CN105158723B - A kind of error evaluation system and method for digitalized electric energy metering system - Google Patents
A kind of error evaluation system and method for digitalized electric energy metering system Download PDFInfo
- Publication number
- CN105158723B CN105158723B CN201510458899.6A CN201510458899A CN105158723B CN 105158723 B CN105158723 B CN 105158723B CN 201510458899 A CN201510458899 A CN 201510458899A CN 105158723 B CN105158723 B CN 105158723B
- Authority
- CN
- China
- Prior art keywords
- electric energy
- error
- parameter
- energy metering
- degeneration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The digitalized electric energy metering system error evaluation method based on many reference amounts degradation model that the invention discloses a kind of, belongs to smart grid equipment on-line status assessment field.The present invention passes through the sensor in each key position installation of substation, successively win access word electric energy metering device running environment data, using based on many reference amounts degradation model appraisal procedure, the error state of digitalized electric energy metering system is assessed, realizes digital electric energy metered system status monitoring truly and maintenance.The present invention includes environmentally sensitive unit, Current Voltage sensing unit, computer and runs on assessment software on computer.The present invention provides the on-line condition monitoring scheme of engineering for digitalized electric energy metering system, assesses the error state of digital electric energy metered system, provides on-line monitoring for the engineering application of digitalized electric energy measurement technology.
Description
Technical field
The invention belongs to smart grid equipment on-line status monitoring fields, are moved back more particularly, to one kind based on many reference amounts
Change the digitalized electric energy metering system error evaluation method of model.
Background technique
With the progress of science and technology, electric system is continued to develop towards digitlization and intelligent direction, Digitized transformation
Station is the important component of smart grid.In digital transformer substation, electrical energy measurement is no longer filled using traditional electrical energy measurement
It sets, but uses totally digitilized electric energy metering device, typical three-phase digital electric energy metering device composition is as shown in Figure 1.Position
It is respectively that A is combined formula electric mutual inductor, B is combined formula electric mutual inductor and C on high-tension side combined electronic transformer 1-3
It is combined formula electric mutual inductor, wherein combined electronic transformer includes the electronic current mutual inductor and electricity fitted together
Minor voltage transformer, synchronized clock 4 is synchronous, acquire respectively A phase bus 5, B phase bus 6 and C phase bus 7 electric current and
Voltage, the data of acquisition are sent to the combining unit 8 of low-pressure side with certain format such as FT3 through optical fiber.Combining unit 8 will receive
Sampled data digital electric energy meter 9 is sent to IEC61850-9-2/LE frame format again.Digital electric energy meter 9 is using certain
Electric energy algorithm calculate electric energy.Digital electric energy meter 9 has RS485 or infrared interface, for being communicated with other equipment.
Digital electric energy metered system is one of key link of intelligent substation, realizes digital electric energy metered system engineering
Using and repair based on condition of component, meet " strong smart grid " requirement.In recent years, the theory and test of digital electric energy metered system are ground
Studying carefully has very much, and main purpose is to promote its stability and reliability, promotes its engineering application.For digital electrical energy measurement system
System, it is most important that error characteristics, in order to grasp the kinematic error rule of digital electric energy metered system, Jiangsu DianKeYuan and Guangdong
The R&D institutions such as DianKeYuan establish digital electrical energy measurement monitoring system, can obtain the running environment ginseng of digital electric energy metered system
Several and error information, but these data are not utilized further.The electric energy metered system thermodynamic state verification that Chongqing DianKeYuan proposes
Strategy makes the repair based on condition of component of electric energy metered system further.Although these work achieve achievement, but be respectively present deficiency
Place: (1) error state for the data assessment metering system for not making full use of monitoring system to obtain；(2) empirical evaluation meter is relied on
The state of amount system rather than scientific and reasonable mathematical model.
There is more perfect research body for the status assessment model of most of power transmission and transforming equipments and appraisal procedure at present
System, but it is substantially at blank for digital electric energy metered system assessment models and method, cause electric energy metered system regular
The overproof interruption maintenance later of measurement error is examined and determine or occurred, power supply reliability is affected, also increases O&M cost.Digital electric energy
The difficult point of metering system error evaluation is: (1) error influence factor is more, and measurement error is the knot of these combined factors effect
Fruit, it is difficult to establish influence model of the single factors to measurement error；(2) the error performance parameter and influence amount of electric energy metered system
Data are more, and between each other without definite relation, it is difficult to using traditional data processing method obtain useful result (3) respectively influence because
Element is shifting to the effect of measurement error, so that its appraisal procedure is significantly different from the performance estimating method of most equipment, does not have
There are mature available assessment models.The stability problem that digitalized electric energy metering system exposes in operation prevents it from a large amount of
Also lack effective ways always for engineering practice, and to the error evaluation of digitalized electric energy metering system.In short, in order to just
The really error of assessment digitalized electric energy metering system needs using new assessment models and uses modern data processing method.
The present invention is directed to digitalized electric energy metering system feature, using many reference amounts degeneration error evaluation model, in conjunction with error
Constraint condition is assessed, reasonable assessment is made to digital electric energy metering systematic error state.Conducive to the digital electrical energy measurement skill of propulsion
Art engineering application, promotes digital electric energy metered system repair based on condition of component, has most important theories and practice significance.
Summary of the invention
Aiming at the problems existing in the prior art, provided by the present application is that the error of digitalized electric energy metering system a kind of is commented
Estimate system and method, wherein being studied and being related to by specific structure to BP neural network and its set-up mode, realizes
To digital electric energy metering directional error state on_line monitoring.
To achieve the above object, according to one aspect of the present invention, a kind of mistake of digitalized electric energy metering system is provided
Poor assessment system, the error evaluation system are used for the online error state assessment of digital electric energy metering device progress, and including
Environmentally sensitive unit, Current Voltage sensing unit and central processing unit, it is characterised in that:
The environmentally sensitive unit includes temperature sensor, humidity sensor, the second microprocessor and the second direct current
Source；Wherein the temperature sensor, humidity sensor are respectively used to locating for the digitalized electric energy metering system as monitoring object
Environment executes temperature, humidity carries out real-time measurement, and gives common transport to second microprocessor, and second DC power supply is then
For providing working power to other component units in described；
The Current Voltage sensing unit includes electromagnetic current transducer, electromagnetic potential transformer, the first micro process
Device, and the first DC power supply for device each in Current Voltage sensing unit power supply；The wherein current transformer, the electricity
Mutual inductor is pressed to be used to carry out real-time measurement to the frequency and harmonic wave of the digitalized electric energy metering system；
In addition, the central processing unit is connected with first, second microprocessor simultaneously, and according to measurement data reality
Shi Jianli and update many reference amounts degradation model assess digital electric energy metered system current and future error state in short-term:
For the integrated data analyzing and processing software of the computer operation as status assessment platform, (1) uses difference normalizing
Change preprocess method to pre-process the measurement data, (2) are again using artificial neural network training degenerate network specifically:
Measurement data through difference normalized is input to the degeneration grid, acquires the corresponding degeneration grid of difference；(3) last
Digital electric energy metered system current and future error in short-term is assessed further according to the measurement data of each sensing unit
State, to realize the error state monitoring of digitalized electric energy metering device.
It is another aspect of this invention to provide that a kind of error evaluation method of digitalized electric energy metering system is provided, it is special
Sign is, the system the following steps are included:
(1) under the premise of same using rate, the influence systematic error to the digitalized electric energy metering system respectively
Factor and corresponding systematic error measure；The factor for influencing systematic error is environment temperature, ambient humidity, electricity
It is gas parameter, frequency and harmonic wave, electromagnetic field, vibration, communication abnormality, decompression, one or more in cutout；
(2) measurement data is pre-processed using difference normalization preprocess method；
(3) learning training is carried out using artificial neural network, and to instruction using artificial neural network training degenerate network again
Practice network to be verified, whether training of judgement result is reasonable, the retraining again if unreasonable, sets until validation error meets
Until value；
(4) number is finally assessed by the degenerate network further according to the measurement data of each sensing unit
Electric energy metered system current and future error state in short-term, to realize the error state monitoring of digitalized electric energy metering device.
Preferably, in the step (3), the measurement data through difference normalized is input to the degeneration net
Lattice are acquired the corresponding degeneration grid of difference, the error of electric energy metered system are assessed with this.
In general, above-mentioned technical concept according to the invention compared with prior art, it is excellent mainly to have technology below
Point:
1, it realizes to digital electric energy metering directional error state on_line monitoring, provides one kind for number electric energy metered system
Engineering grade status monitoring and repair method truly.
Detailed description of the invention
Fig. 1 show typical digitalized electric energy metering device structural schematic diagram in the prior art；
Fig. 2 show single-phase electric energy metering system input and output isoboles；
Fig. 3 show electric current, voltage measurement error schematic diagram；
Fig. 4 show the appraisal procedure figure based on the assessment of many reference amounts degradation model error state；
Fig. 5 show the composition frame of the digitalized electric energy metering system error evaluation method system based on many reference amounts degradation model
Figure；
Fig. 6 show BP neural network structural schematic diagram；
Fig. 7 show temperature raw data plot figure；
Fig. 8 show Temperature pre-treatment data graphs；
Fig. 9 show ECT than poor raw data plot figure；
Figure 10 show ECT than poor preprocessed data curve graph；
Figure 11 show trained the number of iterations and error relationship figure；
Figure 12 show error evaluation result and actual motion error comparison chart.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.As long as in addition, technical characteristic involved in the various embodiments of the present invention described below
Not constituting a conflict with each other can be combined with each other.
The assessment models are many reference amounts degradation model.Specific intension is as follows:
The parameter degradation of one equipment or system generally means that performance declines.It is representational for complex device or system
Can parameter often have it is multiple so that when establishing status assessment model need and meanwhile consider these parameters.Such as in transformer
It is related to oil dissolved gas, insulation resistance, dielectric loss, the parameters such as micro- water in oil in state evaluation；Circuit-breaker status evaluation
In be related to contact electrical wear, divide-shut brake coil current, cut-off the parameters such as stroke.By taking Transformer State Assessment as an example, it is assumed that table
The set for levying the parameter composition of certain transformer performance is defined as { P1, P2 ..., Pn }, a simple and effective Performance Evaluation mould
Type is: working as max | Pi | (i=1,2 ..., n) be more than threshold value when, providing assessment result, " a certain performance of the transformer can decline
To affecting the normal operation of the equipment ".
Above-mentioned Performance Evaluation Model is not suitable for digitalized electric energy metering system assessment.Reason mainly has at 3 points: (1) measuring
The a certain performance parameter of system is degenerated, and not will lead to metering system generally and each parameter of operation troubles (2) metering system integrally occurs
Degeneration there is directionality, i.e. influence of the parameter degradation to electrical energy measurement not absolutely increase measurement error, it is also possible to be in this
That length disappear to not influence whole measurement error, such as voltage transformer becomes larger than difference to positive direction, current transformer than difference to
Negative direction becomes larger, and metering system still can correctly measure electric energy；(3) it under the premise of above-mentioned (1) and (2), cannot be degenerated with maximum
The weighting of amount or each amount of degradation and/or product assess the error of actual digital electric energy metered system.Therefore, for digitlization
Electric energy metered system needs to propose new assessment models.
To simplify modeling, single-phase electric energy metering is only considered.The mean power expression formula of single-phase electric energy metering are as follows:
Single-phase number electric energy metered system can be equivalent to a dual input single output system, input be analog voltage and
Electric current, exports as power, and system equivalent schematic is as shown in Figure 2.
As seen from Figure 2, electrical energy measurement sums up in the point that bottom is measurement voltage and current.The measurement error of voltage and current is straight
Connecing leads to electric energy metering error, therefore the degeneration parameter of electric energy metered system can consider following four: voltage measurement than it is poor,
Voltage measurement angular difference, current measurement are than poor, current measurement angular difference.
Meanwhile to system shown in Fig. 2, substantially it is U (i) and I (i) measurement error than difference and angular difference measurement error, such as schemes
Shown in 3A and Fig. 3 B.
In Fig. 3 A, curve 1 is original input voltage, and curve 2 is the input voltage only than difference, and curve 3 is while having ratio
The input voltage of difference and angular difference.Assuming that the degeneration parameter of single-phase number electric energy metered system initial time is 0, it is being put into operation
Afterwards, they respectively in respective dimension variation, change direction is uncertain.Certain moment, four parameters degenerated to fPT、δPT、fCTAnd δCT,
Discrete power is answered at this time are as follows:
In formula, αiIt is by fPTAnd δPTCaused U (i) error coefficient, αi=h1(fPT,δPT)；βiIt is by fCTAnd δCTIt is caused
I (i) variation coefficient, βi=h2(fCT,δCT), αiAnd βiValue range are as follows: -1≤α≤1, -1≤β≤1.
By formula (2) it is found that working as αiβiWhen=1, electrical energy measurement accuracy is constant, the reason is that the degeneration of each parameter is to electric energy meter
The influence of amount is cancelled out each other, so that measurement error is 0.
Based on the above analysis, digital electric energy metering error assessment has aobvious with the Performance Evaluation Model of general device or system
The otherness of work, the main distinction are that the parameter of digital electric energy metered system is degenerated with directionality, and any parameter degeneration is to meter
The influence of amount error should consider current other parameters variation.Therefore, it proposes herein for electric energy metered system based on more ginsengs
Measure degeneration error evaluation model.
Existing theoretical research thinks that the digital electric energy metered system that substation puts into operation is influenced by various aspects, including ring
Border temperature, humidity；Electric parameter such as load, frequency and harmonic wave；Electromagnetic field, vibration, communication abnormality；The power grids thing such as decompression, cutout
Part.It is divided into reliability and stability influence factor according to whether electric energy metered system normal work is influenced；According to influence time point
For duration and sporadic influence factor.
It is clear and accurate in order to facilitate present invention narration, be further simplified the model, choose influence the duration of stability because
Element is modeled, respectively environment temperature, humidity, mains frequency, harmonic wave.Four influence factor collective effects are in digital electric energy meter
Amount system influences four degeneration parameter fPT, δPT, fCT, δCT, it can be described with following equation group:
In above formula, Temp (t), Hum (t), Freq (t), Harm (t), be respectively temperature, humidity, frequency, harmonic wave at any time
Variation function.It enables:
Solution of equations matrix is denoted as A, then equation (4) can be written as:
Formula (7) is digital electric energy metered system parameter degeneration equation.Wherein,It is parameter degeneration vector,It is to move back
Change parameter, definition A is many reference amounts degenerate network of the degeneration parameter in the case where acting on vector.
The error of electric energy metered system is defined as assessment constraint condition:
In above formula, ηPFor mean power error.Formula (7) and (8) collectively form digitalized electric energy metering system error evaluation
Model.
The error state appraisal procedure is to be assessed based on many reference amounts degradation model error state, and specific intension is as follows:
The present invention is based on the appraisal procedure assessed based on many reference amounts degradation model error state as shown in figure 4, including data
The error evaluation of pretreatment, training degenerate network and metering system.
(1) data prediction.It is unified that sample rate is carried out to input many reference amounts, then does difference normalized, is respectively obtained:
{ΔTemp(1),ΔTemp(2),...,ΔTemp(n)}；
{ΔHum(1),ΔHum(2),...,ΔHum(n)}；
{ΔFreq(1),ΔFreq(2),...,ΔFreq(n)}；
{ΔHarm(1),ΔHarm(2),...,ΔHarm(n)}；
{ΔδCT(1),ΔδCT(2),...,ΔδCT(n)}；
{ΔfCT(1),ΔfCT(2),...,ΔfCT(n)}；
{ΔδPT(1),ΔδPT(2),...,ΔδPT(n)}；
{ΔfPT(1),ΔfPT(2),...,ΔfPT(n)}。
It is loud by the effect of { Δ Temp (i), Δ Hum (i), Δ Freq (i), Δ Harm (i) } composition, it is moved back as shown in Fig. 3
Change the input of network, { Δ δCT(i),ΔfCT(i),ΔδPT(i),ΔfPT(i) } output of Fig. 3 is constituted.
(2) training degenerate network.The input and output sample sequence according to acquired in (1), using BP neural network
Training is practised, and training network is verified, whether training of judgement result is reasonable, the retraining again if unreasonable, until testing
Until card error meets setting value；
(3) metering system error evaluation.The actuating quantity sequence { Δ Temp, Δ Hum, Δ Freq, Δ Harm } of measurement is defeated
Enter to degenerate network, acquire the variable quantity of degeneration parameter, it is each to obtain current electric energy metered system in conjunction with degeneration parameter initial value
State parameter finally assesses the error current of electric energy metered system.
It is provided by the invention based on the digitalized electric energy metering system error evaluation method of many reference amounts degradation model by
The sensor of each key position installation of substation, obtains the degradation effects amount of digital electric energy metered system, including three buses
The measured current and voltage, combination for three bus 5-7 that the accurate electric current of 5-7 and voltage and combined electronic transformer measure
The influence amount is constituted the parameter degeneration shadow of digital electric energy metered system by the ambient temperature and humidity etc. of formula electric mutual inductor 1-3
Ring vector；Many reference amounts degradation model is established by the history run state of digital electric energy metered system again；Finally parameter is degenerated
It influences vector and is input to degradation model, obtain the error state of digital electric energy metered system current and future in short-term.To realize
To digital electric energy metering directional error state on_line monitoring, a kind of work truly is provided for number electric energy metered system
Journey grade status monitoring and repair method.
Fig. 5 show the digitalized electric energy metering system error evaluation based on many reference amounts degradation model of the embodiment of the present invention
The system composition block diagram.As shown in figure 5, the digitalized electric energy metering system error evaluation method system based on many reference amounts degradation model
Including environmentally sensitive unit 10, Current Voltage sensing unit 13 and computer 14.
Environmentally sensitive unit 10 includes temperature sensor 101, humidity sensor 102, first microprocessor 103, Yi Jiwei
First DC power supply 104 of each device power supply in environmentally sensitive unit 10.In embodiments of the present invention, temperature sensor 101 can
To select Digital Measurement of Temperature chip DS18B20；Humidity sensor 102 can select STTS751；First microprocessor 103 selects TI
The MPS430F149 low-power scm of company；First DC power supply 104 selects SD6-S05A1 unit power supply.
Current Voltage sensing unit 11 includes current transformer 131, voltage transformer 132, the 4th microprocessor 133, with
And the 4th DC power supply 134 for device each in Current Voltage sensing unit 13 power supply.In embodiments of the present invention, Current Mutual Inductance
Device 131 and voltage transformer 132 require to select the electromagnetic potential mutual inductance for meeting class of accuracy requirement according to accuracy of measurement
Device and electromagnetic current transducer, measured value can be used as standard primary voltage and primary current；4th microprocessor 133 can
To select the microprocessor MPC860 of PowerPC framework；4th DC power supply 134 selects SD25-S12D1 unit power supply.
Computer 14 obtains the measurement data of each sensing module 10,13.Computer 14 includes data preprocessing module, degenerates
Network training module, error state evaluation module and memory module, it is logical first using the error analysis method based on degeneration thought
Cross digital electric energy metered system history data and establish many reference amounts degradation model, training many reference amounts degenerate network, reuse by
Measurement and calculating gained key parameter composition amount of degradation act on vector, and the effect vector is input to many reference amounts degeneration net
Network, to assess the current and future of digitalized electric energy metering device error state in short-term.Current micro computer hardware is matched
Height is set, processor arithmetic speed is fast, and hard-disk capacity is big, is generally able to satisfy operation and memory requirement.In embodiments of the present invention,
Computer 14 selects Hewlett-Packard HP Pavilion P6-1480CN microcomputer.
The data that digitalized electric energy metering system error evaluation system based on many reference amounts degradation model obtains include number
The operation data of many reference amounts, parameter degeneration amount and the digitalized electric energy metering device of electric energy metering device itself.Ring
Border sensing unit 10, Current Voltage sensing unit 11 constitute sensing and monitoring system, and the continuous online digitalized electric energy that obtains measures dress
The many reference amounts and parameter degeneration amount related data set, comprising: the accurate electric current of three bus 5-7 and voltage and combination
The measured current and voltage and thus obtained electronic current voltage that formula electric mutual inductor measures on three bus 5-7
The ratio difference and angular difference of measurement, the temperature and humidity of environment etc..
The acquisition methods of each sensing unit mounting means and 6 groups of data are specifically described below.
1, three same environmentally sensitive units 10 environmental sensory data: are separately mounted to combined electronic transformer 1-
At 3 pedestal, for obtaining temperature, the humidity of combined electronic transformer 1-3 running environment.
2, three same Current Voltage sensing units 13 accurate primary current voltage sensor data: are installed on three
On bus 5-7, for obtaining the accurate primary current and accurate primary voltage of three bus 5-7, calculating is sent to by network interface
Machine 14.
The computer 14 that Fig. 4 show the embodiment of the present invention carries out the flow chart of error state assessment.It is poor that computer 14 is used
Divide normalization data preprocess method that parameter degeneration amount and parameter are carried out data processing, with historical data to digital electric energy
The parameter degenerate network of metering system has carried out that parameter actuating quantity is finally input to degeneration net based on BP neural network training
Network obtains digital electric energy metered system current and future error state in short-term.Specifically, in embodiments of the present invention, belonging to number
Word electric energy metered system degeneration parameter is ECT measurement error, EVT measurement error etc.；Belong to degeneration parameter actuating quantity is ring
Border temperature, humidity, electric parameter such as harmonic wave, frequency etc..
The pretreated specific intension of the difference normalization data and method are as follows:
The parameter degeneration vector of degradation model includes temperature, humidity, frequency and harmonic wave, is adopted when obtaining these data
Cause data volume different with different sample frequencys, such as temperature and humidity is that sampling in every three minutes is primary, and frequency, harmonic wave, than it is poor,
Angular difference samples (calculating) once per minute, then the sample rate between temperature, wetness action amount and other amounts differs 180 times；It is all kinds of
The radix and dimension of data are also inconsistent, such as temperature radix is 25, and frequency radix is 50.Firstly, being answered for synchrodata
The sample rate to Various types of data carries out unification.Secondly, the radix difference of each input quantity will affect subsequent degradation network training knot
The small input quantity of radix can be flooded, need that Various types of data is normalized by the correctness of fruit, the big input quantity of radix,
So as to uniform variable weight and dimension.
Since epidemic disaster can not mutate, first-order linear interpolation method can be used, epidemic disaster data are carried out
Data volume expand, with reach and other.By taking temperature data is handled as an example, it is assumed that i-th and j point sampling temperature is T (i) and T (j),
Data volume needs to increase to N times, then the data sequence after carrying out data extending, between two temperature sampling values are as follows:
Similarly, after to humidity using the processing of (9) formula, data of the Temperature and Humidity module sample rate is identical as its dependent variable.
Difference normalization is reused to handle Various types of data, it is assumed that Variables Sequence is { V1, V2 ..., Vn, Vn+1 },
Its variable gradient is defined as
Δ V=max { Δ Vi(i=2,3 ..., n, n+1) (10)
Following difference normalized then is carried out to Variables Sequence:
Equation (7) is written as normalized form as a result, are as follows:
In above formula,Vector is acted on for difference,For difference degeneration parameter, B is corresponding difference degenerate network.
The reason of data have first been done with difference processing before above-mentioned normalization is: with non-differential normalization data training
Degenerate network trains resulting degenerate network that can not correctly assess metering system as the result is shown when carrying out case verification
Error.Further data feature discovery, rate of change and the variable quantity difference, such as temperature, humidity etc. of Various types of data become
Change slowly and variable quantity is larger, and frequency variation is then quite small.It is obvious that small frequency fluctuation is to electrical energy measurement
The error influence of system be it is biggish, i.e., the actuating quantity of small weight has the influence degree of big weight, therefore proposes to use difference
Normalization is pre-processed, and the weight of minor change is amplified.
The specific intension and method of many reference amounts degenerate network training are as follows:
The essence for solving many reference amounts degenerate network is the coefficient matrix that equation (12) are solved according to actuating quantity and degeneration parameter
B, it describes effect of the multi-factor comprehensive influence to degeneration parameter, however the element of matrix B does not have resolvability, Bu Nengyong
Constant or elementary function description.In this case, artificial intelligence learning method training degenerate network, approaching to reality are preferably used
The input-output characteristic of coefficient matrix B.
Present example is by taking BP feedforward neural network as an example.Feedforward neural network is one kind of artificial neural network, each mind
Through member since input layer, previous stage input is received, then be input to next stage, until output layer.BP neural network is feedforward mind
The mode of learning of one kind through network, network structure and two-way propagation makes the neural network have a characteristic that 1) distribution
Formula information storage means；2) mode of learning parallel processing；3) self study and adaptivity；4) stronger robustness and fault-tolerance.
For the training of many reference amounts degenerate network, learning network structural schematic diagram is as shown in Figure 6.I, j, k are respectively corresponded
Input layer, hidden layer, output layer.For the degenerate network that this paper is trained, input as actuating quantityOutput is amount of degradation
Coefficient matrix B in trained network (12) approximant to the change action of input.
Further, in order to illustrate the general implementation steps of heretofore described error state appraisal procedure, below with
Specific example describes in detail.The present invention is to install acquired in the digital electrical energy measurement on-line monitoring system of Mr. Yu 110kV substation
Data are analyzed as sample.Steps are as follows for concrete analysis:
(1) data of the interception morning 10 on April 7th, 2015 when afternoon 5 are returned using difference first as analysis object
One change pre-processes data.Actuating quantity is by taking temperature as an example, and amount of degradation is by taking electronic current mutual inductor is than difference as an example, pre- place
Result is managed as shown in Fig. 7, Fig. 8, Fig. 9 and Figure 10.
It is illustrated in figure 7 temperature initial data, the temperature change in one day is gentle；Fig. 8 is that temperature carries out difference normalizing
Data after change processing；ECT is illustrated in figure 9 than poor initial data；Figure 10 be ECT than difference carry out difference normalized it
Data afterwards.It can be seen from Fig. 7, Fig. 8, Fig. 9 and Figure 10 after difference normalized, it is worth between -1 to 1,
Advantageously ensure that the correctness of trained degenerate network.
(2) data being trained using BP neural network, setting training error is 0.01%, degenerate network is obtained, and
It tests to degenerate network, obtained training result is as shown in figure 11.
As seen from Figure 11, after carrying out 50 iteration, error is less than 0.01%, it is believed that study terminates, then uses
Cross check method tests to training result.
Digital electric energy metered system is assessed April 7 in conjunction with amount of degradation initial value according to the degenerate network of above-mentioned training
Error state interior in short-term after when afternoon 5, and analyze on-line monitoring system data acquired in the same period.As a result as schemed
Shown in 12.Correlation curve 1 and curve 2, it can be seen that the assessment result based on many reference amounts degradation model, with the actual running results
Trend substantially conforms to, and for assessment errors compared with actual error, absolute error is not more than 0.2% in short-term.
Understanding deeper to the present invention is as follows: final only to need to the digital electric energy metered system in a certain substation
The sensor of measurement degeneration parameter actuating quantity, i.e., environmentally sensitive unit 10 shown in fig. 5 are set, but sensor is not limited to
It is listed in Fig. 5；Without the sensor for measuring digital electrical energy measurement degeneration parameter is arranged, i.e., circuit current sensing shown in fig. 5
Unit 11.By the way that degeneration parameter actuating quantity is inputted well-drilled degenerate network, digital electric energy metered system can be obtained and miss
Poor state.
As it will be easily appreciated by one skilled in the art that the foregoing is merely illustrative of the preferred embodiments of the present invention, not to
The limitation present invention, any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should all include
Within protection scope of the present invention.
Claims (1)
1. a kind of error evaluation system of digitalized electric energy metering system, which is used for digital electric energy metering
System carries out online error state assessment, and including environmentally sensitive unit, Current Voltage sensing unit and central processing unit,
It is characterized in that:
Under the premise of uniform sampling rate, respectively the factor of the influence systematic error to the digitalized electric energy metering system and
Corresponding systematic error measures；It is described influence systematic error factor be environment temperature, ambient humidity, electric parameter,
It is frequency and harmonic wave, electromagnetic field, vibration, communication abnormality, decompression, one or more in cutout；
The environmentally sensitive unit includes temperature sensor, humidity sensor, the second microprocessor and the second DC power supply；
Wherein the temperature sensor, humidity sensor are respectively used to the digitalized electric energy metering system local environment as monitoring object
Temperature, humidity progress real-time measurement are executed, and the local environment is executed into temperature, humidity measurement results common transport to described
Second microprocessor, second DC power supply are then used to provide working power to other component units in described；
The Current Voltage sensing unit includes electromagnetic current transducer, electromagnetic potential transformer, first microprocessor, with
And the first DC power supply for device each in Current Voltage sensing unit power supply；Wherein the current transformer, the voltage are mutual
Sensor is used to carry out real-time measurement to the frequency and harmonic wave of the digitalized electric energy metering system；
In addition, the central processing unit is connected with first, second microprocessor simultaneously, and built in real time according to measurement data
Vertical and update many reference amounts degradation model assesses digital electric energy metered system current and future error state in short-term, specifically:
The lasting sexual factor for choosing influence stability is modeled, respectively environment temperature, humidity, mains frequency, harmonic wave four
Influence factor collective effect influences four degeneration parameter f in the digital electric energy metered systemPT, δPT, fCT, δCT, can use as follows
Equation group description:
In equation 1 above, Temp (t), Hum (t), Freq (t), Harm (t), be respectively temperature, humidity, frequency, harmonic wave at any time
The function of variation；It enables:
Solution of equations matrix is denoted as A, then formula 1 can be written as:
Formula 4 is digital electric energy metered system parameter degeneration equation；Wherein,It is parameter degeneration vector,It is degeneration parameter, it is fixed
Adopted A is many reference amounts degenerate network of the degeneration parameter in the case where acting on vector；
The error evaluation system further includes computer, and the integrated data analyzing and processing software of the computer operation is as state
Evaluation Platform executes following steps processing:
Step (1) pre-processes the measurement data using difference normalization preprocess method；
It is unified that sample rate is carried out to input many reference amounts, then does difference normalized, is respectively obtained:
{ΔTemp(1),ΔTemp(2),...,ΔTemp(n)}；
{ΔHum(1),ΔHum(2),...,ΔHum(n)}；
{ΔFreq(1),ΔFreq(2),...,ΔFreq(n)}；
{ΔHarm(1),ΔHarm(2),...,ΔHarm(n)}；
{ΔδCT(1),ΔδCT(2),...,ΔδCT(n)}；
{ΔfCT(1),ΔfCT(2),...,ΔfCT(n)}；
{ΔδPT(1),ΔδPT(2),...,ΔδPT(n)}；
{ΔfPT(1),ΔfPT(2),...,ΔfPT(n)}；
Step (2) is again using artificial neural network training degenerate network specifically: by the measurement data through difference normalized
It is input to the degenerate network, acquires the corresponding degenerate network of difference；
By { Δ Temp (i), Δ Hum (i), Δ Freq (i), Δ Harm (i) } composition effect vector as the degenerate network
Input, is exported { Δ δCT(i),ΔfCT(i),ΔδPT(i),ΔfPT(i)}；Wherein integer of the i between 1~n；
The variable quantity for the output degeneration parameter that step (3) is obtained by the step (2), is worked as in conjunction with degeneration parameter initial value
Preceding each state parameter of electric energy metered system assesses error current, to realize the error state prison of digitalized electric energy metering system
It surveys.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510458899.6A CN105158723B (en) | 2015-07-30 | 2015-07-30 | A kind of error evaluation system and method for digitalized electric energy metering system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510458899.6A CN105158723B (en) | 2015-07-30 | 2015-07-30 | A kind of error evaluation system and method for digitalized electric energy metering system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105158723A CN105158723A (en) | 2015-12-16 |
| CN105158723B true CN105158723B (en) | 2019-02-15 |
Family
ID=54799641
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510458899.6A Active CN105158723B (en) | 2015-07-30 | 2015-07-30 | A kind of error evaluation system and method for digitalized electric energy metering system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105158723B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107490778B (en) * | 2017-08-25 | 2019-07-02 | 成都思晗科技股份有限公司 | A kind of multifactor quantitative analysis method of measurement error influence factor |
| CN107861089A (en) * | 2017-09-26 | 2018-03-30 | 国网江西省电力公司电力科学研究院 | A kind of intelligent electric energy meter method of testing for being adapted to exceptional operating conditions |
| CN108088469B (en) * | 2017-11-14 | 2021-08-17 | 中国航空工业集团公司西安飞机设计研究所 | Long-endurance aircraft inertial navigation error compensation method |
| CN108445435B (en) * | 2018-02-05 | 2020-08-25 | 国电南瑞科技股份有限公司 | Online error evaluation method for electric energy meter calibrating device |
| CN108693496A (en) * | 2018-05-07 | 2018-10-23 | 国网山东省电力公司电力科学研究院 | A kind of intelligent electric energy meter error predictor method based on parameter degeneration equation |
| CN109188341A (en) * | 2018-09-18 | 2019-01-11 | 南方电网科学研究院有限责任公司 | Live kinematic error appraisal procedure, device, equipment and the storage medium of electric energy meter |
| CN109597014A (en) * | 2018-11-30 | 2019-04-09 | 国网上海市电力公司 | A kind of electric energy meter error diagnostic method based on artificial intelligence technology |
| CN110658487A (en) * | 2019-10-18 | 2020-01-07 | 国网福建省电力有限公司 | Meter box and system capable of achieving intelligent electric meter error online estimation |
| CN110907885B (en) * | 2019-12-06 | 2021-10-01 | 国网湖北省电力有限公司计量中心 | Method and system for evaluating field operation state of digital electric energy metering system |
| CN113138362A (en) * | 2021-05-11 | 2021-07-20 | 深圳市先行电气技术有限公司 | Calibration method for automatically compensating electric energy metering error of electric energy meter |
| CN113887846B (en) * | 2021-12-07 | 2022-02-22 | 武汉格蓝若智能技术有限公司 | Out-of-tolerance risk early warning method for capacitive voltage transformer |
| CN115693916B (en) * | 2022-09-07 | 2023-05-30 | 国网安徽省电力有限公司宿州供电公司 | Intelligent online monitoring method and system for direct-current power supply of transformer substation |
| CN115480204A (en) * | 2022-09-29 | 2022-12-16 | 武汉格蓝若智能技术有限公司 | Current transformer operation error online evaluation optimization method based on big data deduction |
| CN117233687B (en) * | 2023-11-13 | 2024-02-06 | 威胜集团有限公司 | CVT initial error assessment method, medium and terminal based on historical data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1172261A (en) * | 1997-08-27 | 1998-02-04 | 武汉国测电器公司 | Method and device for checking electric power mutual inductor |
| CN201069470Y (en) * | 2007-07-20 | 2008-06-04 | 太原市优特奥科电子科技有限公司 | Real time online monitoring device for integrated error of high-voltage power calculation device |
| CN102707256A (en) * | 2012-06-20 | 2012-10-03 | 北京航空航天大学 | Fault diagnosis method based on BP-Ada Boost nerve network for electric energy meter |
| CN202758062U (en) * | 2012-07-05 | 2013-02-27 | 广东电网公司电力科学研究院 | Overall metering error detection system of digital transformer station electric energy metering device |
| CN202794496U (en) * | 2012-05-22 | 2013-03-13 | 鹤岗电业局 | Real-time online monitoring instrument for power metering device |
| CN104076319A (en) * | 2014-05-04 | 2014-10-01 | 贵州电力试验研究院 | Online error analysis system of digitized electric energy metering device |
-
2015
- 2015-07-30 CN CN201510458899.6A patent/CN105158723B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1172261A (en) * | 1997-08-27 | 1998-02-04 | 武汉国测电器公司 | Method and device for checking electric power mutual inductor |
| CN201069470Y (en) * | 2007-07-20 | 2008-06-04 | 太原市优特奥科电子科技有限公司 | Real time online monitoring device for integrated error of high-voltage power calculation device |
| CN202794496U (en) * | 2012-05-22 | 2013-03-13 | 鹤岗电业局 | Real-time online monitoring instrument for power metering device |
| CN102707256A (en) * | 2012-06-20 | 2012-10-03 | 北京航空航天大学 | Fault diagnosis method based on BP-Ada Boost nerve network for electric energy meter |
| CN202758062U (en) * | 2012-07-05 | 2013-02-27 | 广东电网公司电力科学研究院 | Overall metering error detection system of digital transformer station electric energy metering device |
| CN104076319A (en) * | 2014-05-04 | 2014-10-01 | 贵州电力试验研究院 | Online error analysis system of digitized electric energy metering device |
Non-Patent Citations (1)
| Title |
|---|
| 人工神经网络在电能计量综合误差分析中的应用;陶锐;《内蒙古科技与经济》;20080430(第7期);第86-87页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105158723A (en) | 2015-12-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105158723B (en) | A kind of error evaluation system and method for digitalized electric energy metering system | |
| CN103268519B (en) | Based on the power-system short-term load forecasting method and the device that improve Lyapunov index | |
| CN102377180B (en) | Power system load modeling method based on electric energy quality monitoring system | |
| CN101441231B (en) | Harmonic electric energy metering error analytical apparatus | |
| CN103020459B (en) | A kind of cognitive method of various dimensions electricity consumption behavior and system | |
| CN104076319A (en) | Online error analysis system of digitized electric energy metering device | |
| CN106786567B (en) | A kind of online load modeling method based on PMU noise like data | |
| CN110031788A (en) | A kind of hollow coil current transformer error environment correlation analysis | |
| CN114201924B (en) | Solar irradiance prediction method and system based on transfer learning | |
| Cui et al. | HVDC transmission line fault localization base on RBF neural network with wavelet packet decomposition | |
| CN110045317A (en) | A kind of mutual inductor measurement error online test method and system | |
| Kong et al. | A remote estimation method of smart meter errors based on neural network filter and generalized damping recursive least square | |
| CN105488335A (en) | Lyapunov exponent based power system load prediction method and apparatus | |
| CN105044649A (en) | Current transformer error horizontal dynamic detection method and system | |
| CN113837432A (en) | Power system frequency prediction method driven by physics-data combination | |
| CN114583767B (en) | Data-driven wind power plant frequency modulation response characteristic modeling method and system | |
| CN111091223A (en) | Distribution transformer short-term load prediction method based on Internet of things intelligent sensing technology | |
| CN111553080B (en) | Closed-loop identification method for load dynamic equivalent non-mechanism model parameters of distribution substation | |
| Zhang et al. | A hierarchical framework for ambient signals based load modeling: Exploring the hidden quasi-convexity | |
| KR100946278B1 (en) | Modeling and parameter test method for wind farm | |
| CN107069710B (en) | Power system state estimation method considering new energy space-time correlation | |
| CN106570345A (en) | Parameter identification method based on graph theory and device | |
| CN114861969A (en) | LSTM-based electric power system oscillation risk prediction method | |
| Chu et al. | A relaxed support vector data description algorithm based fault detection in distribution systems | |
| CN105808833A (en) | Online parameter identification method of parallel synchronous generator based on multi-data sets |
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 |