CN106054104B - A kind of intelligent electric meter failure real-time predicting method based on decision tree - Google Patents

A kind of intelligent electric meter failure real-time predicting method based on decision tree Download PDF

Info

Publication number
CN106054104B
CN106054104B CN201610342042.2A CN201610342042A CN106054104B CN 106054104 B CN106054104 B CN 106054104B CN 201610342042 A CN201610342042 A CN 201610342042A CN 106054104 B CN106054104 B CN 106054104B
Authority
CN
China
Prior art keywords
electricity
failure
ammeter
formula
intelligent electric
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
Application number
CN201610342042.2A
Other languages
Chinese (zh)
Other versions
CN106054104A (en
Inventor
李宁
袁铁江
杨金成
蒋平
王刚
董小顺
罗庆
李国军
薛飞
段志尚
山宪武
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Marketing service center of State Grid Xinjiang Electric Power Co., Ltd. (capital intensive center, metering center)
Xinjiang University
Original Assignee
Xinjiang University
Electric Power Research Institute of State Grid Xinjiang Electric Power Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xinjiang University, Electric Power Research Institute of State Grid Xinjiang Electric Power Co Ltd filed Critical Xinjiang University
Priority to CN201610342042.2A priority Critical patent/CN106054104B/en
Publication of CN106054104A publication Critical patent/CN106054104A/en
Application granted granted Critical
Publication of CN106054104B publication Critical patent/CN106054104B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R35/00Testing or calibrating of apparatus covered by the other groups of this subclass
    • G01R35/04Testing or calibrating of apparatus covered by the other groups of this subclass of instruments for measuring time integral of power or current

Abstract

A kind of intelligent electric meter failure real-time predicting method based on decision tree, step are as follows: 1, to the intelligent electric meter data prediction of power information acquisition system;2, according to intelligent electric meter breakdown judge model, the faulty data of intelligent electric meter of power information acquisition system are screened into intelligent electric meter Mishap Database;3, the historical data in intelligent electric meter Mishap Database is divided into training set and test set, data mining is carried out to training set using decision Tree algorithms, forms intelligent electric meter failure decision tree and preliminary classification rule;4, accuracy assessment is carried out to preliminary classification rule by the data of test set, if accuracy is met the requirements, it is determined that classifying rules returns to training set, re -training if being unsatisfactory for requiring;5, intelligent electric meter failure real-time prediction model is generated by the classifying rules finally determined;6, intelligent electric meter real time fail data basd link is connected to intelligent electric meter failure real-time prediction model to be predicted in real time, obtains the real-time prediction result of intelligent electric meter failure.

Description

A kind of intelligent electric meter failure real-time predicting method based on decision tree
Technical field
The present invention relates to a kind of intelligent electric meter failure prediction method, in particular to a kind of intelligent electric meter event based on decision tree Hinder real-time predicting method.
Background technique
With the deep propulsion of State Grid Corporation of China " three collection five are big " System Construction, the power information that " big marketing " is promoted is adopted Collecting system construction has had scale, especially intelligent electric energy meter and has obtained large-scale popularization and application.The intelligence of such vast number Ammeter directly influences the vital interests of user and the safety and stability of society, therefore predict intelligence in time once breaking down Electric meter fault and to repair or replace to it be highly important.
With the extensive use of intelligent electric meter, operation troubles shows sudden, versatility, complexity more and more Feature, the failure prediction method of traditional ammeter are no longer applicable in.Domestic and foreign scholars mainly study at present neural network, grey are pre- Survey and the prediction techniques such as expert system applied in each field, and achieve good effect, but still Shortcomings it Place.The mass data that power information acquisition system provides has given data mining technology one ample scope for abilities just, and in intelligent electricity Table failure predication field, the correlative study in terms of data mining technology are not yet deeply unfolded.
Summary of the invention
The purpose of the present invention is in view of the above-mentioned problems, proposing a kind of intelligent electric meter failure based on the decision tree side of prediction in real time Method, predicted in real time intelligent electric meter failure in the real reliable situation of now forecast accuracy.
In order to achieve the above objectives, the present invention is using technical solution:
Step 1, data prediction is carried out to the intelligent electric meter data of power information acquisition system;
Step 2, according to intelligent electric meter breakdown judge model, by the faulty number of the intelligent electric meter of power information acquisition system According to screening into intelligent electric meter Mishap Database;
The effect of the intelligent electric meter breakdown judge model is to judge whether intelligent electric meter breaks down.
Step 3, the historical data in intelligent electric meter Mishap Database is chosen, training set and test set are classified as, is used Decision Tree algorithms carry out data mining to training set, form intelligent electric meter failure decision tree, then form preliminary classification rule;
Step 4, accuracy assessment is carried out to preliminary classification rule by the data of test set, if accuracy is met the requirements, Then determine that classifying rules is back to training set if accuracy is unsatisfactory for requiring, and re-starts training;
Step 5, intelligent electric meter failure real-time prediction model is generated by the classifying rules finally determined;
Step 6, intelligent electric meter real time fail data basd link intelligent electric meter failure real-time prediction model is connected to carry out in real time Prediction, obtains the real-time prediction result of intelligent electric meter failure.
The step of step 1 carries out data prediction to the intelligent electric meter data of power information acquisition system is, first It first removes the attribute data unrelated with intelligent electric meter failure, have the data of apparent error and the duplicate data of attribute, it then will be each A attribute continuous data discretization.
The step 2 is faulty by the intelligent electric meter of power information acquisition system according to intelligent electric meter breakdown judge model Method of the data screening into intelligent electric meter Mishap Database it is as follows:
(1) when the not equal fault data screening of progress the sum of ammeter total electricity and each rate electricity, judged using following formula The failure that the sum of ammeter total electricity and each rate electricity do not wait:
In formula: W represents electricity ,+represent the direction of electricity as forward direction ,-represent the direction of electricity be it is reversed, i=1 is represented always Period, W+1Represent positive total electricity, W-1Reversed total electricity is represented, i=2 represents peak period, and i=3 represents usually section, i=4 Low-valley interval is represented, i=5 represents the spike period, and e is ammeter rate number, and ρ sentences for the sum of total electricity and each rate electricity be not equal The disconnected factor.As e=4, expression ammeter is four rate ammeters, takes ρ=0.4;As e=3, expression ammeter is three rate ammeters, Take ρ=0.3;As e=2, expression ammeter is two rate ammeters, takes ρ=0.2;
The four rate ammeters, which refer to, supports peak period, usually section, low-valley interval, spike period 4 period chargings Ammeter;Three rate ammeters refer to support only peak period, usually section, 3 period chargings of low-valley interval ammeter;Two rate ammeters refer to Only support the ammeter of 2 usually section, low-valley interval period chargings;
Judgment principle is as follows:
1. when forward and reverse total electricity of ammeter, forward and reverse peak period electricity, forward and reverse usually section electricity, forward and reverse low ebb Section electricity is all larger than 0, and is not sky;
2. the sum of forward and reverse each rate electricity of ammeter is greater than 0;
3. the absolute value of the sum of ammeter total electricity and each rate electricity difference is greater than some threshold values, threshold rule is as follows: such as Fruit is four rate ammeters, and by 0.4 judgement, three rate ammeters are by 0.3 judgement, and two rate ammeters are by 0.2 judgement;
1. 2. 3. 4. being judged as catastrophe failure if meeting judgment principle simultaneously;
(2) when progress electric energy meter flies away with mutation failure data screening, judge the failure journey that electric energy meter flies away and is mutated Sequence is as follows:
A. the same day maximum electricity W is calculated firstf:
In formula: WfFor same day maximum electricity;ImaxFor maximum current;IbFor fundamental current;
B. ammeter is then calculated to fly away and mutation factor K:
In formula: K flies away for ammeter and mutation factor, WtTo work as daily electricity;
Judgment principle is as follows:
1. being directed to resident's table, maximum current I is taken in 1,2,7,8,9, Decembermax, flown away and dashed forward by 12 hours calculating ammeters Variable factor K;3 times of fundamental current I are taken in 3,4,5,6,10, Novemberb, fly away and mutation factor K by 8 hours calculating ammeters;
3. judging that electric energy meter flies away and dashes forward using formula (4) when progress electric energy meter flies away with mutation failure data screening Accident barrier:
K≥1 (4)
3. being judged as catastrophe failure if meeting criterion 2.;
(3) when reversely active indicating value is greater than zero failure data screening to progress ammeter, judging ammeter reversely using following formula has Function indicating value is greater than zero failure:
In formula: P represents active power, and Q represents reactive power, PIt is totalRepresent ammeter reversely active general power, QIt is totalRepresent ammeter Reversed idle general power;
Judgment principle is as follows:
1. there is reversed active general power in electric energy meter or reversed idle general power is greater than 0;
2. being judged as catastrophe failure if meeting criterion 1.;
(4) when carrying out the screening of electric energy meter inverted walk fault data, the premise that electric energy meter inverted walk judges is exclusion copy reading ammeter Positive and negative total electricity is always empty record, and the failure of electric energy meter inverted walk is judged using following formula:
In formula: W+1yFor the positive total electricity of the previous day, W-1yFor the reversed total electricity of the previous day;
Judgment principle is as follows:
Judged according to daily electricity statistical form, for low pressure resident and single-phase industrial and commercial producer, only judges the previous day positive total electricity Whether amount and reversed total electricity are greater than the meter reading indicating value on the same day, if set up, are judged as catastrophe failure;
(5) when carrying out ammeter clock and not screening to fault data, the breakdown judge program of ammeter clock not pair is as follows:
It first determines whether clock synchronization errors number m, if clock synchronization errors number is greater than 3 times, is directly determined as catastrophe failure, That is:
M > 3 are serious (7)
In formula: m is clock synchronization errors number;
If m is not more than 3, when whether meeting the requirement of Δ t by on-line monitoring inquiry, and generating electric energy meter by the standard of Δ t The overproof grade of clock is then judged using formula (8):
In formula: Δ t is the difference of terminal and clock of power meter, and calculation method is shown in formula (9):
Δ t=| tTerminal-tAmmeter| (9)
In formula: tTerminalIndicate terminal clock, tAmmeterIndicate clock of power meter;
Judgment principle is as follows:
1. being directly judged as catastrophe failure if clock synchronization errors number m is more than 3 times;
2. being judged as generic failure as the difference Δ t of terminal and clock of power meter meets 5min≤Δ t < 15min;
If 15min≤Δ t < 30min is judged as important failure;If Δ t > 30min is judged as catastrophe failure;
(6) when carrying out ammeter electric energy rate setting abnormal failure data screening, ammeter electric energy rate is judged using following formula Abnormal failure is set:
W+5≠ 0 or W-5≠0 (10)
Judgment principle is as follows:
1. only judging the electric energy meter of DLT-2007 specification;
2. judging whether there is positive spike period electricity or reversed spike period electricity, and if it exists, be then judged as serious Failure;
(7) when carrying out the screening of electric energy meter forward direction shunt running fault data, the event of electric energy meter forward direction shunt running is judged using following formula Barrier:
Or
Judgment principle is as follows:
If meeting formula (11) 3 times, it is judged as important failure, if meeting formula (11) 5 times, is judged as serious event Barrier;
(8) when the reversed shunt running fault data screening of progress electric energy meter, the event of the reversed shunt running of electric energy meter is judged using following formula Barrier:
Judgment principle is as follows:
The case where electric energy meter forward direction total electricity and reversed total electricity for clearing class exist simultaneously and be greater than 0.1, then sentence Break as catastrophe failure.
The step 3 chooses the historical data in intelligent electric meter Mishap Database, is classified as training set and test set, Data mining is carried out to training set using decision Tree algorithms, forms intelligent electric meter failure decision tree, then forms preliminary classification rule Then, its step are as follows:
If R is the historical data of intelligent electric meter Mishap Database, and gives training set S for 60% data of R, the 40% of R Data give test set T.Data mining is carried out to training set S for using C5.0 algorithm.
Setting tool has the attribute X of n value that S is divided into n subset S1, S2..., SnIf the sum of sample is in S | S |, freq (Ci, S) is to belong to classification C in SiThe number of samples of (i=1,2 ..., N), some sample belongs to classification C in SiProbability beThe information that it is conveyed are as follows:
The entropy info (S) of training set S is indicated by formula (13);
After training set S is divided into n subset according to attribute X, the comentropy of each subset is calculated, then the expectation of set S Information infoX(S) it is indicated with formula (14);
In order to measure the obtained information of S for carrying out subregion according to inspection by attributes X, using gain standard gain (X), it is selected Selecting enables its value to maximize, i.e., the selection of this standard is to have the attribute of highest information gain to carry out every subzone;
Gain (X)=info (S)-infoX(S) (15)
S is divided into S according to n of attribute X different values1, S2..., SnThe potential information generated after total n subset Split_info (X) is shown in formula (16);
Formula (17) is that X carries out division information ratio of profit increase gain_ratio (X) to S;
It calculates for the first time using selecting the attribute of highest information gain-ratio as the root node of decision tree, then decision tree is every One node the method calculates the information gain-ratio of remaining attribute, throws away and selects the attribute of highest information gain-ratio as decision tree Present node, until ultimately forming entire decision tree;
Preliminary classification rule is formed by the decision tree.
The step 4 carries out accuracy assessment to preliminary classification rule by the data of test set, if accuracy meets It is required that, it is determined that classifying rules is back to training set if accuracy is unsatisfactory for requiring, and re-starts training, detailed step It is as follows:
(1) accuracy assesses formula are as follows:
Z is assessment accuracy in formula, and A is that preliminary classification rule predicts correct number in test set T;B is test set T Total data number;
(2) Z ' is set as the accuracy of setting;If Z >=Z ', it is determined that classifying rules;If Z < Z ', returns to training set again Training decision tree.
Intelligent electric meter failure real-time predicting method of the present invention can accurately reflect intelligent electric meter real time fail situation.
Detailed description of the invention
Fig. 1 is the General Implementing flow diagram of the method for the present invention.
Specific embodiment
With reference to the accompanying drawing and specific embodiment the present invention will be further described.
It is right by taking a collection of model DDZY102-Z intelligent electric meter is in the 2 years operation datas in In A Certain Place of Xinjiang area statistics as an example below The present invention is described further.
As shown in Figure 1, the present invention is based on the intelligent electric meter failure real-time predicting method of decision tree the following steps are included:
Step 1, data prediction is carried out to the DDZY102-Z intelligent electric meter data of power information acquisition system, step is such as Under:
(1) it removes the attribute data unrelated with intelligent electric meter failure, have the data of apparent error and the duplicate data of attribute;
(2) by each attribute continuous data discretization.
Step 2, according to DDZY102-Z intelligent electric meter breakdown judge model, by the intelligent electric meter of power information acquisition system Faulty data screening is into intelligent electric meter Mishap Database, and steps are as follows:
The effect of the intelligent electric meter breakdown judge model is to judge whether intelligent electric meter breaks down.
DDZY102-Z intelligent electric meter breakdown judge model is as follows:
(1) when the not equal fault data screening of progress the sum of ammeter total electricity and each rate electricity, because of DDZY102-Z intelligence Energy ammeter is four rate ammeters, and e=4, equal judgement factor ρ does not take 0.4 for the sum of total electricity and each rate electricity.Sentenced using following formula Power off the failure that the sum of table total electricity and each rate electricity do not wait:
Judgment principle is as follows:
1. when forward and reverse total electricity of ammeter, forward and reverse peak period electricity, forward and reverse usually section electricity, forward and reverse low ebb Section electricity is all larger than 0, and is not sky.
2. the sum of forward and reverse each rate electricity of ammeter is greater than 0;
3. the absolute value of the sum of ammeter total electricity and each rate electricity difference is greater than 0.4.
1. 2. 3. 4. being judged as catastrophe failure if meeting judgment principle simultaneously.
(2) when progress electric energy meter flies away with mutation failure data screening, judge the failure journey that electric energy meter flies away and is mutated Sequence is as follows:
A. the same day maximum electricity W is calculated firstf:
In formula: WfFor same day maximum electricity;ImaxFor maximum current;IbFor fundamental current.
B. ammeter is then calculated to fly away and mutation factor K:
In formula: K flies away for ammeter and mutation factor;WtTo work as daily electricity.
Judgment principle is as follows:
1. taking maximum current I in 1,2,7,8,9, December in two annual datas of DDZY102-Z intelligent electric metermax, It flies away and mutation factor K by 12 hours calculating ammeters;3 times of fundamental current I are taken in 3,4,5,6,10, Novemberb, based on 8 hours Ammeter is calculated to fly away and mutation factor K.
2. judging that electric energy meter flies away and dashes forward using formula (4) when progress electric energy meter flies away with mutation failure data screening Accident barrier:
K≥1 (4)
3. being judged as catastrophe failure if meeting criterion 2..
(3) when reversely active indicating value is greater than zero failure data screening to progress ammeter, judging ammeter reversely using following formula has Function indicating value is greater than zero failure:
In formula: P represents active power, and Q represents reactive power, PIt is totalRepresent ammeter reversely active general power, QIt is totalRepresent ammeter Reversed idle general power.
Judgment principle is as follows:
1. there is reversed active general power in electric energy meter or reversed idle general power is greater than 0.
2. being judged as catastrophe failure if meeting criterion 1..
(4) when carrying out the screening of electric energy meter inverted walk fault data, the premise that electric energy meter inverted walk judges is exclusion copy reading electric energy The forward and reverse active total record for sky of indicating value, the failure of electric energy meter inverted walk is judged using following formula:
In formula: W+1yFor the positive total electricity of the previous day, W-1yFor the reversed total electricity of the previous day.
Judgment principle is as follows:
Judged according to daily electricity statistical form, for low pressure resident and single-phase industrial and commercial producer, only judges the previous day positive total electricity Whether amount and reversed total electricity are greater than the meter reading indicating value on the same day, if set up, are judged as catastrophe failure.
(5) when carrying out ammeter clock and not screening to fault data, the breakdown judge program of ammeter clock not pair is as follows:
It first determines whether clock synchronization errors number m, if clock synchronization errors number is greater than 3 times, is directly determined as catastrophe failure, That is:
M > 3 are serious (7)
In formula: m is clock synchronization errors number.
If m is not more than 3, when whether meeting the requirement of Δ t by on-line monitoring inquiry, and generating electric energy meter by the standard of Δ t The overproof grade of clock is then judged using formula (8):
In formula: Δ t is the difference of terminal and clock of power meter, and calculation method is shown in formula (9):
Δ t=| tTerminal-tAmmeter| (9)
In formula: tTerminalIndicate terminal clock, tAmmeterIndicate clock of power meter.
Judgment principle is as follows:
1. being directly judged as catastrophe failure if clock synchronization errors number m is more than 3 times;
2. being judged as generic failure as the difference Δ t of terminal and clock of power meter meets 5min≤Δ t < 15min;
If 15min≤Δ t < 30min is judged as important failure;If Δ t > 30min is judged as catastrophe failure;
(6) when carrying out ammeter electric energy rate setting abnormal failure data screening, ammeter electric energy rate is judged using following formula Abnormal failure is set:
W+5≠ 0 or W-5≠0 (10)
Judgment principle is as follows:
1. only judging the electric energy meter of DLT-2007 specification.
2. judging whether there is positive spike period electricity or reversed spike period electricity, and if it exists, be then judged as serious Failure.
(7) when carrying out the screening of electric energy meter forward direction shunt running fault data, the event of electric energy meter forward direction shunt running is judged using following formula Barrier:
Or
Judgment principle is as follows:
If meeting formula (11) 3 times, it is judged as important failure, if meeting formula (11) 5 times, is judged as serious event Barrier.
(8) when the reversed shunt running fault data screening of progress electric energy meter, the event of the reversed shunt running of electric energy meter is judged using following formula Barrier:
Judgment principle is as follows:
The case where electric energy meter forward direction total electricity and reversed total electricity for clearing class exist simultaneously and be greater than 0.1, then sentence Break as catastrophe failure.
The faulty storage of the institute that DDZY102-Z intelligent electric meter breakdown judge model discrimination is gone out is electric to DDZY102-Z intelligence In table Mishap Database.
Step 3, the historical data in DDZY102-Z intelligent electric meter Mishap Database is chosen, training set and survey are classified as Examination collection carries out data mining to training set using decision Tree algorithms, forms intelligent electric meter failure decision tree, then forms preliminary point Rule-like, detailed step are as follows:
If R is the historical data of intelligent electric meter Mishap Database, and gives training set S for 60% data of R, the 40% of R Data give test set T.Data mining is carried out to training set S for using C5.0 algorithm.
Setting tool has the attribute X of n value that S is divided into n subset S1, S2..., SnIf the sum of sample is in S | S |, freq (Ci, S) is to belong to classification C in SiThe number of samples of (i=1,2 ..., N), some sample belongs to classification C in SiProbability beThe information that it is conveyed are as follows:
The entropy info (S) of training set S is indicated by formula (13);
After training set S is divided into n subset according to attribute X, the comentropy of each subset is calculated, then the expectation of set S Information infoX(S) it is indicated with formula (14);
In order to measure the obtained information of S for carrying out subregion according to inspection by attributes X, using gain standard gain (X), it is selected Selecting enables its value to maximize, i.e., the selection of this standard is to have the attribute of highest information gain to carry out every subzone;
Gain (X)=info (S)-infoX(S) (15)
S is divided into S according to n of attribute X different values1, S2..., SnThe potential information generated after total n subset Split_info (X) is shown in formula (16);
Formula (17) is that X carries out division information ratio of profit increase gain_ratio (X) to S;
It calculates for the first time using selecting the attribute of highest information gain-ratio as the root node of decision tree, then decision tree is every One node the method calculates the information gain-ratio of remaining attribute, throws away and selects the attribute of highest information gain-ratio as decision tree Present node, until ultimately forming entire decision tree.
Preliminary classification rule is formed by the decision tree.
Step 4, accuracy assessment is carried out to preliminary classification rule by the data of test set, if accuracy is met the requirements, Then determine that classifying rules is back to training set if accuracy is unsatisfactory for requiring, and re-starts training, detailed step is as follows:
(3) accuracy assesses formula are as follows:
Z is assessment accuracy in formula, and A is that preliminary classification rule predicts correct number in test set T;B is test set T Total data number.
(4) Z ' is set as the accuracy of setting.If Z >=Z ', it is determined that classifying rules;If Z < Z ', returns to training set again Training decision tree.
Step 5, DDZY102-Z intelligent electric meter failure real-time prediction model is generated by the classifying rules finally determined;
Step 6, DDZY102-Z intelligent electric meter real time fail database is connected to DDZY102-Z intelligent electric meter failure reality When prediction model predicted in real time, obtain the real-time prediction result of DDZY102-Z intelligent electric meter failure.

Claims (2)

1. a kind of intelligent electric meter failure real-time predicting method based on decision tree, which is characterized in that the real-time predicting method Specific step is as follows:
Step 1, data prediction is carried out to the intelligent electric meter data of power information acquisition system;
Step 2, according to intelligent electric meter breakdown judge model, the faulty data of the intelligent electric meter of power information acquisition system are sieved Choosing is into intelligent electric meter Mishap Database;
The intelligent electric meter breakdown judge model is as follows:
(1) when the not equal fault data screening of progress the sum of ammeter total electricity and each rate electricity, ammeter is judged using following formula The failure that the sum of total electricity and each rate electricity do not wait:
In formula: W represents electricity ,+represent the direction of electricity as forward direction ,-represent the direction of electricity be reversely, when i=1 represents total Section, W+1Represent positive total electricity, W-1Reversed total electricity is represented, i=2 represents peak period, and i=3 represents usually section, i=4 generation Table low-valley interval, i=5 represent the spike period, and e is ammeter rate number, and ρ is the judgement that the sum of total electricity and each rate electricity do not wait The factor;As e=4, expression ammeter is four rate ammeters, takes ρ=0.4;As e=3, expression ammeter is three rate ammeters, takes ρ =0.3;As e=2, expression ammeter is two rate ammeters, takes ρ=0.2;
The four rate ammeters refer to support peak period, usually section, low-valley interval, the 4 period chargings of spike period ammeter; Three rate ammeters refer to support only peak period, usually section, 3 period chargings of low-valley interval ammeter;Two rate ammeters, which refer to, only to be propped up Maintain an equal level the period, 2 period chargings of low-valley interval ammeter;
Judgment principle is as follows:
1. forward and reverse total electricity of ammeter, forward and reverse peak period electricity, forward and reverse usually section electricity, forward and reverse low-valley interval are electric Amount is all larger than 0, and is not sky;
2. the sum of forward and reverse each rate electricity of ammeter is greater than 0;
3. the absolute value of the sum of ammeter total electricity and each rate electricity difference is greater than some threshold values, threshold rule is as follows: if it is Four rate ammeters, by 0.4 judgement, three rate ammeters are by 0.3 judgement, and two rate ammeters are by 0.2 judgement;
1. 2. 3. 4. being judged as catastrophe failure if meeting judgment principle simultaneously;
(2) when carrying out electric energy meter and flying away with mutation failure data screening, judge malfunction routine that electric energy meter flies away and is mutated such as Under:
A. the same day maximum electricity W is calculated firstf:
In formula: WfFor same day maximum electricity;ImaxFor maximum current;IbFor fundamental current;
B. ammeter is then calculated to fly away and mutation factor K:
In formula: K flies away for ammeter and mutation factor, WtTo work as daily electricity;
Judgment principle is as follows:
1. being directed to resident's table, maximum current I is taken in 1,2,7,8,9, Decembermax, by 12 hours calculating ammeters fly away and be mutated because Sub- K;3 times of fundamental current I are taken in 3,4,5,6,10, Novemberb, fly away and mutation factor K by 8 hours calculating ammeters;
2. judging that electric energy meter flies away and be mutated event using formula (4) when progress electric energy meter flies away with mutation failure data screening Barrier:
K≥1 (4)
3. being judged as catastrophe failure if meeting criterion 2.;
(3) when reversely active indicating value is greater than zero failure data screening to progress ammeter, judge that ammeter is reversely active using following formula and show Value is greater than zero failure:
In formula: P represents active power, and Q represents reactive power, PIt is totalRepresent ammeter reversely active general power, QIt is totalIt is reversed to represent ammeter Idle general power;
Judgment principle is as follows:
1. there is reversed active general power in electric energy meter or reversed idle general power is greater than 0;
2. being judged as catastrophe failure if meeting criterion 1.;
(4) when carrying out the screening of electric energy meter inverted walk fault data, the premise that electric energy meter inverted walk judges is positive and negative to exclude copy reading ammeter Total electricity is always empty record, and the failure of electric energy meter inverted walk is judged using following formula:
In formula: W+1yFor the positive total electricity of the previous day, W-1yFor the reversed total electricity of the previous day;
Judgment principle is as follows:
Judged according to daily electricity statistical form, for low pressure resident and single-phase industrial and commercial producer, only judges the previous day positive total electricity W+1y With reversed total electricity W-1yWhether the meter reading indicating value on the same day is greater than,That is the same day positive total electricityW+1 With reversed total electricityW-1 If It sets up, is then judged as catastrophe failure;
(5) when carrying out ammeter clock and not screening to fault data, the breakdown judge program of ammeter clock not pair is as follows:
It first determines whether clock synchronization errors number m, if clock synchronization errors number is greater than 3 times, is directly determined as catastrophe failure, it may be assumed that
M > 3 is serious (7)
In formula: m is clock synchronization errors number;
If m is not more than 3, the requirement of Δ t whether is met by on-line monitoring inquiry, and super by the standard of Δ t generation clock of power meter Poor grade is then judged using formula (8):
In formula: Δ t is the difference of terminal and clock of power meter, and calculation method is shown in formula (9):
Δ t=| tTerminal-tAmmeter| (9)
In formula: tTerminalIndicate terminal clock, tAmmeterIndicate clock of power meter;
Judgment principle is as follows:
1. being directly judged as catastrophe failure if clock synchronization errors number m is more than 3 times;
2. being judged as generic failure as the difference Δ t of terminal and clock of power meter meets 5min≤Δ t < 15min;
If 15min≤Δ t < 30min is judged as important failure;If Δ t > 30min is judged as catastrophe failure;
(6) when carrying out ammeter electric energy rate setting abnormal failure data screening, judge that ammeter electric energy rate is arranged using following formula Abnormal failure:
W+5≠ 0 or W-5≠0 (10)
In formula, W+5For the positive electricity of spike period, W-5For the reversed electricity of spike period;
Judgment principle is as follows:
1. only judging the electric energy meter of DLT-2007 specification;
2. judging whether there is the positive electricity W of spike period+5Or the reversed electricity W of spike period-5, and if it exists, then it is judged as Catastrophe failure;
(7) when carrying out the screening of electric energy meter forward direction shunt running fault data, the failure of electric energy meter forward direction shunt running is judged using following formula:
In formula, W+3For the positive electricity of usually period, W+4For the positive electricity of low-valley interval, W+1For positive total electricity;
Judgment principle is as follows:
If meeting formula (11) 3 times, it is judged as important failure, if meeting formula (11) 5 times, is judged as catastrophe failure;
(8) when the reversed shunt running fault data screening of progress electric energy meter, the failure of the reversed shunt running of electric energy meter is judged using following formula:
In formula, W+1Represent positive total electricity, W-1Represent reversed total electricity;
Judgment principle is as follows:
The case where electric energy meter forward direction total electricity and reversed total electricity for clearing class exist simultaneously and be greater than 0.1, then be judged as Catastrophe failure;
Step 3, the historical data in intelligent electric meter Mishap Database is chosen, training set and test set are classified as, using decision Tree algorithm carries out data mining to training set and forms intelligent electric meter failure decision tree, then forms preliminary classification rule;
Step 4, accuracy assessment is carried out to preliminary classification rule by the data of test set, if accuracy is met the requirements, really Determine classifying rules, if accuracy is unsatisfactory for requiring, is back to training set, re-starts training;
Step 5, intelligent electric meter failure real-time prediction model is generated by the classifying rules finally determined;
Step 6, intelligent electric meter real time fail data basd link intelligent electric meter failure real-time prediction model is connected to be predicted in real time, Obtain the real-time prediction result of intelligent electric meter failure.
2. the intelligent electric meter failure real-time predicting method according to claim 1 based on decision tree, which is characterized in that described Step 1 the step of data prediction is carried out to the intelligent electric meter data of power information acquisition system are as follows: removal and intelligence first The unrelated attribute data of electric meter fault has the data of apparent error and the duplicate data of attribute, then by each attribute consecutive numbers According to discretization.
CN201610342042.2A 2016-05-20 2016-05-20 A kind of intelligent electric meter failure real-time predicting method based on decision tree Active CN106054104B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610342042.2A CN106054104B (en) 2016-05-20 2016-05-20 A kind of intelligent electric meter failure real-time predicting method based on decision tree

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610342042.2A CN106054104B (en) 2016-05-20 2016-05-20 A kind of intelligent electric meter failure real-time predicting method based on decision tree

Publications (2)

Publication Number Publication Date
CN106054104A CN106054104A (en) 2016-10-26
CN106054104B true CN106054104B (en) 2019-01-11

Family

ID=57177407

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610342042.2A Active CN106054104B (en) 2016-05-20 2016-05-20 A kind of intelligent electric meter failure real-time predicting method based on decision tree

Country Status (1)

Country Link
CN (1) CN106054104B (en)

Families Citing this family (59)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106439199B (en) * 2016-11-09 2019-04-09 北京化工大学 A kind of control valve failure monitoring method based on DCS data
CN106567964B (en) * 2016-11-09 2018-10-23 北京化工大学 A kind of flow control valve monitoring method based on DCS data
CN106710160A (en) * 2016-12-28 2017-05-24 湖南坤宇网络科技有限公司 Decision-making tree system-based boiler clausilium smoke temperature early-warning method
CN106681303A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler central control system line fault warning method based on decision tree system
CN106642077A (en) * 2016-12-28 2017-05-10 湖南坤宇网络科技有限公司 Boiler return feeder blockage prewarning method based on decision tree system
CN106682423A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler safety valve blocking early warning method based decision-making tree system
CN106779233A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of heat-insulated fault rupture method for early warning of boiler based on decision tree system
CN106768957A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler flange cut-off valve fault early warning method based on decision tree system
CN106781342A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler air preheater fault early warning method based on decision tree system
CN106781307A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler blow-off stop valve early warning method for failure based on decision tree system
CN106774265A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler small fire defective valve method for early warning based on decision tree system
CN106765294A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler combustion machine insufficient method for early warning of burning based on decision tree system
CN106678055A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Decision tree system based early warning method for faults of boiler circulating pump
CN106682778A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler flue pressure warning method based on decision tree system
CN106779235A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler feed early warning method for failure based on decision tree system
CN106683352A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler fire tube perforating early-warning method based on decision tree system
CN106682422A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler fire tube scale early-warning method based on decision tree system
CN106779236A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler drum deformation method for early warning based on decision tree system
CN106779450A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler water softener efficiency lost resin method for early warning based on decision tree system
CN106774266A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler superheater early warning method for failure based on decision tree system
CN106642726A (en) * 2016-12-28 2017-05-10 湖南坤宇网络科技有限公司 Boiler gas hot-blast stove fault early warning method based on decision tree system
CN106709605A (en) * 2016-12-28 2017-05-24 湖南坤宇网络科技有限公司 Method for early warning of boiler fire tube corrosion based on decision tree system
CN106644445A (en) * 2016-12-28 2017-05-10 湖南坤宇网络科技有限公司 Boiler check valve failure warning method based on decision tree system
CN106646234A (en) * 2016-12-28 2017-05-10 湖南坤宇网络科技有限公司 Boiler main motor fault early-warning method based on decision tree system
CN106845690A (en) * 2016-12-28 2017-06-13 湖南坤宇网络科技有限公司 A kind of boiler water level method for early warning based on decision tree system
CN106774077A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler dusting deashing device fault early warning method based on decision tree system
CN106802646A (en) * 2016-12-28 2017-06-06 湖南坤宇网络科技有限公司 A kind of boiler tube bursting fault early warning method based on decision tree system
CN106709657A (en) * 2016-12-28 2017-05-24 湖南坤宇网络科技有限公司 Decision tree system-based boiler intake pump failure early warning method
CN106845689A (en) * 2016-12-28 2017-06-13 湖南坤宇网络科技有限公司 A kind of boiler feed temperature method for early warning based on decision tree system
CN106781306A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler air duct based on decision tree system blocks method for early warning
CN106682833A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler cotton filter obstruction early warning method based on decision tree system
CN106710162A (en) * 2016-12-28 2017-05-24 湖南坤宇网络科技有限公司 Boiler furnace scaling early warning method based on decision tree system
CN106482943A (en) * 2016-12-28 2017-03-08 湖南坤宇网络科技有限公司 A kind of boiler stop valve based on decision tree system closes failure method for early warning
CN106525433A (en) * 2016-12-28 2017-03-22 湖南坤宇网络科技有限公司 Boiler combustion machine main bearing wear early-warning method based on decision-making tree system
CN106779234A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of coal-saving apparatus for boiler early warning method for failure based on decision tree system
CN106683351A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler flue gas desulfurization equipment fault warning method based on decision tree system
CN106768956A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler down-comer fault early warning method based on decision tree system
CN106781308A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of anti-explosive door for boiler early warning method for failure based on decision tree system
CN106642191A (en) * 2016-12-28 2017-05-10 湖南坤宇网络科技有限公司 Boiler coal dropping pipe blocking early warning method based on decision tree system
CN106679953A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler regulating gate failure early-warning method based on decision tree system
CN106682777A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler flue gas tube blockage early warning method based on decision tree system
CN106710163A (en) * 2016-12-28 2017-05-24 湖南坤宇网络科技有限公司 Boiler air pressure early warning method based on decision-making tree system
CN106778009A (en) * 2016-12-28 2017-05-31 湖南坤宇网络科技有限公司 A kind of boiler furnace incrustation scale method for early warning based on decision tree system
CN106802647A (en) * 2016-12-28 2017-06-06 湖南坤宇网络科技有限公司 A kind of boiler priming fault early warning method based on decision tree system
CN106845084A (en) * 2016-12-28 2017-06-13 湖南坤宇网络科技有限公司 A kind of boiler induced-draft fan based on decision tree system blocks method for early warning
CN106682421A (en) * 2016-12-28 2017-05-17 湖南坤宇网络科技有限公司 Boiler central-air-pipe clogging early-warning method based on decision-making tree system
CN106710161A (en) * 2016-12-28 2017-05-24 湖南坤宇网络科技有限公司 Decision tree system-based boiler water-cooling wall corrosion early warning method
CN106596090A (en) * 2016-12-28 2017-04-26 湖南坤宇网络科技有限公司 Decision-making-tree-system-based early warning method for boiler steam valve fault
CN106842106A (en) * 2017-02-23 2017-06-13 广东电网有限责任公司茂名供电局 Electrical energy meter fault Forecasting Methodology and device
CN108733966A (en) * 2017-04-14 2018-11-02 国网重庆市电力公司 A kind of multidimensional electric energy meter field thermodynamic state verification method based on decision woodlot
CN107346995A (en) * 2017-07-31 2017-11-14 威胜信息技术股份有限公司 Clock synchronizing method based on power information acquisition terminal
CN107911762A (en) * 2017-11-15 2018-04-13 国网安徽省电力公司宿州供电公司 A kind of ONU method for diagnosing faults based on decision tree
CN109017799A (en) * 2018-04-03 2018-12-18 张锐明 A kind of new-energy automobile driving behavior prediction technique
CN108664010A (en) * 2018-05-07 2018-10-16 广东省电信规划设计院有限公司 Generating set fault data prediction technique, device and computer equipment
CN109188196B (en) * 2018-08-23 2021-08-31 宁波三星医疗电气股份有限公司 Electric quantity sudden change fault positioning method based on hamming codes
CN110222914A (en) * 2019-07-02 2019-09-10 国家电网有限公司 A kind of concentrator that accuracy rate is high operation prediction technique
CN110888101A (en) * 2019-12-05 2020-03-17 新奥数能科技有限公司 Electric energy meter abnormity diagnosis method and device
CN111383428B (en) * 2020-05-29 2020-09-11 成都千嘉科技有限公司 Online meter state monitoring and early warning method
CN111781554A (en) * 2020-06-28 2020-10-16 南方电网科学研究院有限责任公司 Metering device fault determination method and system based on four-quadrant electric energy data

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103532745A (en) * 2013-09-25 2014-01-22 国家电网公司 Metering online monitoring system based on power user power consumption information acquisition system
CN103559655A (en) * 2013-11-15 2014-02-05 哈尔滨工业大学 Microgrid novel feeder load prediction method based on data mining
CN104021264A (en) * 2013-02-28 2014-09-03 华为技术有限公司 Defect prediction method and device
CN104111920A (en) * 2013-04-16 2014-10-22 华为技术有限公司 Decision-making tree based prediction method and device
CN104794195A (en) * 2015-04-17 2015-07-22 南京大学 Data mining method for finding potential telecommunication users changing cell phones
CN104915898A (en) * 2015-06-26 2015-09-16 南京易司拓电力科技股份有限公司 Voltage quality reason analysis method and device thereof
CN105182065A (en) * 2015-09-23 2015-12-23 国网山东莒县供电公司 Electricity information acquisition system based on intelligent electric meter

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104021264A (en) * 2013-02-28 2014-09-03 华为技术有限公司 Defect prediction method and device
CN104111920A (en) * 2013-04-16 2014-10-22 华为技术有限公司 Decision-making tree based prediction method and device
CN103532745A (en) * 2013-09-25 2014-01-22 国家电网公司 Metering online monitoring system based on power user power consumption information acquisition system
CN103559655A (en) * 2013-11-15 2014-02-05 哈尔滨工业大学 Microgrid novel feeder load prediction method based on data mining
CN104794195A (en) * 2015-04-17 2015-07-22 南京大学 Data mining method for finding potential telecommunication users changing cell phones
CN104915898A (en) * 2015-06-26 2015-09-16 南京易司拓电力科技股份有限公司 Voltage quality reason analysis method and device thereof
CN105182065A (en) * 2015-09-23 2015-12-23 国网山东莒县供电公司 Electricity information acquisition system based on intelligent electric meter

Also Published As

Publication number Publication date
CN106054104A (en) 2016-10-26

Similar Documents

Publication Publication Date Title
CN106054104B (en) A kind of intelligent electric meter failure real-time predicting method based on decision tree
Diao et al. Decision tree-based online voltage security assessment using PMU measurements
Zhu et al. Time series shapelet classification based online short-term voltage stability assessment
CN106154209B (en) Electrical energy meter fault prediction technique based on decision Tree algorithms
CN106384210B (en) A kind of power transmission and transforming equipment maintenance prioritization method based on maintenance risk income
CN106019084B (en) Medium-voltage power distribution network disconnection fault diagnosis method based on power distribution and utilization data correlation
CN102934312B (en) Energy production system and control thereof
CN107358366B (en) Distribution transformer fault risk monitoring method and system
CN103793853A (en) Overhead power transmission line running state assessment method based on bidirectional Bayesian network
CN104504607A (en) Method for diagnosing photovoltaic power station faults on the basis of fuzzy clustering algorithm
CN108320043A (en) A kind of distribution network equipment state diagnosis prediction method based on electric power big data
CN105300692A (en) Bearing fault diagnosis and prediction method based on extended Kalman filtering algorithm
CN106446016A (en) Parallel association rule mining-based method for predicting running reliability of power distribution network
CN110263172A (en) A kind of evented autonomous classification method of power system monitor warning information
CN104123678A (en) Electricity relay protection status overhaul method based on status grade evaluation model
CN109492790A (en) Wind turbines health control method based on neural network and data mining
CN105956788A (en) Dynamic management control method for cost of power transmission and transformation project
CN106022594B (en) Electric system on-line operation safety risk estimating method based on extreme value theory
CN105678456A (en) Method for automatically assessing electric energy metering device operation status and system thereof
CN108847686A (en) A kind of photovoltaic DC-to-AC converter failure prediction method
CN105678642A (en) Cascading failure risk recognition method for power system
CN109002781A (en) A kind of energy accumulation current converter failure prediction method
CN104036433A (en) Method for evaluating running management level of power distribution network
Hu et al. Operational reliability evaluation method based on Big data technology
CN105741184A (en) Transformer state evaluation method and apparatus

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20210416

Address after: No. 200, Hengda street, Changchun middle road, high tech Zone (new urban district), Urumqi, Xinjiang Uygur Autonomous Region

Patentee after: Marketing service center of State Grid Xinjiang Electric Power Co., Ltd. (capital intensive center, metering center)

Patentee after: XINJIANG University

Address before: 830000, 66, lane two, Changchun Middle Road, Urumqi hi tech Industrial Development Zone (new urban area), the Xinjiang Uygur Autonomous Region

Patentee before: ELECTRIC POWER SCIENCES RESEARCH INSTITUTE OF STATE GRID XINJIANG ELECTRIC POWER Co.

Patentee before: XINJIANG University