CN118152829A - Health state assessment method and system for oil type iron core reactor - Google Patents
Health state assessment method and system for oil type iron core reactor Download PDFInfo
- Publication number
- CN118152829A CN118152829A CN202410573320.XA CN202410573320A CN118152829A CN 118152829 A CN118152829 A CN 118152829A CN 202410573320 A CN202410573320 A CN 202410573320A CN 118152829 A CN118152829 A CN 118152829A
- Authority
- CN
- China
- Prior art keywords
- current
- trend
- value
- values
- iron core
- 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.)
- Granted
Links
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 title claims abstract description 70
- 230000036541 health Effects 0.000 title claims abstract description 60
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000011156 evaluation Methods 0.000 claims abstract description 45
- 230000002159 abnormal effect Effects 0.000 claims abstract description 25
- 230000009466 transformation Effects 0.000 claims abstract description 19
- 230000006870 function Effects 0.000 claims description 16
- 230000003862 health status Effects 0.000 claims description 12
- 230000005856 abnormality Effects 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 8
- 230000001105 regulatory effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000014509 gene expression Effects 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Landscapes
- Housings And Mounting Of Transformers (AREA)
Abstract
The invention relates to the technical field of measuring electric variables, and provides a health state evaluation method and a system of an oil type iron core reactor, wherein the method comprises the following steps: collecting a current signal of an oil type iron core reactor, and obtaining a current time sequence; dividing a distance group, acquiring current transformation tendencies, acquiring current fitting orders, determining current trend values according to the current fitting orders, acquiring first approaching current signals of current signals, further acquiring current trend deviation degrees, and clustering the current signals according to the current trend deviation degrees to acquire cluster clusters; and acquiring a current trend distribution coefficient, acquiring residual errors and trend values of the cluster, further acquiring a health state evaluation coefficient, determining an abnormal judgment threshold value, and realizing the health state evaluation of the oil type iron core reactor according to the numerical value size relation between the health state evaluation coefficient and the abnormal judgment threshold value. The invention solves the problem that the health state evaluation of the oil type iron core reactor is inaccurate due to the fact that only current value differences are considered in the health state evaluation.
Description
Technical Field
The invention relates to the technical field of measuring electric variables, in particular to a health state evaluation method and system of an oil type iron core reactor.
Background
An oil-type iron core reactor, also called an oil-immersed iron core reactor, is a reactive power compensation device for an electric power system. The oil type iron core reactor improves the heat dissipation effect and the insulation performance by immersing the coil and the iron core in insulating oil, can be used for occasions such as power substations, industrial power systems, high-voltage transmission lines and the like, and is used for limiting current, compensating reactive power or improving the stability of a power grid. When the health state of the oil type iron core reactor is problematic, the stability of the power system is damaged, the power supply reliability is affected, and fire disasters are caused when the oil type iron core reactor is serious. Therefore, it is necessary to accurately evaluate the health state of the oil core reactor being used and replace the failed oil core reactor in time.
The current signals of the oil type iron core reactor are clustered by using a clustering algorithm generally, and the health state evaluation of the oil type iron core reactor is realized according to the current signal differences of different clustering clusters in the clustering result. However, when the traditional clustering algorithm divides the current signals of the oil-type iron core reactor, only the numerical difference of different current signals is considered, the trend and the periodic characteristic of the current signals over time are ignored, and the health state evaluation of the oil-type iron core reactor is often inaccurate.
Disclosure of Invention
The invention provides a health state evaluation method and a system of an oil type iron core reactor, which aim to solve the problem that the health state evaluation of the oil type iron core reactor is inaccurate due to the fact that only current value differences are considered in the health state evaluation, and the adopted technical scheme is as follows:
In a first aspect, an embodiment of the present invention provides a method for evaluating a health state of an oil-type iron core reactor, the method including the steps of:
Collecting a current signal of an oil type iron core reactor, and obtaining a current time sequence;
Obtaining a fitting curve of a current time sequence, determining an average extreme distance of adjacent extreme values, dividing a distance group according to the average extreme distance, determining current deflection characteristic values of the adjacent extreme values, obtaining standard deviation values of the distance group, obtaining current transformation tendencies of the current time sequence according to the numerical distribution of the current deflection characteristic values of the adjacent extreme values in the distance group and the standard deviation values of the distance group, obtaining current fitting orders of the current time sequence according to the current transformation tendencies of the current time sequence, fitting current signals in the current time sequence according to the current fitting orders, obtaining a current trend curve, determining current trend values of the current signals, obtaining first approaching current signals of the current signals according to different current signals and differences between the current trend values of the current signals, obtaining current trend deviation degrees of the current signals according to the current signals, the current trend values corresponding to the current signals and the first approaching current signals of the current signals, and clustering clusters according to the current trend deviation degrees;
Defining serial numbers of clusters, acquiring current trend distribution coefficients of the clusters, decomposing the current trend distribution coefficients of the clusters, acquiring residual errors and trend values of the clusters, acquiring health state evaluation coefficients of a current time sequence according to the residual errors, the trend values and the current trend distribution coefficients of all the clusters, determining an abnormal judgment threshold value, and realizing health state evaluation of the oil-type iron core reactor according to the numerical value magnitude relation between the health state evaluation coefficients and the abnormal judgment threshold value.
Further, the determining the average extreme distance of the adjacent extreme values, and dividing the distance group according to the average extreme distance comprises the following specific methods:
Acquiring all extreme values of a fitted curve of the current time sequence in a time period corresponding to the current time sequence, and recording the ratio of the range of two adjacent extreme values to the acquisition time interval of the two adjacent extreme values as the average extreme distance of the two adjacent extreme values;
Adjacent extremum values with the same average extreme distance in the same current time sequence are divided into the same distance group.
Further, the specific method for obtaining the current transformation trend of the current time sequence comprises the following steps:
the sum of the absolute deviation average value of the current deviation characteristic values of all the adjacent extreme values in the distance group and the standard deviation value of the distance group is recorded as the tendency of the distance group;
The sum of the tendencies of all distance groups corresponding to the current time series is referred to as the current transformation tendencies of the current time series.
Further, the specific method for obtaining the current fitting order is as follows:
Taking the product of the exponential function value taking the current transformation trend of the current time sequence as an independent variable and the first order regulating parameter as a first product;
the upward rounding value of the difference value of the first order regulating parameter and the first product is recorded as a first rounding value;
And recording the difference value between the first rounding value and the second order regulating parameter as the current fitting order of the current time sequence.
Further, the fitting is performed on the current signal in the current time sequence according to the current fitting order, a current trend curve is obtained, and a current trend value of the current signal is determined, including the following specific methods:
Using the current fitting order of the current time sequence as the order of a polynomial, performing polynomial fitting on the current signals contained in the current time sequence, obtaining a fitting curve corresponding to the current time sequence, and marking the fitting curve as a current trend curve;
and recording a fitting value corresponding to the acquisition time corresponding to the current signal in the current trend curve as a current trend value of the current signal.
Further, according to the difference between the different current signals and the current trend values of the current signals, a first approach current signal of the current signals is obtained, and according to the current signals, the current trend values corresponding to the current signals and the first approach current signals of the current signals, a current trend deviation degree of the current signals is obtained, which comprises the following specific methods:
the two current signals with the difference between the current trend values smaller than the second parameter and the difference between the current signal values smaller than the second parameter are recorded as a first approaching current signal of the other current signal;
the absolute value of the difference value of the current trend value corresponding to the current signal is recorded as a first difference value of the current signal;
Recording a ratio of the number of first proximity current signals of the current signals to the number of times the value of the current signals appears in the current time sequence as a first ratio of the current signals;
Recording a function value of an exponential function having an inverse number of the number of first approximation current signals of the current signals as an argument as a first function value of the current signals;
And recording the product of the first difference value, the first ratio and the first function value of the current signal as the deviation degree of the current trend of the current signal.
Further, the clustering is carried out on the current signals according to the deviation degree of the current trend to obtain clusters, and the specific method comprises the following steps:
And clustering all current signal uses by taking the absolute value of the difference value of the normalized value of the current trend deviation degree between the current signals as the distance between the current signals to obtain a cluster.
Further, according to the residual errors, the trend values and the current trend distribution coefficients of all the clusters, the health state evaluation coefficients of the current time sequence are obtained, and the corresponding expressions are as follows:
In the method, in the process of the invention, A health state assessment coefficient representing a current time series; /(I)An exponential function based on a natural constant; m represents the number of clusters; /(I)Residual errors of clusters with the sequence number of I are represented; /(I)A trend value representing a cluster with a sequence number I; /(I)And the current trend distribution coefficient of the cluster with the sequence number of I is represented.
Further, the determining the abnormal judgment threshold value, according to the numerical relation between the health state evaluation coefficient and the abnormal judgment threshold value, realizes the health state evaluation of the oil type iron core reactor, and comprises the following specific steps:
When the oil type iron core reactor works normally, manually selecting a preset number of current time sequences, acquiring health state evaluation coefficients of the current time sequences, and taking the sum of the mean value and standard deviation of the health state evaluation coefficients of the manually selected current time sequences as an abnormality judgment threshold;
When the health state evaluation coefficient of the current time sequence is larger than or equal to the abnormality judgment threshold value, the oil type iron core reactor is considered to belong to the health state in the time period corresponding to the current time sequence;
When the health state evaluation coefficient of the current time series is smaller than the abnormality judgment threshold, the oil type iron core electric reactor is considered to have a problem in a time period corresponding to the current time series.
In a second aspect, an embodiment of the present invention further provides a health status assessment system of an oil-type iron core reactor, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements the steps of any one of the methods described above when executing the computer program.
The beneficial effects of the invention are as follows:
According to the application, analysis is carried out from the characteristic that the current signal which is suddenly increased and decreased within the rated current limit value range does not trigger overload alarm, but the current signal which is suddenly increased and decreased still shows the potential unstable condition inside the oil type iron core reactor, the current fitting order of the current time sequence is obtained according to the obvious degree of the current signal fluctuation of the current time sequence and the complexity of the current signal of the oil type iron core reactor, the current trend curve is determined, namely the order of the current trend curve is adaptively determined, so that the fitting model corresponding to the obvious current time sequence of fluctuation is more accurate, the characteristic of the current identification time sequence can be more accurately captured and described, meanwhile, the order of a smaller polynomial is selected for the current time sequence with smaller complexity of the current signal, and the calculation complexity is reduced; then, when the operation state of the oil type iron core reactor is abnormal, analyzing the characteristic that the current trend value and the current signal at the same acquisition time are large in difference, the acquisition time corresponding to the large difference is large, firstly judging the possibility that the current signal is an abnormal signal, acquiring the current trend deviation degree of the current signal, dividing the current signal with consistent current signal fluctuation trend into the same cluster, avoiding the problem that a few unavoidable noises in the current signal are misjudged as abnormal current signals caused by abnormal operation of the oil type iron core reactor, improving the recognition precision of the abnormal current signal, further evaluating the operation state of the oil type iron core reactor represented by a current time sequence according to the characteristic that the current signal has certain stability and the current trend distribution coefficient of the cluster with different continuity is small, determining an abnormal judgment threshold according to the current signal of the oil type iron core reactor under the normal operation state, and realizing the health state evaluation of the oil type iron core reactor according to the numerical value relation between the health state evaluation coefficient and the abnormal judgment threshold, and only solving the problem that the health state of the oil type iron core reactor is not evaluated accurately due to the fact that the health state of the oil type iron core reactor has the current trend distribution coefficient is not considered.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a flow chart of a method for evaluating a health status of an oil-type iron core reactor according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the difference in current variation characteristics;
fig. 3 is a flow chart of current transformation trend acquisition.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of a method for evaluating a health status of an oil-type iron core reactor according to an embodiment of the invention is shown, the method includes the following steps:
and S001, collecting a current signal of the oil type iron core reactor, and obtaining a current time sequence.
A current sensor is arranged in a circuit of the oil type iron core reactor, and a current signal of the oil type iron core reactor is acquired by using the current sensor. The time interval for collecting the current signal in this embodiment is 0.1 seconds, and the total time length of each sampling is 1 minute.
And arranging all current signals in the same sampling total duration according to the sequence of acquisition to obtain a current time sequence.
In order to avoid the influence of missing data in the current time sequence on subsequent analysis, taking the continuity and the local correlation of the data into consideration, the missing data is supplemented to the current time sequence by adopting a mean value interpolation method. The window length of the mean value interpolation method is set to be 3, and the supplementation of missing data to the time sequence by using the mean value interpolation method is a known technology and will not be described again.
Thus, a current time series is obtained.
Step S002, dividing the distance group, determining current deflection characteristic values of adjacent extreme values, obtaining standard deviation values of the distance group, further obtaining current transformation tendencies of the current time sequence, obtaining current fitting orders of the current time sequence, fitting current signals in the current time sequence according to the current fitting orders, obtaining a current trend curve, determining current trend values of the current signals, obtaining a first near current signal of the current signals, further obtaining current trend deviation degree of the current signals, clustering the current signals according to the current trend deviation degree, and obtaining a cluster.
In general, there is a rated current limit corresponding to the oil core reactor, and when the current signal of the oil core reactor is within the rated current limit, the oil core reactor is considered to be in a normal operation state, and an overload alarm is not triggered.
In a short time, if a current signal that increases or decreases sharply occurs, the current signal that increases or decreases sharply shows a potential unstable condition inside the oil core reactor even if the current signal does not trigger an overload alarm within the rated current limit, and the oil core reactor may be damaged.
And using the current signal as a dependent variable and the acquisition time of the current signal as an independent variable, performing polynomial fitting on the current signal contained in the current time sequence by using a least square method, wherein the order value of the polynomial is a first parameter, and obtaining a fitting curve corresponding to the current time sequence. In this embodiment, the value of the first parameter is 7, and the polynomial fitting is performed on the data by using the least square method is a known technique, which is not described in detail.
And obtaining all maximum values and minimum values of the fitting curve in a time period corresponding to the current time sequence.
And (3) recording the ratio of the range of the two adjacent extreme values to the acquisition time interval of the two adjacent extreme values as the average extreme distance of the two adjacent extreme values.
When the range and the acquisition time interval of different adjacent extremum values within the same current time sequence are the same, the average extreme distance of the different adjacent extremum values is the same, but the characteristics of the current variation between the different adjacent extremum values may be different. As shown in the current change characteristic difference schematic diagram of FIG. 2, A and B are two adjacent extreme points, curves 1,2, 3, 4 and 5 are all current change curves with adjacent extreme points A, B as endpoints and identical acquisition time intervals, and the current change characteristics shown by the curves 1,2, 3, 4 and 5 are different.
And taking each group of adjacent extremum as an analysis object, dividing the adjacent extremum with the same average extreme distance in the same current time sequence into the same distance group, and obtaining a plurality of distance groups.
The average value of the current signals of adjacent extreme values is recorded as the first current average value of the adjacent extreme values.
And recording the middle time of a time interval formed by the acquisition time of the current signals of two adjacent extreme values as the first middle time of the adjacent extreme value, and recording the fitting value of the fitting curve corresponding to the current time sequence at the first middle time as the second current average value of the adjacent extreme value.
And (3) recording the ratio of the second current average value to the first current average value of the adjacent extremum as the current deflection characteristic value of the adjacent extremum.
The standard deviation of the current deflection characteristic values of all adjacent extreme values in the same distance group is recorded as a first standard deviation of the distance group, the standard deviation of the current deflection characteristic values of the adjacent extreme values, the numerical value of which is positioned in the middle 50%, in the same distance group is recorded as a second standard deviation of the distance group, and the absolute value of the difference value between the first standard deviation and the second standard deviation of the distance group is recorded as the standard deviation value of the distance group.
And obtaining the current transformation trend of the current time sequence according to the numerical distribution of the current deflection characteristic values of the adjacent extreme values in each distance group and the standard deviation value of the distance group.
In the method, in the process of the invention,A current transformation trend indicative of a current time series; /(I)Representing the number of distance groups; /(I)Representing the number of adjacent extrema within the ith distance group; /(I)Current deflection characteristic values representing the j-th set of adjacent extrema in the i-th distance set; Standard deviation values representing the i-th distance group; /(I) Representing the average of the current bias eigenvalues for all adjacent extrema for the ith distance group.
A current transformation trend acquisition flow chart is shown in fig. 3.
When the current fluctuation of the current time series is more obvious, the difference of current deflection characteristic values of different adjacent extreme values is larger, the deviation value of standard deviation of a distance group is larger, the current transformation trend of the current time series is larger, at the moment, the complexity of a current signal of the oil type iron core reactor is larger, the order of a polynomial which is larger needs to be selected, so that a fitting model corresponding to the current time series is more accurate, and the characteristics of the current identification time series can be captured and described more accurately.
The complexity of the calculation can be reduced by selecting only the order of the larger polynomial for the current time series with the larger complexity of the current signal and selecting the order of the smaller polynomial for the current time series with the smaller complexity of the current signal.
And obtaining the current fitting order of the current time sequence according to the current transformation trend of the current time sequence.
In the method, in the process of the invention,A current fitting order representing a current time series; /(I)Representing a first order adjustment parameter, wherein the value of the implementation is 5; /(I)Representing a second order regulating parameter, wherein the value of the implementation is 1; /(I)An exponential function based on a natural constant; /(I)Representing an upward rounding function; /(I)The current transformation trend of the current time series is shown.
And using the current fitting order of the current time sequence as the order of a polynomial, performing polynomial fitting on the current signals contained in the current time sequence by using a least square method, obtaining a fitting curve corresponding to the current time sequence, and marking the fitting curve as a current trend curve.
And calculating fitting values of the acquisition moments corresponding to all the current signals contained in the current time sequence according to the current trend curve, and recording the fitting values as current trend values corresponding to the current signals.
Comparing the current trend values of the two current signals with the difference smaller than the second parameter, and considering the two current signals as the first approaching current signal of the other current signal when the difference of the current trend values of the two current signals is also smaller than the second parameter.
In this embodiment, the value of the second parameter is 0.01.
The difference between the current trend value and the current signal at the same collection time is the key for evaluating the running state of the oil type iron core reactor, when the running state of the oil type iron core reactor is abnormal, the difference between the current trend value and the current signal at the same collection time is larger, and the collection time corresponding to the larger difference is more.
And acquiring the deviation degree of the current trend of the current signal according to the current signal, the current trend value corresponding to the current signal and the first approaching current signal of the current signal.
In the method, in the process of the invention,A current trend deviation of a kth current signal representing a current time series; /(I)A kth current signal representing a current time sequence; /(I)A current trend value corresponding to a kth current signal representing a current time sequence; /(I)A number of first proximity current signals representing a kth current signal of the current time series; /(I)An exponential function based on a natural constant; /(I)The number of times the value of the kth current signal representing the current time series appears in the current time series.
When the difference between the current trend values corresponding to the current signals is larger, the possibility that the operating state of the oil core reactor at the acquisition time corresponding to the current signals is abnormal is larger, and at this time, the current trend deviation degree of the current signals is larger.
In order to avoid that a few unavoidable noises in the current signals are misjudged as abnormal current signals caused by abnormal operation of the oil type iron core reactor, the identification precision of the abnormal current signals is improved, the ratio of the number of first approaching current signals of the current signals to the number of times of occurrence of the numerical values of the current signals in the current time sequence is used as the basis for distinguishing whether the current signals are abnormal signals or noise signals, when the number of times of occurrence of the first approaching current signals of the current signals relative to the numerical values of the current signals in the current time sequence is larger, the possibility that the value of the current signals is real data generated in the operation process of the oil type iron core reactor is larger, and the possibility that the abnormal operation state of the oil type iron core reactor occurs in the corresponding acquisition time of the current signals is larger, and at the moment, the current trend deviation degree of the current signals is larger. Meanwhile, when the number of first approaching current signals of the current signals is smaller, the possibility that the current signals are abnormal signals is higher, and at this time, the current trend deviation degree of the current signals is higher.
And carrying out normalization processing on the current trend deviation degree of all the current signals of the same current time sequence, obtaining a normalized value of the current trend deviation degree, taking the absolute value of the difference value of the normalized values of the current trend deviation degree between the current signals as the distance between the current signals, clustering all the current signals by using a DBSCAN clustering algorithm, and obtaining a plurality of clustering clusters. The maximum radius of the community of the DBSCAN clustering algorithm is 0.05, the minimum point number is 10, normalization of data and data clustering by using the DBSCAN clustering algorithm are known techniques, and are not repeated.
So far, a cluster of current signals is obtained, and the current signal fluctuation trend of the current signals in the same cluster is consistent.
Step S003, defining serial numbers of clusters, obtaining current trend distribution coefficients of the clusters, obtaining health state evaluation coefficients of current time sequences, determining an abnormal judgment threshold value, and according to the numerical value magnitude relation between the health state evaluation coefficients and the abnormal judgment threshold value, realizing health state evaluation of the oil type iron core reactor.
Counting the number of current signals contained in each cluster, and arranging the clusters from 1 according to the sequence from small to large in the number of the current signals contained in the clusters to obtain the serial number of each cluster.
According to the difference of all current signals of the cluster, fluctuation trend weight of each current signal is obtained, and according to the fluctuation trend weight of the current signal and the difference of all current signals of the cluster, current trend distribution coefficients of the cluster are obtained.
In the method, in the process of the invention,A fluctuation trend weight of a J-th current signal of the cluster with the sequence number of I is represented; /(I)A J-th current signal representing a cluster number I; /(I)The average value of all current signals of the cluster with the sequence number of I is represented; /(I)Indicating the number of current signals contained within the cluster of sequence number I; /(I)Information entropy of current signals contained in a cluster with a sequence number of I is represented; /(I)The current signal contained in the cluster with the sequence number I is represented as extremely bad; /(I)The number of current signals with different values contained in the cluster with the sequence number I is represented; /(I)And the current trend distribution coefficient of the cluster with the sequence number of I is represented.
The greater the current trend distribution coefficient of the cluster, the greater the current signal difference contained in the cluster.
When the oil type iron core reactor works normally, the current signals of the oil type iron core reactor have certain stability and continuity, namely, the current signals fluctuate within a certain range and cannot change greatly, so that the current signal fluctuation trend of the current signals in the same cluster is consistent, and meanwhile, the current signal fluctuation trend difference of the current signals among different clusters is smaller. Therefore, when the oil type iron core reactor works normally, the current trend distribution coefficient difference of different clusters is smaller.
And arranging the current trend distribution coefficient differences of the clusters according to the sequence numbers of the clusters to obtain a current trend distribution coefficient sequence.
And decomposing the current trend distribution coefficient sequence by using an STL algorithm (Seaseal-Trend decomposition procedure based on Loess) to obtain a residual error corresponding to each current trend distribution coefficient of a trend curve of the current trend distribution coefficient sequence. And marking the residual error corresponding to the current trend distribution coefficient as the residual error of the cluster corresponding to the current trend distribution coefficient. And calculating fitting values corresponding to the current trend distribution coefficients according to the trend curves, and recording the fitting values corresponding to the current trend distribution coefficients as trend values of clustering clusters corresponding to the current trend distribution coefficients.
When the current trend distribution coefficient difference of different clusters is smaller, the numerical values in the current trend distribution coefficient sequences are closer, the trend of the current trend distribution coefficient sequences is not obvious, and the trend value of the cluster corresponding to each current trend distribution coefficient is smaller. Meanwhile, when the difference of the current trend distribution coefficients of different clusters is smaller, the residual error corresponding to the current trend distribution coefficient is smaller.
And acquiring a health state evaluation coefficient of the current time sequence according to the residual errors, the trend values and the current trend distribution coefficients of all the clusters.
In the method, in the process of the invention,A health state assessment coefficient representing a current time series; /(I)An exponential function based on a natural constant; m represents the number of clusters; /(I)Residual errors of clusters with the sequence number of I are represented; /(I)A trend value representing a cluster with a sequence number I; /(I)And the current trend distribution coefficient of the cluster with the sequence number of I is represented.
When the health state evaluation coefficient is closer to 1, the trend of the current time sequence is more obvious, and the operating state of the oil core reactor is healthier in the time period corresponding to the current time sequence.
When the oil type iron core reactor works normally, 100 current time sequences are selected manually, health state evaluation coefficients of all the current time sequences are obtained, the mean value and standard deviation of the health state evaluation coefficients of all the current time sequences are calculated, and the sum of the mean value and the standard deviation is used as an abnormality judgment threshold T.
When the health state evaluation coefficient of the current time sequence is greater than or equal to the abnormality judgment threshold T, the oil type iron core electric reactor is considered to belong to a health state in a time period corresponding to the current time sequence; when the health state evaluation coefficient of the current time sequence is smaller than the abnormality judgment threshold value T, the oil type iron core electric reactor is considered to have a problem in a time period corresponding to the current time sequence, and maintenance or replacement is required.
Thus, the health state evaluation of the oil type iron core reactor is realized.
Based on the same inventive concept as the above method, the embodiment of the invention further provides a health state evaluation system of the oil-type iron core reactor, which comprises a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to realize the steps of any one of the above methods for evaluating the health state of the oil-type iron core reactor.
The foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A method for evaluating the health status of an oil core reactor, comprising the steps of:
Collecting a current signal of an oil type iron core reactor, and obtaining a current time sequence;
Obtaining a fitting curve of a current time sequence, determining an average extreme distance of adjacent extreme values, dividing a distance group according to the average extreme distance, determining current deflection characteristic values of the adjacent extreme values, obtaining standard deviation values of the distance group, obtaining current transformation tendencies of the current time sequence according to the numerical distribution of the current deflection characteristic values of the adjacent extreme values in the distance group and the standard deviation values of the distance group, obtaining current fitting orders of the current time sequence according to the current transformation tendencies of the current time sequence, fitting current signals in the current time sequence according to the current fitting orders, obtaining a current trend curve, determining current trend values of the current signals, obtaining first approaching current signals of the current signals according to different current signals and differences between the current trend values of the current signals, obtaining current trend deviation degrees of the current signals according to the current signals, the current trend values corresponding to the current signals and the first approaching current signals of the current signals, and clustering clusters according to the current trend deviation degrees;
Defining serial numbers of clusters, acquiring current trend distribution coefficients of the clusters, decomposing the current trend distribution coefficients of the clusters, acquiring residual errors and trend values of the clusters, acquiring health state evaluation coefficients of a current time sequence according to the residual errors, the trend values and the current trend distribution coefficients of all the clusters, determining an abnormal judgment threshold value, and realizing health state evaluation of the oil-type iron core reactor according to the numerical value magnitude relation between the health state evaluation coefficients and the abnormal judgment threshold value.
2. The method for evaluating the health status of an oil core reactor according to claim 1, wherein the determining the average extreme distance of the adjacent extreme values, dividing the distance groups according to the average extreme distance, comprises the following specific steps:
Acquiring all extreme values of a fitted curve of the current time sequence in a time period corresponding to the current time sequence, and recording the ratio of the range of two adjacent extreme values to the acquisition time interval of the two adjacent extreme values as the average extreme distance of the two adjacent extreme values;
Adjacent extremum values with the same average extreme distance in the same current time sequence are divided into the same distance group.
3. The method for evaluating the health status of an oil-type iron core reactor according to claim 1, wherein the current transformation trend of the current time sequence is obtained by the following specific method:
the sum of the absolute deviation average value of the current deviation characteristic values of all the adjacent extreme values in the distance group and the standard deviation value of the distance group is recorded as the tendency of the distance group;
The sum of the tendencies of all distance groups corresponding to the current time series is referred to as the current transformation tendencies of the current time series.
4. The method for evaluating the health status of an oil-type iron core reactor according to claim 1, wherein the current fitting order is obtained by the specific method of:
Taking the product of the exponential function value taking the current transformation trend of the current time sequence as an independent variable and the first order regulating parameter as a first product;
the upward rounding value of the difference value of the first order regulating parameter and the first product is recorded as a first rounding value;
And recording the difference value between the first rounding value and the second order regulating parameter as the current fitting order of the current time sequence.
5. The method for evaluating the health status of an oil-type iron core reactor according to claim 1, wherein the fitting of the current signal in the current time sequence according to the current fitting order to obtain a current trend curve and determining the current trend value of the current signal comprises the following specific steps:
Using the current fitting order of the current time sequence as the order of a polynomial, performing polynomial fitting on the current signals contained in the current time sequence, obtaining a fitting curve corresponding to the current time sequence, and marking the fitting curve as a current trend curve;
and recording a fitting value corresponding to the acquisition time corresponding to the current signal in the current trend curve as a current trend value of the current signal.
6. The method for evaluating the health status of an oil-type iron core reactor according to claim 1, wherein the steps of obtaining a first approaching current signal of the current signal according to the difference between the different current signals and the current trend values of the current signals, and obtaining the current trend deviation of the current signal according to the current signal, the current trend value corresponding to the current signal and the first approaching current signal of the current signal, comprise the following specific steps:
the two current signals with the difference between the current trend values smaller than the second parameter and the difference between the current signal values smaller than the second parameter are recorded as a first approaching current signal of the other current signal;
the absolute value of the difference value of the current trend value corresponding to the current signal is recorded as a first difference value of the current signal;
Recording a ratio of the number of first proximity current signals of the current signals to the number of times the value of the current signals appears in the current time sequence as a first ratio of the current signals;
Recording a function value of an exponential function having an inverse number of the number of first approximation current signals of the current signals as an argument as a first function value of the current signals;
And recording the product of the first difference value, the first ratio and the first function value of the current signal as the deviation degree of the current trend of the current signal.
7. The method for evaluating the health status of an oil-type iron core reactor according to claim 1, wherein the clustering of the current signals according to the deviation of the current trend to obtain clusters comprises the following specific steps:
And clustering all current signal uses by taking the absolute value of the difference value of the normalized value of the current trend deviation degree between the current signals as the distance between the current signals to obtain a cluster.
8. The method for evaluating the health status of an oil-type iron core reactor according to claim 1, wherein the acquiring the health status evaluation coefficient of the current time series according to the residual errors, the trend values and the current trend distribution coefficients of all clusters corresponds to the following expression:
In the method, in the process of the invention, A health state assessment coefficient representing a current time series; /(I)An exponential function based on a natural constant; m represents the number of clusters; /(I)Residual errors of clusters with the sequence number of I are represented; /(I)A trend value representing a cluster with a sequence number I; /(I)And the current trend distribution coefficient of the cluster with the sequence number of I is represented.
9. The method for evaluating the health state of an oil-type iron core reactor according to claim 1, wherein the determining the abnormality judgment threshold value, according to the magnitude relation between the health state evaluation coefficient and the abnormality judgment threshold value, implements the health state evaluation of the oil-type iron core reactor, comprises the following specific steps:
When the oil type iron core reactor works normally, manually selecting a preset number of current time sequences, acquiring health state evaluation coefficients of the current time sequences, and taking the sum of the mean value and standard deviation of the health state evaluation coefficients of the manually selected current time sequences as an abnormality judgment threshold;
When the health state evaluation coefficient of the current time sequence is larger than or equal to the abnormality judgment threshold value, the oil type iron core reactor is considered to belong to the health state in the time period corresponding to the current time sequence;
When the health state evaluation coefficient of the current time series is smaller than the abnormality judgment threshold, the oil type iron core electric reactor is considered to have a problem in a time period corresponding to the current time series.
10. A health assessment system for an oil core reactor, comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1-9 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410573320.XA CN118152829B (en) | 2024-05-10 | 2024-05-10 | Health state assessment method and system for oil type iron core reactor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410573320.XA CN118152829B (en) | 2024-05-10 | 2024-05-10 | Health state assessment method and system for oil type iron core reactor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118152829A true CN118152829A (en) | 2024-06-07 |
| CN118152829B CN118152829B (en) | 2024-07-23 |
Family
ID=91301620
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410573320.XA Active CN118152829B (en) | 2024-05-10 | 2024-05-10 | Health state assessment method and system for oil type iron core reactor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118152829B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106990763A (en) * | 2017-04-20 | 2017-07-28 | 浙江大学 | A kind of Vertical Mill operation regulator control system and method based on data mining |
| CN107146004A (en) * | 2017-04-20 | 2017-09-08 | 浙江大学 | A kind of slag milling system health status identifying system and method based on data mining |
| CN112381965A (en) * | 2020-11-03 | 2021-02-19 | 浙大城市学院 | Aeroengine health state identification system and method based on data mining |
| 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 |
| CN115993247A (en) * | 2022-12-08 | 2023-04-21 | 中国矿业大学 | Drilling machine spindle bearing health state assessment method based on time sequence decomposition and order preserving regression |
| WO2023130776A1 (en) * | 2022-01-07 | 2023-07-13 | 国网浙江省电力有限公司电力科学研究院 | Method and system for predicting working condition health status of battery in energy storage power station |
| CN117074985A (en) * | 2023-09-05 | 2023-11-17 | 广汽埃安新能源汽车股份有限公司 | Electric vehicle battery health assessment method and device |
| CN117310348A (en) * | 2023-11-23 | 2023-12-29 | 东莞市时实电子有限公司 | Power adapter fault real-time monitoring method and system |
-
2024
- 2024-05-10 CN CN202410573320.XA patent/CN118152829B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106990763A (en) * | 2017-04-20 | 2017-07-28 | 浙江大学 | A kind of Vertical Mill operation regulator control system and method based on data mining |
| CN107146004A (en) * | 2017-04-20 | 2017-09-08 | 浙江大学 | A kind of slag milling system health status identifying system and method based on data mining |
| CN112381965A (en) * | 2020-11-03 | 2021-02-19 | 浙大城市学院 | Aeroengine health state identification system and method based on data mining |
| 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 |
| WO2023130776A1 (en) * | 2022-01-07 | 2023-07-13 | 国网浙江省电力有限公司电力科学研究院 | Method and system for predicting working condition health status of battery in energy storage power station |
| CN115993247A (en) * | 2022-12-08 | 2023-04-21 | 中国矿业大学 | Drilling machine spindle bearing health state assessment method based on time sequence decomposition and order preserving regression |
| CN117074985A (en) * | 2023-09-05 | 2023-11-17 | 广汽埃安新能源汽车股份有限公司 | Electric vehicle battery health assessment method and device |
| CN117310348A (en) * | 2023-11-23 | 2023-12-29 | 东莞市时实电子有限公司 | Power adapter fault real-time monitoring method and system |
Non-Patent Citations (1)
| Title |
|---|
| 代风;纪杨建;万安平;张真;: "矿渣粉磨健康状态识别模型及***设计", 振动.测试与诊断, no. 04, 15 August 2018 (2018-08-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118152829B (en) | 2024-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117093879B (en) | Intelligent operation management method and system for data center | |
| CN112284440B (en) | Sensor data deviation self-adaptive correction method | |
| CN116306320B (en) | Method and system for managing production quality of non-vascular lumen guide wire | |
| CN117235557B (en) | Electrical equipment fault rapid diagnosis method based on big data analysis | |
| CN117668684A (en) | Power grid electric energy data anomaly detection method based on big data analysis | |
| CN116739829B (en) | Big data-based power data analysis method, system and medium | |
| CN113657622B (en) | Multi-dimensional state data fusion method, device, terminal and storage medium for power equipment | |
| CN115664038B (en) | Intelligent power distribution operation and maintenance monitoring system for electrical safety management | |
| CN117748507B (en) | Distribution network harmonic access uncertainty assessment method based on Gaussian regression model | |
| Dong | Combining unsupervised and supervised learning for asset class failure prediction in power systems | |
| CN117150244B (en) | Intelligent power distribution cabinet state monitoring method and system based on electrical parameter analysis | |
| CN110751217A (en) | Equipment energy consumption ratio early warning analysis method based on principal component analysis | |
| CN112416661B (en) | Multi-index time sequence anomaly detection method and device based on compressed sensing | |
| CN118152829B (en) | Health state assessment method and system for oil type iron core reactor | |
| CN117421616A (en) | Mine shaft detection system and method | |
| CN117574275A (en) | Abnormality detection and classification method for power system | |
| CN113672658B (en) | Power equipment online monitoring error data identification method based on complex correlation coefficient | |
| CN116125260A (en) | Breaker electromechanical fault edge diagnosis method based on multi-element data fusion | |
| CN114595767A (en) | Industrial equipment fault detection method with time series state variable | |
| CN116956089A (en) | Training method and detection method for temperature anomaly detection model of electrical equipment | |
| CN112763678A (en) | PCA-based sewage treatment process monitoring method and system | |
| CN118091234B (en) | Current transformer for fault diagnosis processing | |
| CN118244071A (en) | Method and system for detecting partial discharge of dry type iron core reactor | |
| CN117633695B (en) | Transformer operation monitoring method based on electrical parameter time sequence analysis | |
| CN118100446B (en) | Intelligent detection method for node running state of 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 |