KR102412731B1 - Eco-friendly GIS that can detect dry air leakage using high sensitive vibration sensor and the method using it - Google Patents

Abstract

The present invention has been made to solve the above problems, and after receiving the first vibration value from the sensor using a plurality of vibration sensors, data is processed according to the result of detecting another signal after a predetermined time to simply convert the vibration signal to a frequency. The purpose of the analysis is to provide a GIS detection system using a vibration sensor that shows better detection performance than a database.

Description

Eco-friendly GIS that can detect dry air leakage using high sensitive vibration sensor and the method using it

The present invention relates to a GIS capable of detecting dry air leakage of an eco-friendly GIS (GAS INSULATED SWITCHGEAR) using a vibration sensor. A plurality of vibration sensors, a signal detection unit for detecting a signal from the sensor, and the presence or absence of leakage using the detected value are divided by the vibration signal size and established according to a certain standard, and the expected dangerous time is calculated and an alarm is given to the user It relates to an eco-friendly GIS capable of monitoring leakage using a vibration sensor composed of a leakage monitoring unit that can be presented in detail through

In general, GIS is an important device required to supply power generated in a power plant to a load, and it requires high reliability and is a major device that requires continuous maintenance. However, since the economic loss that occurs in case of failure is very large, continuous inspection and preventive measures are required not only in the manufacturing stage but also during installation and operation. Currently, GIS is operated with a similar operating variable monitoring system, but there are cases in which power supply is stopped due to GIS failure. As a result of analyzing the causes, it is estimated that failures due to electrical causes, mechanical causes, and leakage problems account for most.

GIS is in the process of being replaced with dry air, an eco-friendly insulating material, as the GIS insulation gas cannot be used as an environmental pollutant, SF6 gas, according to 2030 carbon neutrality. At this time, dry air, whose insulation performance is lower than that of SF6 gas, is charged with a charging pressure of 1.0~1.5 bar more than that of the existing GIS in order to secure insulation ability, so the problem of dry air leakage in eco-friendly GIS is expected to arise.

In particular, in the case of gas leakage in the existing GIS, it is necessary to detect it at an early stage because it is directly connected to an accident. The leak location cannot be estimated using only the pressure sensor, and sensitive monitoring is insufficient.

Therefore, in the case of GIS insulated with SF6 gas, a portable IR camera equipped with an FT-IR filter that shows the natural absorption wavelength that occurs during gas leakage is searching for its location.

However, since dry air, an eco-friendly gas, has the same components as existing air when it leaks, the location cannot be found with the existing filter-attached camera. Accordingly, it takes a lot of time to find and maintain the leak location of the eco-friendly GIS, which is also a problem for safety.

A common primary leak detection method is to hear a hissing sound, but it is virtually impossible to hear in many environments, and spraying soapy water over a suspected leak takes a lot of time and is messy and slippery to install and analyze on top spacers, etc.

Currently, the main method for finding compressed gas leaks is with ultrasonic acoustic detectors. A general ultrasonic detector is a method of finding a signal using a time difference between two or more sensors.

It is difficult to locate the vibration using the time difference due to the low frequency, and in the case of a rotating machine, the location of the defect is estimated through frequency analysis.

In addition to the time it takes to find leaks using soapy water or ultrasonic detectors, safety issues can arise while using these techniques to find leaks under or above equipment. Climbing ladders or crawling around equipment can be dangerous.

In the past, technologies for remotely monitoring the operating status in relation to various power facilities as well as GIS have been developed, but the main technology of the present invention, which can manage vibration during operation, is missing, or it is said that the optimal operation of power facilities, especially GIS, is realized. Although similar in terms of aspect, mechanical soundness could not be managed because the subject was limited to a small-capacity GIS or the operating variables to be monitored were limited to electrical signals such as temperature, voltage and current (Korean Patent Registration Nos. 0337184 and 0817611).

In other words, there is a disadvantage in that it is not possible to determine in advance the signs of failure caused by mechanical problems as only the electrical characteristics, which are the basic functions of the power equipment (GIS), are monitored/managed. In addition, in the case of some technologies, monitoring and diagnosis can be implemented remotely, but a specific technology for evaluating the mechanical soundness of GIS has not been disclosed (US Patent Publication No. 20070225945, PCT Publication No. WO06/135994)

In particular, frequency analysis of continuously measured vibration signals through the vibration monitoring system that can monitor the vibrations of the enclosure and major devices at all times during GIS operation is converted into a database, and the analyzed vibration signals are graded based on self-developed standards. In this way, the diagnostic system (Korea Patent No. 1081982) that can manage the vibration of GIS was difficult to use in practice because it was difficult to analyze the x, y, and z-axis sensing information according to the P wave and the S wave in detail.

If you look at the operation of the conventional GIS (SF6) pressure sensor, an alarm is generated when operating 80% of the reference pressure (1 out of 4 spacers).

On the other hand, if you look at the operation of the pressure sensor of the eco-friendly GIS, an alarm occurs when 80% of the reference pressure is operating (1EA among 4EA spacer), the dry air insulation linearity is insufficient, and when an alarm occurs, it is impossible to accurately estimate the position by checking soap bubbles. There was a problem.

Korean Patent Registration No. 0337184 Korean Patent Registration No. 0817611 US Patent Publication No. 2007-0225945 PCT Patent Publication WO06/135994 Korean Patent No. 1081982

The present invention has been made to solve the above problems, determine the leaking point of the eco-friendly GIS, determine the frequency band suitable for the vibration characteristics of the GIS enclosure according to the eco-friendly gas leak, obtain a signal, and a plurality of vibration sensors saves the signal magnitude pattern according to the leak point in the database using It aims to operate eco-friendly GIS more safely and facilitate maintenance.

)을 대푯값으로 선정하여 측정값 > 기준값 이상이고 측정시간 > 기준시간 이상이 측정되어야 주변 센서 값을 동시에 저장하여 크기패턴을 비교하고, 상기 측정값 > 기준값과 측정시간 > 기준시간이 아니면 다시 진동센서를 계측하며, 상기 측정패턴이 일정 누설지점과 동일 또는 유사(측정패턴 = SP#1) 이면 상기 일정 누설지점과 가장 인접한 스페이서의 누설을 판정한다.In order to solve the above problems, the present invention provides a sensor unit for measuring a leakage signal by vibration for each of a plurality of points; a signal detection unit for receiving a signal from the sensor unit in a constant time or multiplexer manner; In the signal received from the signal detection unit, a specific frequency is separated into a frequency specific unit, only the specific frequency part is analyzed, a certain leak point is designated, and the vibration signal measured when leaking at each leak point is removed from noise and stored in the database. Leak monitoring consisting of a leak monitoring algorithm that determines whether there is a leak, and a leak location estimation algorithm for determining the leak location pointed to by a vibration signal having the same or similar pattern by comparing and analyzing the patterns of the measured vibration signals Including;, wherein the leak location estimation algorithm integrates the maximum value (Vp-p) in the feature frequency as well as the values in the feature frequency in order to remove noise and maximize the difference of the measured value of the vibration sensor. value(

) as a representative value, the measured value > the reference value and the measurement time > the reference time or more must be measured. If the measurement pattern is the same as or similar to the predetermined leakage point (measurement pattern = SP#1), the leakage of the spacer closest to the predetermined leakage point is determined.

The leak monitoring unit automatically resets the vibration value if there is no detection from another detection sensor unit after detecting a signal from the sensor unit to increase data reliability.

The leak monitoring unit converts the vibration signal magnitude and time difference into a database.

The leak monitoring unit automatically suggests the location of the leak as a result of data analysis of the pattern, and presents the expected risk point.

The present invention comprises the steps of measuring a vibration sensor of the sensor unit; If the measured value of the vibration sensor > reference value, storing the sensor time discrimination measurement pattern after storing the surrounding sensor value; Measuring the vibration sensor again if the measured value > reference value; determining spacer #1 leakage if the measurement pattern is the same as SP#1 (measurement pattern = SP#1); determining spacer #2 leakage if the measurement pattern is the same as SP#2 (measurement pattern = SP#2); and determining the spacer #3 leak if the measurement pattern is the same as SP#3 (measurement pattern = SP#3).

Measuring the vibration sensor may include: turning on the power of the sensor unit; an initialization operation step of the sensor unit; activating the sensor unit; measuring a coordinate value using the MEMS of the sensor unit; calculating a vector value of the sensor unit; and transmitting the vibration value to the leak monitoring unit by the sensor unit via Modbus communication.

automatically presenting a leak location as a result of the leak monitoring unit analyzing the measurement pattern into data; and dividing the amount of leakage measured by the sensor unit and calculated by the leakage monitoring unit by the amplitude of the vibration signal to establish it according to a predetermined standard.

and presenting the leak monitoring unit through an alarm to the user by calculating the expected dangerous time point.

The technology disclosed herein may have the following effects. However, this does not mean that a specific embodiment should include all of the following effects or only the following effects, so the scope of the disclosed technology should not be construed as being limited thereby.

According to the present invention, which is made as described above, data processing can be determined according to the sensed result after a predetermined period of time after receiving the vibration value from the sensor unit, thereby increasing data reliability.

In addition, the present invention establishes according to a certain standard by dividing the amount of leakage (small amount, small amount, weight) by the vibration signal size, and the leakage monitoring unit calculates the expected dangerous time and presents it in detail through an alarm to the user (or the measured waveform can be implemented as a Human Machine Interface (HMI), enabling accurate preparation for risks.

In addition, in the present invention, the leak determination of the spacer can raise the similarity rate to 80% to 85% or more by artificial intelligence-based pattern analysis.

1 is a view showing the prior invention.
FIG. 2 is a view showing a state in which a sensor unit for calculating a vibration value by dividing detection information by a plurality of points is attached according to an embodiment of the present invention.
3 is a view showing the configuration of a spacer according to an embodiment of the present invention.
4 is a diagram illustrating a connection relationship between a sensor unit, a signal processing unit, a monitoring unit, and the like, according to another embodiment of the present invention.
5 is a diagram showing an example of signal magnitude pattern classification for each sensor according to spacer leakage according to another embodiment of the present invention.
6 is a diagram illustrating a connection relationship between a sensor unit and a leak monitoring unit according to another embodiment of the present invention.
7 is a view showing a detailed configuration of a leak monitoring unit according to another embodiment of the present invention.
8 is a view showing in detail the step of determining spacer leakage according to another embodiment of the present invention.
9 is a diagram for explaining in detail the structure of the processor of the leak monitoring unit according to another embodiment of the present invention by dividing it into A part, B part, C part, and the like.
FIG. 10 is a view showing in detail part A of FIG. 9 .
11 is a view showing in detail part B of FIG. 9 .
12 is a view showing in detail part C of FIG. 9 .

In order to fully understand the present invention, preferred embodiments of the present invention will be described with reference to the accompanying drawings. Embodiments of the present invention may be modified in various forms, and the scope of the present invention should not be construed as being limited to the embodiments described in detail below. This example is provided to more completely explain the present invention to those of ordinary skill in the art. Accordingly, the shapes of elements in the drawings may be exaggerated to emphasize a clearer description. It should be noted that the same members in each drawing are sometimes shown with the same reference numerals. In addition, detailed descriptions of well-known functions and configurations determined to unnecessarily obscure the gist of the present invention will be omitted.

As shown in Table 1 below, the present invention provides a GIS using dry air, an eco-friendly material, as an insulating material instead of SF6 gas, which is an insulating material for 29kV and 25.8kV GIS.

Conventional GIS (SF6) 1.5 to 2.0 bar Eco-friendly GIS 2.5 to 3.0 bar

The present invention includes a GIS spacer that supports a conductor in a gas insulated switchgear (GIS) metal container and sets a section for dry air, which is an eco-friendly gas, and insulating materials necessary for insulation of various devices.

In the present invention, partial discharge in the GIS due to the defect of the GIS spacer that cannot attach a sensor to the inside for gas insulation and insufficient quality control during the assembly and installation process has a possibility of easily reaching total destruction. In this case, the problem of partial discharge is very important.

As shown in FIG. 2 , the present invention includes a sensor unit 100 that calculates vibration values for each of a plurality of points in the GIS.

For example, a floating point operation of the addition or subtraction of the vibration value of the sensor unit 100 is executed, and then rounding, normalization, and post-processing of exception detection are executed to obtain the calculation result through the FPU (Floating Point Unit). is transmitted to the leak monitoring unit 202 .

In one embodiment, the present invention includes a sensor unit 100 for measuring a leak signal as vibration for each of a plurality of points; a signal detection unit 102 for receiving a signal from the sensor unit 100 in a constant time or multiplexer manner; A specific frequency is separated from the signal received from the signal detection unit 102 by the frequency specifying unit 104, only the specific frequency part is analyzed, a certain leak point is designated, and the vibration signal measured when leaking at each leak point is converted to noise. After removal, a leak monitoring algorithm that determines whether there is a leak by distinguishing it from the general vibration stored in the database, and a leak for comparing and analyzing the pattern of the measured vibration signal to determine the leak location pointed to by the vibration signal having the same or similar pattern and a leak monitoring unit 202 configured with a location estimation algorithm.

As shown in FIG. 3 , the spacer 3 is composed of an inner conductor 2 and an enclosure 1 , and the enclosure 1 is made of an epoxy mold and electricity can be conducted through the inner conductor 2 .

)을 대푯값으로 선정하여 측정값 > 기준값 이상이고 측정시간 > 기준시간 이상이 측정되어야 주변 센서 값을 동시에 저장하여 크기패턴을 비교하고, 상기 측정값 > 기준값 또는 측정시간 > 기준시간이 아니면 다시 진동센서를 계측하며, 상기 측정패턴이 일정 누설지점(SP#1)과 동일 또는 유사(측정패턴 = SP#1) 이면 상기 일정 누설지점(SP#1)과 가장 인접한 스페이서의 누설을 판정한다.4 a, b, and c, the leak location estimation algorithm removes noise from the measured value of the vibration sensor and maximizes the difference between the maximum value (Vp-p) in the feature frequency as well as the feature frequency The integral value of the sum of all the values in (

) is selected as a representative value, and when the measured value > reference value and measurement time > reference time or more are measured, the values of the surrounding sensors are simultaneously saved and the size patterns are compared, and if the measured value > reference value or measurement time > is not the reference time, the vibration sensor again If the measurement pattern is the same as or similar to the predetermined leakage point SP#1 (measurement pattern = SP#1), the leakage of the spacer closest to the predetermined leakage point SP#1 is determined.

As shown in FIG. 5 , the leak monitoring unit 202 converts the measurement pattern into data and automatically presents the location of the leak as a result of analysis, and the leak monitoring unit converts the amount of leakage (trace, small amount, weight) into the vibration signal size. It can be divided and established according to a certain standard, and the leakage monitoring unit can calculate the expected risk point and present it to the user through an alarm.

A wireless communication device for performing such wireless communication, a wireless Internet communication device using a wireless Internet network, a CDMA (CODE DIVISION MULTIPLE ACCESS) network, or a GSM (GLOBAL SYSTEM FOR MOBILE COMMUNICATION) network can be used. .

In an example of the classification of signal magnitude patterns for each sensor according to spacer leakage as an embodiment, an alarm may be transmitted according to each leakage rate through the monitoring unit 300 for monitoring a specific situation.

Here, the leak rate is alarm data used to detect and alert in advance when a state in which sufficient strength and confidentiality cannot be maintained in excess of an allowed leak. The alarm and alarm data may be displayed on the manager's terminal.

As the alarm control chart technique, a Shewhat chart, a cumulative sum control chart, an Exponentially Weighted Moving Average (EWMA; Exponentially Weighted Moving Average), etc. can be used.

As shown in FIG. 6, a leak monitoring unit 202 that receives the vibration value from the sensor unit 100, analyzes it, and alarms; and a communication processing unit 140 that receives the vibration value and transmits it to a higher-level device.

Specifically, the LED control unit 120 receives the control of the leak monitoring unit 202, and the sensor unit 100 controls the SET, RESET (101, 103) buttons and seven on/off display units, and the power supply unit 130. The leakage monitoring unit 202 receiving power from the communication processing unit 140 controls the communication processing unit 140 to detect in advance a case in which sufficient strength and confidentiality cannot be maintained in excess of the permitted leakage and alarms and alarms used to provide an alarm. The data may be displayed on the manager's terminal.

Looking at the structure of the power supply 130 and the leakage monitoring unit 202 used in the power supply unit 130 as shown in FIG. 7 , a power switch connected to a backup circuitry; Power is controlled by input/output logic (I/O logic) using kernel logic, and converted through PLL, comparator, OP AMP, etc.

As shown in Figure 8, the present invention is a power ON (DC 12 ~ 24V) step of the sensor unit 100; an initialization operation (about 10 seconds) of the sensor unit 100; activating the sensor unit 100 (MEMS self-diagnosis); measuring a coordinate value using the MEMS of the sensor unit 100; calculating a vector value (vibration value) of the sensor unit 100; The sensor unit 100 transmits the vibration value to the leak monitoring unit through Modbus communication.

More specifically, as shown in FIG. 13, the present invention includes the steps of measuring a vibration sensor of the sensor unit 100; If the measurement value of the vibration sensor > reference value, storing the measurement pattern for each sensor size (time difference) after storing the surrounding sensor value; Measuring the vibration sensor again if the measured value > reference value; determining leakage of spacer #1 if the measurement pattern is the same as SP#1 (measurement pattern = SP#1); determining leakage of spacer #2 if the measurement pattern is the same as SP#2 (measurement pattern = SP#2); and determining the spacer #3 leak if the measurement pattern is the same as SP#3 (measurement pattern = SP#3).

At this time, the leakage determination of each spacer can raise the similarity rate to 80% to 85% or more by artificial intelligence-based pattern analysis.

As a method of inference and classification by the classification module included in the processor for the analysis, a condition (If-Then) rule can be simply used, and in the case of a complex system encompassing the information of the vibration sensor according to the present invention, fuzzy logic is used. A fuzzy logic (FL) module for calculation, a fuzzy clustering module for calculation by a fuzzy clustering method, an artificial neural network (ANN) module for calculation by a statistical learning algorithm, etc. may be used.

In one embodiment, the present invention comprises the steps of: if the measurement value of the vibration sensor through the noise removal technique > reference value, storing the sensor time discrimination measurement pattern after storing the surrounding sensor value; Measuring the vibration sensor again if the measured value > reference value; determining leakage of the spacer #1 when the measurement pattern is the same as or similar to the first leakage point SP#1 (measurement pattern = SP#1); determining leakage of spacer #2 if the measurement pattern is the same as or similar to SP#2 (measurement pattern = SP#2); and determining the leakage of spacer #3 if the measurement pattern is the same as or similar to SP#3 (measurement pattern = SP#3).

Here, the measurement pattern uses the average of the measured peak values, or the difference between the measurement pattern and the leakage pattern (reference value) of each leakage point stored in the database as an error value. have.

In addition, each number is an arbitrarily assigned leak point and spacer, and leak determination can be performed repeatedly along the entire spacer in real time.

If the total error value of the measurement pattern is less than 95% of the total reference value, a green LED indicates normal, if it is 95% to less than 100%, a yellow LED indicates caution, and if more than 100%, a red LED displays a warning. It is possible to accurately prepare for danger by calculating the expected time of danger and presenting it to the user in detail through an alarm.

Hereinafter, as shown in FIG. 9, it will be described in detail by dividing it into A part, B part, C part, and the like.

10, a joint test action group (JTAG) for processor debugging of the leak monitoring unit, a nested vector interrupt controller (NVIC) for controlling multiple interrupt operations for the processor, addition according to the instructions of the processor, Floating point unit (FPU) that performs floating-point operations such as subtraction, and then performs post-processing of rounding, normalization, and exception detection to obtain an operation result, and generating a data preparation signal to receive the data output conversion result of the processor A1 part including a DRDY (Data-Ready signal) for;

Part A2 including the 12-bit analog-to-digital converter of the sensor unit 100; consists of.

In addition, as shown in FIG. 11, the B1 part including Power for supplying power from 3.3V to 1.8V; a B2 part including a supply supervisor through a power-on reset (POR) that controls the power to be operated later for a predetermined time; B3 part and B4 part connected to OSC (oscillator) by controlling RC (Resistor/Capacitor) and HS (High Speed Crystal/Resonator); It consists of a real-time clock (RTC) and AWU (Auto Wakeup) function and a B5 part with a backup function.

As shown in Fig. 12, a C1 part including Watchdog timers (independent, window); C2 part composed of a comparator (GP Comparator) and; The C3 part is composed of an operational amplifier and a digital-to-analog converter (DAC).

For example, if there is no detection after a certain period of time (1 second) after receiving the vibration value through the DRDY (Data-Ready signal) of the processor, after resetting the watchdog timers (independent, window), the data output is converted to accept the data output conversion result. Re-generate the ready signal.

The watchdog timers may provide a safety watchdog function through an Emergency Stop Function unit trained through a learning module using a prediction and detection algorithm, and the like.

At this time, the added watchdog timer diagnosis apparatus outputs an invalid trigger signal to the watchdog timer before entering the stop command mode according to a command from an external device, and then determines whether the watchdog timer outputs a reset signal. Accordingly, it is possible to diagnose whether the watchdog timer is malfunctioning.

For example, by applying the LSTM algorithm for correcting the learning error of the RNN technique using data measured in real-time 1 second to 1 minute cycles, it is possible to check the operation status of the Safety Watchdog and provide the setting function.

The sensor unit 100 for calculating a vibration value by classifying sensing information for each of a plurality of points includes: an optical amplifier generating an optical signal in a broadband wavelength region by natural amplification emission and amplifying a feedback input optical signal; a wavelength selection filter for filtering a specific wavelength from the output optical signal of the optical amplifier using one or more slits movable in the horizontal direction and rotating at a constant speed; and an optical coupler that feeds back a portion of the optical signal of the wavelength filtered by the wavelength selection filter to the optical amplifier and outputs the remainder of the optical signal to an output terminal.

In addition, the present invention is an acceleration sensor, a gyroscope sensor, an optical fiber vibration sensor, a laser vibration meter; etc. can be used, including a charge-to-voltage converter that receives the charge output of the vibration-type gyro sensor and converts it into a voltage signal, so that the detection result can be accurately identified.

Terms used in this document are only used to describe specific embodiments, and may not be intended to limit the scope of other embodiments.

The singular expression may include the plural expression unless the context clearly dictates otherwise. Terms used herein, including technical or scientific terms, may have the same meanings as commonly understood by one of ordinary skill in the art described in this document.

Unless explicitly defined in this document, it should not be construed in an idealistic or overly formal sense. In some cases, terms defined in this document cannot be interpreted to exclude various embodiments other than this document.

For example, the spacer 3 consisting of the inner conductor 2 and the enclosure 1 can mix nano fillers in the epoxy mold, and the carbon nanotubes that can be used as nano fillers are multi-walled carbon nanotubes (Grade). : CM-95, Hanwha Nanotech Co.,), single-walled carbon nanotube (Grade: ASP-100F, Hanwha Nanotech Co.,), etc.

100: sensor unit, vibration sensor
102: signal detection unit
104: frequency specifying part
202: leak monitoring unit
300: monitoring unit

Claims (8)

In a measurement method using an eco-friendly GIS (GAS INSULATED SWITCHGEAR) capable of detecting ultrafine dry air leakage using a high-sensitivity vibration sensor,
a sensor unit 100 for measuring a leak signal as vibration for each of a plurality of points;
a signal detecting unit 102 for receiving a signal from the sensor unit 100 in a constant time or multiplexer manner;
A specific frequency is separated from the signal received from the signal detection unit 102 by the frequency specifying unit 104, only the specific frequency part is analyzed, a certain leak point is designated, and the vibration signal measured when leaking at each leak point is converted into noise. After removal, a leak monitoring algorithm that determines whether there is a leak by distinguishing it from the general vibration stored in the database, and a leak for comparing and analyzing the pattern of the measured vibration signal to determine the leak location pointed to by the vibration signal having the same or similar pattern In a measurement method using an eco-friendly GIS that can detect dry air leakage using the sensor unit 100 including; a leak monitoring unit 202 composed of a location estimation algorithm,
measuring a vibration sensor in the sensor unit;
If the measured value of the vibration sensor > reference value, storing the sensor time discrimination measurement pattern after storing the surrounding sensor value;
Measuring the vibration sensor again if the measured value > reference value;
If the measurement pattern is the same as or similar to the predetermined leakage point (SP#n) (measurement pattern = SP#n), determining the leakage of the spacer at the point closest to the predetermined leakage point (SP#n); ,
The leakage monitoring unit automatically presents the leak location as a result of data analysis of the measurement pattern, the leakage monitoring unit divides the amount of leakage by the vibration signal size and establishes it according to a certain standard, and the leakage monitoring unit calculates the expected dangerous time In presenting to the user through an alarm, the alarm according to each leakage rate by the signal magnitude pattern classification for each sensor through the monitoring unit 300 for monitoring a specific situation according to the signal magnitude pattern classification for each sensor according to the spacer leakage and the leakage rate exceeds the allowed leakage, and is alarm data used to detect and alarm in advance when a certain strength and confidentiality cannot be maintained, and the alarm and alarm data are displayed on the manager's terminal becomes,
The leak monitoring algorithm uses an exponentially weighted moving average (EWMA) as a management chart technique for managing and monitoring the real-time frequency of the sensor unit of the alarm data,
The leak determination of each spacer uses a conditional (If-Then) rule as an inference and classification method by a classification module included in a processor for artificial intelligence-based pattern analysis,
The processor includes a fuzzy logic (FL) module for calculating by fuzzy logic, a fuzzy clustering module for calculating by a fuzzy clustering method, and an artificial neural network (ANN) module for calculating by a statistical learning algorithm. used,
If there is no detection after a certain period of time after receiving the vibration value through the DRDY (Data-Ready signal) of the processor, reset the watchdog timer and generate a data preparation signal again to accept the data output conversion result,
The watchdog timer provides a watchdog function trained through a learning module using a leak monitoring algorithm. A measurement method using an eco-friendly GIS capable of detecting ultra-fine leakage of dry air using a high-sensitivity vibration sensor.
The method according to claim 1,
The watchdog timer diagnosis apparatus outputs an invalid trigger signal to the watchdog timer before entering the stop command mode according to a command from an external device, and then outputs a watchdog timer according to whether the watchdog timer outputs a reset signal. After diagnosing whether the dock timer is malfunctioning,
Using the alarm data measured in real time, the LSTM algorithm for correcting the learning error of the RNN technique is applied to check the operation status of the Safety Watchdog and provide a setting function. A measurement method using eco-friendly GIS that can be detected.
The method according to claim 1,
The spacer 3 mixes nano fillers in an epoxy mold, and the carbon nanotubes that can be used as the nano fillers are dried using a high-sensitivity vibration sensor, characterized in that they are multi-wall carbon nanotubes or single-wall carbon nanotubes. A measurement method using eco-friendly GIS that can detect ultra-fine air leaks. delete delete delete delete delete

Images