, and the judgment rules are as follows:

[0032] Except for users with power generation characteristics, the total reverse active power indication of the energy meter is greater than 0, and the reverse power on the day is greater than the threshold K, and K is 1kWh;

[0033] The phase sequence abnormality specifically refers to the voltage and current of the electric energy meter being in a reverse phase sequence state, and the judgment rule is as follows:

[0034]

[0035] P 总电能表 is the total active power of the energy meter, P 终端总 is the total active power of the terminal, K is the threshold, and K is 10%. If the ratio of the absolute value of the difference between the total active power of the electric energy meter and the total active power of the terminal to the total active power of the terminal is greater than the threshold K, the phase sequence is abnormal;

[0036] The power flow reversal is specifically the current or power reversal, and the judgment rule is:

[0037] If the number of negative values ​​in the current or power curve data per day is greater than the set threshold K, it is judged as power flow reverse, and the value of K is 6.

[0038] The clock anomaly of the metering device is determined based on the following factors: abnormality of the terminal clock and abnormality of the electric energy meter clock;

[0039] The terminal clock anomaly is specifically when the terminal clock deviation exceeds the limit, and the judgment rules are as follows:

[0040] If the absolute value of the difference between the standard time and the terminal calendar clock exceeds the threshold K, it is determined that the terminal clock is abnormal, and the value of K is 1 minute;

[0041] The energy meter clock abnormality is that the clock deviation of the energy meter exceeds the limit, and the judgment rules are as follows:

[0042] If the absolute value of (standard time - terminal calendar clock) + (terminal meter reading time - electricity meter calendar clock) exceeds the threshold K, or the absolute value of the difference between the standard time and the electricity meter calendar clock exceeds the threshold K, it is judged that the electricity meter clock is abnormal, and the value of K is 1 minute.

[0043] The abnormal user power consumption behavior is determined based on the following factors: demand overcapacity, load overcapacity and power factor abnormality;

[0044] The demand overcapacity mentioned above specifically refers to the maximum demand exceeding the user's contract capacity, and the determination rules are as follows:

[0045] Calculate whether the ratio of demand to capacity exceeds the limited threshold value K. If it exceeds the threshold value K, it is determined that the demand exceeds the capacity, and the value of K is 130%;

[0046] The load overcapacity is specifically when the user load exceeds the operating capacity. The judgment rule is: calculate whether the ratio of active power to operating capacity exceeds the specified threshold value K. If it exceeds the threshold value K, it is judged as load overcapacity. The value of K is 130%, and it is monitored at least 3 times in one day and for 3 consecutive days.

[0047] The power factor abnormality is that the daily average power factor is too low, and the judgment rules are as follows:

[0048]

[0049] W p is the total daily positive active power, W q is the total daily forward reactive power, so as to judge whether the daily average power factor is not greater than K, where K is the threshold value of 0.5.

[0050] The step S2 comprises:

[0051] S21. Obtaining historical measurement anomaly related data;

[0052] S22. Construct a correlation model based on historical measurement anomaly related data;

[0053] S23, determining a combination of abnormal events that may be associated;

[0054] S24. According to the order in which the two measurement abnormal events occur, their correlation degrees are calculated based on the correlation coefficients.

[0055] The step S22 comprises:

[0056] When the interval between the occurrence dates of two abnormal events exceeds 15 days, the probability of being caused by the same cause is less than 5%, which is a low-probability event. The threshold of the interval between the occurrence dates d is set to 15; that is, when d>15, it is considered unrelated; d≤15, and one abnormality occurs after the other abnormality is recovered, it is also considered unrelated; when d≤15 and one abnormality occurs before the other abnormality is recovered, the correlation coefficient is:

[0057] The step S23 comprises:

[0058] Assume that there are K types of all the measurement abnormal events that have appeared in the historical data, recorded as

[0059] A k (k=1,2,3,...,K) is the total number of occurrences of each metering abnormal event. The abnormal events are combined in pairs, and according to the order of abnormal occurrence, d≤15 is calculated, and the number of times m that one abnormality occurs before another abnormality is recovered and is for the same user is calculated. k ; Then set an appropriate threshold η, when m k>η, it is preliminarily considered that the two abnormal events may be related; the threshold is set according to the principle of low-probability events, and the calculation formula of the threshold η is as follows:

[0060]

[0061] Where n i A i The total number of occurrences in a year, n j A j The total number of occurrences in a year.

[0062] The step S24 comprises:

[0063] Count the number of times two abnormal events occur within 15 days m i , when m i >η, preliminarily determine A i With A j There may be a correlation, and the correlation coefficient r is calculated each time ip (p=1,2,3...,m i );

[0064] According to statistical experience and repeated data experiments, it is found that when the random variable r ip When the probability of appearing in certain areas is greater than 0.7, it can be considered that r ip Mainly concentrated in this area; Order:

[0065]

[0066] In the formula, x l is the correlation coefficient r ip All possible values ​​of n l For r ip The value is x l The number of l is the probability of the correlation coefficient at each value, for p l Arrange them in order from large to small, denoted as p l The corresponding correlation coefficient is x l ', the number of correlation coefficient values ​​is n l ', let a be the minimum value that makes the following formula valid:

[0067]

[0068] Define k as the dispersion coefficient, and the calculation method is as follows:

[0069]

[0070] The calculation formula of correlation is:

[0071]

[0072] The correlation calculated by the above formula ranges from [0,1]. The closer to 1, the stronger the correlation; otherwise, the weaker it is. When the correlation of the abnormal event combination is greater than or equal to 0.6, the strength of the correlation is medium or above, and it is considered that the abnormal event combination is likely to be a group of concurrent abnormal event combinations.

[0073] Compared with the prior art, the present invention has the following advantages:

[0074] By analyzing and mining the abnormal characteristics and corresponding causes of metering devices, we can fully understand the impact of metering anomalies on power data. Build a metering equipment anomaly identification rule base, comprehensively judge metering equipment anomalies, and improve the timeliness and accuracy of anomaly identification. Carry out correlation analysis of metering anomalies and mine the possibility of concurrent anomalies of metering equipment, thereby reducing the subsequent impact of metering anomalies and improving the reliability of metering equipment operation. The application of this model can reduce the labor costs of power companies, provide effective support for the intelligent construction of new power systems, and enhance the operating efficiency and reliability of power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a flow chart of a method for measuring anomaly identification and anomaly correlation analysis based on a rule base of the present invention;

[0076] Figure 2 It is the flow chart of the relevance algorithm of the present invention;

[0077] Figure 3 It is a statistical diagram of the correlation coefficient of the present invention. DETAILED DESCRIPTION

[0078] The present invention will be further described below by describing a preferred specific embodiment in detail in conjunction with the accompanying drawings.

[0079] like Figure 1 As shown, a method for measuring anomaly identification and anomaly correlation analysis based on a rule base includes the following steps:

[0080] Step S1, taking the corresponding abnormal events caused by the failure of the metering device as the abnormality identification target, and establishing a metering abnormality identification rule base;

[0081] Step S2, conduct correlation analysis of measurement abnormal events to analyze the possibility of abnormalities occurring at similar or the same time, indicating that the events caused by the abnormalities may be concurrent and related.

[0082] The metering device failure includes one or more of: functional failure of the metering device, abnormal wiring of the metering device, abnormal clock of the metering device or abnormal electricity consumption behavior of the user.

[0083] Functional failure of the metering device will cause the electric energy meter to be unable to accurately measure or store the on-site electricity consumption data. The functional failure of the metering device is judged as abnormal based on the following factors: uneven electric energy display value, electric energy meter flying away, electric energy meter running backward, electric energy meter stopping, electric energy meter phase failure, voltage imbalance, current loss and current imbalance;

[0084] The uneven electric energy indication value refers to the difference between the total electric energy indication value of the electric energy meter and the sum of the electric energy indication values ​​of each rate. The data source is the daily frozen positive / reverse active electric energy indication data. The judgment rules are as follows:

[0085] If | Positive / reverse active energy indication -∑ 各费率电能示值 If the positive / reverse active energy indication value|>rate number×K, it indicates that the energy indication value is uneven; where K is the index judgment threshold value and is taken as 0.01, and the rate number is specified by the multi-function energy meter communication protocol specification and the maximum is 63;

[0086] The electric energy meter flying away specifically means that the daily power consumption of the electric energy meter significantly exceeds the normal value, and the data source is the daily frozen power consumption. The judgment rules are as follows:

[0087] like This indicates that the energy meter has flown away. In the formula, K is the index judgment threshold value, which is 60%. The calculation method for the possible maximum daily power consumption is the maximum daily current × voltage value × 24;

[0088] The electric energy meter reverses when the current meter reading data is smaller than the previous data. The data source is the daily frozen electricity. The judgment rule is:

[0089] If the current electric energy reading minus the previous day's electric energy reading < 0, it indicates that the electric energy meter is running backwards;

[0090] The specific stop of the electric energy meter is: the electric energy meter stops running under actual power consumption, the data source is the daily frozen positive / reverse active total electric energy indication and the three-phase current daily curve, and the judgment rule is: if the difference of the daily positive / reverse active total electric energy indication of the electric energy meter within 2 days is equal to 0, and 3 points of any phase of the three-phase current monitored during this period are greater than 0.1A, it indicates that the electric energy meter stops running;

[0091] The voltage phase failure is specifically: one or more phases of the metering circuit voltage are disconnected, the data source is the three-phase voltage curve, and the judgment rule is:

[0092] Three-phase three-wire phase loss: the voltage of any phase in the AC phase is less than the K* reference voltage, and the voltage of the other phase is not less than the K* reference voltage; three-phase four-wire phase loss: the voltage of any phase is less than the K* reference voltage, and the voltage of any phase in the other two phases is not less than the K* reference voltage; at least 3 voltage points must be monitored within a day, and a voltage phase loss event is detected on the energy meter or terminal, and a voltage phase loss anomaly is generated;

[0093] The voltage imbalance is specifically the voltage imbalance exceeding the limit, the data source is the three-phase voltage curve and the judgment rule is:

[0094]

[0095] To ensure the phase voltage (U a ,U b ,U c ) are all greater than 0, the voltage unbalance rate is calculated according to the maximum value - minimum value / maximum value of the three-phase voltage to determine whether it exceeds the limit. The value of K is 0.01. At least 10 points need to be monitored in one day, and the data diagnosis requirements are met for 3 consecutive days, then a voltage unbalance exception is generated;

[0096] The current loss is specifically when any one or two phases of the three-phase current is less than the starting current, and the load current of other phases is greater than 5% of the rated current. The judgment rules are as follows:

[0097] The voltage range must be 0.7 to 0.9 times the rated voltage. Three-phase three-wire current loss: the current of any phase in the AC phase is less than 0.5% of the rated (basic) current, and the current of the other phase is not less than 5% of the rated (basic) current;

[0098] Three-phase four-wire current loss: the current of any phase is less than 0.5% of the rated current, and the current of at least one of the other two phases is not less than 10% of the rated current; at least three current points must be monitored within a day, and if a current loss event is detected, a voltage loss abnormality is generated;

[0099] The current imbalance is specifically the current imbalance exceeding the limit, and the judgment rules are as follows:

[0100]

[0101] To ensure that each phase voltage (I a ,I b ,I c ) are all greater than 0, the current imbalance rate is calculated according to the maximum value - minimum value / maximum value of the three-phase current to determine whether it is out of limit. The value of K is 0.01. At least 10 points need to be monitored in one day, and the data diagnosis requirements are met for 3 consecutive days, then a voltage imbalance abnormality is generated.

[0102] The abnormality of the wiring of the metering device directly affects the correct measurement of electric energy by the electric energy metering device. Due to the complexity of the wiring of the metering device, negligence of the staff, and the unauthorized change of the wiring by the electricity thief will lead to abnormal wiring of the metering device, thereby triggering the corresponding abnormal measurement event. The abnormality of the wiring of the metering device is determined by the following factors: abnormal reverse power, abnormal phase sequence and reverse flow;

[0103] The reverse power anomaly specifically refers to the reverse power occurrence in normal electricity users, and the judgment rules are as follows:

[0104] Except for users with power generation characteristics, the total reverse active power indication of the energy meter is greater than 0, and the reverse power on the day is greater than the threshold K, and K is 1kWh;

[0105] The phase sequence abnormality specifically refers to the voltage and current of the electric energy meter being in a reverse phase sequence state, and the judgment rule is as follows:

[0106]

[0107] P 总电能表 is the total active power of the energy meter, P 终端总 is the total active power of the terminal, K is the threshold, and K is 10%. If the ratio of the absolute value of the difference between the total active power of the electric energy meter and the total active power of the terminal to the total active power of the terminal is greater than the threshold K, the phase sequence is abnormal;

[0108] The power flow reversal is specifically the current or power reversal, and the judgment rule is:

[0109] If the number of negative values ​​in the current or power curve data per day is greater than the set threshold K, it is judged as power flow reverse, and the value of K is 6.

[0110] Metering device clock abnormality. The current system collects frozen indication data in a time-stamped manner. If the metering device clock is abnormal, the tasks and data times between the various levels of the system cannot correspond, which will affect the normal collection of data and thus cause metering abnormality events. The metering device clock abnormality is determined based on the following factors: terminal clock abnormality and electric energy meter clock abnormality;

[0111] The terminal clock anomaly is specifically when the terminal clock deviation exceeds the limit, and the judgment rules are as follows:

[0112] If the absolute value of the difference between the standard time and the terminal calendar clock exceeds the threshold K, it is determined that the terminal clock is abnormal, and the value of K is 1 minute;

[0113] The energy meter clock abnormality is that the clock deviation of the energy meter exceeds the limit, and the judgment rules are as follows:

[0114] If the absolute value of (standard time - terminal calendar clock) + (terminal meter reading time - electricity meter calendar clock) exceeds the threshold K, or the absolute value of the difference between the standard time and the electricity meter calendar clock exceeds the threshold K, it is judged that the electricity meter clock is abnormal, and the value of K is 1 minute.

[0115] Abnormal user electricity consumption behavior will have adverse effects on metering devices and power systems. Therefore, it is necessary to detect such abnormalities in a timely manner and prevent them in time to prevent the situation from deteriorating. The abnormal user electricity consumption behavior is determined based on the following factors: demand overcapacity, load overcapacity and power factor abnormality;

[0116] The demand overcapacity is specifically when the maximum demand exceeds the user's contract capacity. The determination rule is: calculate whether the ratio of demand to capacity exceeds a limited threshold value K. If it exceeds the threshold value K, it is determined as demand overcapacity, and the value of K is 130%;

[0117] The load overcapacity is specifically when the user load exceeds the operating capacity. The judgment rule is: calculate whether the ratio of active power to operating capacity exceeds the specified threshold value K. If it exceeds the threshold value K, it is judged as load overcapacity. The value of K is 130%, and it is monitored at least 3 times in one day and for 3 consecutive days.

[0118] The power factor abnormality is that the daily average power factor is too low, and the judgment rules are as follows:

[0119]

[0120] W p is the total daily positive active power, W q is the total daily forward reactive power, so as to judge whether the daily average power factor is not greater than K, where K is the threshold value of 0.5.

[0121] like Figure 2 As shown, the step S2 includes:

[0122] S21. Obtaining historical measurement anomaly related data;

[0123] S22. Construct a correlation model based on historical measurement anomaly related data;

[0124] S23, determining a combination of abnormal events that may be associated;

[0125] S24. According to the order in which the two measurement abnormal events occur, their correlation degrees are calculated based on the correlation coefficients.

[0126] The step S22 comprises:

[0127] Two abnormal events can have a certain correlation because they may be caused by the same abnormal cause. When the interval between the occurrence dates of two abnormal events exceeds 15 days, the probability of being caused by the same cause is less than 5%, which is a low-probability event. The threshold of the interval between the occurrence dates d is set to 15; that is, when d>15, it is considered unrelated; d≤15, and one abnormality occurs after the other abnormality is recovered, it is also considered unrelated; when d≤15 and one abnormality occurs before the other abnormality is recovered, the correlation coefficient is:

[0128]

[0129] The step S23 comprises:

[0130] Assume that there are K types of all measurement abnormal events that have occurred in historical data, recorded as A k (k=1,2,3,...,K) is the total number of occurrences of each metering abnormal event. The abnormal events are combined in pairs, and according to the order of abnormal occurrence, d≤15 is calculated, and the number of times m that one abnormality occurs before another abnormality is recovered and is for the same user is calculated. k ; Then set an appropriate threshold η, when m k >η, it is preliminarily considered that the two abnormal events may be related; the threshold is set according to the principle of low-probability events, and the calculation formula of the threshold η is as follows:

[0131]

[0132] Where n i A i The total number of occurrences in a year, n j A j The total number of occurrences in a year.

[0133] The step S24 comprises:

[0134] A i A occurs first j After that, calculate the correlation R ij Take the example of counting the number of times two abnormal events occur within 15 days. i , when m i >η, preliminarily determine A i With A j There may be a correlation, and the correlation coefficient r is calculated each time ip (p=1,2,3...,m i );

[0135] According to statistical experience and repeated data experiments, it is found that when the random variable r ip When the probability of appearing in certain areas is greater than 0.7, it can be considered that r ipMainly concentrated in this area; Order:

[0136]

[0137] In the formula, x l is the correlation coefficient r ip All possible values ​​of n l For r ip The value is x l The number of l is the probability of the correlation coefficient at each value, for p l Arrange them in order from large to small, denoted as p l The corresponding correlation coefficient is x l ', the number of correlation coefficient values ​​is n l ', let a be the minimum value that makes the following formula valid:

[0138]

[0139] Define k as the dispersion coefficient, and the calculation method is as follows:

[0140]

[0141] The calculation formula of correlation is:

[0142]

[0143] The correlation calculated by the above formula ranges from [0,1]. The closer to 1, the stronger the correlation; otherwise, the weaker it is. When the correlation of the abnormal event combination is greater than or equal to 0.6, the strength of the correlation is medium or above, and it is considered that the abnormal event combination is likely to be a group of concurrent abnormal event combinations.

[0144] Taking power flow reversal and phase sequence abnormality as an example, the historical data of metering abnormality of special transformer users in Shanghai in 2023 are selected for correlation calculation. Let the power flow reversal be A1 and the phase sequence abnormality be A2. The number of occurrences of power flow reversal and phase sequence abnormality are shown in Table 1.

[0145] Table 1

[0146]

[0147] Correlation calculation process: Preliminary judgment on whether A1 and A2 are correlated when the reverse flow occurs first, and obtain:

[0148]

[0149] Since η<167, it is considered that when A1 occurs first, there may be a correlation between A1 and A2.

[0150] Then calculate the correlation coefficient of each of the 167 groups and get the correlation coefficient r lp (p=1,2,...,167). And the result distribution diagram is obtained, such as Figure 3 As shown in Figure 2, it can be seen that the correlation coefficient is mainly distributed between 0.9 and 1.

[0151] Some values ​​of various parameters in the correlation calculation process are shown in Table 2.

[0152] Table 2

[0153]

[0154]

[0155] Since P0′+P1′+P2′=0.76>0.7, so a=2, k=1-0.1×a=0.8, the final correlation is given by the formula Yes, R 12 =0.78.

[0156] Due to R 12 >0.6, it is considered that when the flow reversal occurs first, the flow reversal is associated with the phase sequence anomaly, and the correlation is extremely strong.

[0157] When power flow reversal occurs first and phase sequence abnormality occurs later, the correlation degree is greater than the set threshold of 0.6. That is, when power flow reversal and phase sequence abnormality occur to the same user in the electricity consumption information collection system within the same period of time, it is considered that there is a high possibility that they are a group of concurrent abnormal events, and there is a correlation between the two metering abnormal events.

[0158] In summary, the method for metering anomaly identification and anomaly association analysis based on a rule base of the present invention not only improves the efficiency and accuracy of metering equipment health management, but also provides a new technical means for the construction of a new power system.

[0159] It should be noted that, in the embodiments of the present invention, the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the embodiments, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0160] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0161] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be appreciated that the above description should not be considered as a limitation of the present invention. After reading the above content, it will be apparent to those skilled in the art that various modifications and substitutions of the present invention will occur. Therefore, the protection scope of the present invention should be limited by the appended claims.

Claims

1. A method for measuring anomaly identification and anomaly correlation analysis based on a rule base, characterized in that: The steps include: Step S1, taking the corresponding abnormal events caused by the failure of the metering device as the abnormality identification target, and establishing a metering abnormality identification rule base; Step S2, conduct correlation analysis of measurement abnormal events to analyze the possibility of abnormalities occurring at similar or the same time, indicating that the events caused by the abnormalities may be concurrent and related.

2. The method for measuring anomaly identification and anomaly correlation analysis based on a rule base according to claim 1, characterized in that: The metering device failure includes one or more of: functional failure of the metering device, abnormal wiring of the metering device, abnormal clock of the metering device or abnormal electricity consumption behavior of the user.

3. The method for measuring anomaly identification and anomaly correlation analysis based on rule base according to claim 2, characterized in that: The functional failure of the metering device is determined based on the following factors: uneven electric energy indication value, electric energy meter flying away, electric energy meter running backward, electric energy meter stopping, electric energy meter phase failure, voltage imbalance, current loss and current imbalance; The uneven electric energy indication value refers to the inequality between the total electric energy indication value of the electric energy meter and the sum of the electric energy indication values ​​of each rate. The judgment rule is as follows: If | Positive / reverse active energy indication -∑ 各费率电能示值 If the positive / reverse active energy indication value|>rate number×K, it indicates that the energy indication value is uneven; where K is the index judgment threshold value and is taken as 0.01, and the rate number is specified by the multi-function energy meter communication protocol specification and the maximum is 63; The electric energy meter flying away specifically means that the daily power consumption of the electric energy meter significantly exceeds the normal value, and the judgment rule is: like This indicates that the energy meter has flown away. In the formula, K is the index judgment threshold value, which is 60%. The calculation method for the possible maximum daily power consumption is the maximum daily current × voltage value × 24; The electric energy meter running backwards specifically means that the current meter reading data is smaller than the previous data, and the judgment rule is: If the current electric energy reading minus the previous day's electric energy reading < 0, it indicates that the electric energy meter is running backwards; The electric energy meter stops running specifically when the electric energy meter stops running under actual power consumption. The judgment rule is: if the difference between the daily positive / reverse active total electric energy indications of the electric energy meter within 2 days is equal to 0, and 3 points of any phase of the three-phase current monitored during this period are greater than 0.1A, it indicates that the electric energy meter stops running; The voltage phase failure is specifically: one or more phases of the metering circuit voltage are disconnected, and the judgment rules are as follows: Three-phase three-wire phase loss: the voltage of any phase in the AC phase is less than the K* reference voltage, and the voltage of the other phase is not less than the K* reference voltage; three-phase four-wire phase loss: the voltage of any phase is less than the K* reference voltage, and the voltage of any phase in the other two phases is not less than the K* reference voltage; at least 3 voltage points must be monitored within a day, and a voltage phase loss event is detected on the energy meter or terminal, and a voltage phase loss anomaly is generated; The voltage imbalance is specifically the voltage imbalance exceeding the limit, and the judgment rule is: To ensure the phase voltage (U a ,U b ,U c ) are all greater than 0, the voltage unbalance rate is calculated according to the maximum value - minimum value / maximum value of the three-phase voltage to determine whether it exceeds the limit. The value of K is 0.

01. At least 10 points need to be monitored in one day, and the data diagnosis requirements are met for 3 consecutive days, then a voltage unbalance exception is generated; The current loss is specifically when any one or two phases of the three-phase current is less than the starting current, and the load current of other phases is greater than 5% of the rated current. The judgment rules are as follows: Three-phase three-wire current loss: the current of any phase in the AC phase is less than 0.5% of the rated current, and the current of the other phase is not less than 5% of the rated current; Three-phase four-wire current loss: the current of any phase is less than 0.5% of the rated current, and the current of at least one of the other two phases is not less than 10% of the rated current; at least three current points must be monitored within a day, and if a current loss event is detected, a voltage loss abnormality is generated; The current imbalance is specifically the current imbalance exceeding the limit, and the judgment rules are as follows: The current imbalance rate is calculated according to the three-phase current maximum value - minimum value / maximum value to determine whether it exceeds the limit. The value of K is 0.

01. At least 10 points need to be monitored in one day, and if the data diagnosis requirements are met for 3 consecutive days, a voltage imbalance abnormality will be generated.

4. The method for measuring anomaly identification and anomaly association analysis based on a rule base as claimed in claim 2, characterized in that: The abnormality of the wiring of the metering device is determined by the following factors: abnormal reverse power, abnormal phase sequence and reverse flow; The reverse power anomaly specifically refers to the reverse power occurrence in normal electricity users, and the judgment rules are as follows: Except for users with power generation characteristics, the total reverse active power indication of the energy meter is greater than 0, and the reverse power on the day is greater than the threshold K, and K is 1kWh; The phase sequence abnormality specifically refers to the voltage and current of the electric energy meter being in a reverse phase sequence state, and the judgment rule is as follows: P 总电能表 is the total active power of the energy meter, P 终端总 is the total active power of the terminal, K is the threshold, and K is 10%. If the ratio of the absolute value of the difference between the total active power of the electric energy meter and the total active power of the terminal to the total active power of the terminal is greater than the threshold K, the phase sequence is abnormal; The power flow reversal is specifically the current or power reversal, and the judgment rule is: If the number of negative values ​​in the current or power curve data per day is greater than the set threshold K, it is judged as power flow reverse, and the value of K is 6.

5. The method for measuring anomaly identification and anomaly correlation analysis based on rule base according to claim 2, characterized in that: The clock anomaly of the metering device is determined based on the following factors: abnormality of the terminal clock and abnormality of the electric energy meter clock; The terminal clock anomaly is specifically when the terminal clock deviation exceeds the limit, and the judgment rules are as follows: If the absolute value of the difference between the standard time and the terminal calendar clock exceeds the threshold K, it is determined that the terminal clock is abnormal, and the value of K is 1 minute; The energy meter clock abnormality is that the clock deviation of the energy meter exceeds the limit, and the judgment rules are as follows: If the absolute value of (standard time - terminal calendar clock) + (terminal meter reading time - electricity meter calendar clock) exceeds the threshold K, or the absolute value of the difference between the standard time and the electricity meter calendar clock exceeds the threshold K, it is judged that the electricity meter clock is abnormal, and the value of K is 1 minute.

6. The method for measuring anomaly identification and anomaly correlation analysis based on rule base according to claim 2, characterized in that: The abnormal user power consumption behavior is determined based on the following factors: demand overcapacity, load overcapacity and power factor abnormality; The demand overcapacity is specifically when the maximum demand exceeds the user's contract capacity. The determination rule is: calculate whether the ratio of demand to capacity exceeds a limited threshold value K. If it exceeds the threshold value K, it is determined as demand overcapacity, and the value of K is 130%; The load overcapacity is specifically when the user load exceeds the operating capacity. The judgment rule is: calculate whether the ratio of active power to operating capacity exceeds the specified threshold value K. If it exceeds the threshold value K, it is judged as load overcapacity. The value of K is 130%, and it is monitored at least 3 times in one day and for 3 consecutive days. The power factor abnormality is that the daily average power factor is too low, and the judgment rules are as follows: W p is the total daily positive active power, W q is the total daily forward reactive power, so as to judge whether the daily average power factor is not greater than K, where K is the threshold value of 0.

5.

7. The method for measuring anomaly identification and anomaly correlation analysis based on rule base according to claim 1, characterized in that: The step S2 comprises: S21. Obtaining historical measurement anomaly related data; S22. Construct a correlation model based on historical measurement anomaly related data; S23, determining a combination of abnormal events that may be associated; S24. According to the order in which the two measurement abnormal events occur, their correlation degrees are calculated based on the correlation coefficients.

8. The method for measuring anomaly identification and anomaly association analysis based on a rule base as claimed in claim 7, characterized in that: The step S22 comprises: When the interval between the occurrence dates of two abnormal events exceeds 15 days, the probability of being caused by the same cause is less than 5%, which is a low-probability event. The threshold of the interval between the occurrence dates d is set to 15; that is, when d>15, it is considered unrelated; d≤15, and one abnormality occurs after the other abnormality is recovered, it is also considered unrelated; when d≤15 and one abnormality occurs before the other abnormality is recovered, the correlation coefficient is:

9. The method for measuring anomaly identification and anomaly association analysis based on a rule base as claimed in claim 8, characterized in that: The step S23 comprises: Assume that there are K types of all measurement abnormal events that have occurred in historical data, recorded as A k (k=1,2,3,...,K) is the total number of occurrences of each metering abnormal event. The abnormal events are combined in pairs, and the number of times m that one abnormal event occurs before another abnormal event is recovered and is for the same user is calculated according to the order of abnormal occurrence. k ; Then set an appropriate threshold η, when m k >η, it is preliminarily considered that the two abnormal events may be related; the threshold is set according to the principle of low-probability events, and the calculation formula of the threshold η is as follows: Where n i A i The total number of occurrences in a year, n j A j The total number of occurrences in a year.

10. The method for measuring anomaly identification and anomaly association analysis based on rule base according to claim 9, characterized in that: The step S24 comprises: Count the number of times two abnormal events occur within 15 days m i , when m i >η, preliminarily determine A i With A j There may be a correlation, and the correlation coefficient r is calculated each time ip (p=1,2,3...,m i ); According to statistical experience and repeated data experiments, it is found that when the random variable r ip When the probability of appearing in certain areas is greater than 0.7, it can be considered that r ip Mainly concentrated in this area; Order: In the formula, x l is the correlation coefficient r ip All possible values ​​of n l For r ip The value is x l The number of l is the probability of the correlation coefficient at each value, for p l Arrange them in order from large to small, denoted as p l The corresponding correlation coefficient is x l ', the number of correlation coefficient values ​​is n l ', let a be the minimum value that makes the following formula valid: Define k as the dispersion coefficient, and the calculation method is as follows: The calculation formula of correlation is: The correlation calculated by the above formula ranges from [0,1]. The closer to 1, the stronger the correlation; otherwise, the weaker it is. When the correlation of the abnormal event combination is greater than or equal to 0.6, the strength of the correlation is medium or above, and it is considered that the abnormal event combination is likely to be a group of concurrent abnormal event combinations.

Images