CN117272033B - DC shunt current metering abnormity monitoring method - Google Patents
DC shunt current metering abnormity monitoring method Download PDFInfo
- Publication number
- CN117272033B CN117272033B CN202311567436.4A CN202311567436A CN117272033B CN 117272033 B CN117272033 B CN 117272033B CN 202311567436 A CN202311567436 A CN 202311567436A CN 117272033 B CN117272033 B CN 117272033B
- Authority
- CN
- China
- Prior art keywords
- current
- value
- shunt
- circuit
- degree
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012544 monitoring process Methods 0.000 title claims abstract description 40
- 230000002159 abnormal effect Effects 0.000 claims abstract description 55
- 238000005259 measurement Methods 0.000 claims abstract description 42
- 230000005856 abnormality Effects 0.000 claims abstract description 25
- 238000003066 decision tree Methods 0.000 claims abstract description 16
- 238000012216 screening Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 11
- 230000004044 response Effects 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 11
- 238000001514 detection method Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Measurement Of Resistance Or Impedance (AREA)
Abstract
The invention relates to the technical field of current anomaly monitoring, in particular to a method for monitoring metering anomalies of direct current shunt current. Firstly, obtaining error source interference degree and an abnormality factor through circuit data in a preset reference period, further correcting the initial path length of each current measurement value in sample data obtained through an isolated decision tree according to the error source interference degree and the abnormality factor, and repeating for a plurality of times to obtain a final score of each current measurement value; and finally, screening the sample data by utilizing the final score to obtain abnormal circuit data. The invention fully considers the fluctuation of voltage and current in the running process of the circuit, the difference between different branches and the error influence of influencing factors on the current metering value, and the decision judgment process is adaptively adjusted in real time, so that the final detection result is more accurate, and the accuracy of the monitoring method is improved.
Description
Technical Field
The invention relates to the technical field of current anomaly monitoring, in particular to a method for monitoring metering anomalies of direct current shunt current.
Background
A dc shunt meter is an instrument for measuring the current of each shunt in a dc circuit, and is typically composed of a current sensor and a display device. The current measuring device can accurately measure current values in different branches, and monitor and evaluate the working state of the circuit in real time. Abnormal current values may suggest problems in the circuit, such as overload, short circuits, leakage, etc., so that measures are taken in time to prevent further damage or failure from occurring.
Besides fault factors, error fluctuation data can be generated in the direct current shunt current meter due to element faults, power supply fluctuation, environmental condition change, equipment aging, human errors and the like, so that great interference is caused to the monitoring accuracy of circuit fault abnormality, the error interference to each shunt is dynamic change, in the traditional isolation tree algorithm in the direct current shunt metering abnormality monitoring process, the dynamic error interference cannot be considered, the obtained path length cannot be directly used for accurately representing the abnormality of data information, the situation that abnormal current metering data cannot be identified or missed is caused, and the monitoring method is not accurate enough.
Disclosure of Invention
In order to solve the technical problem that abnormal current metering data cannot be accurately identified by using a traditional isolated tree algorithm for abnormal monitoring, the invention aims to provide a direct current shunt current metering abnormal monitoring method, which adopts the following specific technical scheme:
collecting circuit data of each shunt circuit at a preset collection frequency within a preset reference period; the circuit data comprises influence factors which influence the preset category of the circuit;
analyzing the fluctuation similarity degree of the current measurement value and the total circuit voltage in the circuit data of each shunt circuit to obtain the error source interference degree of each shunt circuit; analyzing the difference between the fluctuation of the voltage difference in the circuit data of each shunt and the fluctuation of the influence factor change curve in the circuit data to obtain the total potential loss degree of all the shunts; obtaining a loss characteristic value of each shunt according to the statistical characteristics of the voltage and current measurement values of each shunt in the circuit data; analyzing fluctuation characteristics of the loss characteristic values of all branches, and weighting by utilizing the total potential loss degree to obtain abnormal factors of all branches;
acquiring real-time monitoring data of all branches as sample data according to preset monitoring time length; constructing an isolated decision tree according to the current metering values in the sample data, and obtaining the initial path length corresponding to each current metering value in the sample data; according to the anomaly factor and the proportion of the error source interference degree of each shunt in all the shunts, the initial path length is adjusted, and the corrected path length is obtained; obtaining an anomaly score value according to the corrected path length; repeatedly constructing an isolated decision tree according to preset times, obtaining new abnormal score values, and obtaining the final score of each current metering value in the sample data according to all the abnormal score values;
and screening the sample data according to the final score to obtain abnormal circuit data.
Further, the method for obtaining the interference degree of the error source comprises the following steps:
acquiring a voltage coefficient of fluctuation of an input total voltage value in the circuit data in each shunt and a current coefficient of fluctuation of a current metering value of each shunt; the ratio of the current coefficient to the voltage coefficient of each shunt is subjected to logarithmic transformation, and an absolute value is taken, so that the corresponding response degree of each shunt is obtained;
acquiring a correlation coefficient between an input total voltage value and each shunt current measurement value in the circuit data; multiplying the response level of each branch by the correlation coefficient, and taking the product as the error source interference level of each branch.
Further, the method for acquiring the total potential loss degree comprises the following steps:
acquiring the total power consumption loss degree according to a total potential loss degree calculation formula; the total potential loss degree calculation formula includes:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicating the degree of total potential loss; />Representing the acquisition sequence number of circuit data in a preset reference period; />Representing the total number of circuit data acquisition times in a preset reference period; />Indicate->Voltage difference values of input and output ends in the data of the acquisition circuits; />Representing the average value of the voltage difference values of the input end and the output end in all the acquired circuit data; />A serial number indicating the type of the influencing factor in the circuit data; />The number of influencing factor types in the circuit data is represented; />Indicate->The first part of the data of the acquisition circuit>An influence value of a seed influence factor; />Representing the +.>The mean value of the influence values of the influence factors.
Further, the method for obtaining the loss characteristic value comprises the following steps:
and taking the ratio of the voltage range and the current metering value range of each branch in the circuit data as the loss characteristic value of each branch.
Further, the method for obtaining the abnormal factor includes:
acquiring the variance of the loss characteristic value as the loss non-uniformity degree of all branches; the degree of loss unevenness is multiplied by the degree of total potential loss to obtain an abnormality factor.
Further, the method for acquiring the corrected path length includes:
correcting the initial path length by using a path length correction formula; the path length correction formula includes:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein->A serial number indicating the current measurement value; />A serial number representing a shunt; />Representing the total number of branches; />Indicate->Initial path lengths of the individual current metric values; />Indicate->The serial numbers of the branches where the current metering values are located; />Indicate->Error source interference degrees corresponding to the branches; />Representing the summation of the error source interference degrees of all the branches; />Representing an anomaly factor; />Indicate->Corrected path lengths for the individual current measurements.
Further, the method for obtaining the final score comprises the following steps:
and taking the average value of all the abnormal score values of each current measurement value in the sample data as the corresponding final score.
Further, the method of screening the sample data according to the final score includes:
and screening out circuit data with the final score of the current metering value in the sample data being greater than a preset abnormal threshold value to obtain abnormal circuit data.
Further, the preset anomaly threshold value is 0.7.
Further, the preset reference period is 10 minutes, the preset monitoring time period is 1 minute, and the preset acquisition frequency is 1 second and 1 time.
The invention has the following beneficial effects:
firstly, analyzing circuit data in a preset reference period, analyzing the fluctuation similarity of a current metering value and the total voltage of the circuit to obtain the interference degree of an error source, the total potential loss degree of analysis potential and the difference of different shunt loss characteristic values to obtain the abnormal factors of all shunts, and providing a basis for adjusting an abnormal detection rule; further utilizing an isolated decision tree to obtain the initial path length of the current measurement value in the sample data, and preparing for subsequent correction and obtaining an evaluation basis; further correcting the initial path length by utilizing the interference degree of the error source and the abnormality factor to obtain a corrected path length which is more in line with the actual running situation of the circuit, further obtaining a more accurate abnormality score value and providing a basis for evaluating the current metering value in the sample data; further repeatedly constructing an isolated decision tree to obtain new abnormal score values, and obtaining more convincing final scores from all the abnormal score values; and finally, screening sample data by utilizing the final score, and monitoring the abnormal current metering value pair. The invention fully considers the fluctuation of voltage and current in the running process of the circuit, the difference between different branches and the error influence of influencing factors on the current metering value, and the decision judgment process is adaptively adjusted in real time, so that the final detection result is more accurate, and the accuracy of the monitoring method is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for monitoring dc shunt current metering abnormality according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of a method for monitoring the metering abnormality of the direct current shunt current according to the invention, which is specific to the implementation, structure, characteristics and effects thereof, with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the method for monitoring the metering abnormality of the direct current shunt current provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for monitoring metering abnormality of a dc shunt current according to an embodiment of the present invention specifically includes:
step S1: collecting circuit data of each shunt circuit at a preset collection frequency within a preset reference period; the circuit data includes influence factors influencing the preset category of the circuit.
In direct current shunt metering, the error interference suffered by each shunt is dynamic, and the dynamic error interference suffered by each measuring point needs to be accurately judged, so that the interference of error fluctuation data is removed, and the monitoring result is more accurate.
In the embodiment of the invention, the reference time period is set to be 10 minutes before the current running time of the circuit, the preset acquisition frequency is 1 second and 1 time, the acquired circuit data comprise the voltages of the input end and the output end, the current measurement values of all branches and the data of the planned influencing factors, and the data comprise the circuit temperature, the impedance and the influence value change curve of the signal to noise ratio of the sensor; in other embodiments of the present invention, the practitioner can set the reference period, the acquisition frequency, and the type and number of influencing factors by himself.
Step S2: analyzing the fluctuation similarity degree of the current measurement value and the total circuit voltage in the circuit data of each shunt circuit to obtain the error source interference degree of each shunt circuit; analyzing the difference between the fluctuation of the voltage difference in the circuit data of each shunt and the fluctuation of the influence factor change curve in the circuit data to obtain the total potential loss degree of all the shunts; obtaining a loss characteristic value of each shunt according to the statistical characteristics of the voltage and current measurement values of each shunt in the circuit data; and analyzing fluctuation characteristics of loss characteristic values of all branches, and weighting by using the total potential loss degree to obtain anomaly factors of all branches.
The circuit to be tested generally operates in a rated voltage range, certain voltage fluctuation can be inevitably caused by various factors, the fluctuation is mainly reflected from the input end and the output end of the circuit, such as power supply noise, line interference, load change and the like of the input end, and potential loss of charges when the current reaches the output end after passing through a plurality of branches and elements, so that the error source interference degree and an abnormal factor are obtained to be used for representing the error characteristics of the circuit operation, so that the interference of circuit error fluctuation data is removed in the subsequent monitoring process, and the monitoring result is more accurate.
In the running process of the circuit, the current metering value is directly subjected to voltage fluctuation to change, but due to factors such as line interference, load change and the like, the change of the current metering value and the fluctuation change of the total voltage of the circuit have certain difference, so that the fluctuation similarity degree of each shunt circuit needs to be analyzed, and the error source interference degree of each shunt circuit is obtained.
Preferably, in one embodiment of the present invention, considering that the coefficient of the kunity may measure the unequal distribution of the circuit data, the degree of response of the current to the voltage fluctuation is reflected by comparing the similarity of the coefficient of the kunity of the current measurement value and the voltage; meanwhile, in order to further improve the reliability of the interference degree of the error source, the response degree is adjusted by using the correlation coefficient, so that the method for acquiring the interference degree of the error source comprises the following steps:
acquiring a voltage coefficient of fluctuation of an input total voltage value in circuit data in each shunt and a current coefficient of fluctuation of a current metering value of each shunt; taking the ratio of the current coefficient and the voltage coefficient of each shunt as the corresponding response degree of each shunt;
acquiring a correlation coefficient between an input total voltage value and each shunt current measurement value in circuit data; mapping and normalizing the correlation coefficient negative correlation to obtain an interference coefficient; the response degree of each branch is converted by logarithm and takes absolute value, and then multiplied by the interference coefficient, and the product is used as the error source interference degree of each branch. The calculation formula of the error source interference degree comprises:
wherein,a shunt number representing the circuit; />Indicate->Error source interference degrees of the branches; />Representing a preset base number, in one embodiment of the invention +.>；/>Indicate->Current-based coefficients of the individual branches; />Representing the voltage coefficient of the base; />Indicate->The degree of response of the individual branches; />Indicate->A set of shunt current measurement values; />A set of voltage values representing an input total voltage; />Representing the covariance of the acquired bracketed data; />Indicate->Covariance of the current measurement value set and the voltage value set of the branches; />Indicate->Standard deviation of the current measurement value sets of the branches; />Standard deviation of a set of voltage values representing an input total voltage; />Indicate->Shunt current measurement valuesInputting a correlation coefficient of the total voltage value; />Indicate->Interference coefficient of each shunt current measurement value and the input total voltage value.
In the calculation formula of the interference degree of the error source, the more similar the current coefficient is to the voltage coefficient, the closer the ratio is to 1, the higher the response degree of the shunt current measurement value to the input voltage fluctuation is,the smaller the portion, the smaller the degree of error source interference; the larger the correlation coefficient is, the higher the fluctuation similarity degree of the current measurement value and the total voltage of the circuit is, the smaller the interference coefficient is, and the smaller the error source interference degree is.
The number of the measuring points in each branch can be adjusted according to actual needs, a plurality of measuring points can be arranged in each branch, the mean value or mode or other characteristics of the measuring points are counted, then the error source interference degree is calculated, and the error source interference degree can be calculated for each measuring point; the coefficient of kunity is a well known technique to those skilled in the art and will not be described in detail herein.
In the running process of the circuit, various influencing factors exist, the fluctuation similarity of the current measurement value to the voltage cannot completely represent the error characteristics of the circuit, and the loss after the potential flows through various branches, elements and lines is needed to be analyzed to obtain the total potential loss degree, so that the abnormality factor is obtained.
Preferably, in an embodiment of the present invention, the method for acquiring the total potential loss degree includes:
acquiring the total power consumption loss degree according to a total potential loss degree calculation formula; the total potential loss degree calculation formula includes:
wherein,indicating the degree of total potential loss; />Representing the acquisition sequence number of circuit data in a preset reference period; />Representing the total number of circuit data acquisition times in a preset reference period; />Indicate->Voltage difference values of input and output ends in the data of the acquisition circuits; />Representing the average value of the voltage difference values of the input end and the output end in all the acquired circuit data; />A serial number indicating the type of the influencing factor in the circuit data; />The number of influencing factor types in the circuit data is represented; />Indicate->The first part of the data of the acquisition circuit>An influence value of a seed influence factor; />Representing the +.>The mean value of the influence values of the influence factors.
In the total potential loss degree calculation formula, the smaller the difference between the voltage difference fluctuation and the influence factor fluctuation is, the greater the potential loss is caused by various influence factors, the smaller the potential loss degree of the shunt is, and the smaller the total potential loss degree is.
In order to obtain more accurate characteristic parameters capable of representing circuit error fluctuation characteristics, loss characteristics of each shunt are required to be analyzed, so that loss differences among all the shunts can be analyzed, the total potential loss degree is subjected to weighted adjustment, an anomaly factor is obtained, the accuracy of subsequent anomaly detection is improved, and the reliability of a monitoring method is improved.
Preferably, in one embodiment of the present invention, considering that the range of the range can reflect the fluctuation range of the data, the voltage range can be used to measure the degree of voltage loss in the circuit, the current measurement range can be used to measure the degree of current loss in the circuit, and the relationship between each shunt voltage loss and current loss can be quickly and roughly estimated through the ratio of the voltage range and the current measurement range, and can be used as the loss characteristic of the shunt, so the loss characteristic value obtaining method comprises:
the ratio of the voltage range and the current metering value range of each branch in the circuit data is taken as the loss characteristic value of each branch.
In other embodiments of the present invention, the practitioner may obtain the loss feature value through other statistical features such as mean value, or may use regression analysis to perform a fitting model to obtain the loss feature value; it should be noted that, the method for obtaining the extremely poor and average values and the regression analysis method are all technical means well known to those skilled in the art, and are not described herein.
After the loss characteristic value is obtained, the total potential loss degree can be adjusted, and the variance is taken into consideration to measure the data dispersion degree so as to reflect the loss non-uniformity degree among the branches, so that in a preferred embodiment of the invention, the variance of the loss characteristic value is obtained as the loss non-uniformity degree of all branches; the degree of loss unevenness is multiplied by the degree of total potential loss to obtain an abnormality factor. The calculation formula of the anomaly factor comprises:
wherein,representing an anomaly factor; />Indicating the degree of total potential loss; />Representing the number of branches; />A shunt number representing the circuit; />Indicate->Loss characteristic values of the individual branches; />Representing the average of the loss characteristics of all the branches;representing the variance of the loss characteristics, i.e., the degree of loss non-uniformity, of all the branches.
In the calculation formula of the anomaly factor, the greater the total potential loss degree is, the greater the error fluctuation of the potential is, the greater the error fluctuation of the current measurement value is caused, and the greater the anomaly factor is; the loss non-uniformity degree is used as an adjustment weight, so that the abnormal factor is more in line with the actual situation of different shunt loss non-uniformity in the circuit, and the abnormal factor can more represent error fluctuation in the operation process of the circuit.
The number of the data monitoring points of each branch circuit can be adjusted by itself, a plurality of measuring points are arranged in each branch circuit, and the average value or the mode or other characteristics of the measuring points are counted and used for representing the characteristics of the branch circuit; the loss characteristic value of each measurement point can be calculated, the loss non-uniformity degree among all measurement points can be obtained, and then the total potential loss degree can be adjusted.
Step S3: acquiring real-time monitoring data of all branches as sample data according to preset monitoring time length; constructing an isolated decision tree according to the current metering values in the sample data, and obtaining the initial path length corresponding to each current metering value in the sample data; according to the anomaly factors and the proportion of the error source interference degree of each shunt in all the shunts, the initial path length is adjusted, and the corrected path length is obtained; obtaining an anomaly score value according to the corrected path length; and repeatedly constructing an isolated decision tree according to the preset times, obtaining new abnormal score values, and obtaining the final score of each current metering value in the sample data according to all the abnormal score values.
The error source interference degree and the abnormality factor obtained in the step S2 can fully represent the error fluctuation of the circuit, can adaptively adjust the abnormality detection rule, remove the interference of circuit error fluctuation data, further accurately evaluate each sample data, obtain the final score of each sample data, and judge and screen the abnormal state of the sample data in the subsequent operation.
Considering that the isolated forest algorithm is a more classical implementation mode of an isolated decision tree, the method for acquiring the initial path length by utilizing the isolated forest algorithm and acquiring the abnormal score of the monitoring data by utilizing the path length of the isolated decision tree has strong self-adaptability, can effectively adapt to different data characteristics, has high detection speed on multidimensional data, allows the abnormality of a plurality of characteristics to be comprehensively considered, and can quickly identify abnormal values, so in the embodiment of the invention, the evaluation basis of sample data is acquired by constructing the isolated decision tree and adjusting the path length.
Adjusting the path length first requires obtaining an initial path length for the sample data: and randomly selecting a value between the maximum value and the minimum value of all the sample data as a target value, dividing all the sample data into two parts by using the target value, and then repeatedly dividing, wherein each dividing process is an isolated decision tree until the current metering values in all the sample data are isolated, and the dividing times are the initial path lengths of the current metering values in the sample data.
After the initial path length of each current measurement value in the sample data is obtained, the initial path length can be corrected by utilizing two error fluctuation parameters, namely an abnormal factor and an error source interference degree, so that the corrected path length after error interference is eliminated is obtained, the final score is convenient to obtain subsequently, and a basis is provided for sample data screening.
Preferably, in one embodiment of the present invention, the initial path length is corrected using a path length correction formula; the path length correction formula includes:
wherein the method comprises the steps ofA serial number indicating the current measurement value; />A serial number representing a shunt; />Representing the total number of branches; />Indicate->Initial path lengths of the individual current metric values; />Indicate->The serial numbers of the branches where the current metering values are located; />Indicate->Error source interference degrees corresponding to the branches; />Representing the summation of the error source interference degrees of all the branches; />Representing an anomaly factor; />Indicate->Corrected path lengths for the individual current measurements.
In the path length correction formula, the larger the abnormality factor is, the larger the error fluctuation of the current measurement value is, the larger the error of the initial path length of the current measurement value in the sample data is, the factor of the abnormality factor is required to be used as weight for correction, the accuracy of the corrected path length is improved, and the accuracy of the monitoring method is further improved; the greater the proportion of the error source interference level of the branch in all branches, the greater the interference level of the circuit branch in all branches, the greater the error of the initial path length, and the greater the weight needs to be given relative to other branches.
After the corrected path length is obtained, an abnormal score value of the current measurement value in the sample data can be obtained, and the greater the abnormal score value is, the greater the abnormal possibility of the sample data is; meanwhile, in order to avoid the accidental of the data, an isolated decision tree is repeatedly constructed on the sample data according to preset times, a plurality of abnormal score values of the electric power metering value are obtained, and further more accurate and convincing final scores are obtained.
Preferably, in one embodiment of the present invention, the dividing data is performed randomly each time in consideration of the algorithm characteristics of the isolated forest algorithm, so that the average value of all abnormal score values of each current measurement value in the sample data is taken as the respective corresponding final score value.
It should be noted that, in one embodiment of the present invention, the preset monitoring time is 1 minute, the acquisition frequency of the circuit data is also set to 1 second for 1 time, the isolated decision tree is repeatedly constructed for 10 times, the average value of the 10 times of scores is used as the final score of the sample data, and the value interval of the abnormal score value is set to 0-1; in other embodiments of the present invention, the implementer may adjust the repetition number by himself; clustering operation can be carried out on multiple scores, and the mean value or the mode in the largest cluster is selected as the final score; the mode in the score may also be selected as the final score. The acquisition of outlier values based on path length is a well known technique to those skilled in the art and is not described in detail herein.
Step S4: and screening the sample data according to the final score to obtain abnormal circuit data.
After the processing in the step S3, the abnormal score of the real-time circuit data in the circuit operation can be obtained, an accurate and reliable basis is provided for screening the abnormal data, and the greater the final score is, the greater the possibility of abnormality is, so that the circuit data with the final score of the current measurement value in the sample data larger than the preset abnormal threshold value is screened out, and the abnormal circuit data is obtained.
In one embodiment of the present invention, the preset anomaly threshold value is 0.7. In other embodiments of the present invention, the practitioner may adjust the preset anomaly threshold value by himself.
After obtaining the abnormal circuit data, the abnormal data can be monitored in a key way, for example, the abnormal circuit data is matched with a preset abnormal data model, the abnormal fault type is determined rapidly, and corresponding plan countermeasures are started.
In summary, the present invention aims at the technical problem that abnormal current metering data cannot be accurately identified by using the method of performing abnormality monitoring by using the conventional isolated tree algorithm, firstly, obtaining the error source interference degree and the abnormality factor by using circuit data in a preset reference period, further correcting the initial path length of each current metering value in sample data obtained by an isolated decision tree according to the error source interference degree and the abnormality factor, and repeating for a plurality of times to obtain the final score of each current metering value; and finally, screening the sample data by utilizing the final score to obtain abnormal circuit data. The invention fully considers the fluctuation of voltage and current in the running process of the circuit, the difference between different branches and the error influence of influencing factors on the current metering value, and the decision judgment process is adaptively adjusted in real time, so that the final detection result is more accurate, and the accuracy of the monitoring method is improved.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (5)
1. A method for monitoring metering anomalies of direct current shunt current, the method comprising:
collecting circuit data of each shunt circuit at a preset collection frequency within a preset reference period; the circuit data comprises influence factors which influence the preset category of the circuit;
analyzing the fluctuation similarity degree of the current measurement value and the total circuit voltage in the circuit data of each shunt circuit to obtain the error source interference degree of each shunt circuit; analyzing the difference between the fluctuation of the voltage difference in the circuit data of each shunt and the fluctuation of the influence factor change curve in the circuit data to obtain the total potential loss degree of all the shunts; obtaining a loss characteristic value of each shunt according to the statistical characteristics of the voltage and current measurement values of each shunt in the circuit data; analyzing fluctuation characteristics of the loss characteristic values of all branches, and weighting by utilizing the total potential loss degree to obtain abnormal factors of all branches;
acquiring real-time monitoring data of all branches as sample data according to preset monitoring time length; constructing an isolated decision tree according to the current metering values in the sample data, and obtaining the initial path length corresponding to each current metering value in the sample data; according to the anomaly factor and the proportion of the error source interference degree of each shunt in all the shunts, the initial path length is adjusted, and the corrected path length is obtained; obtaining an anomaly score value according to the corrected path length; repeatedly constructing an isolated decision tree according to preset times, obtaining new abnormal score values, and obtaining the final score of each current metering value in the sample data according to all the abnormal score values;
screening the sample data according to the final score to obtain abnormal circuit data;
the method for acquiring the error source interference degree comprises the following steps:
acquiring a voltage coefficient of fluctuation of an input total voltage value in the circuit data in each shunt and a current coefficient of fluctuation of a current metering value of each shunt; the ratio of the current coefficient to the voltage coefficient of each shunt is subjected to logarithmic transformation, and an absolute value is taken, so that the corresponding response degree of each shunt is obtained;
acquiring a correlation coefficient between an input total voltage value and each shunt current measurement value in the circuit data; multiplying the response degree of each branch by the correlation coefficient, and taking the product as the error source interference degree of each branch;
the method for acquiring the total potential loss degree comprises the following steps:
acquiring the total power consumption loss degree according to a total potential loss degree calculation formula; the total potential loss degree calculation formula includes:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicating the degree of total potential loss; />Representing the acquisition sequence number of circuit data in a preset reference period; />Representing the total number of circuit data acquisition times in a preset reference period; />Indicate->Voltage difference values of input and output ends in the data of the acquisition circuits; />Representing the average value of the voltage difference values of the input end and the output end in all the acquired circuit data; />A serial number indicating the type of the influencing factor in the circuit data; />The number of influencing factor types in the circuit data is represented; />Indicate->The first part of the data of the acquisition circuit>An influence value of a seed influence factor; />Representing the +.>The average value of the influence values of the seed influence factors;
the loss characteristic value acquisition method comprises the following steps:
taking the ratio of the voltage range and the current metering value range of each branch in the circuit data as the loss characteristic value of each branch;
the method for acquiring the abnormal factors comprises the following steps:
acquiring the variance of the loss characteristic value as the loss non-uniformity degree of all branches; multiplying the loss unevenness degree by the total potential loss degree to obtain an abnormality factor;
the method for acquiring the corrected path length comprises the following steps:
correcting the initial path length by using a path length correction formula; the path length correction formula includes:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein->A serial number indicating the current measurement value; />A serial number representing a shunt; />Representing the total number of branches; />Indicate->Initial path lengths of the individual current metric values; />Indicate->The serial numbers of the branches where the current metering values are located; />Indicate->Error source interference degrees corresponding to the branches; />Representing the summation of the error source interference degrees of all the branches; />Representing an anomaly factor; />Indicate->Corrected path lengths for the individual current measurements.
2. The method for monitoring metering abnormality of direct current shunt current according to claim 1, wherein the method for obtaining the final score value comprises:
and taking the average value of all the abnormal score values of each current measurement value in the sample data as the corresponding final score.
3. The method of claim 1, wherein the step of screening the sample data based on the final score comprises:
and screening out circuit data with the final score of the current metering value in the sample data being greater than a preset abnormal threshold value to obtain abnormal circuit data.
4. A method of monitoring dc shunt current metering anomalies as described in claim 3, wherein said preset anomaly threshold is 0.7.
5. The method for monitoring abnormal measurement of direct current shunt current according to claim 1, wherein the preset reference period is 10 minutes, the preset monitoring time period is 1 minute, and the preset collection frequency is 1 second and 1 time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311567436.4A CN117272033B (en) | 2023-11-23 | 2023-11-23 | DC shunt current metering abnormity monitoring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311567436.4A CN117272033B (en) | 2023-11-23 | 2023-11-23 | DC shunt current metering abnormity monitoring method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117272033A CN117272033A (en) | 2023-12-22 |
| CN117272033B true CN117272033B (en) | 2024-03-01 |
Family
ID=89220057
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311567436.4A Active CN117272033B (en) | 2023-11-23 | 2023-11-23 | DC shunt current metering abnormity monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117272033B (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113673580A (en) * | 2021-07-28 | 2021-11-19 | 贵州电网有限责任公司 | Electricity stealing suspicion judgment method based on big data modeling |
| CN113674106A (en) * | 2021-08-11 | 2021-11-19 | 国网山东省电力公司平阴县供电公司 | Combined positioning method for ground fault of medium-low voltage distribution network |
| CN113704241A (en) * | 2021-10-27 | 2021-11-26 | 国网浙江省电力有限公司信息通信分公司 | Low-business-dependence intelligent energy data auditing method |
| CN114325555A (en) * | 2022-01-04 | 2022-04-12 | 国网上海市电力公司 | Metering equipment abnormity on-line monitoring model and error calculation method |
| WO2022088265A1 (en) * | 2020-10-29 | 2022-05-05 | 株洲中车时代电气股份有限公司 | Method and device for monitoring abnormality of direct-current circuit of converter |
| WO2022252505A1 (en) * | 2021-06-02 | 2022-12-08 | 杭州安脉盛智能技术有限公司 | Device state monitoring method based on multi-index cluster analysis |
| CN116148753A (en) * | 2023-04-18 | 2023-05-23 | 北京京仪北方仪器仪表有限公司 | Intelligent electric energy meter operation error monitoring system |
| CN116304962A (en) * | 2023-05-25 | 2023-06-23 | 湖南东润智能仪表有限公司 | Intelligent anomaly monitoring method for water meter metering data |
| CN116578870A (en) * | 2023-05-19 | 2023-08-11 | 国网福建省电力有限公司罗源县供电公司 | Distribution network voltage abnormal data filling method based on fluctuation cross-correlation analysis |
| CN116718890A (en) * | 2023-05-10 | 2023-09-08 | 沈阳工程学院 | Method and device for detecting circuit faults and computer readable storage medium |
| CN116794385A (en) * | 2023-08-21 | 2023-09-22 | 山东德源电力科技股份有限公司 | High-voltage current monitoring method based on multidimensional data analysis |
| CN116937575A (en) * | 2023-09-14 | 2023-10-24 | 济南高品伟业信息科技有限公司 | Energy monitoring management system for grid system |
| CN116955895A (en) * | 2023-09-21 | 2023-10-27 | 山东博诚电气有限公司 | Transformer operation abnormity monitoring method and system |
| CN116990621A (en) * | 2023-09-26 | 2023-11-03 | 国网吉林省电力有限公司通化供电公司 | Dynamic fault monitoring system for electric power safety tools |
-
2023
- 2023-11-23 CN CN202311567436.4A patent/CN117272033B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022088265A1 (en) * | 2020-10-29 | 2022-05-05 | 株洲中车时代电气股份有限公司 | Method and device for monitoring abnormality of direct-current circuit of converter |
| WO2022252505A1 (en) * | 2021-06-02 | 2022-12-08 | 杭州安脉盛智能技术有限公司 | Device state monitoring method based on multi-index cluster analysis |
| CN113673580A (en) * | 2021-07-28 | 2021-11-19 | 贵州电网有限责任公司 | Electricity stealing suspicion judgment method based on big data modeling |
| CN113674106A (en) * | 2021-08-11 | 2021-11-19 | 国网山东省电力公司平阴县供电公司 | Combined positioning method for ground fault of medium-low voltage distribution network |
| CN113704241A (en) * | 2021-10-27 | 2021-11-26 | 国网浙江省电力有限公司信息通信分公司 | Low-business-dependence intelligent energy data auditing method |
| CN114325555A (en) * | 2022-01-04 | 2022-04-12 | 国网上海市电力公司 | Metering equipment abnormity on-line monitoring model and error calculation method |
| CN116148753A (en) * | 2023-04-18 | 2023-05-23 | 北京京仪北方仪器仪表有限公司 | Intelligent electric energy meter operation error monitoring system |
| CN116718890A (en) * | 2023-05-10 | 2023-09-08 | 沈阳工程学院 | Method and device for detecting circuit faults and computer readable storage medium |
| CN116578870A (en) * | 2023-05-19 | 2023-08-11 | 国网福建省电力有限公司罗源县供电公司 | Distribution network voltage abnormal data filling method based on fluctuation cross-correlation analysis |
| CN116304962A (en) * | 2023-05-25 | 2023-06-23 | 湖南东润智能仪表有限公司 | Intelligent anomaly monitoring method for water meter metering data |
| CN116794385A (en) * | 2023-08-21 | 2023-09-22 | 山东德源电力科技股份有限公司 | High-voltage current monitoring method based on multidimensional data analysis |
| CN116937575A (en) * | 2023-09-14 | 2023-10-24 | 济南高品伟业信息科技有限公司 | Energy monitoring management system for grid system |
| CN116955895A (en) * | 2023-09-21 | 2023-10-27 | 山东博诚电气有限公司 | Transformer operation abnormity monitoring method and system |
| CN116990621A (en) * | 2023-09-26 | 2023-11-03 | 国网吉林省电力有限公司通化供电公司 | Dynamic fault monitoring system for electric power safety tools |
Non-Patent Citations (3)
| Title |
|---|
| Decision tree based weld defect classification using current and voltage signatures in GMAW process;A Sumesh et al.;《Materials Today: Proceedings》;8354–8363 * |
| 李国成.类别不平衡的用电数据辨识方法研究.《中国优秀硕士学位论文全文数据库》.2023,1-62. * |
| 计及用户特征与信用等级的线损分析与评估;胡晓峰;《中国优秀硕士学位论文全文数据库》;1-78 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117272033A (en) | 2023-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111474510B (en) | Error evaluation method and system for voltage transformer with non-stable output | |
| CN116879662B (en) | Transformer fault detection method based on data analysis | |
| US6192317B1 (en) | Statistical pattern analysis methods of partial discharge measurements in high voltage insulation | |
| He et al. | Multivariate measurement system analysis in multisite testing: An online technique using principal component analysis | |
| CN110081923B (en) | Fault detection method and device for automatic acquisition system of field baseline environmental parameters | |
| CN114528721B (en) | Cable intermediate joint insulation state assessment method and system | |
| CN112651119A (en) | Multi-performance-parameter accelerated degradation test evaluation method for space harmonic reducer | |
| CN111007390B (en) | Analog circuit fault diagnosis model based on algebraic method | |
| CN117272033B (en) | DC shunt current metering abnormity monitoring method | |
| CN114252748A (en) | Cable intermediate joint insulation state assessment method, system and device | |
| CN113189513A (en) | Ripple-based redundant power supply current sharing state identification method | |
| CN117630800A (en) | Fault diagnosis method and system for automatic calibrating device of electric energy meter | |
| CN115728672A (en) | Circuit system health condition prediction method based on low-frequency noise and deep learning | |
| CN109308395B (en) | Wafer-level space measurement parameter anomaly identification method based on LOF-KNN algorithm | |
| CN116646281A (en) | Abnormal test structure acquisition method, abnormal test structure verification method and related devices | |
| CN114002574A (en) | Method for testing semiconductor structure | |
| CN116087759B (en) | Method for inspecting conductive path of circuit board and circuit system | |
| CN108318213B (en) | A kind of wind-tunnel measurements diagnosis method for system fault based on many kinds of parameters comprehensive assessment | |
| CN117851960B (en) | Temperature test data management method based on electronic cigarette atomizer | |
| CN118091300B (en) | Patch resistor fault diagnosis method based on data analysis | |
| CN110032758A (en) | Calculate the method, apparatus and computer storage medium of the energy of electric signal | |
| CN110955951B (en) | Product life prediction method and device based on path classification and estimation | |
| Lin et al. | Fault diagnosis of rolling bearings using multifractal detrended fluctuation analysis and Mahalanobis distance criterion | |
| US20230039158A1 (en) | System for calibration management and method of managing calibration | |
| CN118316202A (en) | Real-time inspection device for power system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |