KR20230137023A - diagnosis method and system for diagnose water valve - Google Patents

Abstract

In order to improve the accuracy of failure determination, the present invention is a method of diagnosing a fault of a water control valve provided with a disk that is axially connected to a cap hinged to the upper side of the body connected between fluid transfer pipes and opens and closes the inside of the body, A first step of transmitting first pressure data and second pressure data measured by the first pressure sensor and the second pressure sensor respectively provided at the inlet end and the discharge end to the server unit; A calculation unit communicationally connected to the server unit continuously calculates flow rate data through a preset mathematical equation including a variable value for the pressure difference between the first pressure data and the second pressure data, and transmits and stores the flow data to the server unit. Step 2; And a third step in which the calculation unit compares the flow rate data with the steady-state reference average data and determines whether the mutual agreement rate is outside the preset range, and a third step of transmitting a failure signal to an output unit communicationally connected to the server unit. Failure diagnosis of the water control valve. Provides a method.

Description

Failure diagnosis method and system for water valve {diagnosis method and system for diagnose water valve}

The present invention relates to a method and system for diagnosing a water sluice valve failure, and more specifically, to a method and system for diagnosing a water sluice valve in which the accuracy of failure determination is improved.

In general, a valve is a device that has an opening and closing function to block and pass the flow of fluid, and a function to control flow rate, pressure, etc., and water drain valves for opening and closing are generally used in water supply networks.

At this time, in the water pipe network, the water drain valve is installed at the water pipe joint to control the flow and amount of running water to ensure that the pressure in the pipe is maintained appropriately, and is mainly used to isolate sections of the water pipe network or to cut off and pass water. It is a facility. In addition, it can be said to be a major facility essential for underground water leak exploration.

These water sump valves require replacement when they deteriorate due to causes such as water leakage due to corrosion, inability to open and close, and reduction of the pipe cross-section due to scale. In addition, when the pipe network is isolated, the water system is adjusted, water is cut off, or the valve is suddenly opened or closed (opened and closed) during pipe cleaning, the pressure rises rapidly in the water supply pipe, causing water to hit the pipe wall, causing noise and vibration, resulting in a water hammer effect. This happens. This phenomenon causes damage to the tube. In addition, at this time, corrosion-resistant coating on the inner surface of the drain pipe and rust or foreign substances stagnating in the pipe leak into the customer's water hydrant, causing distrust in tap water.

At this time, most of the conventional water control valves are basically products equipped with a display device that can check the opening state (opening degree rate) on site, an alarm for sudden valve opening and closing, and a communication function that can remotely check the current opening state from the center. As a result, research is being conducted to remotely diagnose whether the water valve is broken.

Here, conventionally, field workers open and close the water sump valve, but there is no information to know the opening state, so they operate by referring to a data sheet that organizes the opening and closing rotations by water sump valve diameter and model. There was a problem of not being able to accurately determine the state of the island.

Accordingly, a display device that allows workers to check the opening state of the valve in the field, an alarm device that allows workers to be aware of sudden valve operation, and a device that can transmit the current opening state to the central server in real time through a wireless communication function. A technology equipped with is required.

Korean Patent No. 10-1535732

In order to solve the above problems, the object of the present invention is to provide a failure diagnosis method and system for a water control valve that improves the accuracy of failure determination.

In order to solve the above problem, the present invention is a method for diagnosing the failure of a water control valve provided with a disk that is axially connected to a cap portion hinged to the upper side of the body portion connected between fluid transfer pipes and opens and closes the inside of the body portion, A first step of transmitting first pressure data and second pressure data measured by the first pressure sensor and the second pressure sensor respectively provided at the inlet end and the discharge end to the server unit; A calculation unit communicationally connected to the server unit continuously calculates flow rate data through a preset mathematical equation including a variable value for the pressure difference between the first pressure data and the second pressure data, and transmits and stores the flow data to the server unit. Step 2; And a third step in which the calculation unit compares the flow rate data with the steady-state reference average data and determines whether the mutual agreement rate is outside the preset range, and a third step of transmitting a failure signal to an output unit communicationally connected to the server unit. Failure diagnosis of the water control valve. Provides a method.

Meanwhile, the present invention includes a body part connected between a fluid transfer pipe through which fluid moves, a cap part hinged to an upper part of the body part, and a shaft connected to the cap part to open and close the inside of the body part according to the degree of movement in the vertical direction. A water control valve including a shaft-connected disk; A first pressure sensor provided at the inlet end of the body part to generate first pressure data; a second pressure sensor provided at the discharge end of the body unit to generate second pressure data; An opening quantity measurement sensor connected to the cap unit and acquiring opening quantity data in response to a rotation angle of the cap unit; a server unit that is connected to the first pressure sensor, the second pressure sensor and the opening quantity measurement sensor and receives and stores the first pressure data, the second pressure data and the opening quantity data; It is connected to the server for communication, extracts the opening amount data, inputs the flow coefficient values according to the opening amount into a plurality of preset mapping data, calculates the flow coefficient as an output value, and calculates the flow coefficient as an output value. Flow rate data according to the opening amount is continuously calculated through a preset mathematical equation including the variable value for the pressure difference between the second pressure data and the flow rate coefficient, and transmitted to the server, and the flow rate data is converted into steady-state reference average data. an operation unit that compares and determines; and is connected to the server unit for communication, where the first pressure data, the second pressure data, and the opening quantity data are displayed, and the calculation unit compares the flow data with the steady-state reference average data to determine that the mutual agreement rate is preset. Provides a fault diagnosis system for a water control valve that includes an output unit that receives a fault signal when it is out of range.

Through the above solutions, the present invention provides the following effects.

First, by measuring the opening rate and pressure drop of the sump valve, the calculation unit compares the flow rate data calculated through a preset mathematical equation with the normal state standard average data, and when the mutual agreement rate is outside the preset range, a failure signal is sent, so the sump valve The fault diagnosis accuracy can be significantly improved.

Second, the flow rate data is calculated through the pressure difference between the first and second pressure data measured from the first and second pressure sensors respectively provided at the inlet and discharge ends of the water sump valve, so there is no need to damage the water sump valve. Economic efficiency can be significantly improved by diagnosing water valve failures.

1 is a flowchart showing a method for diagnosing a failure of a water drain valve according to an embodiment of the present invention.
Figure 2 is a block diagram showing a failure diagnosis system for a water drain valve according to an embodiment of the present invention.
Figure 3 is a perspective view showing a water drain valve in a method for diagnosing a water drain valve failure according to an embodiment of the present invention.
Figure 4 is an example of a partial projection of the inside of the water control valve in the method for diagnosing a failure of the water control valve according to an embodiment of the present invention.
Figures 5 and 6 are graphs showing mapping data in which the flow coefficient value is set according to the opening amount in the failure diagnosis method of the water sump valve according to an embodiment of the present invention.

Hereinafter, a failure diagnosis method and system for a water drain valve according to a preferred embodiment of the present invention will be described in detail with reference to the attached drawings.

Figure 1 is a flowchart showing a method for diagnosing a failure of a water control valve according to an embodiment of the present invention, Figure 2 is a block diagram showing a system for diagnosing a failure of a water control valve according to an embodiment of the present invention, and Figure 3 is a flowchart showing a system for diagnosing a failure of a water control valve according to an embodiment of the present invention. It is a perspective view showing a sump valve in a method for diagnosing a sump valve according to an embodiment of the present invention, and Figure 4 is an exemplary diagram showing a partial projection of the inside of the sump valve in a method for diagnosing a sump valve according to an embodiment of the present invention. Figures 5 and 6 are graphs showing mapping data in which the flow coefficient value is set according to the opening amount in the failure diagnosis method of the water sump valve according to an embodiment of the present invention.

As shown in Figures 1 to 6, the method for diagnosing a failure of a water drain valve according to an embodiment of the present invention uses a first pressure sensor 20 and a second pressure sensor respectively provided at the inlet end and the discharge end of the body part 11. The first pressure data and second pressure data measured by the pressure sensor 30 are transmitted to the server unit 50 (s10), and the calculation unit 60 provides a variable for the pressure difference between the first pressure data and the second pressure data. The flow rate data is continuously calculated through a preset mathematical equation including the value, transmitted and stored in the server unit 50 (s20), and the calculation unit 60 compares the flow rate data stored by time with the steady state reference average data and determines the mutual It includes a series of steps of transmitting a failure signal to the output unit 70 (s30) when the coincidence rate deviates from the preset range. In addition, the failure diagnosis system of the water sump valve according to an embodiment of the present invention includes a water sump valve 10, a first pressure sensor 20, a second pressure sensor 30, an opening quantity measurement sensor 40, and a server unit ( 50), a calculation unit 60, and an output unit 70 are preferably included.

Here, the water control valve 10 according to an embodiment of the present invention has a body part 11 connected between the fluid transfer pipe 1 through which fluid moves, and a hinged connection to the upper part of the body part 11 to rotate. It is preferable to include a cap portion 12 and a disk 13 that is axially connected to the cap portion 12 via a shaft 14 and opens and closes the internal space of the body portion 11 according to the degree of vertical movement. .

In detail, the body portion 11 may be connected between the fluid transfer pipes 1 through which fluid moves. Additionally, the cap portion 12 may be coupled to the shaft 14 that is disposed on the top of the water valve 10 and extends in the vertical direction to adjust the flow rate through rotation. At this time, the cap portion 12 is axially connected to the upper end of the shaft 14, and the cap portion 12 can be rotated along a horizontal direction about an axis disposed in the vertical direction.

In addition, the disk 13 is axially connected to the lower end of the shaft 14 and can move in the vertical direction by rotation of the cap portion 12, and the internal space of the body portion 11 depending on the degree of vertical movement. can be opened and closed. Accordingly, the amount of fluid passing through the body 11 can be adjusted by opening and closing the inside of the body 11 by the disk 13.

Here, the shaft 14 is connected between the cap portion 12 and the disk 13, and the disk 13 moves the inside of the body portion 11 based on the rotation amount of the cap portion 12. The disk 13 can be moved vertically to open or block.

Meanwhile, first pressure data and second pressure data individually measured from the first pressure sensor 20 and the second pressure sensor 30 provided at the inlet and outlet ends of the body portion 11, respectively. It is transmitted to the server unit 50 (s10). At this time, preferably, the first pressure sensor 20 and the second pressure sensor 30 are respectively connected to the fluid transfer pipe 1 on the inlet end and the fluid transfer pipe 1 on the discharge end of the body 11. do.

And, the first pressure sensor 20 and the second pressure sensor 30 communicate wirelessly with the server unit 50 to transmit the first pressure data and the second pressure data to the server unit 50. Wireless communication can be connected to each other through means.

In addition, an opening quantity measurement sensor 40 is connected to the cap part 12 of the water valve 10 to obtain opening quantity data corresponding to the rotation angle of the cap part 12 and transmit it to the server unit 50. desirable. At this time, the opening quantity measurement sensor 40 may be connected to the server unit 50 through wireless communication. In addition, the server unit 50 provided as a server device is connected to the server unit 50 as an arithmetic device, and the output unit 70 is provided as an input/output device such as a terminal to the server unit 50. communication can be established. At this time, the first pressure data, the second pressure data, and the opening amount data may be displayed on the output unit 70.

In addition, the first pressure data measured by the first pressure sensor 20 and the second pressure data measured by the second pressure sensor 30 are sensed in units of sensing time, and the server unit 50 It is desirable to transmit and store it. For example, the sensing time unit refers to the sensing cycle and can be set to 1/1,000 second (1,000 Hz). That is, the first pressure sensor 20 and the second pressure sensor 30 repeatedly sense the first pressure data and the second pressure data once every 1/1,000 second, that is, 1,000 times per second. You can.

And, the step of transmitting the first pressure data and the second pressure data measured by the first pressure sensor 20 and the second pressure sensor 30 to the server unit 50 includes the cap unit 12 It is preferable to include the step of transmitting the opening quantity data obtained in correspondence to the rotation angle of the cap unit 12 from the opening quantity measurement sensor 40 connected to ) to the server unit 50 in units of the sensing time. .

Here, the opening amount measurement sensor 40 includes a rotation measuring means for measuring the rotation angle of the cap portion 12, and maps the rotation angle to preset conversion data as an input value to calculate the opening amount data as a result value. It may include a conversion operation unit. At this time, the opening quantity measurement sensor 40 may further include a storage means for storing the conversion data. Of course, in some cases, the conversion data may be stored in the server unit 50 and transmitted to the opening quantity measurement sensor 40.

In addition, the conversion operation unit may extract the rotation angle measured by the rotation measurement means and input it into the conversion data. As the rotation angle is input into the conversion data, the opening quantity data mapped correspondingly is the result. It can be calculated as a value. For example, the process of calculating the opening amount data of the water control valve 10 based on the rotation amount of the cap portion 12 is specifically, the rotation amount of the cap portion 12 measured by the rotation measurement means by the conversion operation unit. It can be performed by mapping to the pre-stored conversion data and calculating the opening amount data of the water control valve 10. At this time, the pre-stored conversion data is preferably understood as map data that uses the rotation amount of the cap unit 12 as input and the opening amount data as output. Subsequently, the opening quantity data may be transmitted to the server unit 50 and stored in units of the sensing time.

Meanwhile, the calculation unit 60, which is communication-connected to the server unit 50, continuously calculates flow rate data through a preset mathematical equation including a variable value for the pressure difference between the first pressure data and the second pressure data. and transmit and store it in the server unit 50 (s20). At this time, the above equation is as follows.

Here, Cv is the flow coefficient, △P means first pressure data (P1) - second pressure data (P2), and SG means specific gravity. At this time, △P is preferably understood as the difference between the first pressure data (P1) and the second pressure data (P2), that is, the amount of pressure change inside the body portion 11. Additionally, in the case of water, SG is set to 1.0 at 15°C. To this end, in some cases, the water control valve 10 according to an embodiment of the present invention may further include a temperature sensor (not shown) that measures the temperature of the fluid flowing inside the body portion 11, The calculation unit 60 may match and calculate the specific gravity of the fluid based on the temperature data of the fluid measured by the temperature sensor (not shown).

In addition, referring to FIG. 5 and FIG. 6 (a), (b), and (c), the flow coefficient (Cv) is required to calculate the flow rate data, and for this purpose, the calculation unit 60 uses the It is preferable to perform a step of calculating the flow coefficient (Cv) as an output value by extracting the opening amount data and inputting it into a plurality of preset mapping data with different flow coefficient values according to the opening amount.

In detail, a plurality of valve information data corresponding to the internal shape of the body 11 is input and stored in the server unit 50, and each valve information data is individually matched to the opening amount. A plurality of the mapping data, each with different flow coefficient values, may be pre-stored. In other words, the flow coefficient value according to the opening amount may appear differently corresponding to the shape of the orifice pipe, etc. inside the body portion 11. To reflect this, the server portion 50 individually matches each valve information data. Thus, a plurality of the mapping data, each with a different flow coefficient value depending on the opening amount, can be pre-stored.

In addition, the calculation unit 60 extracts the input valve information data, extracts one of the mapping data matching the extracted valve information data, and inputs the opening amount data into the extracted mapping data. , the flow coefficient can be calculated. At this time, the valve information data may be input from the output unit 70 and transmitted to and stored in the server unit 50.

Next, it is preferable that the calculation unit 60 extracts the first pressure data, the second pressure data, and the flow rate coefficient and continuously calculates flow rate data according to the opening amount through the above equation.

Meanwhile, the calculation unit 60 determines by comparing the flow rate data stored by time with the steady-state reference average data, and when the mutual agreement rate deviates from the preset range, a failure signal is sent to the output unit 70 that is communication connected to the server unit 50. transmit (s30). Through this, as the user checks the failure signal displayed on the output unit 70, the user can easily determine whether a failure has occurred remotely and take corresponding measures, thereby improving convenience of use.

In addition, the calculation unit 60 determines by comparing the flow rate data stored by time with the steady state reference average data, and if the mutual agreement rate is within a preset range, the output unit 70 connected to the server unit 50 by communication determines the normal state. A status notification signal can be transmitted. At this time, the mutual agreement rate may be calculated by the calculation unit 60 in correspondence with the ratio between the flow rate data and the steady state reference average data.

In addition, the steady-state reference average data can be pre-stored in the server unit 50 as pressure change data over time when fluid moves within the body unit 11 in a steady state.

Alternatively, the steady-state reference average data may be generated through an unsupervised machine learning process by a machine learning unit included in the calculation unit 60. For example, the machine learning unit included in the calculation unit 60 divides the pressure change data over time obtained from the first pressure sensor 20 and the second pressure sensor 30 into a plurality of preset time units, It may be generated by calculating an autoregressive cumulative moving average of the plurality of divided pressure change data. At this time, the pressure change data is preferably understood as the difference between data measured by the first pressure sensor 20 and the second pressure sensor 30.

At this time, the pressure change data obtained from the first pressure sensor 20 and the second pressure sensor 30 are time series data measured over time, and a source code called Orion is used to detect abnormalities in time series data. Open Python libraries are useful. In detail, the Orion is an artificial intelligence library developed by the Data-to-AI group at the MIT Laboratory for Information and Decision Systems, and is used to analyze signals generated from spacecraft and the temperature of turbines. It is used to detect abnormal signs in various fields, such as computer server usage and soil moisture variations.

Here, to explain the time series anomaly detection algorithm, the machine learning model created after machine learning normal time series data (Train Data) with no anomalies regresses the predicted value, compares this predicted value with the evaluation data (Test Data), and other When there is a part, it is detected as abnormal. At this time, machine learning algorithms range from ARIMA (Auto Regressive Integrated Moving Average) and LSTM (Long Short Term Memory), which are traditionally used for time series prediction, to GAN Generative Adversarial Network for anomaly detection. There is TadGAN (Time-series GAN for Anomaly Detection) used. In addition, the Orion library is implemented so that the algorithm used for such regression analysis can be selected simply by changing the hyperparameter.

Here, the learning data obtained from the first pressure sensor 20 and the second pressure sensor 30 are preprocessed with a timestamp suitable for the Orion library that uses Unix time, and a machine learning model is generated. Then, the previously mentioned regression analysis algorithm is first selected as the hyperparameter. At this time, the abnormality detection performance of the traditional LSTM algorithm was found to be superior to that of the TadGAN algorithm, which is known to be excellent for abnormality detection in other industrial fields, and the calculation speed was also faster. In addition, anomaly detection performance changes sensitively to other hyperparameter-interval of timestamp settings and is also affected by the number of learning times (Epoch), so the task of finding appropriate hyperparameters must be repeated. Through this hyperparameter optimization, an anomaly detection model is created and anomalies are detected by comparing them with abnormal evaluation data.

And, the calculation unit 60 compares and determines the number of times the mutual coincidence rate deviates from the preset range in a preset reference time unit, and if the number of times deviates from the preset range, the server unit 50 In order to update the pre-stored steady-state reference average data, the calculation unit 60 extracts one of a plurality of steady-state reference average data pre-stored in the machine learning library of the server unit 50 to change the previous steady-state reference average data. A step of artificial intelligence learning can be performed by changing the hyperparameters to replace .

That is, if the number of times is outside the preset number range, the calculation unit 60 stores data in the machine learning library of the server unit 50 to replace the steady-state reference average data previously stored in the server unit 50. A step of artificial intelligence learning by extracting one of a plurality of stored steady-state reference average data and changing hyperparameters to replace the steady-state reference average data may be performed.

For example, when the preset number of times range is set to 8 to 10, if the number of times the mutual agreement rate deviates from the preset range is 8 to 10, the steady state reference average pre-stored in the server unit 50 To update the data, the calculation unit 60 extracts one of a plurality of steady-state reference average data previously stored in the machine learning library of the server unit 50 and sets a hyperparameter to replace the previous steady-state reference average data. Steps to change may be performed.

At this time, the steady-state reference average data ranges from ARIMA (Auto Regressive Integrated Moving Average) and LSTM (Long Short Term Memory), which are traditionally used for time series prediction, to GAN Generative Adversarial Network. A plurality of numbers can be generated by a machine learning algorithm such as TadGAN (Time-series GAN for Anomaly Detection) used for anomaly detection and stored in a machine learning library, and the calculation unit 60 changes the hyperparameter. Accordingly, one of a plurality of steady-state reference average data previously stored in the machine learning library can be extracted and replaced with the previous steady-state reference average data just by changing the hyperparameter.

In addition, the preset number range refers to the hyperparameter, interval of timestamp, number of learning times (Epoch), and the number of divisions in the window that indicate how much the entire data should be divided into when determining an abnormality. It is influenced by size (Window Size Portion) and can be found using the trial-and-error method. That is, the steady-state reference average data can be variably set through a trial and error method of continuously changing hyperparameters, generating the steady-state reference average data, and comparing it with the pressure change data.

In addition, the calculation unit 60 converts the variable value data for the pressure difference between the first pressure data and the second pressure data from the time domain to the frequency domain to generate frequency domain converted data and generate wavelet function data. , a step of storing as time series data, and a step of individually comparing and determining the frequency domain converted data, the wavelet function data, and the time series data may be further performed.

In this way, the present invention measures the opening rate and pressure drop of the water control valve 10, and compares the flow rate data calculated by the calculation unit 60 through a preset mathematical equation with the steady-state reference average data. When the coincidence rate deviates from the preset range, a failure signal is transmitted to the output unit 70, so the accuracy of failure diagnosis of the water control valve 10 can be significantly improved.

In addition, the flow rate data is calculated through the pressure difference between the first pressure data and the second pressure data measured from the first and second pressure sensors respectively provided at the inlet and discharge ends of the water sump valve ( Since it is possible to diagnose a failure of the water valve 10 without damaging the 10), economic efficiency can be significantly improved.

In addition, the calculation unit 60 compares and determines the number of times the mutual agreement rate deviates from the preset range, and when the number deviates from the preset range, a plurality of pre-stored The precision of failure determination can be improved by extracting one of the steady-state reference average data and replacing the existing steady-state reference average data.

At this time, terms such as “include,” “comprise,” or “equipped” described above mean that the corresponding component may be present, unless specifically stated to the contrary, excluding other components. It should be interpreted as being able to include other components. All terms, including technical or scientific terms, unless otherwise defined, have the same meaning as generally understood by a person of ordinary skill in the technical field to which the present invention pertains. Commonly used terms, such as terms defined in a dictionary, should be interpreted as consistent with the contextual meaning of the related technology, and should not be interpreted in an idealized or overly formal sense unless explicitly defined in the present invention.

As described above, the present invention is not limited to the above-described embodiments, and modifications and implementations can be made by those skilled in the art without departing from the scope of the claims of the present invention. And such modifications fall within the scope of the present invention.

1: Fluid transfer pipe 10: Water control valve
20: first pressure sensor 30: second pressure sensor
40: Opening quantity measurement sensor 50: Server unit
60: calculation unit 70: output unit

Claims (5)

In the method of diagnosing the failure of a water control valve provided with a disk that is axially connected to a cap part hinged to the upper side of the body part connected between fluid transfer pipes and opens and closes the inside of the body part,
A first step of transmitting first pressure data and second pressure data measured by the first pressure sensor and the second pressure sensor respectively provided at the inlet and discharge ends of the body to the server unit;
A calculation unit communicationally connected to the server unit continuously calculates flow rate data through a preset mathematical equation including a variable value for the pressure difference between the first pressure data and the second pressure data, and transmits and stores the flow data to the server unit. Step 2; and
A failure diagnosis method of a water control valve comprising a third step of the calculation unit comparing the flow rate data with the steady-state reference average data, and transmitting a failure signal to an output unit communicationally connected to the server unit when the mutual agreement rate is outside the preset range. . According to claim 1,
The first step includes transmitting the opening quantity data obtained in correspondence to the rotation angle of the cap unit from the opening quantity measurement sensor connected to the cap unit to the server unit,
The second step includes calculating the flow coefficient as an output value as the calculation unit extracts the opening amount data and inputs it into a plurality of preset mapping data with different flow coefficient values according to the opening amount,
A fault diagnosis method for a water sump valve, wherein the calculation unit extracts the first pressure data, the second pressure data, and the flow coefficient and continuously calculates flow rate data according to the opening amount through the equation. . According to claim 2,
In the second step,
The above equation is

Calculated by, Cv is the flow coefficient, △P is the first pressure data (P1) - second pressure data (P2), SG is specific gravity,
A plurality of valve information data corresponding to the internal shape of the body is input and stored in the server unit, and a plurality of mapping data are individually matched to each valve information data and each has a different preset flow coefficient value according to the opening amount. It is pre-stored,
The operation unit extracts the inputted valve information data, extracts one of the mapping data matching the extracted valve information data, and inputs the opening amount data into the extracted mapping data. Troubleshooting method. According to claim 1,
The third step is,
a step where the calculation unit compares and determines the number of times the mutual agreement rate deviates from a preset range in a preset standard time unit;
If the number of times is outside the preset number range, the calculation unit extracts one of a plurality of steady-state reference average data pre-stored in the machine learning library of the server unit to replace the steady-state reference average data previously stored in the server unit. A fault diagnosis method for a water control valve, characterized in that it includes the step of changing hyperparameters and learning artificial intelligence to replace the steady-state reference average data. A body part connected between the fluid pipes through which the fluid moves, a cap part hinged to the upper part of the body part, and a disk axially connected to the cap part through a shaft to open and close the inside of the body part according to the degree of vertical movement. Includes a water control valve;
A first pressure sensor provided at the inlet end of the body part to generate first pressure data;
a second pressure sensor provided at the discharge end of the body portion to generate second pressure data;
An opening quantity measurement sensor connected to the cap unit and acquiring opening quantity data in response to a rotation angle of the cap unit;
a server unit that is connected to the first pressure sensor, the second pressure sensor and the opening quantity measurement sensor and receives and stores the first pressure data, the second pressure data and the opening quantity data;
It is connected to the server for communication, extracts the opening amount data, inputs the flow coefficient values according to the opening amount into a plurality of preset mapping data, calculates the flow coefficient as an output value, and calculates the flow coefficient as an output value. Flow rate data according to the opening amount is continuously calculated through a preset mathematical equation including the variable value for the pressure difference between the second pressure data and the flow rate coefficient, and transmitted to the server, and the flow rate data is converted into steady-state reference average data. an operation unit that compares and determines; and
Communication is connected to the server unit, and the first pressure data, the second pressure data, and the opening amount data are displayed, and the calculation unit compares the flow rate data with the steady-state reference average data and determines that the mutual agreement rate is within a preset range. A fault diagnosis system for a water control valve that includes an output unit that receives a fault signal when it deviates.