KR20230107920A - Digital power meter capabling adjustment of aberration accuracy and the motion -monitoring - Google Patents

Abstract

The present invention relates to a watt-hour meter capable of adjusting error accuracy by monitoring an operating state. The present invention calculates the amount of power generated from the voltage sensor and the current sensor and stores the measured data, monitors the measured data and transmits the data to the outside according to a self-set cycle, and the main MCU measures. It is characterized in that it includes a sub-MCU that monitors the received data and error precision, compares the error with the watt-hour meter, and when an error occurs, notifies the upper server of a necessary error correction value and sounds an alarm.

Description

A watt-hour meter capable of adjusting the error accuracy by monitoring the operating state {Digital power meter capabling adjustment of aberration accuracy and the motion -monitoring}

The present invention relates to a watt-hour meter capable of adjusting the error accuracy by monitoring the operating state of an MCU, and more particularly, by monitoring the operation of the MCU and adjusting the error accuracy by monitoring the operating state capable of more precisely adjusting the error accuracy. It's about a power meter that can do it.

The error accuracy of a single-phase or three-phase electronic watt-hour meter has different error accuracy depending on the performance of the current/voltage sensor and electronic parts of the watt-hour meter. Currently, the error accuracy of the electronic watt-hour meter is released after testing and calibration through an error tester with a standard meter in the production process. In addition, the self-diagnosis function that can determine device failures such as battery failure, over/under voltage, voltage phase loss, etc. of the watt-hour meter depends on the design of the firmware.

The lifespan of a watt-hour meter is usually 7 to 13 years. After installation, voltage/current/phase are measured incorrectly due to external factors that can affect product performance, such as sensor failure, measurement device failure, secular change, and noise. As a result, problems with error accuracy often occur during the product use period.

If the watt-hour meter has a problem with error precision, problems such as meter failure or meter reading errors may cause consumers to under-charge or over-charge, which damages the reliability and publicity of electricity metering, lowers customer satisfaction, and Because it adversely affects the establishment of a trading environment, power companies are sensitively managing errors and precision defects to prevent them from occurring.

In order to minimize this error precision problem, electric power companies are checking the error precision by using a watt-hour meter error tester at the site through tours or special inspections by deploying dedicated personnel. This method is time-consuming and costly. There is a problem.

Korean Patent Publication No. 2008-0010171 Korean Patent Publication No. 2013-0073659 Korean Patent Publication No. 2020-0095069

Therefore, the present invention provides a main MCU formed inside a single-phase or three-phase watt-hour meter and actively performing functions such as measurement, metering, and communication, and a sub-MCU for monitoring the main MCU and calculating an error correction value for the watt-hour meter. When the sub-MCU calculates an optimal error correction value, the value is transmitted through a firmware update through an upper server, and the error accuracy can be adjusted.

In addition, the present invention uses two MCUs to monitor the MCU periodically to monitor errors and self-diagnosis, and to monitor an operating state capable of presenting an error correction value to a server, to a power meter capable of adjusting error accuracy. it's about

In order to solve this problem, the present invention is a watt-hour meter capable of adjusting the error accuracy by monitoring the operating state of measuring voltage, current, and phase from a current sensor and a voltage sensor, and the amount of power generated from the voltage sensor and the current sensor A main MCU that stores calculated and measured data, monitors the measured data and transmits the data to the outside according to a self-set cycle, and monitors the measured data and error accuracy of the main MCU to reduce the error with the watt-hour meter. It is characterized in that it includes a sub-MCU that notifies the necessary error correction value to the upper server and sounds an alarm when an error occurs by comparison.

In addition, it is characterized in that it further includes an upper server that is formed apart from the watt-hour meter and receives, collects, and stores the data transmitted from the main MCU.

In addition, the upper server is characterized in that it receives the error correction value from the sub MCU and transmits it to the upper system so that the error correction value can be changed through firmware update.

In addition, the sub-MCU is characterized in that the error correction value is automatically calculated using an algorithm embedded therein.

Therefore, the present invention monitors the main MCU inside the watt-hour meter and embeds a sub-MCU for calculating an error correction value, determines the error precision of the main MCU, and if an error occurs, utilizes the sub-MCU By calculating the error correction value and notifying the upper server of the change in the error correction value, the reliability of power measurement is increased, a fair power trading environment is established, and the overall operation cost is reduced through efficient business processing. will be.

1 is an overall configuration diagram of a watt-hour meter capable of adjusting error accuracy by monitoring an operating state according to the present invention.
2 is a flowchart illustrating a method in which a sub MCU monitors a main MCU in real time to detect an error accuracy;
3 is a flowchart illustrating a method in which a sub MCU determines an error correction value and executes a firmware update through an upper server;
4 is a configuration diagram of a system for monitoring an watt-hour meter;
Figure 5 is a block diagram showing the main configuration of the operation server.
Figure 6 is a block diagram showing the main configuration of the smart sensor of Figure 4;
7 is a flowchart of a method for monitoring an electricity meter using an artificial intelligence algorithm;
8 is an overall configuration diagram for explaining a system for diagnosing and monitoring a power meter failure;
9 is a block diagram for explaining a monitoring device;
10 is a view for explaining in detail an example of a monitoring and warning process of an arc analyzer;
11 is a process flow chart of an artificial intelligence-based electrical fire prediction and predictive maintenance method.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. First of all, in adding reference numerals to components of each drawing, it should be noted that the same components have the same numerals as much as possible even if they are displayed on different drawings.

In addition, in the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

In addition, since the terms used in this application are only used to describe specific embodiments, it is not intended to limit the present invention, and it is clear in advance that a singular expression also means a plurality of expressions unless the context clearly indicates otherwise. want to leave

In the present invention, in a watt-hour meter in which a voltage sensor and a current sensor measure voltage, current, and phase and receive the measured data, the amount of power generated by the voltage sensor and the current sensor is calculated and the measured data is stored, A main MCU that monitors the measured data and transmits the data to the outside at all times according to a self-set cycle, and monitors the measured data and error accuracy of the main MCU, compares the error with the watt-hour meter, and if an error occurs, the It consists of a sub-MCU that notifies the main MCU of the required error correction value and sounds an alarm.

In addition, an upper server that is formed apart from the watt-hour meter and receives, collects, and stores the data transmitted from the main MCU is further provided.

1 is an overall configuration diagram of a watt-hour meter capable of adjusting error accuracy by monitoring an operating state according to the present invention, and FIG. 2 is a flowchart showing a method for detecting error accuracy by a sub MCU monitoring a main MCU in real time. 3 is a flowchart showing how the sub-MCU determines the error correction value and executes the firmware update through the upper server, FIG. 4 is a configuration diagram of a system monitoring a power meter, and FIG. 6 is a block diagram showing the main configuration of the smart sensor of FIG. 4, FIG. 7 is a flow chart of a method for monitoring an watt-hour meter using an artificial intelligence algorithm, and FIG. 8 describes a system for diagnosing and monitoring a power meter failure FIG. 9 is a block diagram for explaining a monitoring device, FIG. 10 is a diagram for specifically explaining an example of an arc analyzer monitoring and warning process, and FIG. 11 is an artificial intelligence-based electrical fire prediction. and a process flow diagram of the predictive maintenance method.

Prior to explaining the present invention, the term MCU, which appears in the present invention, is an abbreviation of Micro Controller Unit, and is a microcontroller. As a device, it is an integrated circuit that plays a role in controlling the operation or process of a device or device.

In addition, the main MCU 10 and the sub MCU 20 of the present invention are both embedded inside the watt-hour meter 5, and the watt-hour meter 5 is a normal electric instrument that measures and measures the active power of the consumer. .

1, it is an overall configuration diagram of the present invention. As shown, the voltage sensor 11 and the current sensor 12 measure voltage, current, and phase, and these measurement data are embedded through the watt-hour meter 5. transmitted to the main MCU 10.

Then, the main MCU 10 calculates active, inactive, apparent power or amount of power, calculates, records and stores measurement and measurement data (active/inactive/apparent power and amount of power), and executes an operation. The contents of the operation perform and store operations such as battery (not shown) abnormality, over/low voltage, over/low current, voltage/current phase loss, abnormal temperature, magnetic field detection, and the like.

In addition, the sub MCU 20, which will be described later, receives data from the main MCU 10, collects operation information and metering data, and the sub MCU 20 collects the data at a constant monitoring cycle, and the main MCU ( 10) is monitored, and an error is compared through data received from the main MCU 10.

When the sub MCU 20 detects the difference in error, it calculates and determines the error correction value and transmits it to the upper server 30 formed at a certain distance from the power meter 5 in real time.

The method of calculating the error correction value of the sub MCU 20 can be performed through automatic calculation using an internal algorithm. Since the description of automatic calculation using the algorithm is a well-known technique, a detailed description thereof will be omitted.

In other words, the sub MCU 20 receives data from the main MCU 10, collects operation information and measurement values, and compares them. 10), collection, and comparison.

In addition, the sub MCU 20 also plays a role of notifying the upper server 30 by sounding an alarm when an error occurs while monitoring the main MCU 10 . Here, a detailed description of the configuration and operation of notifying the sub-MCU 20 to sound an alarm and causing the upper server 30 to sound an alarm will be omitted since it is a well-known technique.

In addition, the main MCU 10 always transmits data to the upper server 30 in real time according to a self-set period. For example, data values and measurement values to be stored at 15-minute intervals (adjustable) are transmitted to the upper server 30 .

In addition, the upper server 30 also serves to receive, collect, and store data transmitted from the main MCU 10 or the sub MCU 20 .

Like the main MCU 10, the sub MCU 20 measures the voltage, current, and phase through the voltage sensor 21 and the current sensor 22 and transmits the data to the sub MCU 20, and It calculates effective, reactive, apparent power or amount of power, and records and stores measurement and metering data, which is the same as the main MCU 10.

In addition, the sub MCU 20 detects an excess threshold value through data comparison with the main MCU 10 and notifies the upper server 30 when an abnormal situation occurs, such as defective detection data, etc. is to transmit In other words, it also serves to notify the upper server 30 when an abnormal situation occurs by detecting error accuracy. A description of grasping the error accuracy will be described later.

Here, the form of the upper server 30 may be a normal server or modem, and serves to receive and store data from the main MCU 10, or transmit it to a high-level upper system (not shown) or related organizations. Here, the upper system means a system connected through wireless communication, such as a superior server different from the upper server 30 or a modem.

In other words, the upper server 30 serves to receive an accurate error correction value from the sub MCU 20 and notify and transmit it to the upper system so that the error correction value can be changed through firmware update.

The firmware update refers to an online service capable of correcting an error found in a device or adding a new function to the watt-hour meter 5 through remote program upgrade.

That is, the sub MCU 20 pulls in data from the main MCU 10 and compares errors in operation and the like to check errors in real time.

Since the sub MCU 20 performs functions such as measurement, metering, self-diagnosis, etc. and records and stores data in the same way as the main MCU 10, in case of failure or malfunction of the main MCU 10 , It is to prevent the loss of metering data, thereby minimizing the blank period of measurement until repair and replacement of the watt-hour meter (5).

Hereinafter, the operational relationship of the upper server 30 will be described.

As described above, the sub MCU 20 periodically monitors (monitors) functions such as measurement, measurement, and self-diagnosis of the main MCU 10, and the measurement and measurement data collected from the main MCU 10 By comparing measurement and weighing data, it is detected that the difference set for a certain period of time exceeds the threshold value, and the value for error correction is calculated using an internal algorithm. The error correction value calculated as above is transmitted to the upper server 30 by the sub MCU 20 .

That is, to explain the operational relationship of the sub MCU 20 again, the measurement/measurement data of the main MCU 10 is monitored and an alarm is sent to the upper server 30 when an error value exceeds a set threshold value, An optimal error correction value for the current power state is transmitted to the upper server 30 .

The upper server 30 formed remotely from the watt-hour meter 5 checks the error precision value transmitted from the sub MCU 20 and determines a solution.

Then, the upper server 30 notifies the replacement of the corresponding power meter 50 or updates the firmware to the upper system (not shown) in the remote location.

Therefore, the present invention embeds the main MCU 10 and the sub MCU 20 for monitoring for error correction of the main MCU 10 inside the watt-hour meter 5, and the error of the main MCU 10 When a situation requiring correction occurs, an error correction value is calculated through an algorithm built in the sub MCU 20 .

Then, the sub MCU 20 transmits the error correction value to the upper server 30 to change the error correction value of the energy meter 5 through firmware update.

As such, the present invention has the effect of providing accurate metering services to consumers through accurate error correction, increasing the reliability of data, establishing a fair electricity trading environment, and reducing overall operating costs through efficient work processing.

Hereinafter, with reference to drawings, a description will be given of a flow in which the sub MCU 20 monitors the main MCU 10 in real time to determine the error accuracy. For reference, redundant descriptions in the above-described embodiments will be omitted to some extent.

First, the main MCU 10 measures the voltage from the voltage sensor 11 and the current from the current sensor 12 to calculate the amount of power. And, the sub MCU 20 also measures the voltage from the voltage sensor 21 and the current from the current sensor 22 to calculate the amount of power (first step).

The sub MCU 20 monitors the main MCU 10 in real time (second step).

The sub MCU 20 compares the measured power amount received through the main MCU 10 and determines whether the error range set by itself is out of a threshold value (step 3).

Here, the setting value is set in advance in the sub MCU 20, so that, for example, when an error of 0.1% to 1% occurs, an arbitrary alarm can be generated and other measures can be taken. In addition, it is possible to change the setting value as much as possible.

Furthermore, the sub MCU 20 compares voltage, current, phase, and power with the main MCU 10 in addition to the amount of power. Therefore, the third step is a step in which the sub MCU 20 detects whether or not the error accuracy is poor.

If the value such as the amount of power compared in the third step is out of the set threshold, the sub MCU 20 notifies the upper server 30 of a signal that the error precision of the main MCU 10 is poor, to transmit (step 4).

The upper server 30, notified through the fourth step, checks the error precision, determines an appropriate solution, and takes action (fifth step).

Hereinafter, a description will be given of a method in which the sub MCU 20 monitors the main MCU 10 in real time, determines an error correction value, and transmits the error correction value to the upper server 30 with reference to the drawings. Likewise, redundant descriptions in the above-described embodiments will be omitted to some extent.

First, the main MCU 10 receives current and voltage data from the voltage sensor 11 and the current sensor 12 to measure/measure the amount of power (S1).

Next, the sub MCU 20 executes a monitoring operation for the main MCU 10 in real time (S2).

The sub MCU 20 compares the error in the main MCU 10 and determines whether the error exceeds a certain range (S3).

If, in S 3 , the error exceeds the range or does not match, the sub MCU 20 calculates and determines an error correction value of the main MCU 10 by calculation through an internal algorithm, and the sub MCU 20 determines the error correction value. (20) transmits the data of the calculated error correction value to the upper server 30 (S4). Here, the sub MCU 20 and the upper server 30 have ports (not shown) on both sides, through which they communicate with each other.

The upper server 30, which has received the data through S 4 , requests an external institution and the like to transmit the watt-hour meter 5 so that it can take appropriate measures such as firmware update (S 5).

Hereinafter, a description will be given of a device for monitoring an electric power meter capable of adjusting an error accuracy by monitoring an operating state of the present invention with reference to the drawings by using artificial intelligence.

Referring to FIGS. 4 and 5, the monitoring device is a main built-in control unit to control the driving of the watt-hour meter 5 that measures voltage, current, and phase from the current sensors 12 and 22 and the voltage sensors 11 and 21. It consists of an MCU 10, a communication module 120 for transmitting monitoring information of the watt-hour meter 5, and a smart sensor 50.

Then, a repeater 60 providing monitoring information of the smart sensor 50 to the outside is formed.

In addition, referring to FIG. 5 again, the communication module 120 receives and provides monitoring information of the smart sensor 50 provided from the repeater 60 through the Internet network, and stores and manages the received monitoring information. The DB module 130, the artificial intelligence module 140 for determining the alarm situation by learning the monitoring information of the DB module 130 and the real-time monitoring information provided from the smart sensor 50, and the communication module ( 120), the operation server 100 including the control unit 110 for controlling driving of the artificial intelligence module 140 and the DB module 130 is provided.

In addition, a terminal (90: see FIG. 4) displaying the monitoring information provided from the operation server 100 to the user is formed.

The smart sensor 50 monitors various environments around the watt-hour meter 5 and collects information, and includes a plurality of sensors (not shown). The smart sensor 50 collects information by detecting ambient temperature, humidity, fine dust, etc., and the information detected by the sensor is connected to the Internet through the repeater 60 and transmitted to the real-time operation server 100 .

Referring to FIG. 6 , the main MCU 10 controls a sensor (not shown) of the watt-hour meter 5 to transmit sensed information to the outside.

The communication module 10-1 transmits detection information of the watt-hour meter 5 to the repeater 60 under the control of the main MCU 10, and may be configured with various types of wired/wireless communication means.

The display module 10-2 displays whether the watt-hour meter 5 is normally operating under the control of the main MCU 10, and may be composed of LED lamps for displaying various colors.

The alarm module 10-3 is configured to notify an alarm in the smart sensor 50 itself when an abnormal signal is detected from a specific sensor under the control of the main MCU 10, and is composed of a normal alarm means such as a buzzer or a siren. It can be.

At this time, the main MCU 10 is configured to determine whether the signal detected by the sensor is an abnormal signal. The power module 10-4 is configured to supply power for driving each component of the smart sensor 50 using external power, and may include a power connection means and a charging means.

The repeater 60 receives detection information from the smart sensor 50, transmits it to the external operation server 100, and receives a control signal from the operation server 100 or the terminal 90. To this end, the repeater 60 is configured as a means capable of internet access to the outside and short-distance communication with internal devices, and may be configured as an LTE router or a wired/wireless router, for example.

The operation server 100 uses the information transmitted from the smart sensor 50 to determine whether an alarm situation has occurred inside the energy meter, and provides the internal situation to the terminal 90 so that the user can check it in real time.

As shown in FIG. 6, the operation server 100 for this purpose includes a control unit 110, a communication module 120, a DB module 130, a learning unit 141, an analysis unit 142, and a judgment unit 143. As shown, the learning unit 141, the analysis unit 142 and the judgment unit 143 constitute the artificial intelligence module 140.

The communication module 120 is configured to communicate with the smart sensor 50 and the terminal 100 and may be configured with various communication means such as LTE and 5G. The DB module 130 cumulatively stores and manages information transmitted from the smart sensor 50 .

The information stored and managed in the DB module 130 is provided to the artificial intelligence module 140 and used as data for determining an alarm situation according to repetitive learning.

The artificial intelligence module 140 learns and statistics/analyzes the real-time information provided from the smart sensor 50 and the DB module 130 to control an alarm situation to be determined most efficiently and precisely. The artificial intelligence module 140 performs deep learning on the sensor information monitored by the smart sensor 50 and the sensor information provided by the DB module 130, and through this mechanical learning, the regional power meter 5 is installed in each region. Controls the alarm situation by setting and determining the most appropriate risk display value for the environment and characteristics.

The artificial intelligence module 140 for this may include a learning unit 141, an analysis unit 142, and a judgment unit 143. The learning unit 141 learns the information monitored in real time through deep learning, the analysis unit 142 analyzes the monitored information in conjunction with learning to set a risk display value, and the judgment unit 143 analyzes Depending on the result, it is determined whether there is an alarm situation for the real-time monitored information.

Therefore, in the operation server 100 interworking with the artificial intelligence module 140, it is possible to minimize an error in issuing an alarm due to the characteristics of the local environment or residents.

The control unit 110 controls information transmission with the smart sensor 50 and the terminal 90 through the communication module 120, and controls the artificial intelligence module 140 to learn and judge an alarm situation. In addition, the control unit 110 provides the information monitored by the smart sensor 50 to the terminal 90 in real time so that the user can check the situation even from the outside.

The terminal 90 is a terminal owned by a person having a position to monitor information in the watt-hour meter 5, for example, an external manager, and provides real-time monitoring information transmitted from the operation server 100 to the manager. To this end, a system for receiving and displaying real-time monitoring information is installed in the terminal 90 in the form of an application.

The monitoring system configured as described above is configured to simultaneously collect various information using the smart sensor 50, etc., learn and analyze the collected information cumulatively, and alarm by the environment and characteristics in which the watt-hour meter 5 is installed. Ordering errors can be minimized.

Hereinafter, with reference to FIG. 7, a description will be given of a method for monitoring a power meter using an artificial intelligence algorithm.

The operation server 100 predicts failure of the watt-hour meter 5 through the learned artificial intelligence module 140 (S1). In the above S1, the artificial intelligence module 140 is trained based on big data such as status information of each component collected from the watt-hour meters 5, diagnostic information generated accordingly, and other various information.

The operation server 100 analyzes the information collected from the main MCU 10 of the watt-hour meter 5 to generate diagnostic information indicating whether it is operating normally, whether there is a failure, or whether a failure is expected, and the generated diagnostic information may be fed back to the main MCU 10.

The operation server 100 pre-sets a normal functioning range, analyzes the status information received from the main MCU 10 of the watt-hour meter 5, and if there is a part out of the preset normal range, the corresponding part is broken or the replacement cycle It is determined that has arrived, and warning information including the determination result is generated and notified to the main MCU 10 .

The normal range for each part may be set to a predetermined value by experience or statistical methods or by experts.

That is, the standard deviation range that can be judged normal for each part is determined based on the status information collected from the electricity meters currently operating normally, and the part outside the standard deviation range determined above is compared with the status information of the diagnosis target. In this case, warning information indicating that the corresponding part is out of order or a replacement period has arrived may be generated. The standard deviation range used as a reference may be set differently for each part. In addition, the judgment unit 143 of the artificial intelligence module 140 includes data collected from the same type of watt-hour meters to generate diagnostic information for parts for which the criteria for failure or the normal usable range are not set in advance. use big data

In S2, a failure for the watt-hour meter 5 to be diagnosed is predicted using the learned artificial intelligence module 140.

The artificial intelligence module 140 will input information about the watt-hour meter 5 to be diagnosed, analyze the watt-hour meter, and output a failure prediction result for each part.

In addition, the artificial intelligence module 140 has a learning unit 141 learned based on big data including state information collected from the watt-hour meter 5 and diagnosis results based thereon, and predicts failures for the watt-hour meter to be diagnosed. information can be generated.

That is, as a method of predicting a failure or replacement point using big data collected by the analysis unit 142 of the artificial intelligence module 140 through the main MCU 10 of the watt-hour meter 5, based on the collected information When comprehensively analyzing according to the learned results, failure prediction information is generated by predicting which part will fail and when, and when it is predicted that it will fail or needs to be replaced within a predetermined period of time, corresponding diagnostic information is generated.

In S3, if there is a part in which failure is predicted in the watt-hour meter 5, S4 generates failure prediction information.

For example, if the judgment unit 143 of the artificial intelligence module 140 predicts that the probability of failure of component A is 85% before 2 years in the future for the watt-hour meter to be diagnosed, the corresponding information is included. Failure prediction information to be generated and transmitted to the main MCU 10 of the corresponding watt-hour meter 5.

In S5, the main MCU 10 transmitted in S4 transmits the generated failure prediction information to a management system (not shown) of the corresponding watt-hour meter 5 to refer to and take action.

Hereinafter, a description will be given of a system for diagnosing and monitoring a power meter capable of adjusting error accuracy by monitoring an operating state according to the present invention with reference to the drawings.

As shown in FIG. 8, the monitoring system includes a monitoring device M remotely monitoring a plurality of watt-hour meters 5-1 to 5-N sporadically distributed, a gateway 200, and a management server 300. made up of, etc.

The monitoring device M obtains failure-related information about the watt-hour meter by measuring vibration, voltage, and current failures of the strategic meters 5-1 to 5-N, and the obtained failure-related information about the electrical equipment. It also transmits device identification information.

The wired/wireless communication network ( 250) is passed through.

The management server 300 is configured with failure-related information about the watt-hour meters 5-1 to 5-N of the monitoring device M received from the gateway 200 through the wired/wireless communication network 250. Fault conditions are diagnosed based on the identification information, stored and managed in a database, and fault conditions of the watt-hour meters (5-1 to 5-N) are remotely monitored in real time.

In particular, the monitoring device M measures the failure of the watt-hour meters 5-1 to 5-N, obtains failure-related information of the watt-hour meter, and transmits unique device identification information together with the obtained failure-related information. perform a function

At this time, the unique device identification information of each watt-hour meter (5-1 to 5-N) is, for example, installation location, serial number of the device, manufacturer of the device, MAC (Media Access Control) address of the device, model and device of the device. It is preferable to include at least one of the versions of the information.

The unique device identification information may be provided from at least one of the producers and managers of the watt-hour meters 5-1 to 5-N, and may be previously stored in the storage module 156 in the monitoring device M.

The monitoring device M for monitoring the plurality of watt-hour meters 5-1 to 5-N is connected to the gateway 200 through a low power wide area network (LPWAN) (reference numerals are omitted). At this time, a low-power wide area network (LPWAN) is a low-power long-distance communication technology, and can utilize direct spread spectrum code division multiple access (CDMA) technology capable of simultaneously transmitting data by distributing data over a wide bandwidth using little power.

Such a low-power wide area network (LPWAN) can process low-capacity data with low power, secure maximum security and safety through encryption/decryption communication, and provide various services over a wide area at low cost.

One or more monitoring devices M may be disposed remotely from the place where the plurality of watt-hour meters 5 are installed.

The monitoring device M performs a function of obtaining failure-related information on the corresponding watt-hour meter by measuring a failure of at least one of vibration, voltage, and/or current for the point where the watt-hour meters 5-1 to 5-N are installed. do.

Referring to FIG. 9, the monitoring device M as described above includes a control module 150, a vibration sensing module 151, a voltage sensing module 152, a current sensing module 153, and a communication module 155. , and a power supply module 154 and the like.

As shown, the monitoring device M may further include a storage module 156 and a display module 157.

The vibration sensing module 151 performs a function of detecting vibrations caused by external shocks of various components provided in the watt-hour meter 5 and outputting a vibration sensing signal for vibration generation.

The vibration sensing module 151 senses environmental information of a field to be monitored for vibration and outputs the corresponding sensor signal, such as a shock sensor and a speed sensor, in order to detect vibration indoors and/or outdoors. , an acceleration sensor, a gyroscope sensor, a piezoelectric sensor, and/or a pressure sensor may be used.

The voltage detection module 152 performs a function of detecting voltages of various devices such as the voltage sensors 11 and 21 connected to the watt-hour meter 5.

The current sensing module 153 performs a function of sensing currents of the current sensors 12 and 22 connected to the watt-hour meter 5 .

The voltage sensing module 152 and/or the current sensing module 153 may be integrated into one or provided separately, and are directly and/or indirectly connected to power lines (PL) that are introduced into a customer and distributed to a plurality, respectively. , and performs a function of detecting the current and/or voltage flowing through each power line (PL).

That is, the voltage sensing module 152 and/or the current sensing module 153 are directly and/or indirectly connected around each power line PL, so that each power line PL is detected by an electromagnetic induction phenomenon and/or a current magnetic effect. A module that detects the current and/or voltage flowing through the

The communication module 155 is operated under the control of the control module 150, and transmits unique device identification information together with at least one failure-related information among vibration, voltage, and/or current for the watt-hour meter 5 to a low-power wide-area communication network ( It performs a function of transmitting to the management server 300 through the gateway 200 using LPWAN).

The control module 150 is a module for controlling the monitoring device M as a whole, and according to a specific control signal, the vibration sensing module 151, the voltage sensing module 152, the current sensing module 153, and the communication module ( 155), the power supply module 154, the storage module 156, and/or the display module 157, etc., perform a function of controlling operations.

In addition, the control module 150 detects at least one of vibration, voltage, and/or current for the watt-hour meter from at least one of the vibration sensing module 151, the voltage sensing module 152, and/or the current sensing module 153. Failure-related information is received, and unique device identification information is transmitted to the management server 300 through the gateway 200 along with at least one failure-related information among vibration, voltage, and/or current for the received watt-hour meter (5). It performs a function of controlling the operation of the communication module 155 to be transmitted.

In addition, the power supply module 154 includes each module, that is, vibration sensing module 151, voltage sensing module 152, current sensing module 153, communication module 155, control module 150, storage module 156 ) and/or a function of supplying necessary power to the display module 157, etc., it is preferable to be implemented with a conventional portable battery, but is not limited thereto.

The monitoring device M operates under the control of the control module 150, and identifies a unique device together with at least one failure-related information among vibration, voltage, and/or current for the watt-hour meters 5-1 and 5-N. It may further include a storage module 156 that stores information periodically and/or in real time.

That is, the storage module 156 may store and maintain at least one program code executed through the control unit 150 and at least one data set used by the program code. Such a storage module 156 is, for example, a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg SD or XD memory, etc.), RAM (RAM), SRAM (Static Random Access Memory), It may include at least one type of storage medium among ROM, EPROM, Programmable Read-Only Memory (PROM), magnetic memory, magnetic disk, and optical disk.

The monitoring device (M) operates under the control of the control module 150, and displays failure-related information and/or failure status of at least one of vibration, voltage, and/or current for the watt-hour meter 5 on the screen. A display module 157 may be further included.

The monitoring device M is preferably configured based on an AMI (Advanced Metering Infrastructure) network and operated according to control information received from the outside, such as the management server 300.

Based on such an AMI network, it is possible not only to read the power consumption from the watt-hour meter (metering), but also to provide a charge information service.

In addition, the AMI network base remotely reads the usage of various smart meters such as electricity, gas, water, hot water, and heat installed in consumers consuming energy, while automatically collecting, analyzing, and processing the information in the upper system. After that, it means a two-way communication infrastructure that provides it to the consumer again.

The gateway 200 is connected between the monitoring device M and the remote management server 300 to serve as a relay for data and/or signals, and the gateway 200 connects each power meter through a low power wide area network (LPWAN). It is connected to (5-1 to 5-N) and the monitoring device (M), and is connected to the management server 300 through a wired/wireless communication network 250.

The gateway 200 receives the unique device identification information along with the failure-related information transmitted from the monitoring device M through the LPWAN, and transmits it to the management server 300 through the wired/wireless communication network 250. do.

In addition, the gateway 200 performs a function of transmitting control information transmitted from the management server 300 to the monitoring device M through the wired/wireless communication network 250. On the other hand, it is preferable that the wired/wireless communication network 250 is made of, for example, Ethernet or a mobile communication network.

The management server 300 is connected to the gateway 200 through the wired/wireless communication network 250, and is related to the failure of the watt-hour meter of the monitoring device M received from the gateway 200 through the wired/wireless communication network 250. In addition to diagnosing fault conditions based on unique device identification information along with information and storing and managing them in a database (DB), fault conditions for watt-hour meters (5-1 to 5-N) are remotely real-time and/or periodic. to perform monitoring functions.

In addition, the management server 300 transmits at least one failure-related information of vibration, voltage, and/or current for at least one point of the watt-hour meter transmitted from each watt-hour meter 5-1 to 5-N through the gateway 200. is provided, and based on this, when at least one range of a predetermined normal value of vibration, normal voltage value, and/or normal current value is exceeded, the corresponding watt-hour meter can be diagnosed as a faulty state.

In addition, the management server 300 receives unique device identification information along with failure-related information about the watt-hour meter transmitted from the monitoring device M through the gateway 200, and uses it to determine the corresponding watt-hour meter (5-1 to 5- Search for location information of a preset watt-hour meter corresponding to N), store and manage the failure-related information and/or failure state of the watt-hour meter for each location in the database (DB), and based on this, the watt-hour meter (5) for each location It is possible to provide an energy meter management service so that the failure state of is displayed on the display screen.

Hereinafter, an electric fire prediction device for an artificial intelligence-based watt-hour meter capable of foreseeing and preparing for a fire in the watt-hour meter will be described with reference to the drawings.

As shown in FIG. 10, the arc analyzer 200 is formed inside the watt-hour meters 5, and can sense not only the temperature, but also detection data for overcurrent and leakage current, and surface temperature.

In addition, the arc analyzer 400 includes a load circuit for analyzing components of a load receiving power from the watt-hour meter 5, and measures the resistive component R, the inductive component L, and the capacitive component C of the load. Analysis to detect whether or not an effective arc has occurred, and generating sensing data using this.

These arcs can be divided into effective arcs that cause electrical defects such as fire or deterioration, and invalid arcs that are not related to electrical defects. By analyzing the resistive, inductive, and capacitive components of the load, it is possible to determine whether an effective arc is generated. there will be

The form of the arc can be detected in the form of a fine electric spark or a complex parallel arc, and the complex parallel arc is an arc that can occur before the wire is completely shorted, and the arc intensity can be classified into strong and weak.

At this time, in order to determine overcurrent and leakage current, current data can be collected in real time to analyze the allowable current versus the wire and the allowable current versus the breaker, and a built-in temperature sensor (not shown) is placed to detect the temperature data in real time, It is possible to check whether it is detected more than allowed.

In addition, the sensing data detected by the arc analyzer 400 is transmitted through a gateway (reference numeral omitted) connected to the arc management server 500 in a local network (wireless communication private network, etc.), and the arc management server 500 In the case where the sensing data is detected by the safety manager's management terminal 550 or the operating PC 600 at a threshold or higher, or predicted by learning, a warning pop-up message (text message) is sent or the location of the occurrence point information can be transmitted.

The arc management server 500 receives all the sensing data collected from the sensing device (not shown) of the watt-hour meter 5, stores it in a database (not shown), converts it into big data, and clusters and classifies the stored sensing data according to data properties. and fire occurrence can be predicted through machine learning.

In particular, in the clustering process, clustering is performed by a K-means algorithm to classify sensing data of similar properties, and the K-means algorithm is an algorithm that groups given data into k clusters. It operates in a way that minimizes the variance of the cluster and the distance difference. The similar properties may vary depending on electrical characteristics (voltage, current, ambient temperature, etc.), for example, current abnormality (overcurrent, leakage current, etc.), power abnormality (effective, ineffective, apparent power, power factor, etc.), arc abnormality (effective arc, ambient temperature, etc.).

An average can be defined and calculated for the set of sensing data using the K-means algorithm, and by recognizing the attribute pattern of the collected sensing data, data that is non-ideally farther from the centroid than other data points Clustering can be done in a way that eliminates

Sensing data classified with similar attributes is learned by using a learning model based on a machine learning algorithm for each attribute, and as a result, arcs, fires, or power abnormalities are predicted.

In the case of the machine learning algorithm, various types of machine learning algorithms such as a deep neural network (DNN), a convolutional neural network (CNN), or a recurrent neural network (RNN) may be used. For example, in the present invention, a support vector machine (SVM) technique described later may be used as a prediction technique for accurately distinguishing abnormal patterns of sensing data for each attribute.

The SVM technique is one of supervised learning machine learning methods mainly used for classification, regression, and outliers detection. For example, among several methods for classifying two groups of datasets, it can be said that accurately distinguishing the middle point at the maximum distance of each group is the best way to increase classification accuracy.

In particular, SVM is known as an optimized method for finding an optimal decision boundary capable of distinguishing a plurality of dimensions for data having a plurality of dimensions. In addition, when the SVM technique as described above is applied to the sensing data obtained from the sensors, the number of machine learning training increases as the sensing data is gradually accumulated, and as a result, the accuracy of the modeling obtained through training gradually increases. do.

In addition, in order to predict whether an arc or fire occurs or whether there is a power failure, it may be based on whether sensing data is greater than or equal to a set threshold value, and the threshold value may be determined or changed by learning of a learning model.

For example, if the collected sensing data is sensing data for overcurrent, the overcurrent state is periodically checked based on the overcurrent relative to the circuit breaker capacity to predict whether an arc or fire occurs or a power abnormality occurs according to the overcurrent state.

Similarly, based on the collected sensing data for leakage current, circuit breaker on/off state, current/voltage/power for each phase, single-phase current/voltage/power, power factor, and harmonics, power abnormality can be predicted through learning in the learning model. .

The learning model can create an anomaly detection learning model by converting it into big data through unsupervised learning that does not inform an abnormal state by digitizing it to set a threshold value using the sensing data in a normal state and the change value of the currently collected sensing data.

In addition, even if there is no initial threshold value as a standard, the learning model may collect sensing data in a normal state and generate a sample of sensing data in an abnormal state in contrast with the sensing data sample to determine the threshold value. In addition, the arc management server 500 may predict the subject of occurrence while predicting whether an arc or fire occurs or whether there is a power failure based on the sensing data.

11 is a flowchart showing the process of an artificial intelligence-based electric fire prediction method.

First, through a sensing device (not shown), sensing data capable of checking a power abnormality including at least one of an overcurrent, an effective arc, and a temperature state is collected (S10).

The collected sensing data is transmitted to the arc management server 500 (S20).

Thereafter, the arc management server 500 stores the collected sensing data to convert them into big data, and clusters and classifies the stored sensing data according to data attributes (S30).

In addition, whether a fire or an arc occurs or a power failure is predicted through machine learning (S40).

If a fire, arc occurrence or power abnormality is found in S40, the related private management company or management authority is notified by way of providing an alarm (S50).

Those skilled in the art to which the present invention pertains as described above will understand that it can be implemented in other specific forms without changing the technical spirit or essential characteristics of the present invention. Therefore, it should be understood that the embodiments described above are illustrative and not restrictive.

The scope of the present invention is indicated by the appended claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and equivalent concepts should be construed as being included in the scope of the present invention. do.

5: power meter 10: main MCU
11: voltage sensor 12: current sensor
20: sub MCU 21: voltage sensor
22: current sensor 30: upper server
40: communication module 50: smart sensor
60: repeater 90: terminal
100: operation server 110: control unit
120: communication module 130: DB module
140: artificial intelligence module 141: learning unit
142: analysis unit 143: judgment unit
M: monitoring device 200: gateway
250: wired/wireless communication network 300: management server
400: arc analyzer 500: arc management server
550: manager terminal 600: operating PC

Claims (4)

In the watt-hour meter in which the voltage sensor and the current sensor measure voltage, current, and phase and receive the measured data,
a main MCU that calculates the amount of power generated from the voltage sensor and the current sensor, stores and monitors the measured data, and transmits the data to the outside according to a self-set cycle; and
and a sub-MCU that monitors the measured data and error accuracy of the main MCU, compares the error with the watt-hour meter, and notifies an externally necessary error correction value and sounds an alarm when an error occurs. A watt-hour meter that can adjust the error precision by monitoring the operation status.
According to claim 1,
The watt-hour meter capable of adjusting error accuracy by monitoring an operating state, characterized in that it further comprises an upper server that is formed apart from the watt-hour meter and receives, collects, and stores data transmitted from the main MCU.
According to claim 1 or 2,
The upper server receives the error correction value from the sub MCU and transmits the error correction value to the outside so that the error correction value can be changed through firmware update.
According to claim 1,
The energy meter capable of calculating the error correction value, characterized in that the sub-MCU automatically calculates the error correction value using an internal algorithm.

Images