CN109598052A - Intelligent electric meter life cycle prediction technique and device based on correlation analysis - Google Patents
Intelligent electric meter life cycle prediction technique and device based on correlation analysis Download PDFInfo
- Publication number
- CN109598052A CN109598052A CN201811440265.8A CN201811440265A CN109598052A CN 109598052 A CN109598052 A CN 109598052A CN 201811440265 A CN201811440265 A CN 201811440265A CN 109598052 A CN109598052 A CN 109598052A
- Authority
- CN
- China
- Prior art keywords
- life cycle
- meter
- representing
- manufacturer
- fault
- 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
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000010219 correlation analysis Methods 0.000 title abstract description 4
- 238000013528 artificial neural network Methods 0.000 claims abstract description 17
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 33
- 238000004458 analytical method Methods 0.000 claims description 22
- 238000012937 correction Methods 0.000 claims description 16
- 230000000737 periodic effect Effects 0.000 claims description 14
- 238000003860 storage Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 8
- 230000005611 electricity Effects 0.000 claims description 8
- 238000009434 installation Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 238000013461 design Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 8
- 238000012423 maintenance Methods 0.000 description 7
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 3
- 238000007418 data mining Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013179 statistical model Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of intelligent electric meter life cycle prediction technique and device based on correlation analysis, by the intelligent electric meter life cycle prediction technique based on correlation analysis according to unit factor to affect, manufacturer's factor to affect and fault type factor to affect be major parameter, and using neural network algorithm as parameter training auxiliary.This method process is as follows: firstly, calculating the cycle dependency of three verifying unit, ammeter manufacturer, failure cause factors；Then the prediction model of the prediction replacement cycle of ammeter, the every value occurred in initialization model are obtained；Weight is determined using heuristic iterative search later, is modified using the update weight mode based on neural network algorithm；Finally design Early-warning Model.It realizes the Accurate Prediction to the life cycle of intelligent electric meter and replaces the technical effect of early warning.
Description
Technical Field
The invention relates to the technical field of data mining in computer science, in particular to a method and a device for predicting the life cycle of an intelligent electric meter based on correlation coefficient analysis.
Background
The intelligent electric energy meter is a novel electric energy meter which is popularized to a living application layer in a large quantity in recent years. With the worldwide development of the "smart grid" and Advanced Measurement Infrastructure (AMI) and related technologies, smart meters, which are their basic components and core devices, are attracting a large number of meter manufacturers. The smart meter needs to have high reliability and long service life under normal conditions and can continuously and uninterruptedly work under unattended conditions.
In order to effectively manage the smart meter and maintain the national power, it is very important to be able to accurately predict and estimate the life characteristics of the smart meter. Factors influencing reliability credibility of the intelligent electric meter are as follows: function, complexity, design, manufacturing process, failure criteria, operating conditions, installation maintenance, etc.
In the prior art, some scholars conduct relevant research on maintenance of the intelligent electric meter, wherein a new repairable system maintenance period prediction method based on a fault statistical model is built, and a top measurement method of the repairable system fault statistical model and the maintenance period is provided. The method has practical applicability in predicting the maintenance period of the maintainable system, but cannot be directly applied to equipment such as a smart meter which is more prone to rotation than maintenance. Furthermore, the basic concept of the credibility of electrical measuring devices is set forth in the national standard of the people's republic of China GB 17215.911-200X/IEC/TR 62059-11: 2002. But no relevant analysis is performed on the life cycle of the smart meter.
Disclosure of Invention
In view of the above, the present invention provides a method and an apparatus for predicting a life cycle of a smart meter based on correlation coefficient analysis, so as to solve or at least partially solve the technical problem in the prior art that the life cycle of the smart meter cannot be accurately predicted.
In order to solve the technical problem, a first aspect of the present invention provides a method for predicting a life cycle of a smart meter based on correlation coefficient analysis, including:
step S1: calculating and verifying the periodic correlation among three factors of a unit, an electric meter manufacturer and a fault reason, wherein the periodic correlation specifically comprises the following steps:mu 'and sigma' respectively represent the average value and the variance of the service life of each corresponding ammeter in the same unit with different fault reasons of the same ammeter manufacturer;
step S2: according to the period correlation, a life cycle prediction model of the intelligent electric meter is constructed, and the prediction model specifically comprises the following steps:
wherein M represents the total number of the manufacturers of the electric meter, N represents the total number of the fault reasons, W represents the total number of units, min (M, N) represents the minimum value in M, N,
wherein, ω is1、ω2As a weighting factor, the initial value is 1.00,representing the number weighted life cycle average of all meters produced by meter manufacturer j in all units; omegaDiLife cycle impact weight in units of i; omegaFjMainErrorLife cycle impact weight representing the primary cause of failure for meter manufacturer j; omegaDiCorrelationErrorLife cycle influence weight representing the correlation fault cause of unit i;
step S3: and predicting the life cycle of the intelligent electric meter based on the life cycle prediction model of the intelligent electric meter.
In one embodiment, after step S2, the method further comprises:
updating weight mode pair weight factor omega based on preset neural network algorithm1And ω2The correction is carried out so that the correction is carried out,
and optimizing the prediction model based on the corrected weight factor.
In one embodiment, the method further comprises:
obtaining a predicted life cycle and an average error rate according to a prediction result obtained through the prediction model;
acquiring early warning time corresponding to the intelligent ammeter based on the predicted life cycle and the average error rate;
and setting early warning reminding according to the early warning time.
In one embodiment, the weighting parameters in step S2ωDi、ωFjMainErrorAnd ωDiCorrelationErrorThe calculation method of (c) is as follows:
is calculated byWherein,record, maxN, representing the h-th meter of the manufacturer j that manufactured the smart meterFjRepresents the maximum number of records of the meter factory j, maxError represents the maximum fault rate allowed by a single batch, the default value is 2 percent, and ErrorFjhH-th record, T, representing the number of faulty meters of meter factory jErrorTimeIndicating the time of failure, T, of the watchInstallTimeRepresents the initial installation time of the electricity meter,indicating the number of plant j fault tables;
ωDiis calculated byWherein, TDiThe life cycle in units of i is expressed,
denotes the average life cycle, T, of all unitsDiThe calculation method is as follows:
wherein,the number of the a-th electric meters in the unit i is represented; max NiaThe maximum value of the number of records in the unit i,the a-th life cycle record representing the unit i,the calculation method comprises the following steps: represents the life cycle average of all meters used in unit i, sum represents the total number of units;
ωFjMainErroris calculated by Number weighted life cycle mean, T, representing the major cause of failure for meter manufacturer jallThe number representing all the causes of failure weights the mean of the life cycle averages,the calculation method comprises the following steps:max NFjMainErrorthe maximum number of records of life cycle of the main cause of failure K of meter manufacturer j,a kth record of the number of meters indicating the main causes of failure for meter manufacturer j;
ωDiCorrelationErrorthe calculation method of (2) is as follows:wherein,the quantity representing all the causes of failure is weighted by the life cycle average,NAllErrorthe maximum number of categories representing the cause of the fault,the number representing the cause of the fault i, weights the life cycle, wherein,number weighted life cycle mean value representing correlation fault cause of meter manufacturer jmax N represents the maximum number of records of the correlation failure cause of the manufacturer j,to representThe ith cycle of the failure cause j is rotated.
In one embodiment, the life cycle data of the smart meter of a real unit i-meter manufacturer j is predefinedAs reference data, the weight factor omega is subjected to weight updating based on the updating weight mode of the preset neural network algorithm1And ω2The correction specifically includes:
step S4.1: randomly extracting 2/3 electric meters from all the pre-obtained intelligent electric meter information records to serve as a training set, then randomly extracting 2/3 electric meters from the training set to serve as a training sample set, and taking the rest 1/3 of all the intelligent electric meter information records to serve as a reference set, wherein the number of individuals in the reference set is A and is a positive integer;
step S4.2: using formulas in predictive modelsAnd
performing prediction by taking the prediction object as an individual in the reference set to obtain a predicted value
Step S4.3: determining whether the absolute value of the difference between the predicted value and the reference value is less than a threshold value delta, i.e.If yes, go to step S4.4;
step S4.4: judgment ofIf true, then ω is updated2，If not, update ω1，
Step S4.5: and judging whether each individual in the reference set is predicted to be finished or not, if so, finishing, and if not, continuing iteration and executing the step S4.2.
Based on the same inventive concept, the second aspect of the present invention provides a device for predicting life cycle of a smart meter based on correlation coefficient analysis, comprising:
the calculation and verification module is used for calculating and verifying the periodic correlation among three factors, namely a unit, an electric meter manufacturer and a fault reason, wherein the periodic correlation specifically comprises the following steps:mu 'and sigma' respectively represent the average value and the variance of the service life of each corresponding ammeter in the same unit with different fault reasons of the same ammeter manufacturer;
the model building module is used for building a life cycle prediction model of the intelligent electric meter according to the cycle correlation, and the prediction model specifically comprises the following steps:
wherein M represents the total number of the manufacturers of the electric meter, N represents the total number of the fault reasons, W represents the total number of units, min (M, N) represents the minimum value in M, N,
wherein, ω is1、ω2As a weighting factor, the initial value is 1.00,indicating electric meter manufacturerj number weight life cycle mean of all units of all electricity meters produced; omegaDiLife cycle impact weight in units of i; omegaFjMainErrorLife cycle impact weight representing the primary cause of failure for meter manufacturer j; omegaDiCorrelationErrorLife cycle influence weight representing the correlation fault cause of unit i;
and the prediction module is used for predicting the service life of the intelligent electric meter based on the service life prediction model of the intelligent electric meter.
In one embodiment, the method further comprises a prediction model optimization module for, after building the prediction model:
updating weight mode pair weight factor omega based on preset neural network algorithm1And ω2The correction is carried out so that the correction is carried out,
and optimizing the prediction model based on the corrected weight factor.
In one embodiment, the system further comprises an early warning setting module, specifically configured to:
obtaining a predicted life cycle and an average error rate according to a prediction result obtained through the prediction model;
acquiring early warning time corresponding to the intelligent ammeter based on the predicted life cycle and the average error rate;
and setting early warning reminding according to the early warning time.
In one embodiment, the weight parameters in the model building moduleωDi、ωFjMainErrorAnd ωDiCorrelationErrorThe calculation method of (c) is as follows:
is calculated byWherein,record, maxN, representing the h-th meter of the manufacturer j that manufactured the smart meterFjRepresents the maximum number of records of the meter factory j, maxError represents the maximum fault rate allowed by a single batch, the default value is 2 percent, and ErrorFjhH-th record, T, representing the number of faulty meters of meter factory jErrorTimeIndicating the time of failure, T, of the watchInstallTimeRepresents the initial installation time of the electricity meter,indicating the number of plant j fault tables;
ωDiis calculated byWherein, TDiThe life cycle in units of i is expressed,
denotes the average life cycle, T, of all unitsDiThe calculation method is as follows:
wherein,the number of the a-th electric meters in the unit i is represented; max NiaThe maximum value of the number of records in the unit i,the a-th life cycle record representing the unit i,the calculation method comprises the following steps: represents the life cycle average of all meters used in unit i, sum represents the total number of units;
ωFjMainErroris calculated by Number weighted life cycle mean, T, representing the major cause of failure for meter manufacturer jallThe number representing all the causes of failure weights the mean of the life cycle averages,the calculation method comprises the following steps:max NFjMainErrorthe maximum number of records of life cycle of the main cause of failure K of meter manufacturer j,a kth record of the number of meters indicating the main causes of failure for meter manufacturer j;
ωDiCorrelationErrorthe calculation method of (2) is as follows:wherein,the quantity representing all the causes of failure is weighted by the life cycle average,NAllErrorthe maximum number of categories representing the cause of the fault,the number representing the cause of the fault i, weights the life cycle, wherein,number weighted life cycle mean value representing correlation fault cause of meter manufacturer jmax N represents the maximum number of records of the correlation failure cause of the manufacturer j,the ith cycle representing the cause of the fault j is rotated.
Based on the same inventive concept, a third aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed, performs the method of the first aspect.
One or more technical solutions in the embodiments of the present application have at least one or more of the following technical effects:
the invention provides a life cycle prediction method of an intelligent electric meter based on correlation coefficient analysis. The invention provides a life cycle prediction method of an intelligent ammeter based on correlation coefficient analysis, which can obtain a prediction model of the life cycle of the ammeter according to unit influence factors, manufacturer influence factors and fault type influence factors as main parameters and by taking a neural network algorithm as parameter training assistance.
Furthermore, a heuristic method is adopted to iteratively search and determine the weight, and an updating weight mode based on a neural network algorithm is used for correction, so that the constructed prediction model can be optimized, and the prediction accuracy is further improved.
Furthermore, the invention also provides an early warning mechanism, and when the intelligent ammeter is actually applied, a power management department can utilize the model to carry out regular sampling inspection on the intelligent ammeter when the service life cycle of the intelligent ammeter is close.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flowchart of a life cycle prediction method for an intelligent electric meter based on correlation coefficient analysis according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for determining weights by iterative search using a heuristic approach based on neural networks;
FIG. 3 is a block diagram of an apparatus for predicting life cycle of a smart meter based on correlation coefficient analysis according to an embodiment of the present invention;
fig. 4 is a block diagram of a computer-readable storage medium according to an embodiment of the present invention.
Detailed Description
The invention aims to provide a method and a device for predicting the life cycle of an intelligent electric meter based on correlation coefficient analysis, aiming at the problems of reliability of the intelligent electric meter caused by long-term unattended operation and the problem that the life cycle of the intelligent electric meter cannot be accurately predicted by the conventional method.
In order to achieve the above object, the main concept of the present invention is as follows: by applying the basic principle of data mining, the periodic correlation among three factors of a unit, an ammeter manufacturer and a fault reason is calculated and verified, a prediction model based on relevant parameter analysis is constructed, and the service life of the intelligent ammeter is predicted by using the prediction model.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
The embodiment provides a smart meter life cycle prediction method based on correlation coefficient analysis, please refer to fig. 1, and the method includes:
step S1 is first executed: calculating and verifying the periodic correlation among three factors of a unit, an electric meter manufacturer and a fault reason, wherein the periodic correlation specifically comprises the following steps:the μ 'and σ' respectively represent the average value and the variance of the life of each electric meter corresponding to the same electric meter manufacturer in the same unit with different fault causes.
Specifically, correlation coefficient analysis is a statistical analysis method for studying the correlation between two or more equally positioned random variables. For example, between the height and weight of a person; the correlation between the relative humidity in the air and the rainfall is a problem of relevant analytical research. Correlation analysis can be applied to various fields such as industry and agriculture, hydrology, meteorology, social economy, biology and the like by finding various correlation characteristics among random variables.
In a specific implementation process, the method utilizes the 'operating fault information data' of the intelligent electric meters of each power bureau provided by a certain unit, and analyzes the performance difference factors of the meters caused by the difference of the 'unit using the intelligent electric meters' and the 'electric meter manufacturer'. And the design and manufacturing process are different according to different manufacturers. The "fault type" can express the difference of possible element components of the intelligent electric meter. The "unit using the electricity meter" may cause differences in operating environments such as geographical environment, temperature, humidity, and the like, and differences in installation and maintenance conditions.
Then, step S2 is executed: according to the period correlation, a life cycle prediction model of the intelligent electric meter is constructed, and the prediction model specifically comprises the following steps:
wherein M represents the total number of the manufacturers of the electric meter, N represents the total number of the fault reasons, W represents the total number of units, min (M, N) represents the minimum value in M, N,
otherwise
Wherein, ω is1、ω2As a weighting factor, the initial value is 1.00,representing the number weighted life cycle average of all meters produced by meter manufacturer j in all units; omegaDiLife cycle impact weight in units of i; omegaFjMainErrorLife cycle impact weight representing the primary cause of failure for meter manufacturer j; omegaDiCorrelationErrorLife cycle impact weight, T, representing the cause of a correlation failure in unit iDiFjFor the weight-influencing parameters, it is, in general,when only the primary cause of the fault is considered,
specifically, the step is to initialize a prediction model of the life cycle of the smart meter and calculate various weight parameters influencing the life cycle.
Specifically, the weight parameter in step S2ωDi、ωFjMainErrorAnd ωDiCorrelationErrorThe calculation method of (c) is as follows:
is calculated byWherein,record, maxN, representing the h-th meter of the manufacturer j that manufactured the smart meterFjRepresents the maximum number of records for Table factory j, maxError represents a single batchMaximum allowable failure rate, default 2%, ErrorFjhH-th record, T, representing the number of faulty meters of meter factory jErrorTimeIndicating the time of failure, T, of the watchInstallTimeRepresents the initial installation time of the electricity meter,indicating the number of plant j fault tables;
ωDiis calculated byWherein, TDiThe life cycle in units of i is expressed,
denotes the average life cycle, T, of all unitsDiThe calculation method is as follows:
wherein,the number of the a-th electric meters in the unit i is represented; max NiaThe maximum value of the number of records in the unit i,the a-th life cycle record representing the unit i,the calculation method comprises the following steps: life cycle of all meters expressing usage of unit iMean of the periods, sum represents the total number of units;
ωFjMainErroris calculated by Number weighted life cycle mean, T, representing the major cause of failure for meter manufacturer jallThe number representing all the causes of failure weights the mean of the life cycle averages,the calculation method comprises the following steps:max NFjMainErrorthe maximum number of records of life cycle of the main cause of failure K of meter manufacturer j,a kth record of the number of meters indicating the main causes of failure for meter manufacturer j;
ωDiCorrelationErrorthe calculation method of (2) is as follows:wherein,the quantity representing all the causes of failure is weighted by the life cycle average,NAllErrorthe maximum number of categories representing the cause of the fault,the number representing the cause of the fault i, weights the life cycle, wherein,number weighted life cycle mean value representing correlation fault cause of meter manufacturer jmax N represents the maximum number of records of the correlation failure cause of the manufacturer j,the ith cycle representing the cause of the fault j is rotated.
Step S3 is executed next: and predicting the life cycle of the intelligent electric meter based on the life cycle prediction model of the intelligent electric meter.
Specifically, the life cycle of the smart meter can be predicted through the constructed prediction models mu 'and sigma', and the prediction result is the life cycle of the smart meter, such as 5 years, 10 years and the like.
To improve the accuracy of the prediction, in one embodiment, after step S2, the method further comprises:
updating weight mode pair weight factor omega based on preset neural network algorithm1And ω2The correction is carried out so that the correction is carried out,
and optimizing the prediction model based on the corrected weight factor.
In particular, a heuristic method may be employed to iteratively search for the determined weight factors based on the neural network.
In one embodiment, the life cycle data of the smart meter of a real unit i-meter manufacturer j is predefinedAs reference data, the weight factor omega is subjected to weight updating based on the updating weight mode of the preset neural network algorithm1And ω2The correction specifically includes:
step S4.1: randomly extracting 2/3 electric meters from all the pre-obtained intelligent electric meter information records to serve as a training set, then randomly extracting 2/3 electric meters from the training set to serve as a training sample set, and taking the rest 1/3 of all the intelligent electric meter information records to serve as a reference set, wherein the number of individuals in the reference set is A and is a positive integer;
step S4.2: using formulas in predictive modelsAnd
performing prediction by taking the prediction object as an individual in the reference set to obtain a predicted value
Step S4.3: determining whether the absolute value of the difference between the predicted value and the reference value is less than a threshold value delta, i.e.If yes, go to step S4.4;
step S4.4: judgment ofIf true, then ω is updated2，If not, update ω1，
Step S4.5: and judging whether each individual in the reference set is predicted to be finished or not, if so, finishing, and if not, continuing iteration and executing the step S4.2.
Specifically, referring to fig. 2, a flowchart of a method for determining weights by using heuristic iterative search based on a neural network is shown, where life cycle data of a smart meter of a real unit i-meter manufacturer j is predefinedAs reference data. In specific implementation, an individual variable X may be set, the initial value is 1, the corresponding individual is used as a prediction object, and then through iteration, every time an individual is calculated, X is added by 1, until all the individuals in the reference set are calculated, the iteration is completed. In the iterative process, the updating condition and the updating mode of the specific weight factor are included. Through the updated weight factors, the prediction model can be optimized, and the prediction effect is further improved. In addition, after the weight factor is updated, the periodic correlation is further verified, and a fuzzy early warning function is added, so that the practicability of the method is improved.
In one embodiment, the method further comprises:
obtaining a predicted life cycle and an average error rate according to a prediction result obtained through the prediction model;
acquiring early warning time corresponding to the intelligent ammeter based on the predicted life cycle and the average error rate;
and setting early warning reminding according to the early warning time.
In particular, in order to make the prediction model in the present invention better applicable, it can be realized by: in practical application, fuzzy early warning can be added, the predicted life cycle is set as A, the average error rate is set as B, and when the early warning time of the intelligent ammeter is T-A-B. When the automatic early warning is reminded, the intelligent electric meter is subjected to spot check. If the qualification rate of the spot check reaches 95%, the current batch of the electric meters are continuously used, otherwise, the rotation processing is carried out.
The invention provides a life cycle prediction method of an intelligent electric meter based on correlation coefficient analysis. The invention provides a life cycle prediction method of an intelligent ammeter based on correlation coefficient analysis, which can obtain a prediction model of the life cycle of the ammeter according to unit influence factors, manufacturer influence factors and fault type influence factors as main parameters and by taking a neural network algorithm as parameter training assistance.
Based on the same inventive concept, the application also provides a device corresponding to the method for predicting the life cycle of the smart meter based on the correlation coefficient analysis in the first embodiment, which is detailed in the second embodiment.
Example two
The present embodiment provides a smart meter life cycle prediction method and device based on correlation coefficient analysis, please refer to fig. 3, the device includes:
the calculation and verification module 301 is configured to calculate and verify a periodic correlation among three factors, namely a unit, an electric meter manufacturer, and a fault reason, where the periodic correlation specifically includes:mu 'and sigma' respectively represent the average value and the variance of the service life of each corresponding ammeter in the same unit with different fault reasons of the same ammeter manufacturer;
the model building module 302 is configured to build a life cycle prediction model of the smart meter according to the cycle correlation, where the prediction model specifically includes:
wherein M represents the total number of the manufacturers of the electric meter, N represents the total number of the fault reasons, W represents the total number of units, min (M, N) represents the minimum value in M, N,
otherwise
Wherein, ω is1、ω2As a weighting factor, the initial value is 1.00,representing the number weighted life cycle average of all meters produced by meter manufacturer j in all units; omegaDiLife cycle impact weight in units of i; omegaFjMainErrorLife cycle impact weight representing the primary cause of failure for meter manufacturer j; omegaDiCorrelationErrorLife cycle influence weight representing the correlation fault cause of unit i;
and the predicting module 303 is used for predicting the life cycle of the intelligent electric meter based on the life cycle prediction model of the intelligent electric meter.
In one embodiment, the method further comprises a prediction model optimization module for, after building the prediction model:
updating weight mode pair weight factor omega based on preset neural network algorithm1And ω2The correction is carried out so that the correction is carried out,
and optimizing the prediction model based on the corrected weight factor.
In one embodiment, the system further comprises an early warning setting module, specifically configured to:
obtaining a predicted life cycle and an average error rate according to a prediction result obtained through the prediction model;
acquiring early warning time corresponding to the intelligent ammeter based on the predicted life cycle and the average error rate;
and setting early warning reminding according to the early warning time.
In one embodiment, the weight parameters in the model building moduleωDi、ωFjMainErrorAnd ωDiCorrelationErrorThe calculation method of (c) is as follows:
is calculated byWherein,record, maxN, representing the h-th meter of the manufacturer j that manufactured the smart meterFjRepresents the maximum number of records of the meter factory j, maxError represents the maximum fault rate allowed by a single batch, the default value is 2 percent, and ErrorFjhH-th record, T, representing the number of faulty meters of meter factory jErrorTimeIndicating the time of failure, T, of the watchInstallTimeRepresents the initial installation time of the electricity meter,indicating the number of plant j fault tables;
ωDiis calculated byWherein, TDiThe life cycle in units of i is expressed,
denotes the average life cycle, T, of all unitsDiThe calculation method is as follows:
wherein,the number of the a-th electric meters in the unit i is represented; max NiaThe maximum value of the number of records in the unit i,the a-th life cycle record representing the unit i,the calculation method comprises the following steps: represents the life cycle average of all meters used in unit i, sum represents the total number of units;
ωFjMainErroris calculated by Number weighted life cycle mean, T, representing the major cause of failure for meter manufacturer jallThe number representing all the causes of failure weights the mean of the life cycle averages,the calculation method comprises the following steps:max NFjMainErrorthe maximum number of records of life cycle of the main cause of failure K of meter manufacturer j,a kth record of the number of meters indicating the main causes of failure for meter manufacturer j;
ωDiCorrelationErrorthe calculation method of (2) is as follows:wherein,the quantity representing all the causes of failure is weighted by the life cycle average,NAllErrorthe maximum number of categories representing the cause of the fault,the number representing the cause of the fault i, weights the life cycle, wherein,number weighted life cycle mean value representing correlation fault cause of meter manufacturer jmax N represents the maximum number of records of the correlation failure cause of the manufacturer j,the ith cycle representing the cause of the fault j is rotated.
Since the device described in the second embodiment of the present invention is a device used for predicting the life cycle of the smart meter based on the correlation coefficient analysis in the first embodiment of the present invention, a person skilled in the art can understand the specific structure and deformation of the device based on the method described in the first embodiment of the present invention, and thus the details are not described herein. All the devices adopted in the method of the first embodiment of the present invention belong to the protection scope of the present invention.
EXAMPLE III
Based on the same inventive concept, the present application further provides a computer-readable storage medium 400, please refer to fig. 4, on which a computer program 411 is stored, which when executed implements the method in the first embodiment.
Since the computer-readable storage medium introduced in the third embodiment of the present invention is a computer-readable storage medium used for implementing the correlation coefficient analysis-based smart meter life cycle prediction in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, persons skilled in the art can understand the specific structure and deformation of the computer-readable storage medium, and thus details are not described herein. The computer readable storage medium used in the method according to the embodiment of the present invention is within the scope of the present invention.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made in the embodiments of the present invention without departing from the spirit or scope of the embodiments of the invention. Thus, if such modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to encompass such modifications and variations.
Claims (10)
1. The intelligent ammeter life cycle prediction method based on correlation coefficient analysis is characterized by comprising the following steps:
step S1: calculating and verifying the periodic correlation among three factors of a unit, an electric meter manufacturer and a fault reason, wherein the periodic correlation specifically comprises the following steps:mu 'and sigma' respectively represent the unit of the same electric meter manufacturer with different fault reasons, corresponding to each electric meter service lifeMean and variance;
step S2: according to the period correlation, a life cycle prediction model of the intelligent electric meter is constructed, and the prediction model specifically comprises the following steps:
wherein M represents the total number of the manufacturers of the electric meter, N represents the total number of the fault reasons, W represents the total number of units, min (M, N) represents the minimum value in M, N,
wherein, ω is1、ω2As a weighting factor, the initial value is 1.00,representing the number weighted life cycle average of all meters produced by meter manufacturer j in all units; omegaDiLife cycle impact weight in units of i; omegaFjMainErrorLife cycle impact weight representing the primary cause of failure for meter manufacturer j; omegaDiCorrelationErrorLife cycle influence weight representing the correlation fault cause of unit i;
step S3: and predicting the life cycle of the intelligent electric meter based on the life cycle prediction model of the intelligent electric meter.
2. The method of claim 1, wherein after step S2, the method further comprises:
updating weight mode pair weight factor omega based on preset neural network algorithm1And ω2The correction is carried out so that the correction is carried out,
and optimizing the prediction model based on the corrected weight factor.
3. The method of claim 1, wherein the method further comprises:
obtaining a predicted life cycle and an average error rate according to a prediction result obtained through the prediction model;
acquiring early warning time corresponding to the intelligent ammeter based on the predicted life cycle and the average error rate;
and setting early warning reminding according to the early warning time.
4. The method of claim 1, wherein the weight parameter in step S2ωDi、ωFjMainErrorAnd ωDiCorrelationErrorThe calculation method of (c) is as follows:
is calculated by
Wherein,record, max N, representing the h meter of manufacturer jFjRepresents the maximum number of records of the meter factory j, max Error represents the maximum fault rate allowed by a single batch, the default value is 2 percent, and the Error FjhH-th record, T, representing the number of faulty meters of meter factory jErrorTimeIndicating the time of failure, T, of the watchInstallTimeRepresents the initial installation time of the electricity meter,indicating the number of plant j fault tables;
ωDiis calculated byWherein, TDiThe life cycle in units of i is expressed,
denotes the average life cycle, T, of all unitsDiThe calculation method is as follows:
wherein,the number of the a-th electric meters in the unit i is represented; max NiaThe maximum value of the number of records in the unit i,the a-th life cycle record representing the unit i,the calculation method comprises the following steps: represents the life cycle average of all meters used in unit i, sum represents the total number of units;
ωFjMainErroris calculated by Number weighted life cycle mean, T, representing the major cause of failure for meter manufacturer jallThe number representing all the causes of failure weights the mean of the life cycle averages,the calculation method comprises the following steps:max NFjMainErrorthe maximum number of records of life cycle of the main cause of failure K of meter manufacturer j,a kth record of the number of meters indicating the main causes of failure for meter manufacturer j;
ωDiCorrelationErrorthe calculation method of (2) is as follows:wherein,the quantity representing all the causes of failure is weighted by the life cycle average,NAllErrorthe maximum number of categories representing the cause of the fault,the number representing the cause of the fault i, weights the life cycle, wherein,number weighted life cycle mean value representing correlation fault cause of meter manufacturer jmax N represents the maximum number of records of the correlation failure cause of the manufacturer j,the ith cycle representing the cause of the fault j is rotated.
5. The method of claim 4, wherein a true is predefinedReal smart meter life cycle data of unit i meter manufacturer jAs reference data, the weight factor omega is subjected to weight updating based on the updating weight mode of the preset neural network algorithm1And ω2The correction specifically includes:
step S4.1: randomly extracting 2/3 electric meters from all the pre-obtained intelligent electric meter information records to serve as a training set, then randomly extracting 2/3 electric meters from the training set to serve as a training sample set, and taking the rest 1/3 of all the intelligent electric meter information records to serve as a reference set, wherein the number of individuals in the reference set is A and is a positive integer;
step S4.2: using formulas in predictive modelsAndperforming prediction by taking the prediction object as an individual in the reference set to obtain a predicted value
Step S4.3: determining whether the absolute value of the difference between the predicted value and the reference value is less than a threshold value delta, i.e.If yes, go to step S4.4;
step S4.4: judgment ofIf true, then ω is updated2，If not, update ω1，
Step S4.5: and judging whether each individual in the reference set is predicted to be finished or not, if so, finishing, and if not, continuing iteration and executing the step S4.2.
6. Smart electric meter life cycle prediction device based on correlation coefficient analysis, its characterized in that includes:
the calculation and verification module is used for calculating and verifying the periodic correlation among three factors, namely a unit, an electric meter manufacturer and a fault reason, wherein the periodic correlation specifically comprises the following steps:mu 'and sigma' respectively represent the average value and the variance of the service life of each corresponding ammeter in the same unit with different fault reasons of the same ammeter manufacturer;
the model building module is used for building a life cycle prediction model of the intelligent electric meter according to the cycle correlation, and the prediction model specifically comprises the following steps:
wherein M represents the total number of the manufacturers of the electric meter, N represents the total number of the fault reasons, W represents the total number of units, min (M, N) represents the minimum value in M, N,
wherein, ω is1、ω2As a weighting factor, the initial value is 1.00,representing the number weighted life cycle average of all meters produced by meter manufacturer j in all units; omegaDiLife cycle impact weight in units of i; omegaFjMainErrorLife cycle impact weight representing the primary cause of failure for meter manufacturer j; omegaDiCorrelationErrorLife cycle shadow representing correlation fault cause of unit iA sound weight;
and the prediction module is used for predicting the service life of the intelligent electric meter based on the service life prediction model of the intelligent electric meter.
7. The apparatus of claim 6, further comprising a predictive model optimization module to, after building the predictive model:
updating weight mode pair weight factor omega based on preset neural network algorithm1And ω2The correction is carried out so that the correction is carried out,
and optimizing the prediction model based on the corrected weight factor.
8. The method of claim 6, further comprising an early warning setup module specifically configured to:
obtaining a predicted life cycle and an average error rate according to a prediction result obtained through the prediction model;
acquiring early warning time corresponding to the intelligent ammeter based on the predicted life cycle and the average error rate;
and setting early warning reminding according to the early warning time.
9. The apparatus of claim 8, in which the weight parameters in the model building moduleωDi、ωFjMainErrorAnd ωDiCorrelationErrorThe calculation method of (c) is as follows:
is calculated byWherein,record, maxN, representing the h-th meter of the manufacturer j that manufactured the smart meterFjRepresents the maximum number of records of the meter factory j, maxError represents the maximum fault rate allowed by a single batch, the default value is 2 percent, and ErrorFjhH-th record, T, representing the number of faulty meters of meter factory jErrorTimeIndicating the time of failure, T, of the watchInstallTimeRepresents the initial installation time of the electricity meter,indicating the number of plant j fault tables;
ωDiis calculated byWherein, TDiThe life cycle in units of i is expressed,
denotes the average life cycle, T, of all unitsDiThe calculation method is as follows:
wherein,the number of the a-th electric meters in the unit i is represented; max NiaThe maximum value of the number of records in the unit i,the a-th life cycle record representing the unit i,the calculation method comprises the following steps: represents the life cycle average of all meters used in unit i, sum represents the total number of units;
ωFjMainErroris calculated by Number weighted life cycle mean, T, representing the major cause of failure for meter manufacturer jallThe number representing all the causes of failure weights the mean of the life cycle averages,the calculation method comprises the following steps:max NFjMainErrorthe maximum number of records of life cycle of the main cause of failure K of meter manufacturer j,a kth record of the number of meters indicating the main causes of failure for meter manufacturer j;
ωDiCorrelationErrorthe calculation method of (2) is as follows:wherein,the quantity representing all the causes of failure is weighted by the life cycle average,NAllErrorthe maximum number of categories representing the cause of the fault,the number representing the cause of the fault i, weights the life cycle, wherein,number weighted life cycle mean value representing correlation fault cause of meter manufacturer jmax N represents the maximum number of records of the correlation failure cause of the manufacturer j,the ith cycle representing the cause of the fault j is rotated.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed, implements the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811440265.8A CN109598052B (en) | 2018-11-29 | 2018-11-29 | Intelligent ammeter life cycle prediction method and device based on correlation coefficient analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811440265.8A CN109598052B (en) | 2018-11-29 | 2018-11-29 | Intelligent ammeter life cycle prediction method and device based on correlation coefficient analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109598052A true CN109598052A (en) | 2019-04-09 |
| CN109598052B CN109598052B (en) | 2022-07-05 |
Family
ID=65959833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811440265.8A Active CN109598052B (en) | 2018-11-29 | 2018-11-29 | Intelligent ammeter life cycle prediction method and device based on correlation coefficient analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109598052B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110261811A (en) * | 2019-07-05 | 2019-09-20 | 北京志翔科技股份有限公司 | Intelligent electric meter batch method for early warning and system |
| CN110287640A (en) * | 2019-07-03 | 2019-09-27 | 辽宁艾特斯智能交通技术有限公司 | Life prediction method, apparatus, storage medium and the electronic equipment of lighting apparatus |
| CN110967695A (en) * | 2019-10-28 | 2020-04-07 | 兰州大方电子有限责任公司 | Radar echo extrapolation short-term prediction method based on deep learning |
| CN115358347A (en) * | 2022-09-30 | 2022-11-18 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708306A (en) * | 2012-06-19 | 2012-10-03 | 华北电网有限公司计量中心 | Prediction method for q-precentile life of intelligent meter |
| CN103018706A (en) * | 2012-11-16 | 2013-04-03 | 浙江省电力公司电力科学研究院 | Method and system for forecasting rotation cycle of intelligent electric meter |
| CN103383445A (en) * | 2013-07-16 | 2013-11-06 | 湖北省电力公司电力科学研究院 | System and method for forecasting service life and reliability of intelligent electric meter |
| CN105022019A (en) * | 2015-06-23 | 2015-11-04 | 国家电网公司 | Method of comprehensively estimating reliability of single-phase intelligent electric energy meter |
| CN106054105A (en) * | 2016-05-20 | 2016-10-26 | 国网新疆电力公司电力科学研究院 | Intelligent ammeter reliability prediction correction model building method |
| KR20160140474A (en) * | 2015-05-27 | 2016-12-07 | 내셔날 쳉쿵 유니버시티 | Metrology sampling method with sampling rate decision scheme and computer program product thereof |
| CN108680890A (en) * | 2018-08-23 | 2018-10-19 | 重庆市计量质量检测研究院 | Intelligent electric energy meter life characteristics detection method |
| CN109344967A (en) * | 2018-08-31 | 2019-02-15 | 武汉大学 | A kind of intelligent electric meter life cycle prediction technique based on artificial neural network |
-
2018
- 2018-11-29 CN CN201811440265.8A patent/CN109598052B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708306A (en) * | 2012-06-19 | 2012-10-03 | 华北电网有限公司计量中心 | Prediction method for q-precentile life of intelligent meter |
| CN103018706A (en) * | 2012-11-16 | 2013-04-03 | 浙江省电力公司电力科学研究院 | Method and system for forecasting rotation cycle of intelligent electric meter |
| CN103383445A (en) * | 2013-07-16 | 2013-11-06 | 湖北省电力公司电力科学研究院 | System and method for forecasting service life and reliability of intelligent electric meter |
| KR20160140474A (en) * | 2015-05-27 | 2016-12-07 | 내셔날 쳉쿵 유니버시티 | Metrology sampling method with sampling rate decision scheme and computer program product thereof |
| CN105022019A (en) * | 2015-06-23 | 2015-11-04 | 国家电网公司 | Method of comprehensively estimating reliability of single-phase intelligent electric energy meter |
| CN106054105A (en) * | 2016-05-20 | 2016-10-26 | 国网新疆电力公司电力科学研究院 | Intelligent ammeter reliability prediction correction model building method |
| CN108680890A (en) * | 2018-08-23 | 2018-10-19 | 重庆市计量质量检测研究院 | Intelligent electric energy meter life characteristics detection method |
| CN109344967A (en) * | 2018-08-31 | 2019-02-15 | 武汉大学 | A kind of intelligent electric meter life cycle prediction technique based on artificial neural network |
Non-Patent Citations (2)
| Title |
|---|
| 张强等: "智能电能表全生命周期质量跟踪策略探讨", 《东北电力技术》 * |
| 贺宁: "智能电表故障大数据分析探究", 《中小企业管理与科技(上旬刊)》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287640A (en) * | 2019-07-03 | 2019-09-27 | 辽宁艾特斯智能交通技术有限公司 | Life prediction method, apparatus, storage medium and the electronic equipment of lighting apparatus |
| CN110287640B (en) * | 2019-07-03 | 2023-10-13 | 辽宁艾特斯智能交通技术有限公司 | Lighting equipment service life prediction method and device, storage medium and electronic equipment |
| CN110261811A (en) * | 2019-07-05 | 2019-09-20 | 北京志翔科技股份有限公司 | Intelligent electric meter batch method for early warning and system |
| CN110967695A (en) * | 2019-10-28 | 2020-04-07 | 兰州大方电子有限责任公司 | Radar echo extrapolation short-term prediction method based on deep learning |
| CN115358347A (en) * | 2022-09-30 | 2022-11-18 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
| CN115358347B (en) * | 2022-09-30 | 2023-01-31 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109598052B (en) | 2022-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109598052B (en) | Intelligent ammeter life cycle prediction method and device based on correlation coefficient analysis | |
| JP2021064370A (en) | Method and system for semi-supervised deep abnormality detection for large-scale industrial monitoring system based on time-series data utilizing digital twin simulation data | |
| CN107506868B (en) | Method and device for predicting short-time power load | |
| CN107730127B (en) | Relay storage degradation data prediction method based on output characteristic initial distribution | |
| CN114297036B (en) | Data processing method, device, electronic equipment and readable storage medium | |
| CN109344967B (en) | Intelligent electric meter life cycle prediction method based on artificial neural network | |
| KR20200008621A (en) | Electrical Equipment Safety Evaluation System Using Artificial Intelligence Technique | |
| CN110083518B (en) | AdaBoost-Elman-based virtual machine software aging prediction method | |
| CN107992991A (en) | Annual electricity sales amount Forecasting Methodology based on external environmental factor and Co-integration Theory | |
| KR20190123040A (en) | Solar power generation prediction model management APPARATUS AND METHOD for setting the renewal cycle of the deep run-based solar power generation prediction model | |
| KR102059472B1 (en) | A System and Method for Prediction of Geomagnetic Disturbance Strength based on Solar Coronal Hole Information | |
| Dong et al. | Short-term building cooling load prediction model based on DwdAdam-ILSTM algorithm: A case study of a commercial building | |
| CN111178585A (en) | Fault reporting amount prediction method based on multi-algorithm model fusion | |
| KR20170106686A (en) | Method and apparatus for predicting home energy usage | |
| CN113988210A (en) | Method and device for restoring distorted data of structure monitoring sensor network and storage medium | |
| CN116930609A (en) | Electric energy metering error analysis method based on ResNet-LSTM model | |
| CN116388150A (en) | Matching short-term load prediction method based on digital twin and neural network technology | |
| CN117113852A (en) | GIS life prediction method based on electric, chemical, structural and vibration parameters | |
| CN117408394B (en) | Carbon emission factor prediction method and device for electric power system and electronic equipment | |
| Ghassemi et al. | Optimal surrogate and neural network modeling for day-ahead forecasting of the hourly energy consumption of university buildings | |
| CN114372558A (en) | Residential electricity load prediction method, medium and equipment based on multi-model fusion | |
| CN115563848A (en) | Distributed photovoltaic total radiation prediction method and system based on deep learning | |
| CN109190749A (en) | A kind of prediction technique and device for the intelligent electric meter service life | |
| CN113159395A (en) | Deep learning-based sewage treatment plant water inflow prediction method and system | |
| CN116960962A (en) | Mid-long term area load prediction method for cross-area data fusion |
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 |