CN113177598B - Error electric quantity supplementing method and terminal - Google Patents
Error electric quantity supplementing method and terminal Download PDFInfo
- Publication number
- CN113177598B CN113177598B CN202110491164.9A CN202110491164A CN113177598B CN 113177598 B CN113177598 B CN 113177598B CN 202110491164 A CN202110491164 A CN 202110491164A CN 113177598 B CN113177598 B CN 113177598B
- Authority
- CN
- China
- Prior art keywords
- missing
- data
- target
- value
- maximum likelihood
- 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 32
- 230000001502 supplementing effect Effects 0.000 title claims abstract description 31
- 238000007476 Maximum Likelihood Methods 0.000 claims abstract description 61
- 239000011159 matrix material Substances 0.000 claims abstract description 25
- 230000000295 complement effect Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000004220 aggregation Methods 0.000 claims 2
- 238000004422 calculation algorithm Methods 0.000 abstract description 17
- 238000004364 calculation method Methods 0.000 description 12
- 230000005611 electricity Effects 0.000 description 10
- 230000000694 effects Effects 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention provides an error electric quantity supplementing method and a terminal, which are used for acquiring target supplementing data of a user to be supplemented in a target supplementing period, judging whether missing objects exist in the target supplementing data, and if yes, calculating the maximum likelihood attribute set of each missing object and each un-missing object in the target supplementing data; obtaining a missing value corresponding to each maximum likelihood attribute set corresponding to a missing object, and obtaining the mode of the missing value corresponding to the missing object as a filling value of the missing object; filling the filling value into the missing object to obtain complete target uncompensated data; and performing the deconvolution according to the complete target deconvolution data. The method and the device acquire the mode of the missing value as the filling value of the missing object, can maximize the space dimension of the fluctuation consistent matrix corresponding to the maximum likelihood attribute set, accords with the characteristic that the electric power data such as current, load and the like have time sequence continuity, and strengthen the influence of the neighbor data compared with a global clustering algorithm without distinction.
Description
Technical Field
The invention relates to the field of electric power metering, in particular to an error electric quantity supplementing method and a terminal.
Background
Along with the change of the running time of the power equipment, the change of the placement environment of the equipment and the continuous upgrading and improving of the equipment, the precision requirement on the electric energy metering device is gradually increased, and hidden dangers existing in the electric energy metering device are gradually developed. The running electric energy metering device is likely to have faults such as fuse fusing, screw loosening, electric energy meter black screen, clock out-of-tolerance and the like, and the faults lead to inaccurate electric energy metering, so that the problem of charging errors occurs, and the error electric quantity needs to be repaired after the faults are removed.
However, the existing electric quantity supplementing method has the following problems: 1. time and effort are consumed. The existing method for carrying out the reverse compensation on the error electric quantity mostly adopts a manual execution mode, the average of the skilled manual calculation needs about 30-40 minutes for the user, the method belongs to a normal time range, and even if the user is trained more, the efficiency cannot be greatly improved, and the method cannot be compared with computer operation. 2. The existing computer calculates the data of the electricity consumption information acquisition system (acquisition system) in a data returning and supplementing mode, the data returning and supplementing calculation is required to be carried out, the original data of the acquisition system must be relied on, the acquisition system possibly has the phenomena of data loss, abnormality, error code and the like, and the prior art lacks a processing mode for abnormal data. 3. Calculation errors and errors the manual calculation inevitably causes errors, such as frequent rounding errors in the calculation iteration process, errors in the formula of the reference procedure, calculation errors caused by negligence, instability of the calculation equipment, and the like. Through investigation, the rework caused by manual calculation is more, and the efficiency of the rework flow is low.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: an error electric quantity compensation method and a terminal are provided, and high-precision electric quantity compensation calculation is realized.
In order to solve the technical problems, the invention adopts a technical scheme that:
an error electric quantity supplementing method comprises the following steps:
s1, acquiring target uncompensated data of a user to be uncompensated in a target uncompensated period, judging whether a missing object exists in the target uncompensated data, and if yes, executing S2;
s2, calculating a maximum likelihood attribute set of each missing object and each un-missing object in the target de-complement data;
s3, obtaining a missing value corresponding to each maximum likelihood attribute set corresponding to the missing object, and obtaining the mode of the missing value corresponding to the missing object as a filling value of the missing object;
s4, filling the filling value into the missing object to obtain complete target uncompensated data;
s5, performing the uncompensated data according to the complete target uncompensated data.
In order to solve the technical problems, the invention adopts another technical scheme that:
an error electric quantity supplementing terminal comprising a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the following steps when executing the computer program:
s1, acquiring target uncompensated data of a user to be uncompensated in a target uncompensated period, judging whether a missing object exists in the target uncompensated data, and if yes, executing S2;
s2, calculating a maximum likelihood attribute set of each missing object and each un-missing object in the target de-complement data;
s3, obtaining a missing value corresponding to each maximum likelihood attribute set corresponding to the missing object, and obtaining the mode of the missing value corresponding to the missing object as a filling value of the missing object;
s4, filling the filling value into the missing object to obtain complete target uncompensated data;
s5, performing the uncompensated data according to the complete target uncompensated data.
The invention has the beneficial effects that: the method comprises the steps of calculating a maximum likelihood attribute set of each non-missing object in missing objects and target uncompensated data, and obtaining a missing value corresponding to each maximum likelihood attribute set, wherein one missing object can obtain a plurality of missing values, and the mode of the missing values is obtained as a filling value of the missing objects, so that the space dimension of a fluctuation consistent matrix corresponding to the maximum likelihood attribute set is maximized, the characteristics of time sequence continuity of electric data such as current and load are met, compared with a global clustering algorithm without distinction, the influence of the adjacent data is enhanced, the filling precision of the missing data is improved, and the precision of an electric quantity uncompensated calculation result is further improved.
Drawings
FIG. 1 is a flow chart showing steps of an error electric quantity supplementing method according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an error electric quantity compensation terminal according to an embodiment of the present invention;
FIG. 3 is a flowchart of an algorithm for calculating a padding value according to an embodiment of the present invention;
FIG. 4 is a schematic program flow diagram of an error electric quantity compensating method according to an embodiment of the invention;
FIG. 5 is a system block diagram of an automatic generation system for error electric quantity compensation report according to an embodiment of the present invention;
description of the reference numerals:
1. an error electric quantity supplementing terminal; 2. a processor; 3. a memory.
Detailed Description
In order to describe the technical contents, the achieved objects and effects of the present invention in detail, the following description will be made with reference to the embodiments in conjunction with the accompanying drawings.
Referring to fig. 1 and fig. 3 to fig. 5, an error electric quantity supplementing method includes the steps of:
s1, acquiring target uncompensated data of a user to be uncompensated in a target uncompensated period, judging whether a missing object exists in the target uncompensated data, and if yes, executing S2;
s2, calculating a maximum likelihood attribute set of each missing object and each un-missing object in the target de-complement data;
s3, obtaining a missing value corresponding to each maximum likelihood attribute set corresponding to the missing object, and obtaining the mode of the missing value corresponding to the missing object as a filling value of the missing object;
s4, filling the filling value into the missing object to obtain complete target uncompensated data;
s5, performing the uncompensated data according to the complete target uncompensated data.
From the above description, the beneficial effects of the invention are as follows: the method comprises the steps of calculating a maximum likelihood attribute set of each non-missing object in missing objects and target uncompensated data, and obtaining a missing value corresponding to each maximum likelihood attribute set, wherein one missing object can obtain a plurality of missing values, and the mode of the missing values is obtained as a filling value of the missing objects, so that the space dimension of a fluctuation consistent matrix corresponding to the maximum likelihood attribute set is maximized, the characteristics of time sequence continuity of electric data such as current and load are met, compared with a global clustering algorithm without distinction, the influence of the adjacent data is enhanced, the filling precision of the missing data is improved, and the precision of an electric quantity uncompensated calculation result is further improved.
Further, the S2 specifically is:
calculating a maximum likelihood attribute set C (i, j) =max|k| of each missing object in the missing object and the target complement data;
wherein K is the value of K when the average square residual H ({ I, J }, K) =0, I represents one of the missing objects, I is the set of all missing objects, J is one of the non-missing objects, J is the set of all non-missing objects;
as can be seen from the above description, the maximum likelihood attribute sets of the missing object and each non-missing object are calculated, and the missing value is obtained under the maximum likelihood attribute set, so that the missing value is closer to the true value, and the parameters when the average square residual is 0 are taken to calculate the maximum likelihood attribute set, so as to reduce the prediction error to the greatest extent.
Further, the step S3 specifically includes:
acquiring a fluctuation consistent matrix S corresponding to the target data ij ，S ij For matrix a containing the set of maximum likelihood attributes MN Wherein M represents all objects in the target de-complement data, and N represents the attribute corresponding to the object;
solving a missing value corresponding to each maximum likelihood attribute set corresponding to one missing object:
wherein, miss represents a missing value, a ix An x-th attribute, a, representing an i-th object in the target decompensation data jx Representing an xth attribute of a jth object in the target decompensation data;
and obtaining the mode of the missing value corresponding to the missing object as the filling value of the missing object.
According to the above description, according to the fluctuation consistency matrix and the missing value corresponding to each maximum likelihood attribute set, the influence of the historical data and the neighbor data is integrated, especially the influence of the neighbor data is strengthened, and the characteristic that the relevant data of the electric quantity has time sequence continuity is met, so that the filling result is more accurate.
Further, the step S5 specifically includes:
receiving a grant type, and generating a grant report according to the grant type and the complete target grant data;
performing the mending according to the mending report;
the de-compensation type comprises phase A voltage loss, phase B voltage loss, phase C voltage loss, phase A current loss, phase B current loss, phase C current loss and meter black screen time Zhong Chaocha.
According to the description, multiple compensation types can be set, and targeted compensation is performed according to different compensation types, so that the compensation process and the compensation report can be automatically generated, and the manual workload is further reduced.
Further, the S2 specifically is:
and calculating the maximum likelihood attribute set of each missing object and each un-missing object in the target complement data by taking 24 hours as a period.
The above description shows that, with 24 hours as the period, the maximum likelihood attribute set corresponding to the missing object is calculated in divided periods, which accords with the electricity fluctuation period rule and improves the accuracy of the filling result.
Referring to fig. 2, an error electric quantity compensation terminal includes a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the following steps when executing the computer program:
s1, acquiring target uncompensated data of a user to be uncompensated in a target uncompensated period, judging whether a missing object exists in the target uncompensated data, and if yes, executing S2;
s2, calculating a maximum likelihood attribute set of each missing object and each un-missing object in the target de-complement data;
s3, obtaining a missing value corresponding to each maximum likelihood attribute set corresponding to the missing object, and obtaining the mode of the missing value corresponding to the missing object as a filling value of the missing object;
s4, filling the filling value into the missing object to obtain complete target uncompensated data;
s5, performing the uncompensated data according to the complete target uncompensated data.
The invention has the beneficial effects that: the method comprises the steps of calculating a maximum likelihood attribute set of each non-missing object in missing objects and target uncompensated data, and obtaining a missing value corresponding to each maximum likelihood attribute set, wherein one missing object can obtain a plurality of missing values, and the mode of the missing values is obtained as a filling value of the missing objects, so that the space dimension of a fluctuation consistent matrix corresponding to the maximum likelihood attribute set is maximized, the characteristics of time sequence continuity of electric data such as current and load are met, compared with a global clustering algorithm without distinction, the influence of the adjacent data is enhanced, the filling precision of the missing data is improved, and the precision of an electric quantity uncompensated calculation result is further improved.
Further, the S2 specifically is:
calculating a maximum likelihood attribute set C (i, j) =max|k| of each missing object in the missing object and the target complement data;
wherein K is the value of K when the average square residual H ({ I, J }, K) =0, I represents one of the missing objects, I is the set of all missing objects, J is one of the non-missing objects, J is the set of all non-missing objects;
as can be seen from the above description, the maximum likelihood attribute sets of the missing object and each non-missing object are calculated, and the missing value is obtained under the maximum likelihood attribute set, so that the missing value is closer to the true value, and the parameters when the average square residual is 0 are taken to calculate the maximum likelihood attribute set, so as to reduce the prediction error to the greatest extent.
Further, the step S3 specifically includes:
acquiring a fluctuation consistent matrix S corresponding to the target data ij ，S ij For matrix a containing the set of maximum likelihood attributes MN Wherein M represents all objects in the target de-complement data, and N represents the attribute corresponding to the object;
solving a missing value corresponding to each maximum likelihood attribute set corresponding to one missing object:
wherein, miss represents a missing value，a ix An x-th attribute, a, representing an i-th object in the target decompensation data jx Representing an xth attribute of a jth object in the target decompensation data;
and obtaining the mode of the missing value corresponding to the missing object as the filling value of the missing object.
According to the above description, according to the fluctuation consistency matrix and the missing value corresponding to each maximum likelihood attribute set, the influence of the historical data and the neighbor data is integrated, especially the influence of the neighbor data is strengthened, and the characteristic that the relevant data of the electric quantity has time sequence continuity is met, so that the filling result is more accurate.
Further, the step S5 specifically includes:
receiving a grant type, and generating a grant report according to the grant type and the complete target grant data;
performing the mending according to the mending report;
the de-compensation type comprises phase A voltage loss, phase B voltage loss, phase C voltage loss, phase A current loss, phase B current loss, phase C current loss and meter black screen time Zhong Chaocha.
According to the description, multiple compensation types can be set, and targeted compensation is performed according to different compensation types, so that the compensation process and the compensation report can be automatically generated, and the manual workload is further reduced.
Further, the S2 specifically is:
and calculating the maximum likelihood attribute set of each missing object and each un-missing object in the target complement data by taking 24 hours as a period.
The above description shows that, with 24 hours as the period, the maximum likelihood attribute set corresponding to the missing object is calculated in divided periods, which accords with the electricity fluctuation period rule and improves the accuracy of the filling result.
Referring to fig. 1, 3 and 4, a first embodiment of the present invention is as follows:
an error electric quantity supplementing method comprises the following steps:
s1, acquiring target uncompensated data of a user to be uncompensated in a target uncompensated period, judging whether a missing object exists in the target uncompensated data, and if yes, executing S2;
s2, calculating a maximum likelihood attribute set of each missing object in the missing objects and each un-missing object in the target de-complement data, wherein the maximum likelihood attribute set specifically comprises:
calculating a maximum likelihood attribute set C (i, j) =max|k| of each missing object in the missing object and the target complement data;
wherein K is the value of K when the average square residual H ({ I, J }, K) =0, I represents one of the missing objects, I is a set of all missing objects, J is an undelayed object, J is a set of all undelayed objects, and K represents an attribute set corresponding to one of the objects;
wherein ,aij Represents the elements of the ith row and the jth column after the union of the set I and the set J, a IJ Represents the average value of all elements after union of set I and set J, a Ij Represents the average value of all elements in column j, a iJ Representing the average value of all elements in line i;
specifically, let H (i, j) =0, obtain the value of K;
in an alternative embodiment, the maximum likelihood attribute set of each missing object and each un-missing object in the target retirement data is calculated in a 24 hour period;
s3, obtaining a missing value corresponding to each maximum likelihood attribute set corresponding to the missing object, and obtaining the mode of the missing value corresponding to the missing object as a filling value of the missing object, wherein the method specifically comprises the following steps:
acquiring a fluctuation consistent matrix S corresponding to the target data ij ，S ij For matrix a containing the set of maximum likelihood attributes MN Wherein M represents all objects in the target de-complement data, and N represents the attribute corresponding to the object;
solving a missing value corresponding to each maximum likelihood attribute set corresponding to one missing object:
wherein ,Si Finger fluctuation consistency matrix S ij I th row of S j Finger fluctuation consistency matrix S ij Is the j-th column of (2);
wherein, miss represents a missing value, a ix An x-th attribute, a, representing an i-th object in the target decompensation data jx Representing an xth attribute of a jth object in the target decompensation data;
acquiring the mode of the missing value corresponding to the missing object as a filling value of the missing object;
s4, filling the filling value into the missing object to obtain complete target uncompensated data;
s5, performing deconvolution according to the complete target deconvolution data, wherein the method specifically comprises the following steps:
receiving a grant type, and generating a grant report according to the grant type and the complete target grant data;
performing the mending according to the mending report;
the de-compensation type comprises phase A voltage loss, phase B voltage loss, phase C voltage loss, phase A current loss, phase B current loss, phase C current loss and meter black screen time Zhong Chaocha;
in an alternative embodiment, the Word version of the unsupplemented report is automatically generated using MATHLAB software.
Referring to fig. 3 to 5, a second embodiment of the present invention is as follows:
in the specification, the step of calculating complete data of the back-filling is called ISNC (Incomplete Set Neighbor Clustering, neighbor clustering based on the incomplete data set theory) algorithm from S1 to S4;
referring to fig. 3, an error electric quantity compensation method is different from the first embodiment in that:
s1 further comprises the following steps:
the method comprises the steps that electricity consumption of a target user is derived from an electricity consumption information acquisition system, the integrity of derived data is judged through simple information such as 'user name', 'user number', 'multiplying power' of the target user in the system, if the data is complete, step S5 is directly carried out, and if the data is incomplete, historical data of the target user is obtained;
s1 further comprises:
marking a missing object; the power load 24-point measurement data is imported, and the object is set as x i I=1, 2, n, n is the number of objects, and the attribute is set to a k K=1, 2,..m, m is the number of attributes; marking an object containing a missing value as a missing object, i.e. when one attribute of the object is missing, the object is the missing object, and marking the missing attribute of the missing object as a missing attribute, and marking a missing attribute set consisting of the missing attributes corresponding to the missing object xi as MAS i The set of missing objects consisting of all missing objects is denoted as MOS, i.e
When x is i Is a missing object;
in an alternative embodiment, the object is a date, such as 9.1 day, 9.2 day. Current and voltage data of 21:00;
s1 and S2 are specifically:
s11, taking i=1, obtaining xi, judging whether the xi belongs to MOS, if yes, executing S13, otherwise, executing S12;
s12, judging whether i is smaller than n, if so, taking i=i+1, and returning to execute S11, otherwise, stopping circulation;
s13, taking k=1, acquiring ak corresponding to xi, judging whether ak belongs to MASi, if so, executing S15, otherwise, executing S14;
s14, judging whether k is smaller than m, if yes, taking k=k+1 and returning to execute S13, otherwise, returning to execute S12;
s15, taking j=1, obtaining an undelayed object xj, obtaining a maximum similar attribute set of xi and xj, obtaining a missing value in the maximum similar attribute set, and taking the mode of the missing value as a filling value after obtaining the missing value with each xj.
The third embodiment of the invention is as follows:
the error electric quantity supplementing method is verified:
a phase current data set Y of a certain user is randomly derived from the electricity consumption information acquisition system, and is 24-day measurement data of 31 days from 1 day of 9 months in 2018 to 31 days of 9 months in 2018, and 744 records are recorded in total. In order to show the fairness and objectivity of the result, the current value of the data set is randomly lost, marked as "/", and the filling effect of the verification algorithm under different loss conditions is realized, and the conditions that the loss ratio is 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45% and 50% are respectively constructed for the data set, and the average deviation f is passed ABS Mean deviation root mean square f RMSD Two indices measure the filling accuracy. The smaller the index, the higher the description accuracy, and the closer the padded predicted value and the actual value are.
wherein ,Hi Is true value, H i ' is an estimated value, n is the number of deletions;
the partial data of the missing dataset are shown in table 1:
TABLE 1 partial missing data
Taking a data set with a missing proportion of 20% as an example, an ISNC algorithm is operated, and the predicted value and the actual value are shown in table 2:
TABLE 2 predicted and actual values for the 20% missing ratio
From the operation of the algorithm example, the predicted value of the ISNC algorithm is basically consistent with the true value;
then, at 2018, 9, 5 and 19: 00 is filled with missing data, different missing proportions are randomly constructed, and two measurement indexes of ABS and RMSD are recorded, as shown in Table 3; from an examination of the data in Table 3, it can be seen that the higher the missing proportion of the dataset, the more pronounced the ISNC algorithm is.
TABLE 3 comparison of filling Effect indicators at different deletion ratios
In Table 3, ABS is the mean deviation f ABS RMSD is the mean offset square root f RMSD ；
Through experiments, the ISNC algorithm can accurately estimate the missing data in the mining system, so as to achieve the expected target;
the effect of the verification error electric quantity report automatic generation system on time consumption and accuracy is that:
based on the average annual 157 error electricity quantity, from the acquisition of system data for inquiring electricity to the generation of final feedback report, the statistics of the time used are shown in Table 4;
TABLE 4 time contrast for each link calculated in a single pass
As can be seen from the table 4, the automatic generation system for error electricity quantity feedback report of the running ISNC is utilized to automatically carry out the feedback work, the 4742.5 minute time of enterprise staff can be saved each year, the working efficiency is greatly improved, and the method has long-term significance for enterprise benefits.
After the development of the automatic generation system for error electric quantity report is finished, the past 105 cases are subjected to test comparison, and the accuracy of the report generated by the system is 100% under the condition of 0.1% error;
by integrating the analysis, the time required for calculating the additional electricity charge, auditing and replying to the customer of one user is longer, and the system can reduce the workload of personnel, save the time cost and accelerate the refund degree.
Referring to fig. 2, a fourth embodiment of the present invention is as follows:
an error electric quantity compensation terminal 1 comprises a processor 2, a memory 3 and a computer program stored on the memory 3 and capable of running on the processor 2, wherein the processor 2 realizes the steps in the first or the second embodiment when executing the computer program.
In summary, the present invention provides an error electric quantity supplementing method and terminal, an automatic supplementing system is set, first, target supplementing data to be supplemented is obtained, if no missing object exists in the target supplementing data, the supplementing report is directly generated according to the target supplementing data and the received supplementing type to automatically carry out supplementing, if the missing object exists, a filling value corresponding to the missing object is calculated according to an ISNC algorithm through a known value, and the filling value is filled into the target supplementing data to obtain complete supplementing data to carry out automatic supplementing; the ISNC algorithm calculates the missing values of the missing objects and other objects under the maximum likelihood attribute set, classifies the same maximum likelihood attribute set, and has the characteristics of time sequence continuity of the same maximum likelihood attribute set, if a fluctuation consistent matrix is to be formed, the missing values must be equal, and in order to maximize the space dimension of the fluctuation consistent matrix, the mode of the missing values in the same maximum likelihood attribute set is required to be filled, the ISNC algorithm predicts and fills the missing values by searching the highest-dimensional fluctuation consistent matrix, focuses on the strong relevance of the missing value neighbors, strengthens the influence of neighbor data, accords with the characteristics of time sequence continuity of current, load and other data, and has a filling effect far superior to that of a global clustering algorithm without distinction, so that the ISNC algorithm has remarkable effect in the application of grid data missing.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent changes made by the specification and drawings of the present invention, or direct or indirect application in the relevant art, are included in the scope of the present invention.
Claims (6)
1. An error electric quantity supplementing method is characterized by comprising the following steps:
s1, acquiring target uncompensated data of a user to be uncompensated in a target uncompensated period, judging whether a missing object exists in the target uncompensated data, and if yes, executing S2;
s2, calculating a maximum likelihood attribute set of each missing object and each un-missing object in the target de-complement data;
s3, obtaining a missing value corresponding to each maximum likelihood attribute set corresponding to the missing object, and obtaining the mode of the missing value corresponding to the missing object as a filling value of the missing object;
s4, filling the filling value into the missing object to obtain complete target uncompensated data;
s5, performing deconvolution according to the complete target deconvolution data;
the step S2 is specifically as follows:
calculating a maximum likelihood attribute set C (i, j) =max|k| of each missing object in the missing object and the target complement data;
wherein K is the value of K when the average square residual H ({ I, J }, K) =0, I represents one of the missing objects, I is the set of all missing objects, J is one of the non-missing objects, J is the set of all non-missing objects;
wherein ,aij Represents the elements of the ith row and the jth column after the union of the set I and the set J, a IJ Represents the average value of all elements after union of set I and set J, a Ij Represents the average value of all elements in column j, a iJ Representing the average value of all elements in line i;
the step S3 is specifically as follows:
acquiring a fluctuation consistent matrix S corresponding to the target data ij ，S ij For matrix a containing the set of maximum likelihood attributes MN Wherein M represents all objects in the target de-complement data, and N represents the attribute corresponding to the object;
solving a missing value corresponding to each maximum likelihood attribute set corresponding to one missing object:
wherein, miss represents a missing value, a ix An x-th attribute, a, representing an i-th object in the target decompensation data jx Representing an xth attribute of a jth object in the target decompensation data;
acquiring the mode of the missing value corresponding to the missing object as a filling value of the missing object;
wherein ,Si Finger fluctuation consistency matrix S ij I th row of S j Finger fluctuation consistency matrix S ij Is the j-th column of (2).
2. The method for compensating for error electric quantity according to claim 1, wherein the step S5 is specifically:
receiving a grant type, and generating a grant report according to the grant type and the complete target grant data;
performing the mending according to the mending report;
the de-compensation type comprises phase A voltage loss, phase B voltage loss, phase C voltage loss, phase A current loss, phase B current loss, phase C current loss and meter black screen time Zhong Chaocha.
3. The method for compensating for error electric quantity according to claim 1, wherein the step S2 is specifically:
and calculating the maximum likelihood attribute set of each missing object and each un-missing object in the target complement data by taking 24 hours as a period.
4. An error electric quantity supplementing terminal comprising a memory, a processor and a computer program stored on the memory and capable of running on the processor, characterized in that the processor realizes the following steps when executing the computer program:
s1, acquiring target uncompensated data of a user to be uncompensated in a target uncompensated period, judging whether a missing object exists in the target uncompensated data, and if yes, executing S2;
s2, calculating a maximum likelihood attribute set of each missing object and each un-missing object in the target de-complement data;
s3, obtaining a missing value corresponding to each maximum likelihood attribute set corresponding to the missing object, and obtaining the mode of the missing value corresponding to the missing object as a filling value of the missing object;
s4, filling the filling value into the missing object to obtain complete target uncompensated data;
s5, performing deconvolution according to the complete target deconvolution data;
the step S2 is specifically as follows:
calculating a maximum likelihood attribute set C (i, j) =max|k| of each missing object in the missing object and the target complement data;
wherein K is the value of K when the average square residual H ({ I, J }, K) =0, I represents one of the missing objects, I is the set of all missing objects, J is one of the non-missing objects, J is the set of all non-missing objects;
wherein ,a ij representing a collectionIAggregation and collectionJUnion lastiLine 1jThe elements of the column are arranged such that,a IJ representing a collectionIAggregation and collectionJThe average value of all the elements after the union,a Ij represent the firstjThe average value of all the elements is listed,a iJ represent the firstiAverage value of all elements of row;
the step S3 is specifically as follows:
acquiring a fluctuation consistent matrix S corresponding to the target data ij ，S ij For matrix a containing the set of maximum likelihood attributes MN Wherein M represents all objects in the target de-complement data, and N represents the attribute corresponding to the object;
solving a missing value corresponding to each maximum likelihood attribute set corresponding to one missing object:
wherein, miss represents a missing value, a ix An x-th attribute, a, representing an i-th object in the target decompensation data jx Representing an xth attribute of a jth object in the target decompensation data;
acquiring the mode of the missing value corresponding to the missing object as a filling value of the missing object;
wherein ,Si Finger fluctuation consistency matrix S ij I th row of S j Finger fluctuation consistency matrix S ij Is the j-th column of (2).
5. The error electric quantity compensating terminal according to claim 4, wherein the S5 is specifically:
receiving a grant type, and generating a grant report according to the grant type and the complete target grant data;
performing the mending according to the mending report;
the de-compensation type comprises phase A voltage loss, phase B voltage loss, phase C voltage loss, phase A current loss, phase B current loss, phase C current loss and meter black screen time Zhong Chaocha.
6. The error electric quantity compensating terminal according to claim 4, wherein the S2 is specifically:
and calculating the maximum likelihood attribute set of each missing object and each un-missing object in the target complement data by taking 24 hours as a period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110491164.9A CN113177598B (en) | 2021-05-06 | 2021-05-06 | Error electric quantity supplementing method and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110491164.9A CN113177598B (en) | 2021-05-06 | 2021-05-06 | Error electric quantity supplementing method and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113177598A CN113177598A (en) | 2021-07-27 |
| CN113177598B true CN113177598B (en) | 2023-05-02 |
Family
ID=76928319
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110491164.9A Active CN113177598B (en) | 2021-05-06 | 2021-05-06 | Error electric quantity supplementing method and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113177598B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015184853A (en) * | 2014-03-24 | 2015-10-22 | Kddi株式会社 | Missing data complementing device, missing data complementing method, and program |
| CN111507412A (en) * | 2020-04-20 | 2020-08-07 | 南京工程学院 | Voltage missing value filling method based on historical data auxiliary scene analysis |
| CN112434942A (en) * | 2020-11-24 | 2021-03-02 | 国网河南省电力公司技能培训中心 | Intelligent early warning for preventing electricity stealing and user electricity utilization behavior analysis method |
| CN112633421A (en) * | 2021-03-09 | 2021-04-09 | 国网江西综合能源服务有限公司 | Method and device for detecting abnormal electricity utilization behavior of user |
-
2021
- 2021-05-06 CN CN202110491164.9A patent/CN113177598B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015184853A (en) * | 2014-03-24 | 2015-10-22 | Kddi株式会社 | Missing data complementing device, missing data complementing method, and program |
| CN111507412A (en) * | 2020-04-20 | 2020-08-07 | 南京工程学院 | Voltage missing value filling method based on historical data auxiliary scene analysis |
| CN112434942A (en) * | 2020-11-24 | 2021-03-02 | 国网河南省电力公司技能培训中心 | Intelligent early warning for preventing electricity stealing and user electricity utilization behavior analysis method |
| CN112633421A (en) * | 2021-03-09 | 2021-04-09 | 国网江西综合能源服务有限公司 | Method and device for detecting abnormal electricity utilization behavior of user |
Non-Patent Citations (1)
| Title |
|---|
| 邓聪 ; .退补电量理论算法的优化计算.电世界.2020,(第05期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113177598A (en) | 2021-07-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8356022B2 (en) | Approximate representation and processing of arbitrary correlation structures for correlation handling in databases | |
| CN115169818A (en) | Digital modeling-based technical engineering measurement and calculation method and system | |
| CN113406558A (en) | Electric meter misalignment detection method and device based on linear regression and electronic equipment | |
| CN111737249A (en) | Abnormal data detection method and device based on Lasso algorithm | |
| Yatsyshyn et al. | Technology of relational database management systems performance evaluation during computer systems design | |
| US20100251204A1 (en) | System and method for determining software test cycle effectiveness | |
| CN113177598B (en) | Error electric quantity supplementing method and terminal | |
| US8131653B2 (en) | Method and apparatus for warranty cost calculation | |
| Nadi et al. | Cost-effective optimization strategies and sampling plan for Weibull quantiles under type-II censoring | |
| CN110717244B (en) | Data trust analysis computer simulation method based on average deviation algorithm | |
| CN111882289B (en) | Device and method for measuring and calculating project data auditing index interval | |
| CN113609456A (en) | Reliability analysis method, system, medium and terminal for photovoltaic charging station of electric vehicle | |
| CN111191945A (en) | Method and apparatus for calculating revision completion level of standard system | |
| Makieła et al. | Stochastic frontier analysis with generalized errors: inference, model comparison and averaging | |
| Zhou et al. | Design and development of the reliability prediction software for smart meters | |
| CN112001552B (en) | Monthly electricity sales prediction method based on electricity consumption sequence | |
| CN117764638B (en) | Electricity selling data prediction method, system, equipment and storage medium for power supply enterprises | |
| CN111626586B (en) | Data quality detection method, device, computer equipment and storage medium | |
| Sun et al. | Evaluation and Measurement of Software Testing Process Quality Applicable to Software Testing Laboratory | |
| Liu et al. | Analysis on Risk of Stock Market Via Extreme Value Theory | |
| JP6965864B2 (en) | Electricity business profit and loss analysis system | |
| Raza et al. | Empirical evaluation of the ProcessPAIR tool for automated performance analysis | |
| CN113610567A (en) | Data processing method and system for determining equipment product price | |
| CN117495195A (en) | Wind power plant group capacity value assessment method based on Johnson distribution system | |
| KR20240068423A (en) | Method and system for managing underground distribution cable, and computer recording medium comprising the same |
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 |