KR102403270B1 - Controlling system for reservoir water level based on artificial intelligence - Google Patents

Abstract

Disclosed is a reservoir water level control system based on artificial intelligence. The present invention comprises: a water gauge which measures water level data of a water reservoir; a computing device which generates prediction data for reservoir water level based on the water level data and generates state information of the water gauge based on the water level data; and a pump control unit which controls a pump according to the water level data or the prediction data based on the state information. The reservoir water level control system automatically controls the reservoir water level by using artificial intelligence.

Description

Artificial intelligence-based water level control system {CONTROLLING SYSTEM FOR RESERVOIR WATER LEVEL BASED ON ARTIFICIAL INTELLIGENCE}

The present invention relates to an artificial intelligence-based reservoir water level control system. More particularly, it relates to a system and method capable of controlling a pump based on artificial intelligence in order to automatically adjust the water level of the reservoir.

An artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike the existing rule-based smart system, the machine learns, judges, and becomes smarter by itself. As artificial intelligence systems are used, the recognition rate improves and users can understand user preferences more accurately.

Artificial intelligence technology consists of machine learning (deep learning) and element technologies using machine learning. Machine learning is an algorithm technology that categorizes/learns characteristics of input data by itself, and element technology is a technology that uses machine learning algorithms such as deep learning, such as language understanding, visual understanding, reasoning/prediction, knowledge expression, motion control, etc It consists of technical fields of

A drainage basin refers to a reservoir or storage facility created to distribute tap water to various regions, and may be a facility for stably supplying tap water by forming a constant water pressure.

Existing water level control devices use only the water level data of the reservoir to control the pump. There were problems such as waste.

Accordingly, there is an increasing need for artificial intelligence-based water level control technology that can automatically detect and predict the water level by introducing artificial intelligence technology to promote more efficient water resource management.

Korean Patent Laid-Open Publication No. 10-2005-0007232 (Title of Invention: Water level control device for water supply drainage basin)

It is an object of the present invention to automatically control the water level of a reservoir through artificial intelligence.

In addition, an object of the present invention is to quickly and accurately detect an abnormal state of a water level gauge, utilize predicted data, and thereby more efficiently manage water resources.

In order to solve the above problems, the present invention generates a water level gauge for measuring the water level data of the reservoir, the predicted data for the water level of the reservoir based on the water level data, and the state information of the water level meter based on the water level data Based on the generating computing device and the state information, it may include a pump control unit for controlling the pump according to the water level data or the predicted data.

In addition, the water level gauge periodically measures the water level data based on a predetermined unit period, the computing device extracts a predetermined number of measured values from the water level data, and generates the state information based on the measured values can do.

In addition, the pump control unit may control the pump according to the prediction data when the water level gauge is in an abnormal state, and control the pump according to the water level data when the water level gauge is in a normal state.

Also, the computing device may generate demand forecast data for water stored in the reservoir based on the LSTM model, and generate the forecast data based on the demand forecast data.

In addition, the computing device generates water usage information for an hourly decrease amount of water stored in the reservoir based on the water level data, and pump operation information for on/off hourly on/off of the pump from the pump control unit generating, the computing device generates a training data set using the water level data, the water usage information, and the pump operation information as input values for the Long Short-Term Memory (LSTM) model, and the learning By learning the data set, the demand forecast data may be generated as a result value.

In addition, when the water level gauge is in an abnormal state, the computing device generates the prediction data based on the water level data before a first time, and the first time is a measurement value based on the abnormal state. may include time.

Also, the computing device may verify the prediction data based on the water level data prior to the first time.

The present invention has the effect of automatically controlling the water level of the reservoir through artificial intelligence.

In addition, the present invention has the effect that more efficient water resource management is possible through the use of predicted data by quickly and accurately detecting the abnormal state of the water level gauge.

In addition, the present invention has the effect of providing a water level gauge with strong durability and a water level control system using the same by protecting the pressure sensor through a coating layer that is strong in moisture and has a strong stain resistance.

In addition, the present invention has the effect of generating more accurate prediction data by overcoming the inaccuracy of the existing RNN model by using the LSTM model.

In addition, the present invention can automatically verify prediction data by using a clustering technique as a verification process, so that it can be applied in a variety of environments collectively, thereby reducing cost and time, as well as performing the verification process to reduce errors. It has the effect of significantly reducing the probability of occurrence.

The effects obtainable according to the present invention are not limited to the above-mentioned effects, and other effects not mentioned are clearly understood by those of ordinary skill in the art to which this specification belongs from the external surface of the protection part below. it could be

1 shows an artificial intelligence-based reservoir water level control system according to the present invention.
2 is a diagram illustrating a computing device according to the present invention.
3 is a diagram illustrating a memory including a functional configuration according to the present invention.
4 shows a process of generating state information according to the present invention.
5 shows a process of generating prediction data according to the present invention.
6 shows a process of generating command data according to the present invention according to the present invention.
7 schematically shows a water level meter according to the present invention.
8 shows a prediction data verification process according to the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are included as a part of the detailed description to help the understanding of the present specification, provide embodiments of the present specification, and together with the detailed description, explain the technical features of the present specification.

Hereinafter, the embodiments disclosed in the present specification will be described in detail with reference to the accompanying drawings, but the same or similar components are assigned the same reference numbers regardless of reference numerals, and redundant description thereof will be omitted.

In addition, in describing the embodiments disclosed in the present specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in this specification, the detailed description thereof will be omitted.

In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in the present specification, and the technical idea disclosed herein is not limited by the accompanying drawings, and all changes included in the spirit and scope of the present invention , should be understood to include equivalents or substitutes.

Terms including an ordinal number such as 1st, 2nd, etc. may be used to describe various elements, but the elements are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

The singular expression includes the plural expression unless the context clearly dictates otherwise.

In the present application, terms such as "comprises" or "have" are intended to designate that the external features, numbers, steps, operations, components, parts, or combinations thereof of the protective part exist in the specification, one or It should be understood that this does not preclude the possibility of addition or existence of other features or numbers, steps, operations, components, parts, or combinations thereof.

Hereinafter, an artificial intelligence-based water level control system according to a preferred embodiment of the present specification will be described in detail based on the above-mentioned contents.

1 shows an artificial intelligence-based reservoir water level control system according to the present invention.

Referring to FIG. 1 , the artificial intelligence-based reservoir water level control system according to the present invention may include a water level gauge 100 , a computing device 200 , and a pump control unit 300 . The above-described components may be interconnected through wired or wireless communication to transmit/receive data.

The water level meter 100 according to the present invention may measure water level data of the reservoir. In addition, the water level gauge 100 according to the present invention may measure water level data of water tanks other than the reservoir. The water level meter 100 according to the present invention may include an ultrasonic level meter, a pressure sensing level meter, and the like.

The computing device 200 according to the present invention may refer to any device for performing an operation based on data received from other components. The computing device 200 according to the present invention may mean various user terminals such as a smart phone, a notebook computer, or a computer, or may mean a server, a clouding computer, and the like.

The computing device 200 according to the present invention may generate prediction data for the water level of the reservoir based on the water level data. The prediction data may refer to data on the water level currently predicted through the learning of the conventional water level data and the learning result.

Also, the computing device 200 according to the present invention may generate state information of the water level meter 100 based on water level data received from the water level meter 100 . The state information may refer to information on a normal state or an abnormal state of the water level meter 100 . The abnormal state of the water level gauge 100 may include (1) a case in which the water level gauge 100 stops operating, (2) a case in which communication between the water level gauge 100 and other components is interrupted. The normal state of the water level gauge 100 may mean all cases other than the above-described abnormal state.

The pump control unit 300 according to the present invention may control the installed pump to adjust the water level of the reservoir. A pump may refer to any device that moves water in a reservoir based on power. The pump control unit 300 may receive a command from the computing device 200 and control the pump based on the received command. Control of the pump may include not only on/off of the pump, but also control of the intensity of the pump.

The pump control unit 300 according to the present invention may receive the state information generated by the computing device 200 and control the pump based on the state information. The pump control unit 300 according to the present invention may receive water level data and prediction data from the water level gauge 100 or the computing device 200 . The pump control unit 300 according to the present invention may control the pump according to water level data or predicted data, based on state information.

The water level meter 100 according to the present invention may periodically measure the water level data based on a predetermined unit period, and the computing device 200 according to the present invention extracts a predetermined number of measured values from the water level data, Status information can be generated based on the measured value. A specific method of generating the state information will be described later.

When the water level gauge 100 is in an abnormal state, the pump control unit 300 according to the present invention may control the pump according to prediction data. Also, when the water level gauge 100 is in a normal state, the pump control unit 300 may control the pump according to the water level data. That is, the pump control unit 300 may control the pump based on different data according to the state of the water level gauge 100 . Also, when the pump control unit 300 operates according to a command of the computing device 200 , based on which data the pump is to be controlled may be determined by the computing device 200 .

The computing device 200 according to the present invention may generate demand forecast data for water stored in a reservoir based on the LSTM model. Also, the computing device 200 according to the present invention may generate forecast data based on the generated demand forecast data.

Specifically, the computing device 200 according to the present invention generates water usage information for the hourly decrease amount of water stored in the reservoir based on the water level data, and from the pump control unit 300 to on/off the pump per hour. You can create pump operation information for

Therefore, the computing device 200 according to the present invention learns the training data set by inputting the training data set for the LSTM (Long Short-Term Memory) model, and results in demand forecast data based on the learning result. It can be created as a value. That is, the learning data set may include water level data, water usage information, and pump operation information. In addition, the water level data may mean information about the depth of water stored in the reservoir, the water usage information may mean information about the amount of water used based on the amount of change in depth per hour, and the pump operation information is the pump control unit 300 ) may mean information about on/off per hour of the pump. That is, the computing device 200 according to the present invention generates a learning data set using water level data, water usage information, and pump operation information as input values, and learns the generated learning data set to obtain demand forecast data. It can be created as a value.

However, when the water level gauge 100 is in an abnormal state, the computing device 200 according to the present invention may generate prediction data based on the water level data prior to the first time. In this case, the first time may include a time at which a measurement value, which is the basis of the abnormal state, is measured. That is, the computing device 200 according to the present invention may generate prediction data based on the measurement value before the abnormal state occurs.

Also, the computing device 200 according to the present invention may verify the prediction data based on the water usage information.

2 is a diagram illustrating a computing device according to the present invention.

Referring to FIG. 2 , a computing device 200 according to the present invention may include a processor 210 , a memory 220 , and a communication module 230 .

The processor 210 may execute instructions stored in the memory 220 to control other configurations. The processor 210 may execute an instruction stored in the memory 220 .

The processor 210 is a component capable of performing calculations and controlling other devices. Mainly, it may mean a central processing unit (CPU), an application processor (AP), a graphics processing unit (GPU), or the like. In addition, the CPU, AP, or GPU may include one or more cores therein, and the CPU, AP, or GPU may operate using an operating voltage and a clock signal. However, a CPU or AP may consist of a few cores optimized for serial processing, whereas a GPU may consist of thousands of smaller and more efficient cores designed for parallel processing.

The processor 210 may provide or process appropriate information or functions to the user by processing signals, data, information, etc. input or output through the above-described components or by driving an application program stored in the memory 220 .

The memory 220 stores data supporting various functions of the computing device 200 . The memory 220 may store a plurality of application programs (or applications) driven in the computing device 200 , data for operation of the computing device 200 , and instructions. At least some of these application programs may be downloaded from the external computing device 200 through wireless communication. Also, the application program may be stored in the memory 220 , installed in the computing device 200 , and driven by the processor 210 to perform an operation (or function) of the computing device 200 .

Memory 220 is a flash memory type (flash memory type), hard disk type (hard disk type), SSD type (Solid State Disk type), SDD type (Silicon Disk Drive type), multimedia card micro type (multimedia card micro type) ), card-type memory (such as SD or XD memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read (EEPROM) -only memory), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk may include at least one type of storage medium. Also, the memory 220 may include a web storage that performs a storage function on the Internet.

The communication module 220 transmits/receives information to and from a base station or a camera including a communication function through an antenna. The communication module 220 may include a modulator, a demodulator, a signal processor, and the like.

Wireless communication may refer to communication using a communication facility installed by telecommunication companies and a wireless communication network using a frequency of the communication facility. At this time, the communication module 220 is CDMA (code division multiple access), FDMA (frequency division multiple access), TDMA (time division multiple access), OFDMA (orthogonal frequency division multiple access), SC-FDMA (single carrier frequency division multiple) access), and the like, may be used in various wireless communication systems, and the communication module 220 may also be used in 3rd generation partnership project (3GPP), long term evolution (LTE), and the like. In addition, not only 5G communication that is being commercialized recently, but also 6G, which is scheduled to be commercialized in the future, may be used. However, in the present specification, a pre-installed communication network may be utilized without being limited by such a wireless communication method.

In addition, short range communication technology, Bluetooth (Bluetooth), BLE (Bluetooth Low Energy), beacon (Beacon), RFID (Radio Frequency Identification), NFC (Near Field Communication), infrared communication (Infrared Data Association; IrDA) ), Ultra Wideband (UWB), ZigBee, etc. may be used.

3 is a diagram illustrating a memory including a functional configuration according to the present invention.

Referring to FIG. 3 , the memory 220 according to the present invention may include a data collection module 221 , an abnormal situation detection module 222 , a data prediction module 223 , and a data output module 224 . In addition, the memory 220 according to the present invention may further include an LSTM model generation module 225 . In addition, the memory 220 according to the present invention may further include a prediction data verification module 226 .

The data collection module 221 according to the present invention may receive the measured water level data from the water level meter ( 100 in FIG. 1 ). Also, the data collection module 221 may store the measured water level data in time series. That is, the data collection module 221 may receive and store water level data according to time.

The abnormal situation detection module 222 according to the present invention may generate state information of the water level gauge ( 100 in FIG. 1 ). The abnormal situation detection module 222 may generate information on the abnormal state of the water level gauge ( 100 of FIG. 1 ) according to FIG. 4 to be described later.

The data prediction module 223 according to the present invention may input and learn water level data into the deep learning model. The data prediction module 223 may generate prediction data based on a learning result through the deep learning model. In this case, the deep learning model according to the present invention may be composed of a recurrent neural network (RNN). Specifically, the recurrent neural network may predict the water level at a specific point in time based on the water level data according to time series, and in this case, the predicted value may be referred to as prediction data. However, since the conventional recurrent neural network has a characteristic that the gradient decreases during backpropagation as the distance in the time series of related information increases, the data prediction module 223 according to the present invention may use a Long Short Term Memory Network (LSTM) model.

Specifically, the prediction data according to the present invention may be generated based on the demand forecast data for water stored in the reservoir based on the LSTM model, and the corresponding generation process may be performed in the data prediction model. The data prediction module 223 may generate prediction data according to FIG. 5 to be described later.

The data output module 224 according to the present invention may generate command data to be transmitted to the pump control unit ( 300 in FIG. 1 ) according to whether the received data is predicted data or water level data. That is, the data output module 224 may generate and output command data capable of controlling the pump control unit ( 300 of FIG. 1 ) according to the data.

The LSTM model generation module 225 according to the present invention may configure the training data set to train the LSTM model. The training data set is data for generating prediction data, and as described above, may include water level data, water usage information, and pump operation information.

The prediction data verification module 226 according to the present invention may verify the prediction data based on the water usage information. That is, the prediction data verification module 226 may verify the prediction data by comparing the prediction data according to the learning result of the LSTM model with the water level data before the first time. In this case, the first time may include a time at which a measurement value, which is the basis of the abnormal state, is measured. Specifically, the first time may mean a time initially measured among a plurality of measured values serving as the basis of the abnormal state. Hereinafter, a detailed verification process will be described later with reference to FIG. 8 .

4 shows a process of generating state information according to the present invention. The status information may be generated by the above-described computing device (or the abnormal situation detection module 222 ).

The computing device ( 200 in FIG. 1 ) according to the present invention may receive water level data from the water level gauge ( 100 in FIG. 1 ) ( S110 ). The computing device ( 200 of FIG. 1 ) may generate the 10 most recent measurement values among the water level data as a time series list ( S120 ). When the computing device ( 200 in FIG. 1 ) according to the present invention has at least two zero values in the generated time series list (first condition), it is determined that there is an abnormality in communication of the water level gauge ( 100 in FIG. 1 ). It can be done (S130, S160). In addition, when the difference between the maximum value and the minimum value in the time series list is greater than 50 (second condition), the computing device ( 200 in FIG. 1 ) according to the present invention has an abnormal state in communication of the water level gauge ( 100 in FIG. 1 ) It can be determined as (S140, S160). In addition, when the value in the time series list does not satisfy both the first condition and the second condition (the third condition), the computing device ( 200 in FIG. 1 ) according to the present invention communicates with the water level gauge ( 100 in FIG. 1 ). It can be determined as a normal state with no abnormality (S130, S140, S150).

5 shows a process of generating prediction data according to the present invention. The prediction data may be generated by the above-described computing device (or the data prediction module 223 ).

According to FIG. 5 , when it is determined that the water level gauge ( 100 in FIG. 1 ) is in an abnormal state ( S210 ), the computing device ( 200 in FIG. 1 ) according to the present invention is the data stored in the data collection module 221 at the abnormal state time point. A data set may be generated by extracting the most recent 14,400 pieces of water level data measured at a previous point in time (S220). The corresponding data set may mean water level data for a total of 14,400 minutes.

The computing device ( 200 in FIG. 1 ) according to the present invention may minimize the training time of the LSTM model by resizing the generated data set ( S230 ). The computing device ( 200 in FIG. 1 ) according to the present invention may derive an average of the water level data for 14,400 minutes at a 10-minute period, and train the LSTM model with the derived average value. That is, the computing device ( 200 in FIG. 1 ) according to the present invention may generate prediction data by inputting the resized data into the LSTM model ( S250 ).

As an example, 200 pieces of prediction data are generated as an example of the step (S240) of forming the prediction data by inputting the minimized data into the LSTM model, which is data minimized in units of 10 minutes and must be restored to the original form every minute. When the original restoration process is completed, 2,000 pieces of prediction data are generated, which generates approximately 33 hours of prediction data.

As an example, as a result of the learning, a rule that the change of water level data is relatively small on holidays and that the change of water level data is relatively large on weekdays may be learned. In addition, as a result of the learning, a rule that the change in water level data is relatively larger in the afternoon than in the morning may be learned. Based on these rules, the LSTM model can generate predictive data.

The computing device ( 200 in FIG. 1 ) according to the present invention may transmit the generated prediction data to the pump control unit ( 300 in FIG. 1 ) through the data output module 224 .

6 shows a process of generating command data according to the present invention according to the present invention. The command data may be generated by the above-described computing device (or data output module 224 ).

According to FIG. 6 , the computing device ( 200 in FIG. 1 ) according to the present invention generates command data based on which data of prediction data or water level data ( S310 , S320 ), the pump control unit ( 300 in FIG. 1 ) It is possible to determine whether to transmit the first command data (cnt=cnt+1) or the second command data (cnt=0) with respect to ( S330 , S350 , S360 ).

Referring to FIG. 6 , as an example, the computing device ( 200 in FIG. 1 ) according to the present invention uses a variable called cnt as an Index, and when received data is prediction data, sets the cnt-th value of the prediction data as output data. After that, the cnt value is increased to output the next prediction data at the next output. The computing device (200 of FIG. 1 ) according to the present invention, when an abnormal state is converted to a normal state, such as when communication is normalized thereafter, and when normal water level data is continuously input, the cnt variable is not used anymore without using the prediction data. It can be initialized to 0 so that it can be used as an Index in the next communication abnormal situation (S340, S360).

7 schematically shows a water level meter according to the present invention.

Referring to FIG. 7 , the water level meter 100 according to the present invention may be a water level meter 100 device for measuring the water level by pressure. The water level gauge 100 according to the present invention includes a pressure sensor 120 for detecting pressure, a pressure-level conversion module 140 for converting the sensed pressure data into water level data, and a computing device (Fig. 1 of 200) may include a data transmission interface 150 for transmission. In addition, the water level gauge 100 according to the present invention may further include a pressure detection sensor 120 , a pressure-level conversion module 140 , and a sealed housing 110 for including the data transmission interface 150 therein. have.

At least a portion of the pressure sensor 120 according to the present invention may be exposed to the outside of the housing or may be in direct contact with one surface of the housing in order to detect water pressure. Therefore, it may be corroded by permeating moisture or exposed to external contaminants. Accordingly, for the pressure sensor 120 resistant to moisture, the water level gauge 100 according to the present invention may further include a coating layer 130 for coating the pressure sensor 120 .

Accordingly, the present invention is to solve this problem by proposing the coating layer 130 for coating the pressure sensor (hereinafter, the sensor, 120).

Preferably, the sensor 120 includes an acrylic compound represented by the following Chemical Formula 1 on its surface; It may be coated with a coating composition containing an organic solvent, inorganic particles and a dispersing agent.

[Formula 1]

When the sensor 120 is coated with the coating composition, excellent water repellency and stain resistance can be exhibited, so that even if the sensor 120 installed in an external environment is exposed to a polluted environment and salt for a long time, an accurate pressure value can be sensed.

The inorganic particles may be selected from the group consisting of silica, alumina, and mixtures thereof. The average diameter of the inorganic particles is 70 to 100 μm, but is not limited to the above example. After the inorganic particles are formed as a coating layer 130 on the surface of the sensor 120 , physical strength may be improved, and the viscosity may be maintained within a certain range to increase moldability.

The organic solvent is selected from the group consisting of methyl ethyl ketone (MEK), toluene, and mixtures thereof, and preferably methyl ethyl ketone may be used, but is not limited thereto.

As the dispersant, a polyester-based dispersant may be used, and specifically, as a polyester-based dispersion stabilizer composed of a copolymer of 2-methoxypropyl acetate and 1-methoxy-2-propyl acetate, TEGO-Disperse 670 (manufacturer : EVONIK), but is not limited to the above examples and any dispersant obvious to those skilled in the art can be used without limitation.

The coating composition may further include a stabilizer as other additives, and the stabilizer may include a UV absorber, an antioxidant, and the like, but is not limited to the above examples and may be used without limitation.

For forming the coating layer 130, the coating composition may include an acrylic compound represented by Chemical Formula 1; organic solvents, inorganic particles and dispersants.

The coating composition may include 40 to 60 parts by weight of the acrylic compound represented by Formula 1, 20 to 40 parts by weight of inorganic particles, and 5 to 15 parts by weight of a dispersant based on 100 parts by weight of the organic solvent. In the case of the above range, a synergistic effect is expressed to the extent that the water repellency effect due to the interaction of each component has a critical significance, and when it is out of the above range, the synergistic effect is rapidly reduced or almost absent.

More preferably, the viscosity of the coating composition is 1500 to 1800 cP, and when the viscosity is less than 1500 cP, when applied to the surface of the sensor 120, there is a problem that the formation of the coating layer 130 is not easy because it flows down, and exceeds 1800 cP In this case, there is a problem in that the formation of a uniform coating layer 130 is not easy.

[Preparation Example 1: Preparation of coating layer]

1. Preparation of coating composition

A coating composition was prepared by mixing the acrylic compound represented by the following Chemical Formula 1, inorganic particles and a dispersing agent in methyl ethyl ketone:

[Formula 1]

A more specific composition of the coating composition is shown in Table 2 below.

TX1 TX2 TX3 TX4 TX5 organic solvent 100 100 100 100 100 acrylic compound 30 40 50 60 70 inorganic particles 10 20 30 40 50 dispersant One 5 10 15 20

(unit parts by weight) 2. Preparation of coating layer

The coating composition of DX1 to DX5 was applied to one surface of the sensor 120 and cured to form a coating layer 130 .

[Experimental example]

1. Evaluation of surface appearance

Due to the difference in viscosity of the coating composition, after the coating layer 130 was prepared, a sensory evaluation was performed as to whether a uniform surface was formed. Whether or not a uniform coating layer 130 was formed was evaluated, and evaluation was performed according to the following criteria.

○: uniform coating layer formation

×: Formation of a non-uniform coating layer

TX1 TX2 TX3 TX4 TX5 sensory evaluation Х ○ ○ ○ Х

When the coating layer 130 is formed, if the viscosity is less than a certain level, a flow occurs on the surface of the sensor 120 , and after the curing process, it is difficult to form the uniform coating layer 130 in many cases. Accordingly, there may be a problem in that the production yield is lowered. In addition, even when the viscosity is too high, it is difficult to uniformly apply the composition to form the uniform coating layer 130 .

2. Measurement of water repellency angle

After forming the coating layer 130 on the surface of the sensor 120, the results of measuring the water repellency angle are shown in Table 4 below.

Forward contact angle (“) Stop contact angle (“) receding contact angle (“) TX1 117.1±2.9 112.1±4.1 < 10 TX2 132.4±1.5 131.5±2.7 141.7±3.4 TX3 138.9±3.0 138.9±2.7 139.8±3.7 TX4 136.9±2.0 135.6±2.6 140.4±3.4 TX5 116.9±0.7 115.4±3.0 < 10

As shown in Table 4, after forming the coating layer 130 using the coating composition of TX1 to TX5, the result of measuring the contact angle was confirmed. For TX1 and TX5, the receding contact angle was measured to be less than 10 degrees. That is, when it is out of the optimal range for preparing the coating composition, it was confirmed that the phenomenon of pinning water droplets occurs. On the other hand, it was confirmed that the peening phenomenon did not occur in TX2 to 4, thereby exhibiting an excellent waterproof effect.

3. Pollution resistance evaluation

The sensor 120 having the coating layer 130 formed according to the above embodiment was attached to a reservoir or water tank, and exposed to a humid environment for 40 days. As the comparative example (Con), the same sensor 120 on which the coating layer 130 was not formed was used, and each sensor 120 was attached to the same position in a reservoir or a water tank.

Thereafter, the degree of contamination and corrosion of the sensor 120 before and after the experiment was evaluated as related, and for objective comparison, the result was evaluated as an index of 1 to 10 in comparison with the comparative example in which the coating layer 130 was not formed. Table 5 shows. In the following index, the lower the number, the better the stain resistance.

Con TX1 TX2 TX3 TX4 TX5 stain resistance 10 7 3 3 3 8

(Unit: Index) Referring to Table 5, when the coating layer 130 is formed on the sensor 120, while the sensor 120 is installed in an external environment, the sensor 120 has high pollution resistance even when exposed to the outside. can check that Accordingly, the sensor 120 on which the coating layer 130 is formed according to the present invention can be used for a long period of time and can collect a more accurate pressure value. In particular, in the case of TX2 to TX4, it can be seen that the stain resistance by the coating layer 130 is very excellent.

8 shows a prediction data verification process according to the present invention.

The prediction data verification module 226 according to the present invention may generate K clusters and classify data by determining similarity based on the distance values of the generated clusters.

The K-centric clustering process refers to clustering the given data into K clusters. In this K-centric clustering process, one sample value is randomly selected from a given data sample, and the distance from the sample value to other sample data is measured. At this time, the closest center of each data sample is selected, and the distance from the corresponding sample to the other data sample values is calculated again. By repeating this process, clustering is achieved. That is, a plurality of clusters are created. This clustering technique can be mainly implemented in Python. Clustering is being used as one of the algorithms that can be used in signal analysis.

Referring to FIG. 8 , as a result of performing the K-centered clustering process, it is shown that the measured values are classified into clustering 1 and clustering 2 . In this case, the data to be clustered may be values obtained by coordinating prediction data according to a learning result of the LSTM model and water level data before the first time according to a reference time.

That is, the prediction data verification module 226 according to the present invention coordinates the time (minutes) and water level (cm) extracted from the prediction data and the water level data before the first time, and performs clustering on the coordinated data. can be done

In this case, the reference time is a time determined according to a predetermined standard, and may mean a reference time for comparing or verifying different data. For example, the reference time according to the present invention may mean a time when measurement is started or a specific time on a specific date arbitrarily set.

The prediction data verification module 226 according to the present invention calculates the first central value of clustering 1 (clustering for prediction data) through an average calculation method, and the second of clustering 2 (clustering for water level data before the first time) 2 The central value can be calculated through the average calculation method. The prediction data verification module 226 according to the present invention calculates a final distance value between the first central value and the second central value, and when the size of the final distance value is smaller than a predetermined value, the similarity of each cluster is determined in advance. It can be judged to be higher than the value. That is, when the size of the final distance value is smaller than the predetermined value, since both data may be determined to be similar, the predicted data may be determined to be verified.

Preferably, the computing device (200 in FIG. 1 ) according to the present invention performs a K-centric clustering process on the prediction data and the water level data before the first time, and finally, based on the K-centric clustering process, two A cluster is generated, and a degree of similarity can be derived based on the Euclidean distance between the two generated clusters.

According to FIG. 8 , as an example, the prediction data verification module 226 according to the present invention sets the magnitude (min) of the time change value according to the reference time to Feature 1, and sets the size (cm) according to the water level to Feature 2 It can be set and coordinated.

The prediction data verification module 226 according to the present invention may perform the K-centered clustering process by using feature 1 as an x-axis coordinate and feature 2 as a y-axis coordinate. The prediction data verification module 226 according to the present invention may calculate the first central value of the cluster 1 and the second central value of the cluster 2, and may calculate the Euclidean distance between the first central value and the second central value. The Euclidean distance may be defined by Equation 1 below.

[Equation 1]

(However, d is a distance value, (x, y) is a coordinate value, x corresponds to time (imn), and y corresponds to water level (cm))

The prediction data verification module 226 according to the present invention may extract the final distance value of the cluster 1 and the cluster 2 and automatically verify the prediction data based on the final distance value.

In this case, the final distance value according to the present invention may be extracted in the following order.

(1) Extract the first average value (x1, y1) of cluster 1 and the second average value (x2, y2) of cluster 2

(2) The first average value and the second average value are coordinate values, respectively, and the final distance value (L) between the first average value and the second average value is extracted based on Equation 2 below

[Equation 2]

(However, (x1, y1) is a coordinate value generated based on the average value of cluster 1 (= first central value), and (x2, y2) is generated based on the average value of cluster 2 (= second central value) coordinate value)

In the present invention, when the extracted final distance value L is smaller than a preset distance value, it can be defined that the prediction data is verified.

In general, the water level may be measured differently depending on the shape and depth of the reservoir, as well as external environmental factors such as atmospheric pressure. Therefore, rather than obtaining a uniform final distance value, in order to consider other factors, it is more efficient to multiply the final distance value L by a correction factor and compare it with a preset distance value based on this. This is referred to as the corrected final distance value L' and may be extracted based on Equations 3 and 4 below.

[Equation 3]

(However, (x1, y1) is a coordinate value generated based on the average value of cluster 1, and (x2, y2) is a coordinate value generated based on the average value of cluster 2)

[Equation 4]

In this case, the first Euclidean distance may be a Euclidean distance between the origin serving as a coordinate-based reference point and the first central point, and the second Euclidean distance may be a Euclidean distance between the origin serving as a coordinate-based reference point and the second central point.

The first cross-section of the drain paper may be a cross-section cut in a direction parallel to the ground, and the second cross-section of the drain paper may mean a cross-section cut in a direction perpendicular to the ground.

The average water temperature change amount is a change amount of the temperature of water stored in the reservoir, and may mean an average of water temperature values during a time period when data constituting clusters 1 and 2 are measured.

In addition, α is a correction constant for more accurate determination of an abnormal state, and the unit of α can be neglected.

As such, when the fire occurrence information is generated based on the corrected final distance value (L') using Equations 3 and 4, there is an effect that the predicted data is more accurately verified. same.

[Experimental example]

The results of comparing the accuracy of the case in which the prediction data was verified based on the Euclidean distance generated using the correction constant according to the present invention and the case in which the prediction data was verified based on the Euclidean distance that did not use the correction constant are shown in the table below equal to 1.

In this case, the accuracy is calculated for 10 experts in the relevant field, and may be a numerical value calculated by asking 10 experts for a total of 100 cases in the same environment. The accuracy is quantified by determining the sameness between the predicted data sensed by the system according to the present invention and the actual water level data.

Correction factor α not used Use correction factor α accuracy 87 98

(Unit: %) Table 6 shows the accuracy of the cases according to the present invention. As a result of the experiment, it can be seen that the case in which the correction coefficient α is used has significantly higher accuracy than the case in which the correction coefficient α is not used.

The present invention described above can be modeled as computer-readable code on a medium in which a program is recorded. The computer-readable medium includes all types of recording devices in which data readable by a computer system is stored. Examples of computer-readable media include Hard Disk Drive (HDD), Solid State Disk (SSD), Silicon Disk Drive (SDD), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. It also includes modeling in the form of a carrier wave (eg, transmission over the Internet). Accordingly, the above detailed description should not be construed as restrictive in all respects but as exemplary. The scope of the present specification should be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the present specification are included in the scope of this specification.

Any or other embodiments of the present invention described above are not mutually exclusive or distinct. Any of the above-described embodiments or other embodiments of the present invention may be combined or combined with each configuration or function.

The above detailed description should not be construed as restrictive in all respects and should be considered as illustrative. The scope of the present invention should be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the present invention are included in the scope of the present invention.

100: water level gauge
200: computing device
300: pump control unit
400: pump

Claims (7)

a water level meter that measures water level data in the reservoir;
a computing device that generates prediction data for the water level of the reservoir based on the water level data, and generates state information of the water level gauge based on the water level data; and
On the basis of the state information, the pump control unit for controlling the pump according to the water level data or the prediction data; Containing,
The computing device,
generating demand forecast data for water stored in the reservoir based on the LSTM model, generating the forecast data based on the demand forecast data,
The computing device,
Generates water usage information on the hourly decrease amount of water stored in the reservoir based on the water level data, and generates pump operation information on hourly on/off of the pump from the pump control unit,
The computing device,
For the Long Short-Term Memory (LSTM) model, a training data set is generated using the water level data, the water usage information, and the pump operation information as input values, and the demand forecast by learning the training data set Generate data as a result,
The computing device,
When the water level gauge is in an abnormal state, the prediction data is generated based on the water level data before the first time,
The first time includes a time at which a measurement value that is the basis of the abnormal state is measured,
The computing device,
Based on the prediction data and the water level data before the first time, verifying the prediction data through a K-centered clustering process, an artificial intelligence-based reservoir water level control system.
According to claim 1,
The water level gauge is
Periodically measuring the water level data based on a predetermined unit period,
The computing device,
Extracting a predetermined number of measured values from the water level data, and generating the state information based on the measured values,
Artificial intelligence-based reservoir water level control system.
According to claim 1,
The pump control unit,
When the water level gauge is in an abnormal state, the pump is controlled according to the prediction data, and when the water level gauge is in a normal state, the pump is controlled according to the water level data,
Artificial intelligence-based reservoir water level control system.
According to claim 1,
The computing device,
Finally, two clusters are generated based on the K-centric clustering process, and the two clusters include cluster 1 and cluster 2,
verifying the prediction data based on the final distance value (L') between the cluster 1 and the cluster 2;
The final distance value (L') is,
which is extracted based on the first average value of the cluster 1 and the second average value of the cluster 2,
Artificial intelligence-based reservoir water level control system.
5. The method of claim 4,
The final distance value (L ') will be extracted by the following Equations 3 and 4,
Artificial intelligence-based reservoir water level control system.
[Equation 3]

(However, (x1, y1) is a coordinate value generated based on the first average value, and (x2, y2) is a coordinate value generated based on the second average value)
[Equation 4]

(In this case, the first Euclidean distance is the Euclidean distance between the origin and the first center point as the coordinate-based reference point, the second Euclidean distance is the Euclidean distance between the origin and the second center point as the coordinate-based reference point, and the first center point is The center point of the cluster 1, the second center point is the center point of the cluster 2, the first cross-section of the drain paper is a cross-section cut in a direction parallel to the ground, and the second cross-section of the drain paper is a direction perpendicular to the ground It is a cross-section cut by
According to claim 1,
The water level gauge includes a pressure sensor for detecting the pressure,
The pressure sensor is coated with a coating composition comprising an acrylic compound represented by the following formula (1), an organic solvent, inorganic particles and a dispersant,
Artificial intelligence-based reservoir water level control system.
[Formula 1]

According to claim 1,
The data to be clustered in the K-centered clustering process are,
The prediction data and the water level data before the first time is data coordinated according to a reference time,
Artificial intelligence-based reservoir water level control system.

Images