CN115841278B - Method, system, equipment and medium for evaluating running error state of electric energy metering device - Google Patents
Method, system, equipment and medium for evaluating running error state of electric energy metering device Download PDFInfo
- Publication number
- CN115841278B CN115841278B CN202310000732.XA CN202310000732A CN115841278B CN 115841278 B CN115841278 B CN 115841278B CN 202310000732 A CN202310000732 A CN 202310000732A CN 115841278 B CN115841278 B CN 115841278B
- Authority
- CN
- China
- Prior art keywords
- phase
- feature
- curve
- physical quantity
- cosine similarity
- 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 68
- 238000012544 monitoring process Methods 0.000 claims abstract description 61
- 238000003062 neural network model Methods 0.000 claims abstract description 38
- 238000012549 training Methods 0.000 claims abstract description 33
- 238000011156 evaluation Methods 0.000 claims abstract description 25
- 238000013210 evaluation model Methods 0.000 claims abstract description 18
- 238000010276 construction Methods 0.000 claims abstract description 10
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000012163 sequencing technique Methods 0.000 claims description 46
- 239000013598 vector Substances 0.000 claims description 46
- 238000004422 calculation algorithm Methods 0.000 claims description 43
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 34
- 239000010931 gold Substances 0.000 claims description 34
- 229910052737 gold Inorganic materials 0.000 claims description 34
- 230000016507 interphase Effects 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 15
- 238000010187 selection method Methods 0.000 claims description 13
- 238000012216 screening Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 238000000546 chi-square test Methods 0.000 claims description 4
- 230000008030 elimination Effects 0.000 claims description 4
- 238000003379 elimination reaction Methods 0.000 claims description 4
- 208000012661 Dyskinesia Diseases 0.000 claims description 3
- 230000000977 initiatory effect Effects 0.000 claims description 2
- 241000272184 Falconiformes Species 0.000 description 34
- 238000013528 artificial neural network Methods 0.000 description 22
- 238000001514 detection method Methods 0.000 description 6
- 210000002569 neuron Anatomy 0.000 description 6
- 230000015556 catabolic process Effects 0.000 description 3
- 238000006731 degradation reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- OLEWMKVPSUCNLG-UHFFFAOYSA-N 4-methylthioamphetamine Chemical group CSC1=CC=C(CC(C)N)C=C1 OLEWMKVPSUCNLG-UHFFFAOYSA-N 0.000 description 1
- 241000272473 Aquila chrysaetos Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 210000004205 output neuron Anatomy 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/82—Energy audits or management systems therefor
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to a method, a system, equipment and a medium for evaluating the running error state of an electric energy metering device, wherein the method comprises the following steps: collecting historical operation monitoring data of a plurality of electric energy metering devices, wherein each historical operation monitoring data comprises a plurality of different types of physical quantities generated in the operation process of the electric energy metering devices; extracting features of each collected historical operation monitoring data, wherein the feature extraction comprises two steps of feature construction and feature selection, and a final feature set is obtained; adding an error state label to the final feature set of each historical operation monitoring data to form a training sample, constructing a neural network model, and training the neural network model through the training sample to obtain an operation error state evaluation model of the electric energy metering device; and acquiring a current final feature set of the target electric energy metering device, and inputting the current final feature set into a trained electric energy metering device operation error state evaluation model to obtain an error state evaluation result of the target electric energy metering device.
Description
Technical Field
The invention relates to a method, a system, equipment and a medium for evaluating the running error state of an electric energy metering device, and belongs to the technical field of state evaluation of electric energy metering devices.
Background
The electric energy metering device is an important device for measuring electric energy, is quite widely used in all links of an electric power system, and can provide perfect basis for power supply fee accounting, electric quantity measurement and the like. The normal use of the power metering device is affected by various factors and risks, such as wiring errors, poor communication, parameter setting errors, etc., and the above or other faults or defects may cause serious problems to the normal operation of the power metering device, and ultimately affect the accuracy of the power metering device. The state evaluation of the electric energy metering device is an important means for researching the electric energy metering device, and the accurate state evaluation result can clearly reflect the running state of the electric energy metering device, so that the change from traditional period control to future intelligent control of the electric energy metering device of a company is realized.
The invention patent with the prior art publication number of CN114065605A discloses an intelligent electric energy meter running state detection and evaluation system and method, and the method comprises the following steps: s1: acquiring a plurality of error state data of the intelligent electric energy meter, and recording the error state data as history data of the intelligent electric energy meter; s2: respectively carrying out quantization processing and normalization processing on the error state data; s3: carrying out normalization evaluation weighting on the data subjected to normalization processing, carrying out state evaluation according to a preset threshold value, and marking an evaluation result as performance degradation data; s4: performing data preprocessing on the performance degradation data and the intelligent ammeter historical data to obtain training data; s5: establishing a detection evaluation model, and carrying out model training by training data to obtain an optimal detection evaluation model; s6: substituting the error state data of the intelligent electric energy meter to be detected into an optimal detection evaluation model to obtain the detection evaluation result of the running state of the intelligent electric energy meter to be detected. The prior art can be used for scientifically evaluating the running state and performance degradation failure prediction of the intelligent electric energy meter.
The evaluation indexes in the prior art, namely error state data of the electric energy meter, adopt a large amount of human intervention data comprising phenotype selection errors, basic errors, operation time errors, operation fault rate errors, operation monitoring event errors, operation monitoring abnormal errors, error dispersibility errors, full-detection return rate errors, operation quality sampling inspection errors, lead sealing state errors, installation environment errors, user reputation errors and the like, so that error state evaluation of the electric energy meter is strongly related to human intervention, and the electric energy meter is not objective and reliable.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a method, a system, equipment and a medium for evaluating the running error state of an electric energy metering device.
The technical scheme of the invention is as follows:
in one aspect, the invention provides a method for evaluating an operation error state of an electric energy metering device, comprising the following steps:
collecting historical operation monitoring data of a plurality of electric energy metering devices, wherein each historical operation monitoring data comprises a plurality of different types of physical quantities generated in the operation process of the electric energy metering devices;
extracting features of each collected historical operation monitoring data, wherein the feature extraction comprises two steps of constructing features and selecting features;
The construction characteristic steps specifically include: for each historical operation monitoring data, calculating kurtosis sequencing energy characteristics, curve complexity characteristics and inter-phase cosine similarity characteristics corresponding to each physical quantity through each kind of physical quantity;
the characteristic selection step specifically comprises the following steps: screening the feature set obtained in the step of constructing the features for any historical operation monitoring data to obtain a final feature set;
adding an error state label to the final feature set of each historical operation monitoring data to form a training sample, constructing a neural network model, and carrying out iterative training on the neural network model through the training sample to obtain a trained electric energy metering device operation error state evaluation model;
the method comprises the steps of obtaining current operation monitoring data of a target electric energy metering device, extracting features according to the current operation monitoring data to obtain a current final feature set of the target electric energy metering device, and inputting the current final feature set to a trained electric energy metering device operation error state evaluation model to obtain an error state evaluation result of the target electric energy metering device.
As a preferred embodiment, the plurality of different kinds of physical quantities generated during the operation of the electric energy metering device specifically includes:
Voltage data, current data, active power data, reactive power data, power factor data, three-phase imbalance data, and load factor data.
In a preferred embodiment, in the step of calculating the kurtosis ranking energy feature, the curve complexity feature and the inter-cosine similarity feature of each physical quantity by each kind of physical quantity, the calculation method of the kurtosis ranking energy feature of each physical quantity is as follows:
drawing corresponding physical quantity curves according to various physical quantities, wherein the corresponding physical quantity curves comprise a current curve, a voltage curve, an active power curve, a reactive power curve, a power factor curve, a three-phase unbalance curve and a load factor curve;
for each physical quantity curve, the numerical statistic of the random variable distribution characteristic of each physical quantity curve is reflected through kurtosis, and the expression of the kurtosis K is as follows:
；
wherein N represents the signal length of the physical quantity curve;representing the ith signal value in the physical quantity curve; μ represents the signal average value of the physical quantity curve; sigma represents the standard deviation of the signal of the physical quantity curve;
EMD empirical mode decomposition is performed on each signal in each physical quantity curve, kurtosis is calculated on the decomposed signals, and kurtosis sorting energy E is calculated according to the following formula after the kurtosis is sorted in descending order according to the calculated kurtosis:
；
Thereby obtaining current kurtosis sequencing energy characteristics, voltage kurtosis sequencing energy characteristics, active power kurtosis sequencing energy characteristics, reactive power kurtosis sequencing energy characteristics, power factor kurtosis sequencing energy characteristics, three-phase imbalance kurtosis sequencing energy characteristics and load rate kurtosis sequencing energy characteristics.
In a preferred embodiment, in the step of calculating the kurtosis sequencing energy feature, the curve complexity feature and the inter-cosine similarity feature of each physical quantity by each kind of physical quantity, the calculating method of the curve complexity feature of each physical quantity is as follows:
definition of fuzzy entropy is:
；
wherein m is the dimension of the phase space, r is the similarity tolerance, N is the dimension of the time series,is a fuzzy membership function;
respectively calculating complexity characteristics of the curves according to defined fuzzy entropySecond Voltage Curve complexity characteristic +.>Second active Power Curve complexity characteristic +.>Second reactive power curve complexity characteristic +.>Second Power factor Curve complexity order feature +.>Second three-phase imbalance profile complexity characteristic +.>Second load-factor curve complexity characteristic ∈>The method comprises the following steps of:
；
；
；
；
；
；
wherein,,、、respectively representing the current of A phase, B phase and C phase; / >、、The voltages of the phase A, the phase B and the phase C are respectively;、、、respectively representing A phase, B phase, C phase and total active power;、、、Respectively representing A phase, B phase, C phase and total reactive power;、、、Respectively representing A phase, B phase, C phase and total current power factors;Three-phase imbalance data;
R load factor Representing the load factor.
In a preferred embodiment, in the step of calculating the kurtosis ranking energy feature, the curve complexity feature and the inter-phase cosine similarity feature corresponding to each physical quantity by each kind of physical quantity, the calculating method of the inter-phase cosine similarity feature of each physical quantity is as follows:
according to cosine similarity formula, respectively calculating cosine similarity characteristics of A-phase and B-phase currentsCosine similarity feature of B-phase and C-phase currents>Cosine similarity feature of currents of phase A and phase C>Cosine similarity feature of voltages of phase A and phase B>Cosine similarity feature of B-phase and C-phase voltages>Cosine similarity feature of voltages of phase A and phase C>Cosine similarity feature of active power of phase A and phase B>Cosine similarity feature of active power of B phase and C phase>Cosine similarity feature of active power of phase A and phase C>Cosine similarity characteristics of reactive power of phase A and phase B>Cosine similarity characteristics of reactive power of B phase and C phase >Cosine similarity characteristics of reactive power of phase A and phase C>Cosine similarity feature of power factors of phase A and phase B>Cosine similarity feature of B-phase and C-phase power factors>Cosine similarity feature of power factors of phase A and phase C>；
The cosine similarity formula is as follows:
；
wherein, X, Y represents X, Y vector;、respectively representing the modulus of the X, Y vector.
As a preferred embodiment, the method for screening the feature set obtained in the step of constructing the feature set to obtain a final feature set specifically includes:
taking kurtosis ordering energy characteristics of each physical quantity as a first characteristic set, curve complexity characteristics of each physical quantity as a second characteristic set, inter-phase cosine similarity characteristics of each physical quantity as a third characteristic set, and combining the first characteristic set, the second characteristic set and the third characteristic set into a fourth characteristic set;
the fourth feature set is subjected to feature selection by using a first feature selection algorithm, a second feature selection algorithm and a … … nth feature selection algorithm, so that n feature subsets are obtained;
merging the n feature subsets into a fifth feature set;
performing feature selection on the fifth feature set by using an n+1 feature selection algorithm to obtain a sixth feature set, wherein the sixth feature set is used as a final feature set;
The first feature selection algorithm, the second feature selection algorithm, the … … nth feature selection algorithm and the n+1th feature selection algorithm are specifically selected from a variance selection method, a correlation coefficient method, a chi-square test method, a relief algorithm, a recursive feature elimination method, a feature selection method based on penalty terms and a feature selection method based on a tree model.
As a preferred implementation manner, the neural network model adopts a nonlinear time sequence prediction NARX neural network model, and adopts a gold hawk algorithm to optimize the neural network model, and the specific steps are as follows:
initializing the number of eagle individuals in a eagle population, wherein each eagle contains weight coefficient information and threshold parameter information of different NARX neural network models;
calculating fitness value of the eagle individuals and initializing group memory according to the fitness value; the fitness value calculation formula of the gold eagle individual is as follows:
；
wherein value represents fitness value, N is total number of gold eagle individuals,for the predicted value of NARX neural network model obtained according to the weight coefficient information and threshold parameter information of NARX neural network model contained in ith gold eagle, < +.>The sample true value corresponding to the ith gold eagle;
Initiating challenge with gold eagleAnd cruise tendency->；
Updating attack trends according to the following formulaAnd cruise tendency->：
；
Wherein,,and->Is->Initial and final values, +.>And->Is->Is a final value and an initial value of (1); t is the T iteration, and T is the total iteration number.
Randomly selecting prey from attack vectors calculated from the memory of the population:
；
wherein,,for the i-th gold eagle's attack vector, < >>For the best place to be reached by the current eagle, < > for>Is the current position of the ith gold eagle;
calculating a cruising vector d:
；
wherein,,as normal vector +.>Is a decision variable vector;
is provided withAny point on the hyperplane, then:；
handleViewed as a normal to the hyperplane, the hyperplane is expressed as:
；
wherein,,in order for the vector of the attack,in order to make a decision vector,is the selected prey location;representing an attack vector at the t-th iteration;
then:
；
wherein,,k is the number of the fixed variable and is the kth element of the target point;
the target point on the cruise hyperplane is expressed as:
the step size vector of the hawk iteration is defined as:
；
wherein,,and->Is [0,1]The random vector in the iteration is added to the position in the iteration by the step size vector in the iteration, and the position of the gold eagle in the iteration is calculated:
wherein,, The first thing is golden eagleThe secondary position is used for the control of the position,the first thing is golden eagleThe position of the secondary part is determined,the step size of the abnormal movement of the gold hawk;
calculating a fitness value according to the updated new position of the eagle individual, and updating the optimal solution and the optimal position;
judging whether the maximum iteration times are reached, and outputting weight coefficient information and threshold parameter information contained in the eagle individuals at the current optimal position as weight coefficients and threshold parameters of the NARX neural network model if the maximum iteration times are reached; and continuing iteration if the maximum iteration number is not reached.
On the other hand, the invention also provides an operation error state evaluation system of the electric energy metering device, which comprises the following components:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring historical operation monitoring data of a plurality of electric energy metering devices, and each historical operation monitoring data comprises a plurality of different physical quantities generated in the operation process of the electric energy metering devices;
the feature extraction module is used for extracting features of each collected historical operation monitoring data and comprises two steps of feature construction and feature selection;
the construction characteristic steps specifically include: for each historical operation monitoring data, calculating kurtosis sequencing energy characteristics, curve complexity characteristics and inter-phase cosine similarity characteristics corresponding to each physical quantity through each kind of physical quantity;
The characteristic selection step specifically comprises the following steps: screening the feature set obtained in the step of constructing the features for any historical operation monitoring data to obtain a final feature set;
the model training module is used for adding an error state label to the final feature set of each historical operation monitoring data to form a training sample, constructing a neural network model, and carrying out iterative training on the neural network model through the training sample to obtain a trained electric energy metering device operation error state evaluation model;
the evaluation module is used for acquiring current operation monitoring data of the target electric energy metering device, extracting features according to the current operation monitoring data to obtain a current final feature set of the target electric energy metering device, and inputting the current final feature set to the trained electric energy metering device operation error state evaluation model to obtain an error state evaluation result of the target electric energy metering device.
In still another aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the method for evaluating an operation error state of the electric energy metering device according to any embodiment of the present invention when the processor executes the program.
In still another aspect, the present invention further provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method for evaluating an operation error state of an electric energy metering device according to any of the embodiments of the present invention.
The invention has the following beneficial effects:
1. according to the method, the system, the equipment and the medium for evaluating the running error state of the electric energy metering device, the objective and reliable monitoring data are ensured by collecting a plurality of different physical quantities generated in the running process of the electric energy metering device as the original data, then the characteristic construction and selection are carried out on the monitoring data, the dimension of the characteristic data is reduced, and the selected characteristic data is used for training to obtain the running error state evaluation model of the electric energy metering device, so that the running error state of the electric energy metering device can be evaluated according to the objective monitoring data.
2. According to the method, the system, the equipment and the medium for evaluating the running error state of the electric energy metering device, kurtosis sequencing energy characteristics, curve complexity characteristics and inter-phase cosine similarity characteristics are calculated through various physical quantities and serve as characteristic data, and the distribution characteristics of collected monitoring data can be well reflected.
3. According to the method, the system, the equipment and the medium for evaluating the running error state of the electric energy metering device, the NARX neural network model is optimized through the gold eagle algorithm, the whole parameter scale of the network can be limited, the generalization performance of the NARX neural network is finally improved, and the accuracy of error state evaluation is finally improved.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the invention;
FIG. 2 is a flowchart illustrating feature selection according to an embodiment of the present invention;
fig. 3 is a diagram illustrating an example network structure of a NARX neural network model in an embodiment of the present invention.
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.
It should be understood that the step numbers used herein are for convenience of description only and are not limiting as to the order in which the steps are performed.
It is to be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The terms "comprises" and "comprising" indicate the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The term "and/or" refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Embodiment one:
referring to fig. 1, the embodiment provides a method for evaluating an operation error state of an electric energy metering device, including the following steps:
s100, collecting historical operation monitoring data of a plurality of electric energy metering devices, wherein each historical operation monitoring data comprises a plurality of different types of physical quantities generated in the operation process of the electric energy metering devices. Specifically, the electric energy metering device comprises a voltage transformer, a current transformer, a corresponding secondary circuit and an electric energy meter, and typical faults comprise faults of the electric energy meter such as voltage loss, current loss, pressure drop out and the like. The operation monitoring data of the electric energy metering device mainly comprise more than ten physical quantities such as voltage, current, split-phase power and the like.
As a preferred implementation manner of the present embodiment, the plurality of different types of physical quantities generated during the operation of the electric energy metering device collected in the present embodiment specifically includes:
voltage data, current data, active power data, reactive power data, power factor data, three-phase imbalance data, and load factor data, i.e., historical operation monitoring dataThe method comprises the following steps:
wherein,,for three-phase current data>For three-phase voltage data, ">For active power data, +.>For reactive power data, +.>For power factor data, ++>For three-phase imbalance data, +.>As load rate data, further:
；
；
；
；
；
wherein,,、、respectively representing the current of A phase, B phase and C phase;、、Respectively representing voltages of A phase, B phase and C phase;、、、Respectively representing A phase, B phase, C phase and total active power;、、、Respectively representing A phase, B phase, C phase and total reactive power;、、、Respectively representing phase A, phase B, phase C and total power factors.
S200, extracting features of each collected historical operation monitoring data, wherein the steps comprise a step S210 of constructing features and a step S220 of selecting features;
wherein, S210 specifically comprises:
for each historical operation monitoring data collected Kurtosis sequencing energy characteristics, curve complexity characteristics and phase cosine similarity characteristics corresponding to various physical quantities (namely voltage data, current data, active power data, reactive power data, power factor data, three-phase unbalance degree data and load factor data) are calculated through various physical quantities. The method comprises the following specific steps:
s211, constructing a first feature set (kurtosis sequencing energy feature): operating monitoring data based on historyDrawing a current curve, a voltage curve, an active power curve, a reactive power curve, a power factor curve, a three-phase imbalance curve and a load factor curve;
for each physical quantity curve, the kurtosis is a numerical statistic reflecting the random variable distribution characteristic of each physical quantity curve and is a normalized 4-order central moment, and the expression of the kurtosis K is as follows:
；
wherein N represents the signal length of the physical quantity curve;representing the ith signal value in the physical quantity curve; μ represents the signal average value of the physical quantity curve; sigma represents the standard deviation of the signal of the physical quantity curve;
when the electric energy metering device is under different working conditions, the kurtosis value of each IMF (intrinsic mode functions, intrinsic mode function) signal after signal decomposition has a large difference. EMD empirical mode decomposition is performed on each signal in each physical quantity curve, kurtosis is calculated on the decomposed signals, and kurtosis sorting energy E is calculated according to the following formula after the kurtosis is sorted in descending order according to the calculated kurtosis:
；
And obtaining a first characteristic set comprising current kurtosis sequencing energy characteristics, voltage kurtosis sequencing energy characteristics, active power kurtosis sequencing energy characteristics, reactive power kurtosis sequencing energy characteristics, power factor kurtosis sequencing energy characteristics, three-phase imbalance kurtosis sequencing energy characteristics and load rate kurtosis sequencing energy characteristics.
S212, constructing a second feature set (curve complexity feature): according to definition of fuzzy entropy, calculating complexity characteristics of a current curve, voltage curve, active power curve, reactive power curve, power factor curve, three-phase imbalance, load factor and load factor respectively;
the fuzzy entropy (fuzzylen) is also measured as the probability of new pattern generation, and the larger the measure, the greater the probability of new pattern generation, i.e., the greater the sequence complexity. The fuzzy entropy is described as follows:
；
wherein m is the dimension of the phase space, r is the similarity tolerance, N is the dimension of the time series,is a fuzzy membership function;
respectively calculating complexity characteristics of the curves according to defined fuzzy entropySecond Voltage Curve complexity characteristic +. >Second active Power Curve complexity characteristic +.>Second reactive power curve complexity characteristic +.>Second Power factor Curve complexity order feature +.>Second three-phase imbalance profile complexity characteristic +.>Second load-factor curve complexity characteristic ∈>The method comprises the following steps of:
；
；
；
；
；
；
。
s213, constructing a third feature set (inter-phase cosine similarity feature): according to cosine similarity formula, respectively calculating cosine similarity characteristics of A-phase and B-phase currentsCosine similarity feature of B-phase and C-phase currents>Cosine similarity feature of currents of phase A and phase C>Cosine similarity feature of voltages of phase A and phase B>Cosine similarity feature of B-phase and C-phase voltages>Cosine similarity feature of voltages of phase A and phase C>Cosine similarity feature of active power of phase A and phase B>Cosine similarity feature of active power of B phase and C phase>Cosine similarity feature of active power of phase A and phase C>Reactive power of phase A and phase BPower cosine similarity feature->Cosine similarity characteristics of reactive power of B phase and C phase>Cosine similarity characteristics of reactive power of phase A and phase C>Cosine similarity feature of power factors of phase A and phase B>Cosine similarity feature of B-phase and C-phase power factors>Cosine similarity feature of power factors of phase A and phase C >；
The cosine similarity formula is as follows:
；
wherein, X, Y represents X, Y vector;、respectively representing the modulus of X, Y vectors;
the following characteristics were obtained:
。
the step S220 specifically includes: monitoring data for any historical operationSelecting/screening the kurtosis sequencing energy characteristic, curve complexity characteristic and inter-phase cosine similarity characteristic set of the voltage data, current data, active power data, reactive power data, power factor data, three-phase unbalance degree data and load rate data constructed in the step S201 to obtain a final characteristic set; referring specifically to fig. 2, step S220 specifically includes:
s221, taking kurtosis ordering energy characteristics of each physical quantity as a first characteristic set, curve complexity characteristics of each physical quantity as a second characteristic set, inter-phase cosine similarity characteristics of each physical quantity as a third characteristic set, and combining the first characteristic set, the second characteristic set and the third characteristic set into a fourth characteristic set;
s222, performing feature selection on the fourth feature set by using a first feature selection algorithm, a second feature selection algorithm and a … … nth feature selection algorithm to obtain n feature subsets;
s223, converging the n feature subsets into an intersection to obtain a fifth feature set;
S224, performing feature selection on the fifth feature set by using an n+1 feature selection algorithm to obtain a sixth feature set, wherein the sixth feature set is used as a final feature set;
the first feature selection algorithm, the second feature selection algorithm, the … … nth feature selection algorithm and the n+1th feature selection algorithm are specifically selected from feature selection methods such as a variance selection method, a correlation coefficient method, a chi-square test method, a relief algorithm, a recursive feature elimination method, a feature selection method based on punishment items, a feature selection method based on a tree model and the like.
S300, adding an error state label to a final feature set of each historical operation monitoring data to form a training sample, constructing a nonlinear time sequence prediction (Nonlinear Autoregressive models with Exogenous Inputs, NARX) neural network model, and performing iterative training on the NARX neural network model through the training sample to obtain a trained electric energy metering device operation error state evaluation model;
referring specifically to fig. 3, fig. 3 is a schematic diagram of a reference structure of a NARX neural network, which is a dynamic neural network with a clear structure based on a BP neural network, and is introduced into an input vector in an external feedback manner after a part of data of an output vector is stored on the basis of the BP neural network.
Referring to fig. 3, assuming that the current time is t, nx is the order of the neural network input, ny is the output order of the neural network,weight between the ith input vector representing the neural network and the jth hidden layer neuron,/for the neural network>Representing weights between the ith hidden layer neuron and the jth output layer neuron of the neural network,/for the neural network>Hiding the offset value of the nth neuron of the layer for the neural network,>for the offset value of the neural network output layer neuron, X (t) is the input vector, Y (t) is the input vector of the feedback output,/for>Transfer function representing hidden layer of neural network, using Tan-sigmoid function, +.>Neuron output value representing the nth hidden layer,/->Representing the transfer function of the neural network output neurons.
The structural parameters of the NARX neural network model are adjusted by adopting a simple mean square error performance function, and the effect of the NARX neural network model is equivalent to that of enabling the NARX neural network model to obtain the optimal fitting effect on training set data, so that the problem of over-fitting is introduced. The "overfitting" problem is manifested in that the NARX neural network model after training is not strong in generalization ability, i.e., only shows good performance in the training set, but has poor prediction effect when applied to the actual test set. Under the condition of fixed training set, the generalization capability of the neural network is closely related to the network parameter scale. Therefore, the gold eagle algorithm is introduced in the embodiment to limit the whole parameter scale of the network by optimizing the weight coefficient and the threshold parameter in the NARX neural network, and finally improve the generalization performance of the NARX neural network.
The eagle algorithm is the intelligence of tuning speed of eagle at different stages of spiral track, and is used for hunting. They show more tendency to patrol and find hunting in the initial stages of hunting and more tendency to attack in the final stages. A gold eagle adjusts both components to capture the best prey in the shortest time possible. The method comprises the following specific steps:
s311, initializing the number of eagle individuals in the eagle population, wherein each eagle contains weight coefficient information and threshold parameter information of different NARX neural network models;
s312, calculating fitness values of the eagle individuals and initializing group memory according to the fitness values; the fitness value calculation formula of the gold eagle individual is as follows:
；
wherein value represents fitness value, N is total number of gold eagle individuals,for NARX neural network model included according to ith gold eaglePredicted value of NARX neural network model obtained by weight coefficient information and threshold parameter information, < >>The sample true value corresponding to the ith gold eagle;
s313, initializing attack tendency of gold hawkAnd cruise tendency->；
S314, updating attack tendency according to the following formulaAnd cruise tendency->：
；
Wherein,,and- >Is->Initial and final values, +.>And->Is->Is a final value and an initial value of (1); t is the T-th iteration, and T is the total iteration number;
s315, randomly selecting prey from attack vectors calculated by the memory of the population:
；
wherein,,for the i-th gold eagle's attack vector, < >>For the best place to be reached by the current eagle, < > for>Is the current position of the ith gold eagle;
s316, calculating a cruising vector d:
；
wherein,,as normal vector +.>Is a decision variable vector;
is provided withAny point on the hyperplane, then:；
handleViewed as a normal to the hyperplane, the hyperplane is expressed as:
；
wherein,,in order for the vector of the attack,in order to make a decision vector,is the selected prey location;representing an attack vector at the t-th iteration;
then:
；/>
wherein,,k is the number of the fixed variable and is the kth element of the target point;
the target point on the cruise hyperplane is expressed as:
the step size vector of the hawk iteration is defined as:
；
wherein,,and->Is [0,1]The random vector in the iteration is added to the position in the iteration by the step size vector in the iteration, and the position of the gold eagle in the iteration is calculated:
wherein,,the first thing is golden eagleThe secondary position is used for the control of the position,the first thing is golden eagleThe position of the secondary part is determined,the step size of the abnormal movement of the gold hawk;
s317, calculating a fitness value according to the updated new position of the golden eagle individual, and updating the optimal solution and the optimal position;
S318, judging whether the maximum iteration times are reached, and outputting weight coefficient information and threshold parameter information contained in the eagle individuals at the current optimal position as weight coefficients and threshold parameters of the NARX neural network model if the maximum iteration times are reached; if the maximum number of iterations is not reached, the process returns to step S314 and the iteration is continued.
Based on the neural network optimization method provided by the embodiment, the iterative optimization process of the hawk algorithm is an updating process of weight parameters and threshold parameters in the NARX neural network training process, wherein the fitness corresponds to an NARX neural network error function (value), and the error is minimized by changing the position of the hawk. Each eagle comprises weight coefficient information and threshold parameter information in the NARX neural network, and when the eagle reaches the optimal position, the weight coefficient information and the threshold parameter information contained in the eagle are updated optimal parameters.
S400, acquiring current operation monitoring data of the target electric energy metering device, extracting features according to the current operation monitoring data to obtain a current final feature set of the target electric energy metering device, and inputting the current final feature set to a trained electric energy metering device operation error state evaluation model to obtain an error state evaluation result of the target electric energy metering device.
Embodiment two:
the embodiment provides an operation error state evaluation system of an electric energy metering device, which comprises:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring historical operation monitoring data of a plurality of electric energy metering devices, and each historical operation monitoring data comprises a plurality of different physical quantities generated in the operation process of the electric energy metering devices; the module is used for implementing the function of step S100 in the first embodiment, and will not be described here again;
the feature extraction module is used for extracting features of each collected historical operation monitoring data and comprises two steps of feature construction and feature selection; the module is used for realizing the function of step S200 in the first embodiment;
the construction characteristic steps specifically include: for each historical operation monitoring data, calculating kurtosis sequencing energy characteristics, curve complexity characteristics and inter-phase cosine similarity characteristics corresponding to each physical quantity through each kind of physical quantity;
the characteristic selection step specifically comprises the following steps: screening the feature set obtained in the step of constructing the features for any historical operation monitoring data to obtain a final feature set;
the model training module is used for adding an error state label to the final feature set of each historical operation monitoring data to form a training sample, constructing a neural network model, and carrying out iterative training on the neural network model through the training sample to obtain a trained electric energy metering device operation error state evaluation model; the module is used for realizing the function of step S300 in the first embodiment;
The evaluation module is used for acquiring current operation monitoring data of the target electric energy metering device, extracting features according to the current operation monitoring data to obtain a current final feature set of the target electric energy metering device, and inputting the current final feature set into the trained electric energy metering device operation error state evaluation model to obtain an error state evaluation result of the target electric energy metering device; this module is used to implement the function of step S400 in the first embodiment.
Embodiment III:
the embodiment provides an electronic device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the method for evaluating the running error state of the electric energy metering device according to any embodiment of the invention when executing the program.
Embodiment four:
the present embodiment proposes a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method for evaluating an operation error state of an electric energy metering device according to any of the embodiments of the present invention.
In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "and/or", describes an association relation of association objects, and indicates that there may be three kinds of relations, for example, a and/or B, and may indicate that a alone exists, a and B together, and B alone exists. Wherein A, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of the following" and the like means any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c may represent: a, b, c, a and b, a and c, b and c or a and b and c, wherein a, b and c can be single or multiple.
Those of ordinary skill in the art will appreciate that the various elements and algorithm steps described in the embodiments disclosed herein can be implemented as a combination of electronic hardware, computer software, and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In several embodiments provided herein, any of the functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (hereinafter referred to as ROM), a random access Memory (Random Access Memory) and various media capable of storing program codes such as a magnetic disk or an optical disk.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes or direct or indirect application in other related technical fields are included in the scope of the present invention.
Claims (5)
1. The method for evaluating the running error state of the electric energy metering device is characterized by comprising the following steps of:
collecting historical operation monitoring data of a plurality of electric energy metering devices, wherein each historical operation monitoring data comprises a plurality of different types of physical quantities generated in the operation process of the electric energy metering devices;
extracting features of each collected historical operation monitoring data, wherein the feature extraction comprises two steps of constructing features and selecting features;
the construction characteristic steps specifically include: for each historical operation monitoring data, calculating kurtosis sequencing energy characteristics, curve complexity characteristics and inter-phase cosine similarity characteristics corresponding to each physical quantity through each kind of physical quantity;
the characteristic selection step specifically comprises the following steps: screening the feature set obtained in the step of constructing the features for any historical operation monitoring data to obtain a final feature set;
Adding an error state label to the final feature set of each historical operation monitoring data to form a training sample, constructing a neural network model, and carrying out iterative training on the neural network model through the training sample to obtain a trained electric energy metering device operation error state evaluation model;
acquiring current operation monitoring data of a target electric energy metering device, extracting features according to the current operation monitoring data to obtain a current final feature set of the target electric energy metering device, and inputting the current final feature set of the target electric energy metering device into a trained electric energy metering device operation error state evaluation model to obtain an error state evaluation result of the target electric energy metering device;
the plurality of different types of physical quantities generated during operation of the electric energy metering device include:
voltage data, current data, active power data, reactive power data, power factor data, three-phase imbalance data, and load factor data;
in the step of calculating the kurtosis sequencing energy characteristics, the curve complexity characteristics and the inter-phase cosine similarity characteristics of each physical quantity through the physical quantity of each kind, the calculation method of the kurtosis sequencing energy characteristics of each physical quantity is as follows:
drawing corresponding physical quantity curves according to various physical quantities, wherein the corresponding physical quantity curves comprise a current curve, a voltage curve, an active power curve, a reactive power curve, a power factor curve, a three-phase unbalance curve and a load factor curve;
For each physical quantity curve, the numerical statistic of the random variable distribution characteristic of each physical quantity curve is reflected through kurtosis, and the expression of the kurtosis K is as follows:
；
wherein N represents the signal length of the physical quantity curve;representing the ith signal value in the physical quantity curve; μ represents the signal average value of the physical quantity curve; sigma represents the standard deviation of the signal of the physical quantity curve;
EMD empirical mode decomposition is performed on each signal in each physical quantity curve, kurtosis is calculated on the decomposed signals, and kurtosis sorting energy E is calculated according to the following formula after the kurtosis is sorted in descending order according to the calculated kurtosis:
；
thereby obtaining current kurtosis sequencing energy characteristics, voltage kurtosis sequencing energy characteristics, active power kurtosis sequencing energy characteristics, reactive power kurtosis sequencing energy characteristics, power factor kurtosis sequencing energy characteristics, three-phase imbalance kurtosis sequencing energy characteristics and load rate kurtosis sequencing energy characteristics;
in the step of calculating the kurtosis sequencing energy feature, the curve complexity feature and the inter-phase cosine similarity feature of each physical quantity through the physical quantity of each kind, the calculating method of the curve complexity feature of each physical quantity is as follows:
Definition of fuzzy entropy is:
；
wherein m is a phaseThe spatial dimension, r, is the similarity tolerance, N is the dimension of the time series,is a fuzzy membership function;
respectively calculating complexity characteristics of the curves according to defined fuzzy entropySecond Voltage Curve complexity characteristic +.>Second active Power Curve complexity characteristic +.>Second reactive power curve complexity characteristic +.>Second Power factor Curve complexity order feature +.>Second three-phase imbalance profile complexity characteristic +.>Second load-factor curve complexity characteristic ∈>The method comprises the following steps of:
；
；
；
；
；
；
；
wherein,,、、respectively representing the current of A phase, B phase and C phase;、、The voltages of the phase A, the phase B and the phase C are respectively;、、、respectively representing A phase, B phase, C phase and total active power;、、、Respectively representing A phase, B phase, C phase and total reactive power;、、、Respectively representing A phase, B phase, C phase and total current power factors;Three-phase imbalance data;
R load factor Representing the load factor;
in the step of calculating the kurtosis sequencing energy feature, the curve complexity feature and the inter-phase cosine similarity feature of each physical quantity through the physical quantity of each kind, the calculating method of the inter-phase cosine similarity feature of each physical quantity is as follows:
according to cosine similarity formula, respectively calculating cosine similarity characteristics of A-phase and B-phase currents Cosine similarity feature of B-phase and C-phase currents>Cosine similarity feature of currents of phase A and phase C>Cosine similarity feature of voltages of phase A and phase B>Cosine similarity feature of B-phase and C-phase voltages>Cosine similarity feature of voltages of phase A and phase C>Cosine similarity feature of active power of phase A and phase B>Cosine similarity feature of active power of B phase and C phase>Cosine similarity feature of active power of phase A and phase C>Cosine similarity characteristics of reactive power of phase A and phase B>Cosine similarity characteristics of reactive power of B phase and C phase>Cosine similarity characteristics of reactive power of phase A and phase C>Cosine similarity feature of power factors of phase A and phase B>Cosine similarity feature of B-phase and C-phase power factors>Cosine similarity feature of power factors of phase A and phase C>；
The cosine similarity formula is as follows:
；
wherein, X, Y represents X, Y vector;、respectively representing the modulus of X, Y vectors;
the method for screening the feature set obtained in the step of constructing the features to obtain the final feature set specifically comprises the following steps:
taking kurtosis ordering energy characteristics of each physical quantity as a first characteristic set, curve complexity characteristics of each physical quantity as a second characteristic set, inter-phase cosine similarity characteristics of each physical quantity as a third characteristic set, and combining the first characteristic set, the second characteristic set and the third characteristic set into a fourth characteristic set;
The fourth feature set is subjected to feature selection by using a first feature selection algorithm, a second feature selection algorithm and a … … nth feature selection algorithm, so that n feature subsets are obtained;
merging the n feature subsets into a fifth feature set;
performing feature selection on the fifth feature set by using an n+1 feature selection algorithm to obtain a sixth feature set, wherein the sixth feature set is used as a final feature set;
the first feature selection algorithm, the second feature selection algorithm, the … … nth feature selection algorithm and the n+1th feature selection algorithm are specifically selected from a variance selection method, a correlation coefficient method, a chi-square test method, a relief algorithm, a recursive feature elimination method, a feature selection method based on penalty terms and a feature selection method based on a tree model.
2. The method for evaluating the operation error state of an electric energy metering device according to claim 1, wherein the neural network model adopts a nonlinear time sequence prediction NARX neural network model, and optimizes the neural network model by adopting a gold hawk algorithm, and the specific steps are as follows:
initializing the number of eagle individuals in a eagle population, wherein each eagle contains weight coefficient information and threshold parameter information of different NARX neural network models;
Calculating fitness value of the eagle individuals and initializing group memory according to the fitness value; the fitness value calculation formula of the gold eagle individual is as follows:
；
wherein value represents fitness value, N is total number of gold eagle individuals,for the predicted value of NARX neural network model obtained according to the weight coefficient information and threshold parameter information of NARX neural network model contained in ith gold eagle, +.>The sample true value corresponding to the ith gold eagle;
initiating challenge with gold eagleAnd cruise tendency->；
Updating attack trends according to the following formulaAnd cruise tendency->：
；
Wherein,,and->Is->Initial and final values, +.>And->Is->Is a final value and an initial value of (1); t is the T-th iteration, and T is the total iteration number;
randomly selecting prey from attack vectors calculated from the memory of the population:
；
wherein,,attack of ith gold hawkClick vector (s)/(s)>For the best place to be reached by the current eagle, < > for>Is the current position of the ith gold eagle;
calculating a cruising vector d:
；
wherein,,as normal vector +.>Is a decision variable vector;
is provided withAny point on the hyperplane, then:；
HandleViewed as a normal to the hyperplane, the hyperplane is expressed as:
；
wherein,,for attack vector, ++>In order to make a decision vector, Is the selected prey location;Representing an attack vector at the t-th iteration;
then:
；
wherein,,k is the number of the fixed variable and is the kth element of the target point;
the target point on the cruise hyperplane is expressed as:
the step size vector of the hawk iteration is defined as:
；
wherein,,and->Is [0,1]The random vector in the iteration is added to the position in the iteration by the step size vector in the iteration, and the position of the gold eagle in the iteration is calculated:
wherein,,first part of gold hawk>Secondary position(s) (i.e. the position of the person)>First part of gold hawk>Secondary location,/->The step size of the abnormal movement of the gold hawk;
calculating a fitness value according to the updated new position of the eagle individual, and updating the optimal solution and the optimal position;
judging whether the maximum iteration times are reached, and outputting weight coefficient information and threshold parameter information contained in the eagle individuals at the current optimal position as weight coefficients and threshold parameters of the NARX neural network model if the maximum iteration times are reached; and continuing iteration if the maximum iteration number is not reached.
3. An electric energy metering device operation error state evaluation system, characterized by comprising:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring historical operation monitoring data of a plurality of electric energy metering devices, and each historical operation monitoring data comprises a plurality of different physical quantities generated in the operation process of the electric energy metering devices;
The feature extraction module is used for extracting features of each collected historical operation monitoring data and comprises two steps of feature construction and feature selection;
the construction characteristic steps specifically include: for each historical operation monitoring data, calculating kurtosis sequencing energy characteristics, curve complexity characteristics and inter-phase cosine similarity characteristics corresponding to each physical quantity through each kind of physical quantity;
the characteristic selection step specifically comprises the following steps: screening the feature set obtained in the step of constructing the features for any historical operation monitoring data to obtain a final feature set;
the model training module is used for adding an error state label to the final feature set of each historical operation monitoring data to form a training sample, constructing a neural network model, and carrying out iterative training on the neural network model through the training sample to obtain a trained electric energy metering device operation error state evaluation model;
the evaluation module is used for acquiring current operation monitoring data of the target electric energy metering device, extracting features according to the current operation monitoring data to obtain a current final feature set of the target electric energy metering device, and inputting the current final feature set into the trained electric energy metering device operation error state evaluation model to obtain an error state evaluation result of the target electric energy metering device;
The plurality of different types of physical quantities generated during operation of the electric energy metering device include:
voltage data, current data, active power data, reactive power data, power factor data, three-phase imbalance data, and load factor data;
in the step of calculating the kurtosis sequencing energy characteristics, the curve complexity characteristics and the inter-phase cosine similarity characteristics of each physical quantity through the physical quantity of each kind, the calculation method of the kurtosis sequencing energy characteristics of each physical quantity is as follows:
drawing corresponding physical quantity curves according to various physical quantities, wherein the corresponding physical quantity curves comprise a current curve, a voltage curve, an active power curve, a reactive power curve, a power factor curve, a three-phase unbalance curve and a load factor curve;
for each physical quantity curve, the numerical statistic of the random variable distribution characteristic of each physical quantity curve is reflected through kurtosis, and the expression of the kurtosis K is as follows:
；
wherein N represents the signal length of the physical quantity curve;representing the ith signal value in the physical quantity curve; μ represents the signal average value of the physical quantity curve; sigma represents the standard deviation of the signal of the physical quantity curve;
EMD empirical mode decomposition is performed on each signal in each physical quantity curve, kurtosis is calculated on the decomposed signals, and kurtosis sorting energy E is calculated according to the following formula after the kurtosis is sorted in descending order according to the calculated kurtosis:
；
Thereby obtaining current kurtosis sequencing energy characteristics, voltage kurtosis sequencing energy characteristics, active power kurtosis sequencing energy characteristics, reactive power kurtosis sequencing energy characteristics, power factor kurtosis sequencing energy characteristics, three-phase imbalance kurtosis sequencing energy characteristics and load rate kurtosis sequencing energy characteristics;
in the step of calculating the kurtosis sequencing energy feature, the curve complexity feature and the inter-phase cosine similarity feature of each physical quantity through the physical quantity of each kind, the calculating method of the curve complexity feature of each physical quantity is as follows:
definition of fuzzy entropy is:
；
wherein m is the dimension of the phase space, r is the similarity tolerance, N is the dimension of the time series,is a fuzzy membership function;
respectively calculating complexity characteristics of the curves according to defined fuzzy entropySecond Voltage Curve complexity characteristic +.>Second active Power Curve complexity characteristic +.>Second reactive power curve complexity characteristic +.>Second Power factor Curve complexity order feature +.>Second three-phase imbalance profile complexity characteristic +.>Second load-factor curve complexity characteristic ∈>The method comprises the following steps of:
；
；
；
；
；
；
；
wherein,,、、respectively representing the current of A phase, B phase and C phase;、、The voltages of the phase A, the phase B and the phase C are respectively; 、、、Respectively representing A phase, B phase, C phase and total active power;、、、Respectively representing A phase, B phase, C phase and total reactive power;、、、Respectively representing A phase, B phase, C phase and total current power factors;Three-phase imbalance data;
R load factor Representing the load factor;
in the step of calculating the kurtosis sequencing energy feature, the curve complexity feature and the inter-phase cosine similarity feature of each physical quantity through the physical quantity of each kind, the calculating method of the inter-phase cosine similarity feature of each physical quantity is as follows:
according to cosine similarity formula, respectively calculating cosine similarity characteristics of A-phase and B-phase currentsCosine similarity feature of B-phase and C-phase currents>Cosine similarity feature of currents of phase A and phase C>Cosine similarity feature of voltages of phase A and phase B>Cosine similarity feature of B-phase and C-phase voltages>Cosine similarity feature of voltages of phase A and phase C>Cosine similarity feature of active power of phase A and phase B>Cosine similarity feature of active power of B phase and C phase>Cosine similarity feature of active power of phase A and phase C>Cosine similarity characteristics of reactive power of phase A and phase B>Cosine similarity characteristics of reactive power of B phase and C phase>Cosine similarity characteristics of reactive power of phase A and phase C >Cosine similarity feature of power factors of phase A and phase B>Cosine similarity feature of B-phase and C-phase power factors>Cosine similarity feature of power factors of phase A and phase C>；
The cosine similarity formula is as follows:
；
wherein, X, Y represents X, Y vector;、respectively representing the modulus of X, Y vectors;
the method for screening the feature set obtained in the step of constructing the features to obtain the final feature set specifically comprises the following steps:
taking kurtosis ordering energy characteristics of each physical quantity as a first characteristic set, curve complexity characteristics of each physical quantity as a second characteristic set, inter-phase cosine similarity characteristics of each physical quantity as a third characteristic set, and combining the first characteristic set, the second characteristic set and the third characteristic set into a fourth characteristic set;
the fourth feature set is subjected to feature selection by using a first feature selection algorithm, a second feature selection algorithm and a … … nth feature selection algorithm, so that n feature subsets are obtained;
merging the n feature subsets into a fifth feature set;
performing feature selection on the fifth feature set by using an n+1 feature selection algorithm to obtain a sixth feature set, wherein the sixth feature set is used as a final feature set;
the first feature selection algorithm, the second feature selection algorithm, the … … nth feature selection algorithm and the n+1th feature selection algorithm are specifically selected from a variance selection method, a correlation coefficient method, a chi-square test method, a relief algorithm, a recursive feature elimination method, a feature selection method based on penalty terms and a feature selection method based on a tree model.
4. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for evaluating the operational error status of the electric energy metering device according to any one of claims 1 to 2 when executing the program.
5. A computer-readable storage medium having stored thereon a computer program, wherein the program when executed by a processor implements the electric energy metering device operation error state evaluation method according to any one of claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310000732.XA CN115841278B (en) | 2023-01-03 | 2023-01-03 | Method, system, equipment and medium for evaluating running error state of electric energy metering device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310000732.XA CN115841278B (en) | 2023-01-03 | 2023-01-03 | Method, system, equipment and medium for evaluating running error state of electric energy metering device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115841278A CN115841278A (en) | 2023-03-24 |
| CN115841278B true CN115841278B (en) | 2023-07-18 |
Family
ID=85579426
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310000732.XA Active CN115841278B (en) | 2023-01-03 | 2023-01-03 | Method, system, equipment and medium for evaluating running error state of electric energy metering device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115841278B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116126945B (en) * | 2023-03-30 | 2023-06-20 | 创域智能(常熟)网联科技有限公司 | Sensor running state analysis method and system based on data analysis |
| CN116028822B (en) * | 2023-03-30 | 2023-07-18 | 国网福建省电力有限公司 | Electric energy meter error state evaluation method, system, equipment and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110874685A (en) * | 2018-09-04 | 2020-03-10 | 中国计量科学研究院 | Intelligent electric energy meter running state distinguishing method and system based on neural network |
| CN112698261A (en) * | 2020-11-25 | 2021-04-23 | 国网湖南省电力有限公司 | Fault early warning method and device for intelligent electric energy meter |
| CN114696340A (en) * | 2022-02-11 | 2022-07-01 | 南京理工大学 | Wind power frequency modulation gradual inertia control method based on deep learning |
| CN114399032B (en) * | 2022-03-23 | 2022-06-21 | 武汉格蓝若智能技术有限公司 | Method and system for predicting metering error of electric energy meter |
| CN115392393A (en) * | 2022-09-13 | 2022-11-25 | 中国核动力研究设计院 | Temperature measuring instrument state detection method |
-
2023
- 2023-01-03 CN CN202310000732.XA patent/CN115841278B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115841278A (en) | 2023-03-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115841278B (en) | Method, system, equipment and medium for evaluating running error state of electric energy metering device | |
| Karaboga et al. | An adaptive and hybrid artificial bee colony algorithm (aABC) for ANFIS training | |
| CN116757534B (en) | Intelligent refrigerator reliability analysis method based on neural training network | |
| CN115060497B (en) | Bearing fault diagnosis method based on CEEMD energy entropy and optimized PNN | |
| CN113486578A (en) | Method for predicting residual life of equipment in industrial process | |
| CN111144552B (en) | Multi-index grain quality prediction method and device | |
| CN116028822B (en) | Electric energy meter error state evaluation method, system, equipment and storage medium | |
| CN111784061B (en) | Training method, device and equipment for power grid engineering cost prediction model | |
| CN112288137A (en) | LSTM short-term load prediction method and device considering electricity price and Attention mechanism | |
| CN114399032A (en) | Method and system for predicting metering error of electric energy meter | |
| CN116542382A (en) | Sewage treatment dissolved oxygen concentration prediction method based on mixed optimization algorithm | |
| CN116068481A (en) | Method, system, equipment and medium for quantitatively evaluating error of current transformer | |
| CN114595884A (en) | Genetic intelligent optimization neural network wind power generation equipment temperature prediction method | |
| CN115758187A (en) | Coal mine mechanical motor state prediction method based on K-CNN-N-GRU | |
| CN112149896A (en) | Attention mechanism-based mechanical equipment multi-working-condition fault prediction method | |
| CN117633688A (en) | Large-scale power data anomaly detection method based on ridge regression-k-means clustering-LOF-LSTM fusion algorithm | |
| CN118351970A (en) | Prediction method for PM2.5 concentration in ecological environment | |
| CN118134046A (en) | Wind farm power prediction method and system based on machine learning | |
| CN117973852A (en) | Chemical safety assessment method based on improved radial basis function Bayesian network | |
| CN116776209A (en) | Method, system, equipment and medium for identifying operation state of gateway metering device | |
| CN117096860A (en) | Overhead transmission line current-carrying capacity interval prediction method and equipment based on LSSVM model | |
| CN115275977A (en) | Power load prediction method and device | |
| CN114298413A (en) | Hydroelectric generating set runout trend prediction method | |
| CN113379149A (en) | Air micro-station concentration prediction method based on LSTM neural network | |
| Rahmat et al. | Lake Toba Water Quality Prediction Using Extreme Machine Learning |
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 |