KR100983800B1 - Simplicity waterworks monitoring control system and the method of using Neural-Network - Google Patents

Abstract

The present invention relates to a liver abnormality monitoring control system using a neural network, and a control method thereof, and more particularly, a control device having a motor pump for pumping raw water, a water supply pipe for supplying raw water pumped by the control device, In a simple water supply monitoring control system including a water tank for storing the raw water supplied to the water pipe and a disinfectant storage tank for disinfecting liquid into the water tank, the indoor and outdoor temperature of the well and the pressure, flow rate, flow rate of the water pipe And sensor means for measuring the current value of the independent variable variable including the disinfectant input amount of the disinfectant storage tank; Collecting data of the independent variable current value detected from the sensor means, arithmetic processing into physical quantity data, and comparing the current value of the measured characteristic by a neural network control program using a learning optimization algorithm, the normal interval of the abnormal frequency Computer means for determining an inspection section (available) and an unusable section, and adjusting the independent variable values according to the determined result to perform arithmetic processing on analog and digital control output values; It is a technical gist of the control device comprising a control means for controlling the control device, the water supply pipe and the disinfectant storage tank as the control output value.

Neural Network, Liver Number, Monitoring, Learning Optimization, Algorithm, Independent Variable

Description

Simplicity waterworks monitoring control system and the method of using Neural-Network

The present invention relates to a liver abnormality monitoring control system using a neural network and a control method thereof. More particularly, the liver abnormality monitoring is performed in real time with accurate determination by applying a neural network to determine the water supply situation of liver abnormalities. A control system and a control method thereof.

In Korea, granite is widely distributed in geology, so you can collect and eat water wherever it is not contaminated. Doing.

However, in a situation where the environmental and air pollution is intensifying, simple tap water is used in places where water supply such as mountain wallpaper is not spread.

In general, a simple tap water is used to draw groundwater into hills or mountains in a remote village or place where supply lines cannot be connected to the water supply line. Before being supplied, it is required to go through a disinfection process to sterilize various bacteria and E. coli in the water.

Therefore, in simple tap water, the strong oxidizing action of generator oxygen produced by decomposing residual chlorine by dissolving residual chlorine (HClO) by producing some residual chlorine (HClO) by reacting with water by using chlorine which has excellent sterilizing power and is easy to purchase and inexpensive. The chlorine disinfection method to disinfect water is generally applied.

However, in the case of a simple water supply facility, workers have been directly injecting a disinfectant solution regularly into a water supply tank, but the waste of manpower and management costs are high. As it was not possible to maintain constant water by relying on experience, not only was it impossible to manage sanitary water, but also had to be disinfected with disinfectant during disinfection. In addition, due to its characteristics, it is difficult to disinfect a water supply system because of its characteristics.

An example of a chlorine disinfection method currently applied to solve such inconvenience is shown in FIG. 1.

As shown in FIG. 1, the conventional chlorine injecting device has a method in which the chlorine dilution water from the chemical tank 100 is quantitatively injected into the water storage tank 3 as the main motor of the well 1 is operated. If the water supply is large because the concentration can not be adjusted, the disinfection effect is reduced due to the lack of chlorine concentration, there is a problem that the excess of chlorine concentration occurs when the water supply is small.

In addition, since the automatic chlorine injector according to the current temperature or humidity has not been developed, there is a problem that the disinfection effect is lowered because the amount of chlorine that is added to the water in real time is difficult to become optimal.

Accordingly, the present invention has been made to solve the above-mentioned problems of the prior art, and derives the correlation using neural networks with existing data rather than relying on the driver's experience in the water supply situation of the abnormal abnormality. It is an object of the present invention to provide a liver abnormality monitoring control system and a control method using neural networks that allow the liver abnormality to be controlled in real time according to a situation.

In addition, the present invention provides a liver abnormality monitoring control system and a control method using a neural network to determine and control the water supply situation of liver abnormality by six kinds of pressure, indoor and outdoor temperature, flow rate, flow rate and disinfectant input amount. There is another purpose.

Liver abnormal water supply monitoring control system using a neural network according to the present invention for solving the above problems, and a test device having a motor pump for pumping raw water, a water pipe for supplying the raw water pumped by the test device, and A simple water supply monitoring control system including a water tank for storing raw water supplied to a water pipe and a disinfectant storage tank for disinfecting liquid into the water tank, the indoor and outdoor temperature of the well, the pressure, flow rate, flow rate and Sensor means for measuring a current value of the independent variable including the disinfectant input amount of the disinfectant storage tank; Collecting data of the independent variable current value detected from the sensor means, arithmetic processing into physical quantity data, and comparing the current value of the measured characteristic by a neural network control program using a learning optimization algorithm, the normal interval of the abnormal frequency Computer means for determining an inspection section (available) and an unusable section, and adjusting the independent variable values according to the determined result to perform arithmetic processing on analog and digital control output values; And control means for controlling the control device, the water supply pipe, and the disinfectant storage tank with the control output value.

In addition, liver abnormality monitoring control method using a neural network according to another preferred embodiment of the present invention, the temperature and flow rate of the pipe, the pressure, flow rate, flow rate of the water pipe and the disinfectant solution input amount of the disinfectant storage tank Measuring a current value of the variable independent variable; Collecting data of the measured independent variable current value by a computer and performing arithmetic processing on physical quantity data; Comparing the current value of the measured characteristic according to a neural network control program using a backpropagation algorithm in the computer, and determining the normal section, the check section (usable) and the unusable section of the liver abnormality frequency; Adjusting the independent variable values of the pressure, flow rate, flow rate, and disinfection amount of the independent variables according to the determined result and calculating them as analog and digital control output values; And controlling the fixed pump of the motor pump, the solenoid valve of the water pipe, and the disinfecting liquid storage tank of the well as the obtained control output value.

According to the present invention constituted as described above, the nonlinear learning ability of neural networks can appropriately cope with the nonlinear characteristics of liver and liver water, and the pressure, indoor and outdoor temperature, flow rate, flow rate, and disinfectant are changed in real time through fast response and robustness. Active and automatic response to changing conditions such as the input amount is possible, and at the same time, it is effective to provide clean water with the correct amount of chlorine input.

In addition, a certain ratio of disinfectant solution can be added regardless of the characteristics of the influent, flow rate, etc., and it is necessary to establish a simple abnormal water monitoring control system that can artificially intelligentize and automate the input process and stabilize the efficiency, and to build a database and automate the neural network. The optimum target value for securing the constantness, stability of the sensor means 100, the stability of the automatic control system, etc., and maintaining the dosage of the disinfectant solution regardless of the temperature, water pressure, flow rate and flow rate of the four seasons is obtained.

In addition, according to the present invention, by adjusting each variable according to the learned situation to produce a situation, such as applying the operating range according to the actual situation of the abnormal number, there is an effect that can always supply the ideal state of the ideal number. .

In addition, according to the present invention there is an effect that can reduce the factors that increase the operating cost, such as labor costs, power costs due to the inefficiency of the operation management of the abnormal water supply.

Hereinafter, with reference to the accompanying drawings will be described in detail an embodiment of a liver abnormality monitoring control system and a control method using a neural network according to the present invention. However, terms to be described below are terms defined in consideration of functions in the present invention, and may be changed according to intentions or customs of users or operators. Therefore, the definition should be made based on the contents throughout the specification.

2 is a view showing an overview of the liver abnormality monitoring control system using a neural network according to the present invention, Figure 3 is a block diagram showing the overall configuration of FIG.

As shown in Figures 2 and 3, the liver abnormality monitoring monitoring system using the neural network according to the present invention, the well device 10 having a motor pump 12 for pumping raw water, and the well device 10 Water supply pipe 20 for supplying the raw water pumped by the water), the water tank 30 for storing the raw water supplied to the water pipe 20 and the disinfectant storage tank 40 for quantitatively disinfecting the disinfectant in the water tank (30) It includes a plurality of to measure a complex current value of the indoor and outdoor environmental characteristics of the well device 10, the water inflow characteristics of the water pipe 20, and the state characteristics of the disinfectant storage tank 40 It is obtained from the sensor means 100.

Here, the indoor and outdoor environmental characteristics of the crystal device 10 is an indoor temperature and an outdoor temperature, the water inflow characteristics of the water pipe 20 is pressure, flow rate and flow rate, the state characteristic of the disinfectant storage tank 40 is the disinfectant input amount to be. These are preferably sensed through a predetermined sensor and the data is transmitted to a programmable logic controller (PLC) which is a control means 300 after analog-to-digital conversion.

That is, a temperature sensor capable of measuring an indoor temperature and an outdoor temperature may be used inside or outside the well device 10. The indoor temperature detected by the temperature sensor is a temperature where the motor pump 12 of the well 10 is located, and the outdoor temperature is an atmospheric temperature outside the well 10.

In addition, the pressure of the water pipe 20 may be detected from a pressure sensor seated in the water pipe 20. For example, depending on the amount of water flowing in, the flow sensor provides a corresponding current value as an analog signal, and the provided analog signal is converted into a digital signal by the analog-digital converter 110 to be digitally processed. Is converted. The current value measured by the flow sensor converted into the digital signal is input to the control means 300 to be described later to determine whether the inflow of water, for example, if the measured value is 4mA or more, inflow of water through the water pipe 20 It is determined that there is, and if it is 4mA or less, it is determined that there is no inflow of water. Therefore, the pressure sensor can determine whether the motor pump 12 operates according to the pressure of the water pipe 20. In addition, the flow rate and flow rate of the water pipe 20 may be detected through a flow sensor and a speed sensor attached to the inlet of the water pipe 20.

In addition, the disinfectant storage tank 40 automatically supplies the disinfectant solution supplied to the water tank 30 using the metering pump 42 to measure the input amount of the sanitizing solution injected by the metering pump 42. .

These sensor means 100 may be any installation position as long as it does not interfere with the movement of water or disinfectant solution.

On the other hand, according to an embodiment of the present invention, the sensor means 100 is preferably configured to further include an analog-to-digital converter 110 and the periodic signal extraction unit 120.

The analog-digital converter 110 samples a predetermined analog signal and converts the analog signal into a digital signal. In this case, the sampling may be performed at a predetermined time or at a random time. In addition, the predetermined analog signal may mean a detected detection signal or a periodic signal which will be described later.

As shown, the periodic signal extractor 120 extracts the periodic signal from the digital signal. However, this is for convenience of description and the present invention is not limited thereto. The periodic signal extractor 120 may extract one or more periodic signals from the digital signal converted by the analog-digital converter 110. For convenience of explanation, it is assumed that the periodic signal extractor 120 extracts m periodic signals. M is a positive integer greater than one. That is, the converted digital signal or the detected detection signal may represent various periodic signals as time passes, and the periodic signal extractor 120 extracts m periodic signals. In the present invention, it is more preferable to extract the periodic signal by taking six sequential values in real time.

The data measured by the sensor means 100 is collected in the storage device 310 of the control means 300, the collected data is a neural network control program through the interface 350 in the central processing unit 320 It is transmitted to the interface 210 of the computer means 200, such as a PC having a, where the computer means 200 performs remote control and monitoring in both directions and in real time by the TCP / IP (250).

The control means 300 receives an analog signal or a digital signal from the computer means 200 and rotates the motor pump 12 of the tablet device 10 by the D / A card 330 and the digital card 340. And a role of controlling the solenoid valve 22 of the water supply pipe 20, and controlling a metering pump 42 of the disinfectant storage tank 40.

This data constitutes a monitoring control step using a neural network according to the flowchart shown in FIG. 4.

As shown in FIG. 4, when the program is started and initialized (S201), the existing data is collected (S202) and the data preprocessing step (S203) passes. The indoor and outdoor temperatures of the well 10 and the well 10 are defined. The raw water pumped by the motor pump 12 of) is input to the characteristics of the water transferred by the water pipe 20, that is, the pressure, flow rate, flow rate, and the like (S204) and goes through the step (S205) of the current state.

The determined current value is subjected to control output calculations (S207, S208) simultaneously with the optimum target value input (S206), converted into physical quantities, and then control outputs (S210, S211) as analog and digital signals, respectively, and at the same time control means. 300 is stored in the storage device 310 of the PLC (300) (S212).

In addition, through the RS232C communication means shown in FIG. 3 (S213), the Internet transmission control protocol / Internet protocol (hereinafter referred to as TCP / IP) 250 is used for bidirectional communication and computer means 200. Displayed on the screen 230 (S214).

The computer means 200 then links each data obtained from the previous step (temperature, pressure, flow rate, flow rate, disinfection state) with other five data independently determined by a neural network control program using a learning optimization algorithm. Finally, the status judgment (S215) is performed as "normal / check required / unusable". According to the determination result, the independent variable values of the pressure, flow rate, flow rate, and disinfectant amount of the independent variables are adjusted (S216), and the digital control signal processing (S217) reflecting these is performed.

5 shows a flowchart of the overall learning and application process according to the neural network control program using the learning optimization algorithm used in the present invention.

As shown in FIG. 5, after the neural network control program using the learning optimization algorithm is started and initialized, the indoor temperature and the outdoor temperature of the controller 10, the pressure, flow rate and flow rate of the water pipe 20 and In operation S401, data such as a disinfectant solution input amount of the disinfectant storage tank 40 is collected and acquired, and the data is analyzed based on the characteristics of each variable and a data rule is created (S402). In this case, the physical quantity data is subjected to arithmetic processing as independent data.

This data rule is learned by the neural network (S403), where the neural network uses the back-propagation algorithm, the Levenberg-Marquardt method, the Isaac Newton method, and the steepest descent method, as in Equation 1. A possible learning optimization algorithm is used, the operation efficiency after operation of the actual abnormal frequency is calculated (S404), and short-term tuning (S406) is performed until the desired efficiency is obtained by acquiring new data having high efficiency. . Through this re-learning process (S407, S409), a new database is constructed (S408).

Where ω _i ; i-th input data, ω _{i + 1} ; link weight data, and I stands for Identity Martrix. Λ = 0: Issac Newton method, λ → ∞: superb descent method , λ is dynamically adjusted.)

In the neural network control using the learning optimization algorithm, as an example of input learning data, the indoor temperature of the controller 10 is in the range of 0 to 40 ° C. and the outdoor temperature is minus 20 ° C. to 40 ° C., and the water pipe 20, the pressure is 0 ~ 35kgf / ㎠, flow rate is 0 ~ 10㎥ / d, flow rate 0 ~ 2.0m / sec, the user sets the flow rate, etc. to control the input amount of the disinfectant according to the set value.

The arithmetic formula for setting the disinfectant solution amount of the disinfectant storage tank 40 is as follows.

Flow rate (ℓ) × Set chemical concentration (PPM) / (Concentration of liquid product) = Input amount (ml) × 1000

Equation 2 as described above is calculated by calculating the flow rate of the raw water when the user inputs the desired set residual disinfectant concentration. At this time, the concentration of the disinfectant product is 6% (60,000PPM) or 12% (120,000PPM), because the existing liquid disinfectant solution is used in the concentration of 6% and 12%. The input amount in Equation 2 is the amount of the disinfectant that is input when the metering pump 42 of the disinfectant storage tank 40 is injected once to disinfect the solution, and the disinfectant input of the metering pump is the amount of the dispensing solution 42 The amount can be controlled by adjusting the driving time.

The learning data presented above is automatically controlled when the value of the outdoor temperature is changed or when the magnitude of the flow rate and the flow rate are changed even in the same indoor temperature range of the controller 10. For example, if the pressure or flow rate of the water pipe 20 increases, the speed of the motor pump 12 is decreased, and if the pressure or flow rate is decreased, the speed of the motor pump 12 is increased. Accordingly, when the flow rate of the water pipe 20 increases, the amount of disinfectant solution is increased, and when the flow rate is reduced, the amount of disinfectant solution is controlled to be reduced.

6 shows a scheme of the backpropagation algorithm according to the present invention.

The backpropagation algorithm preferably uses a multi-layer perceptron (MLP) model. In the MLP model, layers are input layers that are input from the environment, output layers that output to the environment, and hidden layers that exist between the input and output layers without directly interacting with the environment. Each layer has a structure that receives signals from the previous layer, processes them, and delivers them to the next layer.

As shown in Figs. 6A to 6F, X means neurons that form the entire layer. Xij refers to a weight indicating the degree of coupling of the i-th neuron of the previous layer and the j-th neuron of the next layer. In addition, Xti is a back propagation algorithm model for the room temperature, Xto is the outdoor temperature, Xp is the pressure, Xq is the flow rate, Xv is the flow rate, Xs is the disinfectant input amount.

Learning begins with the input layer and proceeds through the hidden layer to the output layer. The neuron sums the values of the inputs it accepts, and passes the inputs to the next node in the Hidden Layer if the sum is high. Weights are assigned as inputs are passed, or weighting and weakening of weights occur, and this process continues until the final output layer where the model predicts the result. In this case, activation occurs to add specific gravity to the high value pattern and ignore the low value pattern.

In the present invention, since the output layer must predict and output various target values, only the linear sum is used as an active function. In the hidden layer, a nonlinear activation function is used, and the neural network accumulates and stores information obtained from the learning process in weights, but this weight is also used to obtain an optimal disinfectant dose.

In the hidden layer of the present invention shown in Fig. 6, the number of hidden layers is determined according to the expert's empirical knowledge, but in the present invention, the number of hidden layers is preferably two. In this hidden layer, the data input, output, and weight of the input layer and the output layer are determined according to Equations 3 to 5.

(F is the formula that calculates the F that fits the total error as the rate of change of the weight, and Gradient i is the i-th weight.)

(error e is the difference between the target value and the neural network output, k is the kth sample)

(Hessian matrix, find the squared derivative of the weights of F)

As described above, each data (temperature, pressure, flow rate, flow rate, and disinfection state) obtained from the above-described steps is linked with other five data which are independently determined, and finally the output data is " normal section, inspection section (usable). , "Unavailable interval". Through the process of collecting the determinations independently, each data value is adjusted by adjusting six variables such as indoor temperature, outdoor temperature, pressure, flow rate, flow rate, and disinfectant input amount, and at this time, each independent value is adjusted according to the learned situation. The variable value is adjusted and digital control signal processing is performed.

After the above step, the control output is continuously calculated through the adjusted independent variable values, and transmitted as analog signals and digital signals using the D / A card 330 and the digital card 340, and newly transmitted signals. By the computer means 200, the motor pump 12, the solenoid valve 22 of the water supply pipe 20, and the metering pump 42 of the disinfectant storage tank 40 are directly controlled by the computer means 200. The efficiency of the liver abnormality which has undergone the above control process can be compared and analyzed and recorded in the recording device 240 of the computer means 200. The present invention thus facilitates storing, recalling and managing a series of data obtained from the previous step in a single file.

Accordingly, the present invention enables a person to supply a simple constant in an optimal state at all times by producing a situation such as applying the operating range according to the actual abnormal water supply situation, and monitors and controls data from a remote site in real time in real time. A series of steps to build a monitoring control system is also possible.

The present invention has been described with reference to specific embodiments of the present invention, but it is apparent that various modifications and changes can be made as described above without departing from the technical spirit described in the claims of the present invention. Accordingly, the detailed description of the invention and the accompanying drawings should be construed as illustrative rather than limiting to the spirit of the invention.

1 is a system configuration of a conventional chlorine dosing device.

2 is a view showing an overview of the liver abnormality monitoring control system using a neural network according to the present invention.

3 is a block diagram showing the overall configuration of FIG.

4 is a flowchart illustrating a monitoring control step using a neural network according to the present invention.

Figure 5 is a flow chart of the overall learning and application process according to the neural network control program using the learning optimization algorithm used in the present invention.

6 is a diagram illustrating a system of a learning optimization algorithm according to the present invention.

* Explanation of symbols on main parts of the drawings *

10. Control device 12. Motor pump

20. Water line 22. Solenoid valve

30. Water tank 40. Disinfectant storage tank

42. Metering pump 100. Sensor means

110. Analog-to-digital converter 120. Periodic signal detector

200. Computer means 210. Interface

Central processing unit 230. Screen

240. Recording device 300. Control means

310. Memory 320. Central Processing Unit

330. D / A Card 340. Digital Card

350. Interface

Claims (16)

A well device 10 having a motor pump 12 for pumping raw water, a water pipe 20 for supplying raw water pumped by the well device 10, and raw water supplied to the water pipe 20. In a simple abnormality water supply monitoring control system comprising a water tank 30 to disinfect and the disinfectant storage tank 40 to inject the disinfectant into the water tank 30, Sensor means (100) for measuring the current value of the independent variable variable including the indoor and outdoor temperature of the crystal device 10, the pressure, flow rate, flow rate of the water pipe 20, and the disinfectant input amount of the disinfectant storage tank 40 )and; Collecting data of the current value of the independent variable detected from the sensor means 100, arithmetic processing into physical quantity data, and comparing the current value of the measured characteristic by a neural network control program using a learning optimization algorithm, the simple abnormality Computer means (200) for determining a normal section, a check section (available) and an unusable section of the apparatus, and adjusting the independent variable values according to the determined result to perform arithmetic processing on analog and digital control output values; Control means 300 for controlling the control device 10 and the water pipe 20 and the disinfectant storage tank 40 to the control output value; The input learning data used for the neural network control program is the indoor / outdoor temperature of the control device 10, the pressure, flow rate and flow rate of the water pipe 20, and the disinfectant input amount of the disinfectant storage tank 40. The optimum pressure, flow rate, flow rate and the optimal disinfectant input amount of the disinfectant storage tank 40 are calculated based on the weights obtained, The system of the neural network control program includes an input layer, a hidden layer, and an output layer, the number of hidden layers is determined in a range of 2 to 6, and data input, output, and weights from the hidden layer to the input layer and output layer are determined. Liver abnormality monitoring control system using a neural network characterized in that. delete delete The method of claim 1, In the output layer, a linear sum is used as an active function, and in the hidden layer, a nonlinear function is used as an active function, and the weight is accumulated and stored in the weight by storing and obtaining information obtained from the learning process. Liver abnormality monitoring control system using a neural network, characterized in that used to calculate the flow rate, flow rate and the optimum disinfectant solution input amount of the disinfectant storage tank (40). The method of claim 1, The range of input learning data of the neural network control program is The input range is in the range of the indoor temperature of the video device 10 is 0 ~ 40 ℃, the outdoor temperature is minus 20 ℃ ~ 40 ℃ ℃, the pressure of the water pipe 20 is 0 ~ 35kgf / ㎠, the flow rate is 0 ~ 10㎥ / d, flow rate 0 ~ 2.0m / sec range, the disinfectant input amount of the disinfectant storage tank 40 is the flow rate (L) × set chemical concentration (PPM) / (concentration of the liquid product) = input amount (ml) Liver abnormality monitoring control system using a neural network, characterized in that × 1000. According to claim 5, The neural network control program, If the pressure or flow rate of the water pipe 20 is increased to reduce the speed of the motor pump 12, if the pressure or flow rate is reduced to increase the speed of the motor pump 12, Liver abnormality monitoring control system using a neural network, characterized in that to increase the amount of disinfectant solution when the flow rate of the water pipe 20 increases, and to reduce the amount of disinfectant solution when the flow rate is reduced. delete A well device 10 having a motor pump 12 for pumping raw water, a water pipe 20 for supplying raw water pumped by the well device 10, and raw water supplied to the water pipe 20. In the method for monitoring and controlling the abnormal abnormality of water comprising a water tank 30 and a disinfectant storage tank 40 for disinfecting the disinfecting solution into the water tank 30, Measuring a current value of the independent variable variable including the indoor / outdoor temperature of the well 10, the pressure, flow rate and flow rate of the water pipe 20 and the disinfectant input amount of the disinfectant storage tank 40; Collecting data of the measured independent variable current value by a computer and performing arithmetic processing on physical quantity data; Comparing the current value of the measured characteristic according to a neural network control program using a learning optimization algorithm in the computer, and determining the normal section, the check section (usable) and the unusable section of the abnormal abnormality frequency; Adjusting the independent variable values according to the determined result and performing arithmetic processing to analog and digital control output values; Controlling the control device 10, the water pipe 20 and the disinfectant storage tank 40 to the obtained control output value; The input learning data used in the neural network control program is an indoor / outdoor temperature of the control device 10, a pressure, a flow rate and a flow rate of the water pipe 20, and a disinfectant input amount of the disinfectant storage tank 40. The optimum pressure, flow rate, flow rate and the optimal disinfectant input amount of the disinfectant storage tank 40 are obtained from the weights obtained from the water supply pipe 20 as target data. The system of the neural network control program includes an input layer, a hidden layer, and an output layer, the number of hidden layers is determined in a range of 2 to 6, and data input, output, and weights from the hidden layer to the input layer and output layer are determined. Liver abnormality monitoring control method using a neural network, characterized in that. delete delete The method of claim 8, In the output layer, a linear sum is used as an active function, and in the hidden layer, a nonlinear function is used as an active function, and the weight is accumulated and stored in the weight by storing and obtaining information obtained from the learning process. Liver abnormality monitoring control method using a neural network, characterized in that it is used to calculate the flow rate, flow rate and the optimum disinfectant input amount of the disinfectant storage tank (40). The method of claim 8, The range of input learning data of the neural network control program is The input range is in the range of the indoor temperature of the video device 10 is 0 ~ 40 ℃, the outdoor temperature is minus 20 ℃ ~ 40 ℃ ℃, the pressure of the water pipe 20 is 0 ~ 35kgf / ㎠, the flow rate is 0 ~ 10㎥ / d, flow rate 0 ~ 2.0m / sec range, the disinfectant input amount of the disinfectant storage tank 40 is the flow rate (L) × set chemical concentration (PPM) / (concentration of the liquid product) = input amount (ml) Liver abnormality monitoring control method using a neural network, characterized in that × 1000. The method of claim 12, wherein the neural network control program, If the pressure or flow rate of the water pipe 20 is increased to reduce the speed of the motor pump 12, if the pressure or flow rate is reduced to increase the speed of the motor pump 12, If the flow rate of the water pipe 20 is increased, the input amount of the disinfectant solution is increased, and if the flow rate is reduced liver abnormality monitoring control method using a neural network, characterized in that to control to reduce the input amount. The method of claim 13, wherein the neural network control program, Liver abnormality monitoring control method using a neural network, characterized in that the automatic control of the flow rate and the disinfectant amount of the water supply pipe 20 when the value of the outdoor temperature is different even in the same indoor temperature range of the control device (10). The method of claim 8, Storing and managing the data required in the measurement and operation processing step in a file; liver abnormality monitoring control method using a neural network, characterized in that further comprises. The method of claim 15, The data is controlled in real time by two-way communication with a computer using a transmission control protocol / Internet protocol (TCP / IP) 250, and after the control process is performed by a computer using a multi-processing function, the state of the abnormal frequency Comparative analysis and recording in a file; liver abnormality monitoring control method using a neural network, characterized in that further comprises.

Images