CN111273212B - Data-driven electric quantity sensor error online evaluation closed-loop improvement method, system and medium - Google Patents
Data-driven electric quantity sensor error online evaluation closed-loop improvement method, system and medium Download PDFInfo
- Publication number
- CN111273212B CN111273212B CN202010112437.XA CN202010112437A CN111273212B CN 111273212 B CN111273212 B CN 111273212B CN 202010112437 A CN202010112437 A CN 202010112437A CN 111273212 B CN111273212 B CN 111273212B
- Authority
- CN
- China
- Prior art keywords
- electric quantity
- quantity sensor
- iteration
- data
- error
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000011156 evaluation Methods 0.000 title claims abstract description 33
- 230000006872 improvement Effects 0.000 title claims abstract description 17
- 238000012937 correction Methods 0.000 claims abstract description 85
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 6
- 230000006870 function Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 7
- 230000005611 electricity Effects 0.000 claims description 6
- 238000009825 accumulation Methods 0.000 claims description 2
- 230000008859 change Effects 0.000 abstract description 6
- 238000012360 testing method Methods 0.000 abstract description 2
- 238000012986 modification Methods 0.000 description 14
- 230000004048 modification Effects 0.000 description 14
- 230000008569 process Effects 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 7
- 238000009795 derivation Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000013499 data model Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/04—Testing or calibrating of apparatus covered by the other groups of this subclass of instruments for measuring time integral of power or current
Abstract
The invention discloses a data-driven electric quantity sensor error online evaluation closed-loop improvement method, a system and a medium, and the method comprises the steps of establishing an n-th iteration electric quantity sensing data correction model set; collecting electric quantity sensing data to generate a sensing data set; generating an electric quantity sensing data correction coefficient set after the nth iteration according to the electric quantity sensing data correction model set, and correcting the electric quantity sensing data set to generate a corrected data set; carrying out electric quantity and electric quantity sensor error deduction on the correction data set according to an open-loop data driving algorithm to form a deduction result; if the end condition is not met, updating the iteration number n and the electric quantity sensing data correction model, and continuing iteration after accumulating the electric quantity sensing data correction coefficient set; otherwise, outputting an iteration result. The method does not need field test, does not need to change the topology of the physical system, does not influence physical operation, can effectively improve the result deduction accuracy, improves the generalization of a deduction model, realizes high efficiency and safety, and does not influence the physical system.
Description
Technical Field
The invention relates to an error evaluation method during the operation of an electric quantity sensor, in particular to a data-driven electric quantity sensor error online evaluation closed-loop improvement method, a data-driven electric quantity sensor error online evaluation closed-loop improvement system and a data-driven electric quantity sensor error online evaluation closed-loop improvement medium.
Background
The sensor is an important device for measuring physical parameters and is a key medium for connecting a physical world and an information world. With the continuous development and wide application of big data, internet of things and 5G communication technologies, the demand for information physical fusion is increasing day by day, the basic architecture function of a sensor network is more important, the reliability and accuracy of the sensor network directly determine the effectiveness of the information physical fusion process, and error evaluation needs to be carried out. In particular, the electric quantity sensor relates to the purposes of trade settlement, safety protection and the like, and is a measuring instrument which is required to be regularly verified by the state.
The traditional error evaluation method of the electric quantity sensor is detection, calibration and verification and is divided into an actual load mode and a virtual load mode. The method has the advantages of accurate, reproducible and traceable evaluation result, and has the defect that the operation mode of the physical system corresponding to the measured electric quantity sensor needs to be changed, and the change may not have the realization condition. For example, the field real load detection is carried out on the electric energy meter, the electric energy is synchronously measured on the field by using the electric energy measurement standard, the electric energy is compared with the electric energy measured by the measured electric energy meter, and the difference value of the two electric energies, namely the measurement error of the measured electric energy meter is calculated; the electric energy measurement standard can be traced to the national standard, the results of error measurement for multiple times under the same environment are consistent, and the evaluation result of the error of the tested electric energy meter is accurate; the intervention of the electric energy metering standard changes the physical operation mode of the original electric power system, namely, the electric energy metered by the electric energy meter alone is converted into the electric energy metered by the electric energy metering standard and the electric energy meter, and the intervention of the electric energy metering standard needs to disconnect a main electric energy supply loop or a secondary loop of a current transformer in a short time, so that serious safety accidents or power failure accidents can be caused.
At present, an electric quantity sensor network comprises a large number of electric quantity sensor units, a large amount of time is needed for evaluating errors through a traditional method, extremely large resources need to be invested, and the installation positions and the operation modes of part of electric quantity sensors do not support calibration and verification, so that the limitation of the traditional evaluation method can be seen. In recent years, researchers and engineers have proposed state evaluation and error deduction of electric quantity sensors based on data driving. For example, chinese patent application No. 201410413185.9 discloses an online evaluation method and system for the entire metering error of an electric energy metering device, in which a metering device state evaluation method combining electric energy data and electric power system physical topology is described, and chinese patent application No. 201611092500.8 discloses an electric energy metering device error detection method and system based on big data deduction, in which a metering device error deduction method combining bus imbalance rate, electric energy data and electric power system physical topology is described. The methods are based on the existing electric quantity sensor network, data communication network and data platform, and function modules are added on the technical system to realize algorithm operation, so that the actual operation topology does not need to be changed, and no influence is generated on an operation system. Both of the methods of deducting the electric quantity sensor error based on data driving represented by the above two techniques are in an open loop operation manner, i.e., "data" → "deduction" → "result". The open-loop operation mode is limited by data quality, reasonability of a deduction model, matching degree of a data model and the like, generalization is not strong, and the model needs to be further optimized.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the invention provides a closed-loop improved method, a system and a medium for online evaluation of errors of a data-driven electric quantity sensor, aiming at the defects of the existing open-loop type metering error deduction technology.
In order to solve the technical problems, the invention adopts the technical scheme that:
a closed loop improvement method for online error evaluation of a data-driven electric quantity sensor comprises the following implementation steps:
1) initializing iteration times n and error deduction times Q;
2) creating a sensing data correction model set S of the nth iterationn;
3) Synchronously acquiring sensing DATA of the electric quantity sensor group according to a time sequence and generating a sensing DATA set DATA;
4) correcting model set S according to sensing data of nth iterationnGenerating a sensing data correction coefficient set A after the nth iterationnModifying the set of coefficients A based on the sensed datanAnd generating a correction DATA set DATA' (n) from the sensing DATA set DATA;
5) updating the value of the error deduction times Q, and carrying out electric quantity sensor error deduction on the correction DATA set DATA' (n) according to an open-loop DATA driving algorithm to form a deduction result E (Q);
6) judging whether a preset iteration ending condition is met, if the preset iteration ending condition is not met, updating the iteration number n and the sensing data correction model SnCumulative sensing data correction coefficient set AnSkipping to execute the step 4); otherwise, outputting an iteration result A of the nth iterationnAnd then, the process is ended.
Optionally, the sensing data modification model set SnThe array paradigm of (a) is shown as follows:
Sn=[S1n,S2n,S3n,…,SPn]
in the above formula, n is the number of iterations, P represents the total number of the electric quantity sensors of the electric quantity sensor group, and S1n~SPnAnd the sensing data correction models respectively represent the nth iteration of the 1 st to the P th electric quantity sensors in the electric quantity sensor group.
Optionally, the sensing data modification model is a dynamic step modification model, and a function expression of the dynamic step modification model is as follows:
in the above formula, SLnDynamic step-by-step correction model, alpha, representing the nth iteration of any Lth electrical quantity sensorLFor a fixed step-size coefficient, ELnIs the result of the nth error deduction implemented by the L electric quantity sensor error deduction model, n is the iteration number, m is the iteration evolution scale of the correction model, ELAAnd deducing a result for a certain error of the L-th electric quantity sensor.
Optionally, the sensing data modification model is a PID-like modification model, and the functional expression of the PID-like modification model is as follows:
in the above formula, SLnPID-like correction model representing nth iteration of any Lth electric quantity sensor, ELnIs the result of the nth error deduction implemented by the L electric quantity sensor error deduction model, EL(n-1)Is the result of the error deduction model of the L electric quantity sensor for n-1 timesa、Kb、KcAnd the coefficient is a PID coefficient, n is iteration times, and m is an iteration evolution scale of the modified model.
Optionally, the functional expression of the sensing DATA set DATA generated in step 3) is as follows:
in the above formula, D11~DP1Respectively representing the sensing data of the 1 st to P electric quantity sensors in the electric quantity sensor group at the 1 st time sequence point; d12~DP2Respectively representing the sensing data of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the 2 nd time sequence point; d1T~DPTRespectively represents the measured values of the 1 st to the P th electric quantity sensors in the electric quantity sensor group at the T-th time sequence point.
Optionally, the sensing data correction coefficient set a after the nth iteration is generated in step 4)nThe function expression of (a) is as follows:
An=[A1n,A2n,…,APn]
in the above formula, AnRepresents the set of correction coefficients of the sensed data after the nth iteration, A1n~APnThe function expression of the correction coefficients after the nth iteration of the 1 st to P th electric quantity sensors in the electric quantity sensor group is shown as the following formula:
in the above formula, APnIs the correction coefficient after the nth iteration corresponding to any P-th electric quantity sensor, n is the iteration number, m is the parameter of 1-n and is used for the accumulation or the multiplication operation, SPmA sensing data correction model representing the mth iteration of the pth electrical quantity sensor in the electrical quantity sensor group, EnRepresenting a set of electric quantity sensor error deduction results after the nth iteration;
correcting the coefficient set A according to the sensing data in the step 4)nAnd the sensing DATA set DATA to generate the correction DATA set DATA' (n) as a function of the expression:
in the above formula, DATA' (n) represents sensing correction DATA, A1n~APnRepresenting a sensing data correction coefficient set A after the nth iterationnCorrection coefficient of middle 1-P electric quantity sensor, D11~DP1Respectively representing the sensing data of the 1 st to P electric quantity sensors in the electric quantity sensor group at the 1 st time sequence point; d12~DP2Respectively representing the sensing data of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the 2 nd time sequence point; d1T~DPTRespectively represents the measured values of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the T th time sequence point, EnAnd representing a set of electric quantity sensor error deduction results after the nth iteration.
Optionally, the functional expression of the deduction result e (q) formed by the deduction of the error of the developed electric quantity sensor in step 5) is as follows:
E(Q)=[E1Q E2Q … EPQ]
in the above formula, E1Q~EPQRespectively are error deduction results after Q-th iteration of the 1 st electric quantity sensor to the P-th electric quantity sensor.
Optionally, the iteration ending condition preset in step 6) is that an error threshold is met or the iteration number n reaches a set iteration number upper limit.
In addition, the invention also provides a data-driven electricity sensor error online evaluation closed-loop improvement system, which comprises a computer device, wherein the computer device is programmed or configured to execute the steps of the data-driven electricity sensor error online evaluation closed-loop improvement method, or a computer program which is programmed or configured to execute the data-driven electricity sensor error online evaluation closed-loop improvement method is stored on a memory of the computer device.
Additionally, the present invention provides a computer readable storage medium having stored thereon a computer program programmed or configured to perform the data-driven method for closed-loop improved on-line assessment of electrical quantity sensor error.
Compared with the prior art, the invention has the following advantages:
1. the traditional electric quantity sensor error deduction method is an open-loop operation mode, namely a mode of 'data' → 'deduction' → 'result', the open-loop operation mode is limited by data quality, deduction model rationality, data model matching degree and the like, and generalization is not strong; the closed-loop operation mode is 'data' → 'correction' → 'deduction' → 'preliminary result' → 'iteration correction model' → 'data iteration deduction' → 'final result', so that the result deduction accuracy can be effectively improved, and the deduction model generalization is improved.
2. The electric quantity sensor error online evaluation closed-loop improvement method provided by the invention does not need field test, does not need to change the topology of a physical system, does not influence physical operation, can solve the problems of weak generalization and inaccurate derivation result of an open-loop metering error derivation method to a certain extent, can effectively improve the accuracy of result derivation and improve the generalization of a derivation model, and has the advantages of high efficiency and safety of the implementation method and no influence on the physical system.
Drawings
FIG. 1 is a schematic diagram of a basic process of an embodiment of the present invention.
Fig. 2 is a flowchart of a metering error closed-loop iterative deduction gradual process based on a step correction model according to an embodiment of the present invention.
Fig. 3 is a metering error closed-loop iterative deduction gradual change process based on a PID-like correction model according to a second embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the drawings and specific preferred embodiments of the description, without thereby limiting the scope of protection of the invention.
The first embodiment is as follows:
as shown in fig. 1, the implementation steps of the data-driven closed-loop improved method for online error evaluation of an electrical quantity sensor of this embodiment include:
1) initializing iteration times n and error deduction times Q;
2) creating a sensing data correction model set S of the nth iterationn;
3) Synchronously acquiring sensing DATA of the electric quantity sensor group according to a time sequence and generating a sensing DATA set DATA;
4) correcting model set S according to sensing data of nth iterationnGenerating a sensing data correction coefficient set A after the nth iterationnModifying the set of coefficients A based on the sensed datanAnd generating a correction DATA set DATA' (n) from the sensing DATA set DATA;
5) updating the value of the error deduction times Q, and carrying out electric quantity sensor error deduction on the correction DATA set DATA' (n) according to an open-loop DATA driving algorithm to form a deduction result E (Q);
6) judging whether a preset iteration ending condition is met, if the preset iteration ending condition is not met, updating the iteration number n and the sensing data correction model SnCumulative sensing data correction coefficient set AnSkipping to execute the step 4); otherwise, outputting an iteration result A of the nth iterationnAnd then, the process is ended.
In this embodiment, step 1) initializes the number of iterations n to 0 and the number of error deductions Q to 0;
in this embodiment, the sensing data modification model set SnThe array paradigm of (a) is shown as follows:
Sn=[S1n,S2n,S3n,…,SPn]
in the above formula, n is the number of iterations, P represents the total number of the electric quantity sensors of the electric quantity sensor group, and S1n~SPnAnd the sensing data correction models respectively represent the nth iteration of the 1 st to the P th electric quantity sensors in the electric quantity sensor group.
As an optional implementation manner, the sensing data modification model in this embodiment is a dynamic step modification model, and a functional expression thereof is shown as follows:
in the above formula, SLnDynamic step-by-step correction model, alpha, representing the nth iteration of any Lth electrical quantity sensorLFor a fixed step-size coefficient, ELnIs the result of the nth error deduction implemented by the L electric quantity sensor error deduction model, n is the iteration number, m is the iteration evolution scale of the correction model, ELAAnd deducing a result for a certain error of the L-th electric quantity sensor. In this embodiment, a fixed step coefficient α is setLAt 0.5, the iterative evolution scale m is set to 5. In particular, when n is 0, SLn1. The iteration evolution scale m is set manually, the larger the value of m, the longer the historical iteration window period considered by the model is, the more historical iteration results are used for calculation, and the longer the dynamic process of the considered model iteration evolution is, the larger the calculation amount is; the smaller the m value is, the shorter the historical iteration window period considered by the model is, the fewer the historical iteration results used for calculation are, the shorter the dynamic process of the iterative evolution of the considered model is, and the smaller the calculation amount is.
In the functional expression of the dynamic stepping correction model:
the absolute value of the sum of the error deduction results of the L-th electric quantity sensor for m times is expressed as:
the absolute value of the sum of the results of the n-th deduction errors of the L-th electric quantity sensor is represented as:
the sum of the absolute values of the error deduction results of the L-th electric quantity sensor for m times is expressed as:
the sum of the absolute values of the error deduction results of the L-th electric quantity sensor for the last n times is expressed as:
and is defined as the dynamic step coefficient under different conditions as:
if n > m (n-m >0) then the dynamic step size coefficient is selected as:
if n < ═ m (n-m < ═ 0), then the dynamic step size coefficient is chosen to be:
in particular, when n is 0, SLn=1。
In this embodiment, when the sensing DATA of the electric quantity sensor group is synchronously acquired according to the time sequence in step 3), the acquired sensing DATA is arranged from each electric quantity sensor according to the time sequence to form a group of P rows and T columns of matrix, and a sensing DATA set DATA is generated corresponding to the sensing DATA of the P electric quantity sensors at the T time sequence points. The functional expression of the sensing DATA set DATA generated in step 3) of this embodiment is shown as follows:
in the above formula, D11~DP1Respectively representing the sensing data of the 1 st to the P th electric quantity sensors in the electric quantity sensor group at the 1 st time sequence point;D12~DP2Respectively representing the sensing data of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the 2 nd time sequence point; d1T~DPTRespectively represents the measured values of the 1 st to the P th electric quantity sensors in the electric quantity sensor group at the T-th time sequence point.
In this embodiment, the sensing data correction coefficient set a after the nth iteration is generated in step 4)nThe function expression of (a) is as follows:
An=[A1n,A2n,…,APn]
in the above formula, AnRepresents the set of correction coefficients of the sensed data after the nth iteration, A1n~APnThe function expression of the correction coefficients after the nth iteration of the 1 st to P th electric quantity sensors in the electric quantity sensor group is shown as the following formula:
in the above formula, APnThe correction coefficient after the nth iteration corresponding to any P electric quantity sensor, n is the iteration frequency, m is the error deduction frequency, SPmA sensing data correction model representing the mth iteration of the pth electrical quantity sensor in the electrical quantity sensor group, EnRepresenting a set of electric quantity sensor error deduction results after the nth iteration;
the embodiment corrects the coefficient set A according to the sensing data in step 4)nAnd the sensing DATA set DATA to generate the correction DATA set DATA' (n) as a function of the expression:
in the above formula, DATA' (n) represents sensing correction DATA, A1n~APnRepresenting a sensing data correction coefficient set A after the nth iterationnMiddle 1-P electric quantity sensorCorrection coefficient of D11~DP1Respectively representing the sensing data of the 1 st to P electric quantity sensors in the electric quantity sensor group at the 1 st time sequence point; d12~DP2Respectively representing the sensing data of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the 2 nd time sequence point; d1T~DPTRespectively represents the measured values of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the T th time sequence point, EnAnd representing a set of electric quantity sensor error deduction results after the nth iteration.
In this embodiment, the functional expression of the deduction result e (q) formed by the deduction of the error of the developed electric quantity sensor in step 5) is shown as follows:
E(Q)=[E1Q E2Q … EPQ]
in the above formula, E1Q~EPQRespectively are error deduction results after Q-th iteration of the 1 st electric quantity sensor to the P-th electric quantity sensor. In this embodiment, the deduction algorithm disclosed in chinese patent document having application number 201611092500.8 and entitled "method and system for detecting errors of electric energy metering device based on big data deduction" is adopted in step 5) to perform electric quantity sensor error deduction.
If the deduction iteration is carried out for the Q-th time, Q groups of error deduction results are formed in total, and an error deduction data iteration matrix E shown as the following formula is formed.
In the above formula, E11~EP1Respectively the error deduction results after 1 st iteration of the 1 st to P electric quantity sensors, E12~EP2Respectively the error deduction results after the 2 nd iteration of the 1 st to P th electric quantity sensors, E1Q~EPQRespectively are error deduction results after Q-th iteration of the 1 st electric quantity sensor to the P-th electric quantity sensor.
As an optional implementation manner, the iteration ending condition preset in step 6) is that an error threshold is met, and an iteration ending error valve can be set manuallyThe value ε ═ ε1,ε2]For example, the iteration termination error threshold ε is set to [ 0.1%, 0.01%]When the formula is as follows:
When true, the iteration ends. Wherein P is the number of the electric quantity sensors, EpQFor the error deduction results after the qth iteration of the pth electric quantity sensor, respectively, the functional expression of E is:
in the above formula, P is the number of the electric quantity sensors, EpQAnd respectively deducing the error after the Q-th iteration of the p-th electric quantity sensor. Output iteration result AnThe method is based on the error deduction result of the electric quantity sensor group obtained by the method of the embodiment.
As another optional implementation manner, the iteration ending condition preset in step 6) is that the iteration number n reaches a set iteration number upper limit, for example, 100 times.
Finally, if the iteration end condition is not met, updating the sensing data correction model set SnCumulative sensing data correction coefficient set AnAnd adding 1 to the assignment of n, and circularly entering the step 4) until the condition of finishing the iteration is met. Wherein the accumulated sensing data correction coefficient set AnSpecifically, the correction coefficients of the respective electric quantity sensors are accumulated, for example, the correction coefficient a is accumulated for an arbitrary electric quantity sensor PPn。
Fig. 3 is a gradual change process of the closed-loop iterative deduction of the metering error based on the step correction model in the present embodiment, and referring to fig. 2, it can be seen that the more the iteration times, the more stable the deduction result of the metering error tends to be, and the closer to the actual error.
In addition, the present embodiment also provides a data-driven closed-loop improved system for online evaluation of electrical quantity sensor error, which includes a computer device programmed or configured to execute the steps of the aforementioned method for online evaluation of closed-loop improved method for data-driven electrical quantity sensor error, or a computer program stored in a memory of the computer device and programmed or configured to execute the aforementioned method for online evaluation of closed-loop improved method for data-driven electrical quantity sensor error.
In addition, the present embodiment also provides a computer readable storage medium having stored thereon a computer program programmed or configured to perform the aforementioned data-driven method for closed-loop improved on-line assessment of electrical quantity sensor error.
Example two:
the present embodiment is basically the same as the first embodiment, and the main differences are as follows: the sensing data correction model in this embodiment is a PID-like correction model, and the functional expression thereof is shown as follows:
in the above formula, SLnPID-like correction model representing nth iteration of any Lth electric quantity sensor, ELnIs the result of the nth error deduction implemented by the L electric quantity sensor error deduction model, EL(n-1)Is the result of the error deduction model of the L electric quantity sensor for n-1 timesa、Kb、KcAnd the coefficient is a PID coefficient, n is iteration times, and m is an iteration evolution scale of the modified model. In this embodiment, the PID coefficient K is seta=5、Kb=0.1、Kc0.5. In particular, when n is 0, SLn1. Fig. 3 is a metering error closed-loop iteration deduction gradual change process based on the PID-like correction model in the present embodiment, and fig. 3 also shows that the more the iteration times, the more stable the deduction result of the metering error tends to be, the more it approaches to the actual error. Wherein, the iterative evolution scale m of the correction model is set artificially, the larger the value of m is, the longer the historical iterative window period considered by the model is, and the history used for calculation isThe more iteration results, the longer the dynamic process of the iterative evolution of the considered model is, and the larger the calculated amount is; the smaller the m value is, the shorter the historical iteration window period considered by the model is, the fewer the historical iteration results used for calculation are, the shorter the dynamic process of the iterative evolution of the considered model is, and the smaller the calculation amount is.
Referring to the first embodiment and the second embodiment, the sensing data modification model can flexibly select a dynamic stepping modification model and a PID-like modification model according to a use scene.
It should be noted that the technical principle and idea corresponding to the present invention can also be applied to error deduction of electric quantity sensor networks in other fields, such as error deduction of flow meter groups in oil/water/natural gas pipe networks, error deduction of information flow meters in mobile/broadband information networks, error deduction of thermal work meters in operation, and the like.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may occur to those skilled in the art without departing from the principle of the invention, and are considered to be within the scope of the invention.
Claims (7)
1. A closed loop improvement method for online error evaluation of a data-driven electric quantity sensor is characterized by comprising the following implementation steps:
1) initializing iteration times n and error deduction times Q;
2) creating a sensing data correction model set S of the nth iterationn;
3) Synchronously acquiring sensing DATA of the electric quantity sensor group according to a time sequence and generating a sensing DATA set DATA;
4) correcting model set S according to sensing data of nth iterationnGenerating a sensing data correction coefficient set A after the nth iterationnModifying the set of coefficients A based on the sensed datanAnd generating a correction DATA set DATA' (n) from the sensing DATA set DATA;
5) updating the value of the error deduction times Q, and carrying out electric quantity sensor error deduction on the correction DATA set DATA' (n) according to an open-loop DATA driving algorithm to form a deduction result E (Q);
6) judging whether a preset iteration ending condition is met, if the preset iteration ending condition is not met, updating the iteration number n and the sensing data correction model SnCumulative sensing data correction coefficient set AnSkipping to execute the step 4); otherwise, outputting an iteration result A of the nth iterationnAnd ending;
the sensing data correction model set SnThe array paradigm of (a) is shown as follows:
Sn=[S1n,S2n,S3n,…,SPn]
in the above formula, n is the number of iterations, P represents the total number of the electric quantity sensors of the electric quantity sensor group, and S1n~SPnThe sensing data correction model respectively represents the nth iteration of the 1 st to the P th electric quantity sensors in the electric quantity sensor group;
the sensing data correction model is a dynamic stepping correction model, and the function expression of the dynamic stepping correction model is shown as the following formula:
in the above formula, SLnDynamic step-by-step correction model, alpha, representing the nth iteration of any Lth electrical quantity sensorLFor a fixed step-size coefficient, ELnIs the result of the nth error deduction implemented by the L electric quantity sensor error deduction model, n is the iteration number, m is the iteration evolution scale of the correction model, ELADeducing a result for a certain error of the L-th electric quantity sensor;
or the sensing data correction model is a PID-like correction model, and the functional expression of the PID-like correction model is shown as the following formula:
in the above formula, SLnPID-like correction model representing nth iteration of any Lth electric quantity sensor, ELnIs the result of the nth error deduction implemented by the L electric quantity sensor error deduction model, EL(n-1)Is the result of the error deduction model of the L electric quantity sensor for n-1 timesa、Kb、KcAnd the coefficient is a PID coefficient, n is iteration times, and m is an iteration evolution scale of the modified model.
2. The method for improving the online evaluation closed loop of the error of the DATA-driven electric quantity sensor according to claim 1, wherein the function expression of the sensing DATA set DATA generated in the step 3) is shown as follows:
in the above formula, D11~DP1Respectively representing the sensing data of the 1 st to P electric quantity sensors in the electric quantity sensor group at the 1 st time sequence point; d12~DP2Respectively representing the sensing data of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the 2 nd time sequence point; d1T~DPTRespectively represents the measured values of the 1 st to the P th electric quantity sensors in the electric quantity sensor group at the T-th time sequence point.
3. The data-driven closed-loop improvement method for online error evaluation of electric quantity sensor according to claim 1, characterized in that the sensing data correction coefficient set A after the nth iteration is generated in the step 4)nThe function expression of (a) is as follows:
An=[A1n,A2n,...,APn]
in the above formula, AnRepresents the set of correction coefficients of the sensed data after the nth iteration, A1n~APnRepresents the correction coefficient after the nth iteration of the 1 st to the P th electric quantity sensors in the electric quantity sensor group, and the P th electric quantity sensorThe functional expression of the correction coefficient after the nth iteration corresponding to the electric quantity sensor is shown as the following formula:
in the above formula, APnThe correction coefficient after the nth iteration corresponding to any P-th electric quantity sensor is obtained, n is the iteration number, m is a parameter from 1 to n and is used for accumulation operation, SPmA sensing data correction model representing the mth iteration of the pth electrical quantity sensor in the electrical quantity sensor group, EnA set of electric quantity sensor error deduction results after the nth iteration is shown, and a set E of electric quantity sensor error deduction results after the nth iterationnIs an absolute error;
correcting the coefficient set A according to the sensing data in the step 4)nAnd the sensing DATA set DATA to generate the correction DATA set DATA' (n) as a function of the expression:
in the above formula, DATA' (n) represents sensing correction DATA, A1n~APnRepresenting a sensing data correction coefficient set A after the nth iterationnCorrection coefficient of middle 1-P electric quantity sensor, D11~DP1Respectively representing the sensing data of the 1 st to P electric quantity sensors in the electric quantity sensor group at the 1 st time sequence point; d12~DP2Respectively representing the sensing data of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the 2 nd time sequence point; d1T~DPTRespectively represents the measured values of the 1 st to P th electric quantity sensors in the electric quantity sensor group at the T th time sequence point, EnAnd representing a set of electric quantity sensor error deduction results after the nth iteration.
4. The method for improving the online evaluation closed loop of the error of the data-driven electric quantity sensor according to claim 1, wherein the functional expression of the deduction result e (q) formed by deduction of the error of the electric quantity sensor in the step 5) is as follows:
E(Q)=[E1Q E2Q … EPQ]
in the above formula, E1Q~EPQRespectively are error deduction results after Q-th iteration of the 1 st electric quantity sensor to the P-th electric quantity sensor.
5. The data-driven electric quantity sensor error online evaluation closed-loop improvement method according to claim 1, characterized in that the preset iteration ending condition in step 6) is that an error threshold is met or the iteration number n reaches a set iteration number upper limit.
6. A data-driven electricity sensor error online evaluation closed-loop improvement system, comprising a computer device, wherein the computer device is programmed or configured to execute the steps of the data-driven electricity sensor error online evaluation closed-loop improvement method according to any one of claims 1 to 5, or a memory of the computer device has a computer program stored thereon, the computer program being programmed or configured to execute the data-driven electricity sensor error online evaluation closed-loop improvement method according to any one of claims 1 to 5.
7. A computer readable storage medium having stored thereon a computer program programmed or configured to perform the method for closed loop improved on-line error profiling of a data-driven electrical quantity sensor according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010112437.XA CN111273212B (en) | 2020-02-24 | 2020-02-24 | Data-driven electric quantity sensor error online evaluation closed-loop improvement method, system and medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010112437.XA CN111273212B (en) | 2020-02-24 | 2020-02-24 | Data-driven electric quantity sensor error online evaluation closed-loop improvement method, system and medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111273212A CN111273212A (en) | 2020-06-12 |
CN111273212B true CN111273212B (en) | 2022-03-11 |
Family
ID=70997262
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010112437.XA Active CN111273212B (en) | 2020-02-24 | 2020-02-24 | Data-driven electric quantity sensor error online evaluation closed-loop improvement method, system and medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111273212B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113468729B (en) * | 2021-06-15 | 2022-05-24 | 国网湖南省电力有限公司 | Method and system for measuring and calculating operation errors of electric power station metering device without calibration mode |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101968629A (en) * | 2010-10-19 | 2011-02-09 | 天津理工大学 | PID (Proportional Integral Derivative) control method for elastic integral BP neural network based on RBF (Radial Basis Function) identification |
CN203825181U (en) * | 2014-02-24 | 2014-09-10 | 广东东方电讯科技有限公司 | Electric-energy-meter production test system |
CN106443566A (en) * | 2016-12-01 | 2017-02-22 | 国网湖南省电力公司计量中心 | Big data deduction based electric energy metering device error detection method and system |
CN108732528A (en) * | 2018-05-28 | 2018-11-02 | 国网福建省电力有限公司电力科学研究院 | A kind of digitalized electrical energy meter method for diagnosing faults based on depth confidence network |
CN109670580A (en) * | 2018-12-21 | 2019-04-23 | 浙江工业大学 | A kind of data recovery method based on time series |
CN110531308A (en) * | 2019-10-11 | 2019-12-03 | 国网福建省电力有限公司电力科学研究院 | A kind of measurement error and its correction of temperature drift validity check method of electric energy meter calibration variation of ambient temperature |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017001950A1 (en) * | 2015-06-29 | 2017-01-05 | Abb Technology Ltd. | A method for correcting effect of saturation in current transformer and an intelligent electronic device therefor |
-
2020
- 2020-02-24 CN CN202010112437.XA patent/CN111273212B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101968629A (en) * | 2010-10-19 | 2011-02-09 | 天津理工大学 | PID (Proportional Integral Derivative) control method for elastic integral BP neural network based on RBF (Radial Basis Function) identification |
CN203825181U (en) * | 2014-02-24 | 2014-09-10 | 广东东方电讯科技有限公司 | Electric-energy-meter production test system |
CN106443566A (en) * | 2016-12-01 | 2017-02-22 | 国网湖南省电力公司计量中心 | Big data deduction based electric energy metering device error detection method and system |
CN108732528A (en) * | 2018-05-28 | 2018-11-02 | 国网福建省电力有限公司电力科学研究院 | A kind of digitalized electrical energy meter method for diagnosing faults based on depth confidence network |
CN109670580A (en) * | 2018-12-21 | 2019-04-23 | 浙江工业大学 | A kind of data recovery method based on time series |
CN110531308A (en) * | 2019-10-11 | 2019-12-03 | 国网福建省电力有限公司电力科学研究院 | A kind of measurement error and its correction of temperature drift validity check method of electric energy meter calibration variation of ambient temperature |
Non-Patent Citations (2)
Title |
---|
Data-driven compensation method for sensor drift faults in digital PID systems with unknown dynamics;Jun-Sheng Wang;《Journal of Process Control》;20171106;全文 * |
几类系统的迭代学习控制;马法力;《中国优秀硕士学位论文全文数据库 信息科技辑》;20110315;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111273212A (en) | 2020-06-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105740991A (en) | Climate change prediction method and system for fitting various climate modes based on modified BP neural network | |
CN103389472B (en) | A kind of Forecasting Methodology of the cycle life of lithium ion battery based on ND-AR model | |
CN107741578B (en) | Original meter reading data processing method for remote calibration of running error of intelligent electric energy meter | |
JP2009294969A (en) | Demand forecast method and demand forecast device | |
KR100867938B1 (en) | Prediction method for watching performance of power plant measuring instrument by dependent variable similarity and kernel feedback | |
CN102880907B (en) | Wind speed correction method and apparatus | |
CN107862170B (en) | Finite element model correction method based on dynamic polycondensation | |
CN103994334B (en) | Pipeline Leak flow estimation unit based on KPCA-RBF curve matching and method | |
CN103942461A (en) | Water quality parameter prediction method based on online sequential extreme learning machine | |
CN112949930B (en) | PA-LSTM network-based road motor vehicle exhaust high-emission early warning method | |
CN107229824A (en) | Photovoltaic power station power generation cell power curve modeling method and device | |
CN110728008B (en) | Method and device for determining expected service life of intelligent ammeter | |
CN111273212B (en) | Data-driven electric quantity sensor error online evaluation closed-loop improvement method, system and medium | |
CN114091317A (en) | Photovoltaic power station power prediction method based on NWP irradiance correction and error prediction | |
CN114692981A (en) | Medium-and-long-term runoff forecasting method and system based on Seq2Seq model | |
CN105046203A (en) | Satellite telemeasuring data self-adaptive hierarchical clustering method based on intersection angle DTW distance | |
CN103279030B (en) | Dynamic soft measuring modeling method and device based on Bayesian frame | |
CN114357737A (en) | Agent optimization calibration method for time-varying parameters of large-scale hydrological model | |
CN110657882B (en) | Bridge real-time safety state monitoring method utilizing single-measuring-point response | |
CN112539810A (en) | Intelligent water meter operation error calibration method | |
CN114565186A (en) | Method and system for correcting predicted wind speed based on generalized regression neural network | |
CN116663727A (en) | Photovoltaic power prediction method and system | |
CN111612648A (en) | Training method and device of photovoltaic power generation prediction model and computer equipment | |
CN115561697A (en) | Intelligent ammeter error analysis method | |
CN112581311B (en) | Method and system for predicting long-term output fluctuation characteristics of aggregated multiple wind power plants |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |