KR20160023357A - System and method for verifying metering data based on balancing clustering - Google Patents

Abstract

Disclosed are a system and a method to verify metering data based on balancing clustering, capable of forming a master node and a balancing node in a clustering mode, and minimizing the degradation of a processing speed, caused by a supply increase of a smart meter, by balancing the collection and verification of metering data in the balancing node. The disclosed system to verify metering data based on balancing clustering includes: the balancing node verifying the metering data, included in balancing work, and generating collection quantity and meter property information by estimating the metering data, determined to be abnormal; and the master node generating the balancing work based on a message, received from an enterprise service bus, allocating the generated balancing work to the balancing node, and reallocating the balancing work to another balancing node when malfunction occurs in multiple balancing nodes.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system and method for verifying data on the basis of distributed clustering,

The present invention relates to a distributed clustering-based metric data verification system and method, and more particularly, to a distributed clustering-based metric data verification system and method for verifying and aggregating metric data collected through a smart meter.

The meter reading data is the basic data to charge. In order to drive efficient energy use for low-carbon green growth in the smart grid environment, real-time electricity usage meter data must be collected.

The collected meter reading data can be used as basic data of consulting that can induce various energy saving to consumers and can be utilized as basic information for constructing economical energy source by reducing peak load to the supplier. It is important.

In the past, since the number of households in automatic remote meter reading was small, most of the surveyors visited the site for collection of monthly usage data. Therefore, there was no need to verify the reliability of the meter reading data. However, as the number of remote meter reading households increases and the time-of-use (TOU) -based Critical Peak Pricing (CPP) or Real Time Pricing (RTP) pricing policy is implemented, Since the meter reading data must be collected, the number and amount of meter reading data will explosively increase. As a result, there is a limit to verifying meter reading data by manpower. Therefore, it is required to secure technology for securing the reliability of automated meter reading data.

Meanwhile, in Korea, 170,000 high-pressure customers and 550,000 low-pressure customers are in operation, and approximately 22 million smart meters are being phased out in accordance with the mid- to long-term intelligent power meter infrastructure (AMI) As well.

As the spread of smart meters spreads, the number of metering data is expected to increase continuously, and the number of meter reading data that is divided will increase from about 137,500 as of 2014 to about 3,467,900 in 2022 by about 25 times Is expected to do.

As the spread of the smart meter increases rapidly, the load on the system that processes the verification and aggregation of the meter reading data in order to secure the reliability is increased. As a result, the processing speed of the verification system deteriorates.

Korean Patent Laid-Open No. 10-2013-0034391 (titled " meter reading data verifying apparatus and method)

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide a method and apparatus for managing a master node and a distributed node in a clustering manner and minimizing degradation in processing speed due to an increase in the spread of a smart meter, And an object of the present invention is to provide a data clustering-based metering data verification system and method.

In order to achieve the above object, a data clustering-based metering data verification system according to an embodiment of the present invention includes: a data verification unit that verifies metering data included in a distributed operation, estimates metering data determined to be abnormal, Generating distributed nodes; And a master node that generates a distributed job based on the messages received from the enterprise service bus, distributes the distributed job to the distributed nodes, and redistributes the distributed job to the other distributed nodes when a plurality of distributed nodes fail do.

The distributed node includes a power usage verification unit for verifying power usage included in the distributed work distributed from the master node; A power usage estimator for estimating an amount of power consumption determined to be abnormal by the power usage verification unit; A power usage tally unit for calculating an aggregate amount for each instrument based on the power usage amount determined to be normal by the power usage verification unit or the power usage amount estimated by the power usage estimation unit; And a gauge characteristic information generation unit for generating gauge characteristic information based on the power consumption determined to be normal in the power usage verification unit or the power usage estimated in the power usage estimation unit.

The power usage verification unit may check at least one of a missing interval check, a remote meter reading system flag check, a maximum demand check, a spike check, a continuous zero usage check, a zero power use count check, a reactive power check, a maximum / minimum power usage check, To verify power usage.

The power consumption estimating unit estimates power consumption using at least one of an inactive meter estimation method, a zero meter reading estimation method, a linear interpolation method, a history estimation method, a class load profile estimation method, and a register meter scaling estimation.

The power consumption aggregation unit calculates an aggregation amount including the sum of the power consumption amount, the sum of the number of the normal inspection and the number of the abnormal inspection.

The gauge characteristic information generation unit may include a gauge characteristic information generation unit that generates a gauge characteristic information including a maximum power usage amount, an average power usage amount, a minimum power usage amount, a maximum power usage amount increase rate, a cumulative sum of the maximum power usage update count, a cumulative sum of zero power usage amount, And generates characteristic information.

The master node includes a distributed work allocation unit for generating a distributed work based on the instrument number detected from the message received from the enterprise service bus and distributing the distributed work to the distributed nodes; A work state checking unit for monitoring an operation state of a distributed node to which a distributed work is allocated in the distributed work allocating unit; And a distributed job redistributing unit for redistributing the distributed job distributed to the distributed node determined as a failure occurrence to another distributed node by the job status checking unit.

The distributed job distribution unit generates a distributed job of a job structure consisting of a job number, a distribution node number, a job header including a gauge record length, and a gauge record including a gauge record header and a gauge data record, The instrument ID, the SDP ID, and the instrument data record length, and the instrument data record includes the power consumption classification code and the power consumption amount on the meter reading date.

The job status checking unit determines that a failure of the distributed node occurs when the distributed node is stopped due to a system failure or if the distribution job is not completed within the set time after the distribution of the distributed job.

The distributed job redistributing unit reallocates the devices assigned to the distributed nodes determined as having failed in the job status checking unit to the other distributed nodes and distributes the distributed jobs assigned to the distributed nodes determined as the failed to the reassigned other pus nodes Re-distribute.

According to another aspect of the present invention, there is provided a method for generating and distributing a distributed task using a data verification system based on a distributed clustering based on a distributed clustering method, Generating and distributing, as a method of generating and distributing, a distributed operation by a master node based on a message received from an enterprise service bus; Distributing a distributed job to the distributed node based on the distributed node number included in the generated distributed job by the master node; Monitoring, by the master node, the operational status of the distributed node to which the distributed operation is allocated; And redistributing the distributed work distributed by the master node to the distributed nodes determined as a failure in the monitoring step to the other distributed nodes.

The step of creating a distributed operation comprises: receiving, by the master node, a message containing metric data from an enterprise service bus; Checking the syntax and grammar of the received message by the master node to determine whether the message is normally received; Transmitting, by the master node, a response signal for a message determined to be normal in the step of determining, to the enterprise service bus; Detecting an instrument number from a message determined to be normal by the master node in the determining step; And generating a distributed operation corresponding to each distributed node based on the detected instrument number by the master node, wherein the distributed operation generated in the step of generating the distributed operation includes a job number, a distributed node number, And a gauge record including a gauge record header and a gauge data record, wherein the gauge record header includes a gauge ID, an SDP ID, a gauge data record length, Usage classification codes and power usage.

The step of distributing the distributed work includes the steps of: detecting, by the master node, the distributed node number included in the job header from the generated distributed work; Distributing a distributed job to the distributed node based on the detected distributed node number by the master node; And storing the job execution history information based on the distribution result of the distributed job by the master node. In the step of storing the job execution history information, the master node sets the distribution job status to distribution And stores execution history information.

In the step of monitoring the operation state of the distributed node, it is determined that a failure of the distributed node occurs when the distributed node is stopped due to a system failure or if the distributed operation is not completed within the set time after allocation of the distributed operation.

A step of reallocating a device assigned to a distributed node determined as a failure by a master node to another distributed node in a step of redistributing the distributed operation to another distributed node; Rearranging the distributed work distributed to the distributed nodes judged as a failure by the master node to the reallocated other distributed nodes; And updating the previously stored job execution history information based on the redistribution result of the redistributing step, by the master node.

According to another aspect of the present invention, there is provided a method of verifying and aggregating meter reading data using a distributed clustering based meter reading data verification system, the method comprising: A method of data verification and aggregation processing, comprising: extracting power usage from a distributed operation distributed from a master node by a distributed node; Verifying the extracted power usage by the distributed node; Estimating, by the distributed node, an amount of power usage determined to be abnormal in a verification step; Storing the estimated power consumption in a database management system in a step of estimating power consumption and power consumption determined to be normal by the distributed node in the verifying step; Calculating an aggregate amount of each of the instruments based on the power consumption estimated at the step of estimating power consumption and power consumption determined as normal by the distributed node; And generating instrument characteristic information based on the estimated power consumption in the step of estimating the power consumption and the power consumption determined as normal in the verification step by the distributed node.

In the step of verifying the power usage, the distribution node checks the gap between the missing interval test, the remote meter reading system flag test, the peak demand test, the spike test, the continuous zero usage test, the zero power usage amount test, The method of estimating the power consumption using at least one of the inspection and the total inspection, and the step of estimating the amount of power consumption by the distributed node, the inactive meter estimation method, the zero meter reading estimation method, the linear interpolation method, The profile estimation method, and the register metering scaling estimation.

In the step of calculating the aggregate weights, the aggregate including the sum of the power usage, the sum of the number of normal probing and the number of abnormal probing is calculated by the distributed node, and in the step of generating the instrument characteristic information, And generates the gauge characteristic information including the power consumption, the average power consumption, the minimum power consumption, the maximum power increase rate, the maximum power consumption update count cumulative sum, the zero power consumption count cumulative total, and the continuous zero power consumption count cumulative total.

Further comprising the step of storing, by the distributed node, the instrument characteristic information generated in the step of generating the aggregate and instrument characteristic information calculated in the step of calculating the aggregate amount, in the storing step, Only the aggregate calculated within the set period is stored.

And backing up the aggregated amount and the instrument characteristic information stored in the storing step by the distributed node to the database management system at intervals of a set period.

According to the present invention, a system and method for verifying meteorological data based on distributed clustering can be used for verification of meteorological data by constructing a master node and a distributed node in a clustering manner, thereby rapidly verifying and collecting metering data, It is possible to prevent a reduction in processing speed.

In addition, the system and method for verifying meteorological data based on distributed clustering can configure a master node and a distributed node in a clustering manner in order to verify the metering data, so that even if the spread of the smart meter continues, an inexpensive small server is connected in parallel instead of upgrading to a high- Therefore, it is possible to reduce the cost of introducing and maintaining system equipment and realizing the data processing function of real time meter reading at a relatively low cost.

In addition, the data clustering-based metering data verification system and method constitute a system for verifying metering data by connecting small servers in parallel, thereby increasing the supply of smart meters at a lower cost than upgrading to a high performance server (super computer) It is possible to prevent a reduction in processing speed.

In addition, the system and method of the meteorological data verification based on the distributed clustering can be realized by a clustering method between the master node and the distributed node for verifying meteorological data, thereby realizing a real-time charge system based on a metering data management system (MDMS) Introduction of TOU, expansion of demand management and demand response, real-time power failure management of distribution, and real-time electricity quality management are effective.

In addition, the system and method for verifying meteorological data based on distributed clustering can reduce the inspection labor cost by introducing all customer remote meter reading in electric power sales field by configuring the master node and the distributed node in a clustering manner for verifying meter reading data, (Human Error) can be prevented and the accuracy of the meter reading can be improved.

In addition, the system and method of the meteorological data verification based on the distributed clustering can configure the master node and the distributed node in a clustering manner in order to verify the meter reading data, thereby enabling the electric use customer to grasp the power consumption in real time, So that it is possible to provide an environment in which energy can be efficiently consumed.

In addition, the system and method for the verification of the meteorological data based on the distributed clustering, by configuring the master node and the distributed node in a clustering manner in order to verify the metering data, the reserve power can be secured by expanding the demand management program in the system operation field, It is possible to reduce the investment cost in all fields of power generation, transmission and distribution, and also to construct an environment capable of supplying power stably.

Brief Description of the Drawings Fig. 1 is a diagram for explaining a data clustering-based metering data verification system according to an embodiment of the present invention; Fig.
FIGS. 2 to 4 are diagrams for explaining the master node of FIG. 1; FIG.
5 to 9 are views for explaining a distribution node of Fig. 1; Fig.
10 is a flowchart for explaining a distributed work generation and allocation method in the method of verifying meter reading data based on the distributed clustering according to the embodiment of the present invention.
FIG. 11 is a flowchart for explaining a method of verifying meter reading data and an aggregation processing method in the method of verifying meter reading data based on the distributed clustering according to the embodiment of the present invention. FIG.
12 is a flowchart for explaining the power consumption estimation step of FIG.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings in order to facilitate a person skilled in the art to easily carry out the technical idea of the present invention. . In the drawings, the same reference numerals are used to designate the same or similar components throughout the drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a distributed clustering-based metering data verification system according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings. 1 is a view for explaining a data clustering-based metering data verification system according to an embodiment of the present invention. FIGS. 2 to 4 are views for explaining the master node of FIG. 1, and FIGS. 5 to 9 are views for explaining the distribution node of FIG.

As shown in FIG. 1, the distributed clustering-based metering data verification system includes a master node 100 and a distributed node 200. At this time, the enterprise service bus (ESB) 300 receives the power usage meter data message from 28 total number of remote meter reading servers operating in each of 14 regional headquarters, and transmits the data message to the master node 100 in a load balancing manner It is a message-based bus system that performs the role of The database management system 400 stores the processed result in the distributed node 200 and provides the distributed node 200 with reference information for distributed processing.

The master node 100 creates a distributed job based on the message received from the enterprise service bus 300. The master node 100 distributes the generated distributed work to the distributed node 200. [ The master node 100 distributes the distributed operation to the distributed node 200 based on the instrument number assigned to the smart meter. To this end, the master node 100 allocates and manages instrument numbers for each of the distributed nodes 200. For example, as shown in FIG. 2, when the SmartMeter is assigned to each of the first to 22,000,000 meter numbers, the master node 100 transmits the smart meters corresponding to the meter numbers 1 to 500,000 to the first dispersion Allocating the smart meters corresponding to the instrument numbers 500, 001 to 1,000,000 to the second distributed node 200 and assigning the smart meters corresponding to the first to the third distributed nodes 200, And allocates and manages the smart meters corresponding to the instrument numbers 21, 500, 001 to 22,000,000 to the Nth distributed node 200.

The master node 100 always monitors whether or not the distributed work distributed to the distributed node 200 is normally performed in order to prevent the distributed work suspended due to the system failure of the distributed node 200 from being lost. At this time, if the distribution node 200 stops performing the distribution job due to a system failure, the master node 100 redistributes the distribution job to another distribution node 200 so that the distribution job is normally completed. That is, when the system failure is detected in the distributed node 200, the master node 100 reallocates and manages the smart meters allocated to the failed distributed node 200 to the other distributed nodes 200. The master node 100 reassigns the smart meters to the distributed nodes 200 when the new distributed node 200 is added. In this way, the master node 100 distributes the smart meter equally to the distributed nodes 200. [

3, the master node 100 includes a message receiving unit 110, a distributed work distribution unit 120, a work distribution log unit 130, a history information storage unit 140, (150), a distributed work re-distribution section (160), and a work completion log section (170).

The message reception unit 110 receives a message from the enterprise service bus 300. The message receiving unit 110 checks the syntax and grammar of the received message. The message receiving unit 110 transmits a response signal (i.e., 'OK' reply) to the enterprise service bus 300 for a message determined to be normal as an inspection result. The message receiving unit 110 transmits a message determined to be normal to the distributed work allocating unit 120.

The distributed work distribution unit 120 generates a distributed work based on the message received from the message receiving unit 110. [ That is, the distributed task allocation unit 120 detects at least one instrument number from the received message. The distributed work allocating unit 120 generates a distributed work corresponding to each of the distributed nodes 200 based on the detected number of the tool. At this time, as shown in FIG. 4, the distributed work distribution unit 120 generates a distributed operation of a task structure including a job header and an instrument record.

A task header is information representing basic information about a distributed job and includes a task number, a node number, and a meter record LEN. Here, the job number is a sequential number generated sequentially, indicating the number of the distributed job. The distributed node number indicates the number of the distributed node 200 to which the task should be allocated. The gauge record length represents the number of meter records (meter reading data) included in the distribution job.

The meter record is a record generated by one meter (that is, per smart meter), and consists of a meter record header and a meter data record.

The instrument record header represents basic information about the instrument (ie, the smart meter). The instrument record header consists of a instrument ID, an SDP ID, and a meter data record length. Here, the instrument ID represents a unique identification number of the instrument (i.e., the smart meter). The SDP ID represents the unique identification number of the point where the meter is installed and powered. The instrument data record length indicates the number of instrument data records.

Instrument data recording represents metering data by instrument. Instrument data record consists of meter usage date, power usage classification code, and power consumption. Here, the meter reading date and time indicate the date and time when the power consumption from the meter is measured. The power usage classification code is a code that distinguishes what amount of power the amount of power used among the effective power amount, ground reactive power amount, true phase reactive power amount, and apparent power amount corresponds to. The power consumption represents the power consumption value read from the meter.

The distributed work distribution unit 120 distributes the generated distributed work to each of the distributed nodes 200. [ That is, the distributed work distribution unit 120 distributes the distributed work to each of the distributed nodes 200 based on the distributed node number of the job header included in the distributed work. The distributed work distribution unit 120 transmits the distribution work distribution result to the work distribution log unit 130. [

The job allocation logger 130 transmits the distribution job distribution result received from the distribution job allocation unit 120 to the history information storage unit 140 to request the storage of the job execution history information. The work distribution logger 130 transmits the redistribution result received from the distributed work redistribution unit 160 to the history information storage unit 140 to request the redistribution result storage.

The history information storage unit 140 stores the instrument numbers assigned to the distributed nodes 200 in association with each other. That is, the history information storage unit 140 stores the instrument numbers assigned to the respective distributed nodes 200 in association with each other. At this time, the history information storage unit 140 updates the instrument numbers allocated to the previously stored distributed nodes 200 according to the result of the redistribution when receiving the redistribution result from the task allocation logger 130.

The history information storage unit 140 stores job execution history information. That is, the history information storage unit 140 stores the job execution history information based on the distribution result of the distributed job received from the job allocation logger 130. At this time, the history information storage unit 140 sets the state of the distributed operation as 'distribution' and stores the state of the distributed operation when the distributed operation is initially distributed. Here, the history information storage unit 140 updates the task execution history information based on the redistribution result received from the task allocation logger 130. [

The history information storage unit 140 updates the history information based on the operation result received from the operation completion log unit 170. [ That is, if the job completion message is included in the job result, the history information storage unit 140 sets and stores the status of the distributed job as "completed ".

The work status checking unit 150 monitors the operation status of the distributed node 200. That is, the work status checking unit 150 monitors the system status of the distributed node 200 at all times to monitor whether the distributed node 200 is in a normal operation state or is stopped due to a system failure. At this time, the work status checking unit 150 determines that a system failure has occurred in the distributed node 200 if the distribution job is not completed within the set time after the allocation of the distributed job. The work status checking unit 150 may determine that a failure has occurred in the distributed node 200 by distributing a distributed job redistribution request for redistributing the distributed job distributed to the corresponding distributed node 200 to another distributed node 200, To the redistribution section 160.

Upon receiving the distributed job redistribution request from the job status checking unit 150, the distributed job redistributing unit 160 reallocates the devices assigned to the corresponding distributed node 200 to the other distributed nodes 200. [ When the reallocation of the distributed node 200 is completed, the distributed work re-distribution unit 160 redistributes the distributed work distributed to the corresponding distributed node 200 to the reallocated distributed node 200. [ The distributed work redistributing section 160 transmits the redistribution result to the work distribution log section 130. [

The job completion log section 170 receives the job result for the distributed job from the distributed node 200. [ When the job completion recorder 170 receives the job result including the job completion message of the distributed job from the distribution node 200, the job completion recorder 170 transmits the job result to the history information storage unit 140. [

The distributed node 200 performs distributed work distributed from the master node 100. [ That is, the distributed node 200 refines and verifies the metering data included in the distributed job, and then performs the estimation of the metering data in the case of abnormal metering data. The distributed node (200) stores the estimated inspection data in the database management system (400). The distribution node 200 aggregates the meter reading data by time (for each smart meter) and by day using the meter reading data for which verification is completed. The distribution node 200 generates and stores characteristic information for each meter based on the collected meter reading data, and notifies the master node 100 of the completion of the job when the distribution job is completed. At this time, since the distributed node 200 lacks the built-in memory space, the distributed node 200 periodically backs up the characteristic information of each stored instrument to the database management system 400. Accordingly, the distributed node 200 can be recovered in the event of a system failure.

5, the distribution node 200 includes a power consumption refining unit 210, a power consumption verification unit 220, a power consumption estimation unit 230, a power consumption storage unit 240, A usage characteristic aggregation unit 250, a gauge characteristic information generation unit 260, a distributed node storage unit, a distributed node transfer unit 270, and a distributed node update unit 280.

The power usage refinement unit 210 extracts the power usage amount from the distributed work distributed from the master node 100. [ The power usage refinement unit 210 transmits the extracted power usage amount to the power usage verification unit 220.

The power usage verification unit 220 verifies the amount of power usage received from the power usage refinement unit 210. The power usage verification unit 220 determines whether an abnormality occurs by checking the power consumption. At this time, the power consumption verification unit 220 checks whether there is a missing interval test to check whether the power consumption in a specific time period is missing, an automatic meter system flag test to check a flag of the remote meter reading system, a maximum demand Spike test to check if power consumption is higher than immediately before or after the test, continuous zero usage test to check whether zero value is continuously applied, zero power usage test to check how much zero power usage has occurred, And the reactive power amount inspection for checking the amount of power using the relational expression between the ground reactive power amount, the phase reactive power amount and the apparent power amount, and the maximum / minimum power for checking the power usage using the maximum value and the minimum value of the power usage, Usage check, Interval check for the same time zone The verification of the total electric power amount and the like through the check sum to check the difference between the registers (Register) meter reading. Here, the remote meter-reading system flag check is performed by checking whether or not the instrument is in the test mode for checking whether the instrument operates in the test mode, checking whether a pulse overflow occurs to check whether an overflow has occurred in the instrument pulse value, Check whether it is applied or not, check whether or not the meter time is changed, check the amount of back-up power to check whether transmission has occurred in the backward direction, and determine whether or not the power consumption value is measured and measured by the remote meter reading system. A remote error check for checking whether a remote meter reading error has occurred, a phase error check for checking whether a phase error has occurred, a CRC error check for checking whether a CRC error has occurred in a communication process, 15 minutes, 1 hour, and so on). And a check for whether or not the needle has been inspected.

The power consumption estimating unit 230 estimates the amount of power consumption determined as the abnormal power consumption in the power consumption verifying unit 220. [ At this time, the power usage estimating unit 230 estimates power usage using one of an inactive meter estimation method, a zero meter reading estimation method, a linear interpolation method, a history estimation method, a class load profile estimation method, and a register meter scaling estimation.

The power consumption estimating unit 230 determines whether the customer is normally connected to the customer and determines whether the customer is normally connected and estimates the amount of power consumption using an inactive channel estimating method when the customer is not operating normally. The inactive meter estimation method estimates the power consumption to be '0' for the inactive meter.

The power usage estimating unit 230 estimates the power consumption using the zero meter reading estimation method when the zero meter reading estimated value is set to "Yes" with respect to the target device. The zero meter reading estimation method estimates the power consumption to be '0'.

The power usage estimating unit 230 estimates power consumption using a linear interpolation method when the linear interpolation method can be estimated. The linear interpolation method can be estimated only for a short interval (for example, about 1 hour) defined by the system. If the interval required for estimation is less than 1 hour, linear interpolation is applied. The linear interpolation estimates the power consumption using Equation (1) below.

는 전력사용량 추정값이고, X₀는 직전 정상적인 전력사용량 값이고, X_n은 직후 정상적인 전력사용량 값이고, t는 직전과 직후의 시차이다.here,

X ₀ is a normal power consumption value immediately before, X _n is a normal power consumption value right after the time t, and t is a time difference between immediately before and after the power consumption amount.

 When there is a similar day in the past, the power usage estimating unit 230 applies a history estimation method. In the estimation method using the history, the power consumption value of the same time zone of the immediately preceding 5 days is selected. When selecting the immediately preceding five days, the day of the week is selected in consideration of the characteristic (see FIG. 6). A value exceeding ± 25% of the average power usage value of the selected 5 days is excluded as an abnormal value. Next, a maximum of four selected power usage values are averaged to estimate the power consumption.

If there is no similar day in the past, the power usage estimation unit 230 can not apply the estimation method based on the history, and the class load profile estimation method should be applied. The class load profile estimation method is a method of estimating the power consumption by applying an average power consumption in a user-defined period (for example, one year).

The power usage estimation unit 230 performs register metric scaling estimation when the register metric scaling estimation is set. The meter reading value is divided into an interval meter reading and a register reading meter reading. The interval meter reading means the power consumption amount divided by each interval, and the register reading means the cumulative power consumption do. Register meter reading is a meter reading in 1 hour or 1 day. The difference between the immediately preceding and following register meter reading values and the sum of the section power consumption amounts within the same time range is checked. If the sum is different, the amount of power consumption is estimated as shown in Equation (2) below by correcting the difference.

는 레지스터 검침 스케일링 추정에 의해 보정된 전력사용량 추정값이고,

는 이력에 의한 추정 또는 클래스 로드 프로필에 의한 전력사용량 추정값이고, R_n은 t시간의 직후 Register 검침값이고, R₀은 t시간의 직전 Register 검침값이고, I_i는 Interval 검침값이다.here,

Is a power usage estimation value corrected by register metering scaling estimation,

R _n is a register read value immediately after t time, R ₀ is a register read value immediately before t time, and I _i is an interval read value.

The power usage amount storage unit 240 stores the power usage amount in the power usage history information table of the database management system 400. That is, the power usage amount storage unit 240 stores the power usage amount determined by the power usage verification unit 220 as normal and the power usage amount estimated by the power usage estimation unit 230 in the power usage history information table of the database management system 400 . At this time, the structure of the power usage history information table of the database management system 400 is as shown in FIG.

The power usage amount aggregation unit 250 calculates the aggregation amount using the power usage amount of each instrument. That is, the power usage amount aggregation unit 250 calculates the sum of the power usage amounts of the respective units based on the power usage amount determined by the power usage verification unit 220 as normal and the power usage amount estimated by the power usage estimation unit 230 , The total number of normal meter readings that calculated the total of the number of power meter readings of each meter by time / day, and the number of abnormal meter readings of power / Calculate aggregate counts that include the number of meter readings.

The instrument characteristic information generation unit 260 generates instrument characteristic information used in the power consumption verification unit 220 and the power usage estimation unit 230 using the power consumption of each instrument. That is, the instrument characteristic information generation unit 260 generates instrument characteristic information based on the power consumption amount determined by the power consumption verification unit 220 as normal and the power consumption amount estimated by the power usage estimation unit 230. At this time, the gauge characteristic information generation unit 260 generates a gauge characteristic information 260 based on the maximum power consumption amount, which is the maximum power consumption amount of each gauge, the average power use amount which is the average power use amount of each gauge, the minimum power use amount which is the minimum power use amount of each gauge, The maximum power usage increase rate which is the maximum value of the increase / decrease rate, the cumulative sum of the maximum power consumption update counts, which is the cumulative sum of the number of renewals exceeding the maximum power consumption of each instrument, and the sum of the zero power consumption counts And a cumulative sum of the number of consecutive zero power consumption times, which is a sum that is the sum of consecutive zero power consumption counts.

The distributed node storage unit is composed of a memory built in the distributed node 200 and stores the gauge master information, the gauge characteristic information, and the power usage aggregation information (i.e., the power usage aggregation unit 250). At this time, the power usage aggregation amount information stores only the power usage aggregation amount information corresponding to a certain period (for example, within about 7 days) because the amount of data is large.

The distributed node transfer unit 270 periodically transfers the instrument property information and the power usage aggregation information stored in the distributed node storage unit to the database management system 400 for the purpose of recovery in case of a system failure of the distributed node 200, . At this time, the power usage aggregation information table structure of the database management system 400 is as shown in FIG. 8, and the instrument characteristic information table structure is as shown in FIG.

When the power usage aggregation information stored in the distributed node storage unit exceeds a predetermined period (for example, about 7 days) set by the administrator, the distributed node transfer unit 270 transmits the exceeded power usage aggregation information to the database Stored in the management system 400 and deleted from the distributed node storage unit. At this time, if the distributed usage information of the distributed node storage unit is updated recently but is not stored in the database management system 400, the distributed node transfer unit 270 sets a period (for example, about one minute To be stored in the database management system 400 so that the user can inquire the power consumption aggregation amount information.

When the instrument master information of the database management system 400 is added, changed, or deleted, the distributed node updating unit 280 updates the instrument master information stored in the distributed node storage unit with the added, changed, or deleted instrument master information.

The distributed node update unit 280 receives the instrument master information from the database management system 400 when another instrument is allocated due to a system failure of the distributed node 200 and updates the instrument master information stored in the distributed node storage unit.

The distributed node update unit 280 transmits the instrument characteristic information and the power usage aggregation amount information to the database management system 400 for the canceled instrument number when the allocation range of the instrument number is changed for each distributed node. The distributed node update unit 280 receives the instrument master information, the instrument characteristic information, and the power usage aggregation amount information from the database management system 400 for the newly added instrument number, and updates the distributed node storage unit.

Hereinafter, a method for generating and distributing distributed tasks in the method of verifying data based on distributed clustering according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings. 10 is a flowchart illustrating a distributed work generation and allocation method in the method of verifying data on the basis of the distributed clustering according to the embodiment of the present invention.

The enterprise service bus 300 receives power usage meter data messages from a total of 28 remote meter reading servers operated by each of the 14 regional headquarters. The enterprise service bus (300) transmits the received power usage meter data message to the master node (100) in a load balancing manner. The master node 100 generates a distributed job based on the message received from the enterprise service bus 300 (S110). The master node 100 checks the syntax and grammar of the received message and transmits a response signal (i.e., 'OK' reply) to the enterprise service bus 300 when the master node 100 determines that the received message is normal. The master node 100 detects at least one instrument number from the message determined to be normal. The master node 100 generates a distributed job corresponding to each of the distributed nodes 200 based on the detected agent number.

The master node 100 distributes the generated distributed work to the distributed node 200 (S120). In other words, the master node 100 distributes the distributed work created in operation Sl 10 to each of the distributed nodes 200. At this time, the master node 100 distributes the distributed job to each of the distributed nodes 200 based on the distributed node number of the job header included in the distributed job. The master node 100 stores the job execution history information based on the distribution result of the distributed job. At this time, the master node 100 sets the state of the distributed operation as 'distribution' and stores the state of the distributed operation when the distributed operation is initially distributed.

The master node 100 monitors the distributed job execution status of the distributed node 200 (S130). That is, the master node 100 monitors the system state of the distributed node 200 at all times and monitors whether the distributed node 200 is in a normal operation state or is stopped due to a system failure.

The master node 100 determines that a system failure has occurred in the distributed node 200 if the distribution task is not completed within the set time after the allocation of the distributed task. If a failure occurs in the distributed node 200 in operation S140, the master node 100 redistributes the distributed work distributed to the corresponding distributed node 200 to another distributed node 200 in operation S150. That is, the master node 100 reassigns the instrument assigned to the failed distributed node 200 to another distributed node 200. When the reallocation of the distributed node 200 is completed, the master node 100 redistributes the distributed work distributed to the corresponding distributed node 200 to the reallocated distributed node 200. The master node 100 updates the previously stored job execution history information based on the redistribution result.

Thereafter, when the execution of all the distributed tasks generated in step S110 is completed (S160; YES), the master node 100 ends the verification of the meter reading data. At this time, the master node 100 receives the job result for the distributed job from the distributed node 200. The master node 100 updates the stored job execution history information based on the received job result. Here, if the job completion message is included in the job result, the master node 100 updates the job execution history information by setting the status of the distributed job as "completed ".

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to the accompanying drawings. FIG. 11 is a flow chart for explaining a meter reading data verification and aggregation processing method of the meteorological data verification method based on the distributed clustering according to the embodiment of the present invention, and FIG. 12 is a flowchart for explaining the power use amount estimation step of FIG.

The distributed node 200 extracts the power usage amount from the distributed work distributed from the master node 100 (S205). That is, the distributed node 200 extracts the power consumption amount from the instrument data record included in the instrument record of the power consumption.

The distributed node 200 verifies the detected power usage (S210). In other words, the distribution node 200 determines whether or not it is abnormal by checking the detected power usage. At this time, the distribution node 200 checks whether there is a missing interval test to check whether a power usage amount in a specific time period is missing, a flag test of an automatic meter reading system for checking a flag of the remote meter reading system, a maximum demand test to check whether a contracted maximum power is exceeded, Spike test to check if power consumption is higher than before and after, Continuous zero usage test to check whether zero value is continuously applied, Zero power usage test to check how much zero power usage has occurred, A reactive power amount inspection for checking the amount of power using the relational expression between the reactive power amount and the phase reactive power amount and the apparent power amount, a maximum / minimum power usage inspection for checking the power consumption using the maximum value and the minimum value of the power usage measured in the past , The total of the interval meter reading for the same time zone It verifies the power consumption over the total inspection for checking the difference of the register (Register) meter reading. Here, the remote meter-reading system flag check is performed by checking whether or not the instrument is in the test mode for checking whether the instrument operates in the test mode, checking whether a pulse overflow occurs to check whether an overflow has occurred in the instrument pulse value, Check whether it is applied or not, check whether or not the meter time is changed, check the amount of back-up power to check whether transmission has occurred in the backward direction, and determine whether or not the power consumption value is measured and measured by the remote meter reading system. A remote error check for checking whether a remote meter reading error has occurred, a phase error check for checking whether a phase error has occurred, a CRC error check for checking whether a CRC error has occurred in a communication process, 15 minutes, 1 hour, and so on). And a check for whether or not the needle has been inspected.

If it is determined that the abnormal power consumption is the result of the verification (S215; Yes), the distribution node 200 estimates the power consumption determined to be abnormal (S220). That is, the distributed node 200 estimates power consumption using one of an inactive meter estimation method, a zero meter reading estimation method, a linear interpolation method, a history estimation method, a class load profile estimation method, and a register meter scaling estimation. This will be described with reference to FIG. 12 attached hereto.

The distribution node 200 determines whether the user is normally connected to the user. If it is not a normal operation instrument (S221: NO), the dispersion node 200 estimates the power consumption through the inactivity estimation method for the power consumption (S222). That is, the distributed node 200 estimates the power usage to be '0' through the non-active instrument estimation method for an instrument not operating normally.

The distribution node 200 determines whether or not the zero meter reading is estimated based on the setting value of the zero meter reading estimate. At this time, the distribution node 200 estimates the power usage through the zero meter reading estimation method when it is determined that the zero meter reading estimation is set to 'Yes' (S223; Yes) ). That is, the distributed node 200 estimates the power consumption to be '0' through the zero meter reading estimation method.

If it is possible to estimate the linear interpolation method for the abnormal power consumption (S225; YES), the dispersion node 200 estimates the power consumption using the linear interpolation method (S226). That is, the distributed node 200 estimates the power consumption by using the linear interpolation method when the section requiring estimation of the power consumption is within one hour. At this time, the distribution node 200 estimates the power consumption through the linear interpolation using the normal power consumption immediately before, the normal power consumption immediately after, and the parallaxes between immediately before and after the normal power consumption.

If a similar date to the measured power usage amount is present in the past (S227; YES), the distribution node 200 estimates the power consumption using the power consumption estimation method by history (S228). That is, the distributed node 200 selects the power usage value in the same time zone of the last 5 days in consideration of the day-to-day characteristic. The distributed node 200 excludes a value exceeding ± 25% from the average of the power usage values of the selected five days as an abnormal value. The distributed node 200 averages the selected maximum four power usage values to estimate the power usage.

If there is no similar day in the past, the distribution node 200 estimates the power usage using the class load profile estimation method (S229). That is, the distributed node 200 estimates the average value of the power consumption in the period (for example, one year) determined by the user as the power consumption through the class load profile estimation.

If the register metric scaling estimation is set (S230; YES), the distribution node 200 estimates the power consumption through the register metric scaling estimation method (S231). That is, if the register metric scaling estimation is set, the distribution node 200 performs register metric scaling estimation and corrects the power usage estimated in steps S228 and S229. At this time, the distribution node 200 corrects the power usage estimated by the history estimation or the class load profile estimation, the power usage immediately after the register measurement, the previous register measurement value, and the interval measurement value.

The distribution node 200 stores the power consumption amount in the power consumption history information table of the database management system 400 (S240). That is, the distributed node 200 stores the power usage amount determined to be normal in step S210 or the power usage amount estimated in step S220 in the power usage history information table of the database management system 400. [

The distributed node 200 calculates the aggregate amount based on the power consumption (S245). That is, the distributed node 200 calculates the aggregate amount using the power consumption of each instrument. At this time, the distribution node 200 calculates a sum of power consumption calculated by summing the power consumption of each instrument by time and day, based on the power consumption determined as normal in step S210 and the power consumption estimated in step S220, / Calculate the aggregate amount including the sum of the number of normal meter readings that calculate the sum of the number of normal power meter readings per day and the total number of abnormal meter readings per hour / day of each meter.

The distribution node 200 generates instrument characteristic information based on the power consumption (S250). That is, the distributed node 200 generates the instrument characteristic information used in the verification and estimation of the power consumption using the power consumption of each instrument. At this time, the dispersed node 200 generates the instrument characteristic information based on the normal power consumption amount and the power consumption amount estimated in S220 in operation S210. At this time, the distribution node 200 calculates the maximum power consumption amount, which is the maximum power consumption amount of each instrument, the average power consumption amount which is the average power consumption amount of each instrument, the minimum power consumption amount which is the minimum power consumption amount of each instrument, A total cumulative sum of the maximum power consumption renewal counts and the number of zero power consumption counts, which is a cumulative sum of the number of renewals exceeding the maximum power consumption of each instrument, Which is the sum of the number of times of use of zero power and the sum of the number of times of continuous use of power that is zero.

The distributed node 200 stores the aggregation amount and the instrument characteristic information (S255). That is, the distributed node 200 stores the aggregate information calculated in step S245 and the instrument characteristic information generated in step S250 in the embedded memory. At this time, the distributed node 200 stores the gauge master information, the gauge characteristic information, and the power usage aggregation amount information for the gauge allocated to each of the distributed nodes 200. Here, since the distributed node 200 has a limitation in the built-in memory, only the power usage aggregation information corresponding to a certain period (for example, within about 7 days) is stored.

The distributed node 200 backs up the aggregated amount and the instrument characteristic information to the database management system 400 at intervals of a set period (S260). That is, the distributed node 200 periodically escalates the instrument characteristic information and the power consumption aggregation information stored in the internal memory to the database management system 400 for the purpose of recovery in case of a system failure. When the power usage aggregation information stored in the internal memory exceeds a predetermined period (for example, about 7 days) set by the administrator, the distributed node 200 transmits the exceeded power usage aggregation amount information to the database management system 400 ) And then delete it. At this time, if the distribution node 200 has been recently updated among the power consumption aggregation information of the internal memory but is not stored in the database management system 400, And stores it in the management system 400 so that the user can inquire the power consumption aggregation amount information.

On the other hand, when the addition, change, or deletion of the instrument master information occurs (S265; YES), the distribution node 200 updates the instrument master information (S270). That is, when the instrument master information of the database management system 400 is added, changed, or deleted, the distributed node 200 updates the instrument master information stored in the internal memory with the added, changed, or deleted instrument master information. The distribution node 200 receives the instrument master information from the database management system 400 when another instrument is assigned, and updates the instrument master information stored in the internal memory. The distribution node 200 transmits the instrument characteristic information and the power consumption aggregation information to the database management system 400 for the canceled instrument number when the allocation range of the instrument number is changed for each of the distributed nodes 200. [ The distributed node 200 receives and updates the instrument master information, instrument characteristic information, and power usage aggregation amount information from the database management system 400 with respect to the newly added instrument number.

As described above, the distributed clustering-based metering data verification system and method are configured to cluster the master node and the distributed node in order to verify the metering data, so that the verification and aggregation of the metering data can be processed quickly, It is possible to prevent a reduction in processing speed.

In addition, the system and method for verifying meteorological data based on distributed clustering can configure a master node and a distributed node in a clustering manner in order to verify the metering data, so that even if the spread of the smart meter continues, an inexpensive small server is connected in parallel instead of upgrading to a high- Therefore, it is possible to reduce the cost of introducing and maintaining system equipment and realizing the data processing function of real time meter reading at a relatively low cost.

In addition, the data clustering-based metering data verification system and method constitute a system for verifying metering data by connecting small servers in parallel, thereby increasing the supply of smart meters at a lower cost than upgrading to a high performance server (super computer) It is possible to prevent a reduction in processing speed.

In addition, the system and method of the meteorological data verification based on the distributed clustering can be realized by a clustering method between the master node and the distributed node for verifying meteorological data, thereby realizing a real-time charge system based on a metering data management system (MDMS) Introduction of TOU, expansion of demand management and demand response, real-time power failure management of distribution, and real-time electricity quality management are effective.

In addition, the system and method for verifying meteorological data based on distributed clustering can reduce the inspection labor cost by introducing all customer remote meter reading in electric power sales field by configuring the master node and the distributed node in a clustering manner for verifying meter reading data, (Human Error) can be prevented and the accuracy of the meter reading can be improved.

In addition, the system and method of the meteorological data verification based on the distributed clustering can configure the master node and the distributed node in a clustering manner in order to verify the meter reading data, thereby enabling the electric use customer to grasp the power consumption in real time, So that it is possible to provide an environment in which energy can be efficiently consumed.

In addition, the system and method for the verification of the meteorological data based on the distributed clustering, by configuring the master node and the distributed node in a clustering manner in order to verify the metering data, the reserve power can be secured by expanding the demand management program in the system operation field, It is possible to reduce the investment cost in all fields of power generation, transmission and distribution, and also to construct an environment capable of supplying power stably.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but many variations and modifications may be made without departing from the scope of the present invention. It will be understood that the invention may be practiced.

100: Master node 110: Message receiver
120: distributed work distribution unit 130: work distribution log
140: history information storage unit 150: work status checking unit
160: distributed work re-distribution section 170: work completion log section
200: Distributed node 210: Power consumption refining unit
220: power consumption verification unit 230: power consumption estimation unit
240: Power consumption storage unit 250: Power consumption aggregation unit
260: Instrument characteristic information generation unit 270: Distributed node transfer unit
280: Distributed node update unit 300: Enterprise service bus
400: Database Management System

Claims (20)

A distribution node that verifies the meter reading data included in the distributed operation and generates aggregate and meter characteristic information by estimating the meter reading data determined to be abnormal; And
A master node that generates a distributed job based on a message received from the enterprise service bus, distributes the distributed job to the distributed node, and distributes the distributed job to another distributed node when the distributed node fails And a data clustering-based metering data verification system. The method according to claim 1,
The distributed node comprises:
A power usage verifying unit for verifying a power usage amount included in the distributed work distributed from the master node;
A power usage estimator for estimating an amount of power consumption determined to be abnormal by the power consumption verification unit;
A power usage tally unit for calculating an aggregate amount for each instrument based on the power usage amount determined to be normal by the power usage verification unit or the power usage amount estimated by the power usage estimation unit; And
And a gauge characteristic information generator for generating gauge characteristic information based on the power consumption determined to be normal by the power usage verification unit or the power usage estimated by the power usage estimation unit. . The method of claim 2,
Wherein the power usage verification unit comprises:
Power consumption by using at least one of missing interval test, remote meter reading system flag test, peak demand test, spike test, continuous zero usage test, zero power usage test, reactive power test, maximum / minimum power usage test, Based on the result of the verification. The method of claim 2,
The power consumption estimating unit may calculate,
Wherein the power usage is estimated using at least one of an inactive meter estimation method, a zero meter reading estimation method, a linear interpolation method, a history estimation method, a class load profile estimation method, and a register meter scaling estimation method. system. The method of claim 2,
The power consumption aggregation unit,
A total amount of electric power used, a total number of normal inspection counts, and a total number of abnormal inspection counts are calculated. The method of claim 2,
Wherein the instrument characteristic information generation unit comprises:
Generating gauge characteristic information including a maximum power usage amount, an average power usage amount, a minimum power usage amount, a maximum power increase rate, a cumulative sum of maximum power usage update count, a cumulative sum of zero power usage counts, A data clustering - based metric data verification system. The method according to claim 1,
The master node,
A distributed work allocation unit for generating a distributed work based on the detected number from the message received from the enterprise service bus and distributing the distributed work to the distributed nodes;
A work status confirmation unit for monitoring an operation status of a distributed node to which a distributed job is distributed in the distributed job distribution unit; And
And a distributed job re-distribution unit for re-distributing the distributed work distributed to the distributed nodes determined as a failure occurrence to the other distributed nodes by the job status checking unit. The method of claim 7,
The distributed work allocating unit,
Wherein the gauge record header includes a gauge ID, a gauge record header, and a gauge record including gauge data records, wherein the gauge record header includes a gauge ID, an SDP ID and a meter data record length, and the meter data record includes a power use classification code and a power consumption amount on a meter reading date. The method of claim 7,
The work status verifying unit,
Wherein the distributed clustering-based metering data verification system determines that a failure of the distributed node occurs if the distributed node is stopped due to a system failure or if the distributed operation is not completed within the set time after the distributed work is distributed. The method of claim 7,
The distributed work re-
The task state check unit reassigns the devices assigned to the distributed nodes determined as having failed to another distributed node and redistributes the distributed tasks assigned to the distributed node determined as the occurrence of the failure to the reassigned other bus nodes A data clustering - based metric data verification system. A distributed job creation and allocation method using a distributed clustering-based metering data verification system including a master node and a distributed node,
Generating, by the master node, a distributed operation based on a message received from the enterprise service bus;
Distributing the distributed job to the distributed node based on the distributed node number included in the generated distributed job by the master node;
Monitoring, by the master node, an operation state of the distributed node to which the distributed operation is allocated; And
And distributing, by the master node, the distributed work distributed to the distributed nodes determined as a failure occurrence in the monitoring step to the other distributed nodes. The method of claim 11,
Wherein the generating the distributed job comprises:
Receiving, by the master node, a message including metric data from the enterprise service bus;
Checking the syntax and grammar of the received message by the master node to determine whether the message is normally received;
Transmitting, by the master node, a response signal for a message determined to be normal in the determining step to the enterprise service bus;
Detecting, by the master node, an instrument number from a message determined to be normal in the determining step; And
And generating, by the master node, a distributed job corresponding to each distributed node based on the detected agent number,
Wherein the distributed job generated in the generating of the distributed job comprises a job record including a job number, a distribution node number, a job header including a gauge record length, and a gauge record header and gauge data record, And the meter data record includes a meter ID, an SDP ID, and a meter data record length, and the meter data record includes a power consumption classification code and a power consumption amount on a meter reading date. The method of claim 11,
Wherein distributing the distributed operation comprises:
Detecting, by the master node, the distributed node number included in the job header from the generated distributed job;
Distributing the distributed job to the distributed node based on the detected distributed node number by the master node;
And storing the job execution history information based on the distribution result of the distributed job by the master node,
In the step of storing the job execution history information,
And the job execution history information in which the status of the distributed job is set as a distribution is stored by the master node. The method of claim 11,
In the step of monitoring the operation status of the distributed node,
Wherein the master node determines that a failure has occurred in the distributed node if the distributed node is stopped due to a system failure or if the distributed operation is not completed within the set time after the allocation of the distributed operation. The method of claim 11,
In the step of redistributing the distributed job to another distributed node,
Reallocating, by the master node, an instrument assigned to a distributed node determined to have failed, to another distributed node;
Rearranging, by the master node, the distributed work distributed to the distributed node judged as the occurrence of the fault to the reassigned other distributed node; And
And updating the previously stored job execution history information based on the redistribution result of the redistributing step by the master node. A method of verifying and collecting metric data using a distributed clustering based metric data verification system including a master node and a distributed node,
Extracting a power usage amount from the distributed work distributed from the master node by the distributed node;
Verifying the extracted power usage by the distributed node;
Estimating, by the distributed node, an amount of power consumption determined to be abnormal in the verifying step;
Storing, in the database management system, a power usage amount determined by the distribution node as normal and a power usage amount estimated in the step of estimating the power usage amount;
Calculating an aggregate amount of each instrument based on the power usage amount determined by the distribution node as normal and the power usage estimated in the step of estimating the power usage amount; And
And generating instrument characteristic information on the basis of the power consumption amount determined by the distribution node as normal and the power consumption amount estimated in the step of estimating the power consumption amount by the distribution node, Aggregate processing method. 18. The method of claim 16,
In the step of verifying the power consumption,
At least one of a missing interval check, a remote meter reading system flag check, a maximum demand check, a spike check, a continuous zero usage check, a zero power use amount check, a reactive power amount check, a maximum / minimum power usage check, To verify power usage,
In the step of estimating the power consumption,
Wherein the power consumption is estimated by the distributed node using at least one of an inactive meter estimation method, a zero meter reading estimation method, a linear interpolation method, a hysteresis estimation method, a class load profile estimation method, and a register metering scaling estimation, Meter reading data verification and aggregation processing method. 18. The method of claim 16,
In the step of calculating the aggregate amount,
Calculating an aggregate amount including the sum of the power usage amount, the sum of the number of normal inspection probes and the number of abnormal inspection probes by the distributed node,
In the step of generating the instrument characteristic information,
And a total sum of cumulative zeros power consumption counts and a cumulative total of zero power consumption counts, and a cumulative total of cumulative zeros power consumption counts, by the distributed node, based on a maximum power usage amount, an average power usage amount, a minimum power usage amount, a maximum power usage amount increase / And the characteristic information is generated. 18. The method of claim 16,
Further comprising the step of storing, by the distributed node, the instrument characteristic information generated in the step of generating the aggregate amount and the instrument characteristic information calculated in the step of calculating the aggregate amount,
In the storing step,
And only the aggregated amount calculated by the distributed node within the set period is stored in the aggregated data. 18. The method of claim 16,
Further comprising the step of backing up, by the distribution node, the aggregation amount and the instrument characteristic information stored in the storing step to the database management system at a set periodic interval.

Images