KR20220063065A - System and Method for Controlling of Intelligent Fish Feeder - Google Patents

Abstract

The present invention relates to an intelligent fish farm feed supply system which supplies feed in conjunction with an intake degree of fish. The system comprises: an intake degree extraction unit which measures the behavior of fish at each specific time, wherein the specific time is classified by dividing a measurement start time and a measurement end time of a fish trajectory into a plurality of periods, extracts a trajectory image of the fish, and classifies the extracted trajectory image using a learning classifier to calculate multiple levels of intake desire of fish, wherein the learning classifier is a horizontal distance and a vertical distance moved by the fish; a supplied feed amount management system which calculates feed about data about a feed amount to be supplied to a water tank by using, as inputs, the multiple levels of intake desire of fish calculated by the intake degree extraction unit and water tank data measured by a plurality of sensors installed on a fish farm; and a feed supply device which receives the calculated feed amount data to measure the weight of feed to be supplied and supplies the feed to the water tank.

Description

Intelligent Fish Feeder Control System and Method

The present invention relates to an apparatus and method for feeding fish in an aquaculture farm, and more particularly, to an intelligent feeder interlocked with a fish feed level.

Recently, due to the negative effects of climate change such as typhoons, red tides, high temperatures, and large influx of fresh water and pollution of the marine landscape, the conversion from offshore cage culture to land culture and the aquaculture of expensive high-performance fish species and related businesses through large-scale land aquaculture have begun. is in progress

For the success of onshore aquaculture, large-scale facility investment must be made, and steady profits must be achieved for the continuation of the business. To this end, it is necessary to maximize the realization of profits by reducing feed waste and mortality, and by controlling the shipping time of high-quality fish. Therefore, the system of the onshore farm should be intelligent to achieve the above purpose, and it should have an interface that is easy to understand and easy for the actual operator to operate.

In order to reduce the waste and mortality of feed and to control the shipping time of high-quality fish, it is necessary to supply the optimal feed in an environment where the growth of fish is optimal. Here, the meaning of the optimal feed supply is to supply the optimal amount at the optimal time to speed up the growth of the fish and to ensure that there is no feed that is not eaten and wasted.

Recently, as a method for optimal feed supply, studies are being actively conducted on how to measure fish movement using artificial intelligence and convert it into a feeding desire to feed when the feeding desire is active. There is no system that provides a neural network model that can learn the fish activity by applying it in the actual field and does not specifically disclose the relationship between the islands.

In case of large-scale aquaculture, due to incorrect judgment, water quality may be contaminated due to oversupply of feed, which may cause disease and death, or it may be difficult to ship fish with the maximum weight due to insufficient feed. can have

The reason for causing such a problem is that there is no exact criterion for determining the feeding needs of fish, and the amount of feed is controlled without reflecting the tank environment that affects the growth of fish. For example, when dissolved oxygen is low, the metabolism of feed is low, high water temperature can induce fish growth rate, and sunlight affects the maturity rate of fish. In other words, the water temperature is between 18 and 22 degrees Celsius, where fish growth is most active. Even when the current tank water temperature is 15, even if the fish feed is recognized as the maximum, if the feed amount is reduced to the maximum, there is a high probability that the feed will remain, and diseases may occur due to decreased metabolism. Therefore, when the water temperature is between 18 and 22 degrees and the fish intake is the maximum, the maximum amount of feed can be supplied to maximize the growth of fish. That is, there is a need for a model and system for controlling the optimal feed amount reflecting the environment of the tank as described above.

In general, in order to derive the optimal feeding amount, a multi-task neural network model using multiple inputs (which may include, for example, movement level, tank environment data, weight, etc.) and a higher-order hidden layer is used to extract the results. do. However, it has a disadvantage in that it is difficult for the end user to fully understand the extracted results, and thus it is difficult to apply it to the actual field.

In addition, in actual aquaculture sites, the metabolism of fish is significantly different depending on the water temperature in the tank, dissolved oxygen, and the weight of the fish. Therefore, a decision-making model of feed supply using empirical data from these actual sites is required.

An object of the present invention for solving the above problems is to provide an intelligent feeding system and method capable of preventing waste of feed and preventing contamination of water.

Another object of the present invention is to provide an intelligent feeding system and method capable of preventing the occurrence of diseases in fish.

Another object of the present invention is to provide an intelligent feeding system and method capable of accurately supplying feed to a corresponding aquaculture tank.

According to one aspect of the present invention for solving the above-mentioned problems, in the intelligent farm feeding system for supplying food in conjunction with the feed level of the fish, the behavior of the fish is measured for each specific time - the specific time is the fish trajectory classifying by dividing the measurement start time and measurement end time into a plurality of periods -, extracting the trajectory image of the fish and learning the extracted trajectory image of the fish into a learning classifier - The learning classifier is the horizontal distance the fish moved and From the vertical distance - from a plurality of sensors installed in a fish farm and a sub-eido extraction unit that classifies and calculates the feed desire of a plurality of stages of fish and a plurality of levels of fish desires calculated in the sub-eido extraction unit and a feed supply management system for calculating feed amount data to be supplied to the tank by inputting the measured tank data, and a feed supply device for receiving the calculated feed amount data, measuring the weight of feed to be supplied, and supplying the feed amount to the tank.

In addition, the susceptibility extraction unit uses machine learning to classify the extracted trajectory image of the fish using the learning classifier to calculate the supination desire of the fish of the plurality of stages, and as a plurality of inputs of the machine learning It is characterized in that the image of the trajectory of the fish measured for each specific time.

In addition, as an n-th input among the plurality of inputs of the machine learning, it is characterized in that it is a trajectory image of a fish measured at an n-th time among the predetermined specific time.

In addition, the susceptibility extracting unit further includes a trajectory tracking unit that tracks a specific fish, assigns an object identification number, and generates a trajectory image of the fish using a plurality of algorithms.

In addition, the plurality of algorithms includes at least two of YOLOv3, YOLOv2, Farneback, Pyflow, EpicFlow, and FlowNet algorithms.

In addition, the plurality of subi desires are determined on the basis of subi level labeling classified by an expert in advance.

In addition, in the type of feeding degree labeling, the trajectories of the plurality of time-based fish are classified into a plurality of types of patterns, and divided into low-speed famous, high-speed swimming, stationary, and abnormal swimming activity states, and if the low-speed swimming, sub Desire is defined as full, if the high-speed swimming, the sub defines a need as hunger, if the swim in the stationary state, the sub defines a need as very full, and if the abnormal swimming, the condition of the fish is defined as disease or stress.

In addition, the feed amount management system calculates the amount of feed data to be supplied to the tank using a fuzzy inference engine, and the fuzzy inference engine uses a neurofuzzy reasoning system to determine the fish’s feeding desire, measured from the plurality of sensors. The first feed amount data is derived by inputting tank data and expert knowledge.

In addition, the fuzzy inference engine includes a Crisp input layer, an Input Membership function, a fuzzy rule layer, an Input Membership function, and a defuzzification layer. and an output layer.

In addition, the input of the fuzzy inference engine includes the water temperature, dissolved oxygen amount, the weight of the fish, and (?) the activity state of the fish.

The feed amount management system further includes a correction calculator configured to calculate the second feed amount data by calculating the first feed amount data and the feed supply schedule data input by the aquaculture manager.

In addition, the first feed amount may have a positive feed weight or a negative feed weight according to the feeding desire of the fish, measured tank data from the plurality of sensors, and expert knowledge, and the correction calculator is The second feed amount is derived by calculating the first feed amount and the feed weight included in the feed supply schedule data.

In addition, the second feed amount may be the same as, more than, or less than the weight of the feed included in the feeding schedule.

In addition, the feed supply device is connected to the manager terminal, and the manager terminal receives the feed amount data received from the feed amount management system and controls the feed supply device using the MQTT protocol.

In addition, the feed supply device accumulates the second feed amount in the feed box by using the opening/closing operation of the slide and then drops it and supplies it to the water tank using a high-pressure blower.

In addition, the feeding device includes a slide for moving the feed container in which the second feed amount is accumulated to be coupled to a feed pipe connected to a tank to which the second feed amount is to be fed.

In addition, the tank environment data includes at least one of temperature data in the tank, dissolved oxygen data, and fish weight data.

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

According to the intelligent feeding system and method according to the embodiments of the present invention, in the operation of a large aquaculture, the feeding point is reflected in the aquatic environment affecting the fish feed and metabolism recognized by the expert in the actual field. By determining the overfeed, it is possible to provide a method for preventing the actual waste of feed and the occurrence of different types of fish.

According to the intelligent feeding system and method according to the embodiments of the present invention, by correcting the inferred feeding time and feeding amount using the weight of the fish and a pre-determined feeding schedule, the growth of fish and high quality You can provide a way to ship the fish.

1 is a diagram showing a configuration diagram of an intelligent feeding device according to an embodiment of the present invention and a flow of data.
2 is a view showing a fish activity data set for labeling and fish behavior learning for completing a trajectory tracking model according to an embodiment of the present invention.
3 is a diagram for explaining the tracking and display of the trajectory of a fish using the trajectory tracking algorithm of the fish behavior state according to an embodiment of the present invention.
4 is a view showing an intelligent fuzzy inference engine of the feed amount management system according to an embodiment of the present invention.
5 is a view showing an affiliate function editor for fuzzyizing an input into a membership function according to an embodiment of the present invention.
6 is a diagram illustrating a fuzzy rule viewer showing fuzzy rules of input and output according to an embodiment of the present invention.
7 is a view showing the configuration of an intelligent feeder according to an embodiment of the present invention.

1 is a diagram showing a configuration diagram of an intelligent feeding device according to an embodiment of the present invention and a flow of data.

Referring to FIG. 1 , the intelligent feeding system 1 is composed of a feed extraction unit 100 , a feed amount management system 200 , a feed supply device 300 , a sensor 400 , and a manager input device 500 . do.

The angle extraction unit 100 may include a trajectory tracking unit 110 and a trajectory tracking model 120 .

The trajectory tracking unit 110 generates trajectory data 1 from the video of the fish and transmits it to the trajectory tracking model 120, and the sensor 400 measures the water temperature, oxygen, and fish weight in the tank to provide temperature data (3) ), dissolved oxygen data 4 and weight data 5 are generated and the generated data is transmitted to the feed amount management system 200 , and the manager input device 500 inputs feed feed schedule data 7 from the manager. and transmits the data to the feed amount management system 200 .

The trajectory tracking model 120 categorizes the behavioral patterns of fish and derives the final susceptibility, and trajectory data 1 ), the degree of subtance data (2) is calculated.

The feed amount management system 200 may include a fuzzy inference engine 210 and a correction calculator 220 . The fuzzy inference engine 200 receives the temperature data (3), the dissolved oxygen data (4), and the weight data (5) from the sensor 400 and the temperature data (2) that is the result of the trajectory tracking model 120 as inputs. to derive the first feed amount data 6 , and the correction calculator 220 uses the first feed amount data 6 to correct the feed supply schedule data 7 to calculate the second feed amount data 8 . and transmitted to the feed supply device 300 .

The feed supply device 300 controls the internal device to achieve the second feed amount feed amount data 8 derived from the feed amount management system 200 . For example, it is controlled by combining the limit of the measuring cylinder and the motor speed of the high-pressure blower. The feed supply device 300 can increase/decrease up to 0 to 2 times in 1/100 of the target feed amount.

The trajectory tracking model 120 may be updated every feeding cycle of a specific period to complete a more accurate model. In particular, reinforcement learning (reinforcement learning) does not require exact pairs of inputs and outputs, and a balance is formed between the utilization of known data and the search for unknown information, so it is suitable for problem solving that involves short-term versus long-term rewards. A more accurate model is created through learning through a plurality of images of

By deriving the feed amount using the trajectory tracking model 120 and the fuzzy inference engine 210, which are neural network models that are learned by reflecting the expert's knowledge, the user can easily understand the derivation process and can secure trust.

The overall feed supply frame is determined by the feed supply schedule data 7 that is directly input by the manager or calculated through a separate program, and the feed supply management system 200 includes the first feed amount data in the feed supply schedule data 7 . By adding or subtracting (6), the second feed amount data (8) is calculated. For example, according to the feed schedule data 7, the feed amount is 5 kg at a specific time, but when a weight loss instruction is derived from the first feed amount data 6, the feed reduced by the weight loss indicated from 5 kg is supplied. . This process may be performed automatically, or may require further confirmation from the administrator.

2 is a view showing a fish activity data set for labeling and fish behavior learning for completing a trajectory tracking model according to an embodiment of the present invention.

Referring to FIG. 2 , the labeling 610 for the completion of the fish behavior state trajectory tracking model 120 of the present invention is classified into four types. Here, the labeling 610 refers to a pair of input and output that is the correct answer for learning in machine learning. That is, behavior TYPE A (611) is defined as slow swimming (input) and usually full (output), behavior TYPE B (612) is fast swimming (input) and defined as very hungry (output), behavior TYPE C (613) is static swimming (mouth) and is defined as very full (output), behavior TYPE D (614) is abnormal swimming (input) and is defined as having disease or stress (output).

The data set 630 for fish behavior image learning may include Type A 631 , Type B 632 , Type C 633 , and Type D 634 .

The data set 630 for fish behavior image learning includes an Aim step (step 1) for measuring the fish trajectory 620 for the first time, a Creep step for measuring the second time (step 2), and an attack step for measuring the third cycle. (Step 3) and the Return step (Step 4) measured during the fourth cycle, and the learning classifier of the neural network model classifies the fish trajectory measured at each step using the moved horizontal distance and vertical distance.

For example, if the swimming direction of fish from the Aim stage (1st stage) to the Atack stage (3rd stage) increases in the vertical direction, and the return stage (4th stage) shows a descending trajectory after a long horizontal movement, It is labeled as Behavior Type B (612). In the case of behavior TYPE D (614), after the ATTACK stage (3rd stage), it does not show the complete descent seen in the RETURN stage (4th stage), and continues to swim horizontally and shows a new ATTACK stage (3rd stage).

FIG. 3 is a diagram showing the trace of a fish by using the fish behavior state trace tracking algorithm according to an embodiment of the present invention.

Referring to FIG. 3 , the trajectory tracking unit 110 tracks the fish from the image data 700 and assigns object IDs (eg, object ID 1 and object ID 2 ). The trajectory tracking unit 110 may track the fish using various algorithms, for example, a YOLOv3 or YOLOv2 algorithm. In the example shown, the YOLOv3 algorithm was used.

The trajectory tracking unit 110 may generate the trajectory data 1 by extracting the optical flow for each object ID from the image data 700 . Optical flow extraction may extract optical flows using various algorithms, for example, the Farneback algorithm, the Pyflow algorithm, the EpicFlow algorithm, or the FlowNet algorithm. In addition, the trajectory tracking unit 110 may measure the size and swimming speed of the fish.

4 is a diagram illustrating an intelligent fuzzy inference engine of a feed amount management system according to an embodiment of the present invention.

Intelligent fuzzy inference is a computation method that applies the parallel operation and learning ability of artificial neural networks, and human knowledge expression and explanation capabilities of fuzzy systems. , but it is difficult for users to understand, while intelligent fuzzy inference supplements the weaknesses of neural network models so that users can easily understand them through expert information. Therefore, it is possible to use the intelligent fuzzy inference engine to learn high-level data from experts and to output judgments based on the experiences of field experts.

Referring to FIG. 4 , the fuzzy inference engine 210 includes a Crisp input layer 801 , an Input Membership function 802 , a fuzzy rule layer 803 , and an output membership It may include an Input Membership function 804 , a defuzzification layer 805 , and an output layer 806 . According to an embodiment, the temperature data 3, dissolved oxygen data 4, weight data 5, and degree data 2 are applied to x1 and x2 of the crisp input layer 801, and the input membership function ( 802 are membership functions associated with the inputs (x1, x2) and the output (y). Outputs of the membership functions may be applied to nodes R(R1, R2, R3, R4, R5, R6) of the fuzzy rule layer 803 . Nodes R of the fuzzy rule layer 803 may apply outputs according to the learned rule to nodes C of the output belonging function layer 804, and nodes C(C1, C2) of the output belonging function layer 804 ) may be applied to a node of the defuzzy layer 805 . A node in the output layer 806 may generate an output y based on inputs applied from the defuzzy layer 805 . According to one embodiment, the node of the output layer 806 may be a single node, and may add the inputs applied from the defugerization layer 805 to generate an output y, which, as described above, is a layer of the fuzzy inference engine. The structures and aspects of the nodes and nodes may be designed by employing various methods.

Fuzzy reasoning is a formula used when true and false are ambiguous. For example, there is no exact answer to the amount of feeding a fish. Feeding amount may vary depending on external circumstances, and it is not efficient to give a specific amount to farmed fish every time. In other words, it is to provide an appropriate amount to the farmed fish according to the situation. Therefore, it is possible to bring optimization of the result value through fuzzy inference in the feed amount management system 200 . Crisp enters a specific reference value and compares and analyzes it according to the fuzzy rule to find an appropriate amount as a result. This makes it possible to manage the appropriate feed supply for ambiguous results such as feed amount.

5 is a diagram illustrating a membership function editor for fuzzyizing an input into a membership function according to an embodiment of the present invention.

Referring to FIG. 5 , graphs 940 and 950 showing the results obtained by grading the input “input 1” 910 and “input 2” 920 by the membership function in the membership function editor 900 . . Here, the input is written based on the expert's knowledge data in a linguistic form. Based on this linguistic expert knowledge data, an “if-then rule” is configured and applied to the input and output of the fuzzy inference engine 210 to generate feed amount information.

Neuro-fuzzy reasoning system is an artificial intelligence programming method that can infer the number of different cases according to various variables and produce meticulous results by applying the neuro-networking method, which is an artificial intelligence element, to the fuzzy reasoning system, which is an existing intelligent program.

Neurofuzzy inference systems can produce optimal results according to situations based on the number of complex and numerous cases like human neural networks.

In the case of the neurofuzzy reasoning system applied to the automatic feed management system, four variables are applied to manage the optimal automatic feed supply.

Based on four variables such as water temperature, dissolved oxygen amount, individual weight, and activity state, it is possible to calculate the optimal and appropriate result value from the result value for which the correct answer is not determined. As a result, the optimal result can be output by synthesizing the four variables, and a neurofuzzy inference system can be applied using a membership function editor tool such as MATLAB. When the range of each variable is input, the result value according to each numerical value is calculated according to the applied function standard. Since there is no fixed answer for the automatic feed amount, when each variable is input as a range, the numerical value is comprehensively displayed in the form of a graph, and the optimal result value is calculated from the graph as shown in FIG. 5 .

As a result, the optimal value for the amount of feed is calculated by counting the number of cases according to the water temperature, dissolved oxygen amount, individual weight, and activity state, and it is calculated as the optimal value and result value shown in the graph.

6 is a diagram illustrating a fuzzy rule viewer showing fuzzy rules of input and output according to an embodiment of the present invention.

Referring to FIG. 6 , the fuzzy rule is a collection of expert knowledge in “if-then format” and shows the relationship between input and output, and explains the overall system operation. It is a screen showing the result of the artificial intelligence neurofuzzy inference system that shows the optimal result value when four input values (1000) are input. Each variable is input from Input1 to Input 4 (1000), and the optimal output value 1100 is shown according to a comprehensively determined function.

7 is a view showing the configuration of an intelligent feeding device according to an embodiment of the present invention.

Referring to FIG. 7 , the feed supply device 300 includes a feed container 310 , a first slide 320-1, a first limit switch 330-1, a measurement container 340, and a second slide 320-1). , the second limit switch 330-2, the third slide 320-3, the feed pipe 350, the high-pressure blower 360, the control panel 370 (hereinafter referred to as the constituent device), and is composed of the manager terminal ( 380) is connected.

The manager terminal 380 is connected to the feed amount management system 200 to receive feed supply related information from the feed amount management system 200 .

The manager terminal 380 includes a PC or mobile means, and the mobile monitors the operating state of the intelligent feeding device 300 in the form of a responsive web app, and also the feed received from the feed amount management system 200 . The feeding device 300 is controlled using the Mqtt protocol using the supply information. That is, the component devices are controlled according to the received feed amount, feed time, feed stop, etc. instructions.

As an embodiment of the method for controlling the amount of feed, first, the third slider 320-3 slides according to the command of the feed amount management system 200 to determine the direction of feed (or the tank to be fed), and the Thereafter, the first slide 320-1 is opened until it comes into contact with the first limit switch 330-1, so that the feed falls from the feed tank 310 to the measuring tube 340. When the amount of feed received by the feed amount management system 200 is accumulated in the measuring container 340, the first slider 320-1 is closed, and the second slider 320-2 is opened at the same time. The feed accumulated in the second slider 320 - 2 is dropped, and the fallen feed is fed to the water tank through the feed pipe 350 by the high-pressure blower 360 . When the feed is finished, the second slider 320 - 2 is closed to end the feed.

1: Intelligent feeding system
100: supido extraction unit
110: trajectory tracking unit
120: trajectory tracking model
200: feed supply management system
210: Fuzzy inference engine
220: correction calculator
300: intelligent feeding device
320-1, 320-2, 320-3: 1st, 2nd, 3rd slide
330-1, 330-2: 1st, 2nd limit switch
340: measuring tube
360: high pressure blower
380: manager terminal
400: sensor
500 : admin input device
900 : membership function editor

Claims (17)

By measuring the behavior of the fish for each specific time - For each specific time, the measurement start time and the measurement end time of the fish trajectory are divided into a plurality of periods and classified -, the trajectory image of the fish is extracted and the trajectory image of the extracted fish a learning classifier - the learning classifier is a horizontal distance and a vertical distance moved by the fish - to classify the learning classifier using , and an extracting unit to calculate the desire to eat fish of a plurality of stages;
A feed supply management system for calculating the amount of feed data to be supplied to the tank by inputting the feeding desire of the fish at a plurality of stages calculated by the feeding level extraction unit and the tank data measured by a plurality of sensors installed in the farm; and
and a feed supply device for receiving the calculated feed amount data, measuring the weight of feed to be supplied, and supplying the feed to the tank. According to claim 1,
The feeding degree extraction unit classifies the extracted trajectory image of the fish using the machine learning classifier to calculate the feeding desire of the fish of the plurality of stages, and the plurality of inputs of the machine learning include the specific Intelligent feeding system, characterized in that it is the trajectory image of the fish measured by time. 3. The method of claim 2,
An n-th input among the plurality of inputs of the machine learning is an intelligent feeding system, characterized in that it is a trajectory image of a fish measured at an n-th time among the predetermined specific time. According to claim 1,
Intelligent feeding system further comprising a trajectory tracking unit for generating trajectory images of the fish by using a plurality of algorithms, after the susceptibility extractor tracks a specific fish and assigns an object identification number. 5. The method of claim 4,
The plurality of algorithms includes at least two of YOLOv3, YOLOv2, Farneback, Pyflow, EpicFlow and FlowNet algorithms. According to claim 1,
The plurality of feeding desires are intelligent feeding system that is determined based on feeding level labeling classified by an expert in advance. 5. The method of claim 4,
In the type of subdivision labeling, the trajectories of the plurality of time-based fish are classified into a plurality of types of patterns to classify the activity states of low-speed famous, high-speed swimming, stationary, and abnormal swimming, and if the low-speed swimming, the desire for subs Intelligent feeding is defined as full, if the high-speed swimming, the sub's desire is defined as hunger, if the stationary swimming, the sub's need is defined as very full, and if the abnormal swimming, the fish's condition is defined as disease or stress. system. According to claim 1,
The feed amount management system calculates feed amount data to be supplied to the tank by using a fuzzy inference engine, and the fuzzy inference engine uses a neurofuzzy reasoning system to feed the fish, and measured tank data from the plurality of sensors and an intelligent feeding system that derives the first feed amount data by inputting the expert's knowledge. 9. The method of claim 8,
The fuzzy inference engine includes a Crisp input layer, an input membership function, a fuzzy rule layer, an input membership function, a defuzzification layer, and an output. An intelligent feeding system with an output layer. 9. The method of claim 8,
An intelligent feeding system including water temperature, dissolved oxygen amount, fish weight, and (?) fish activity state as inputs of the fuzzy inference engine. 9. The method of claim 8,
The feed amount management system further comprises a correction calculator configured to calculate the second feed amount data by calculating the first feed amount data and the feed supply schedule data input by the aquaculture manager. 12. The method of claim 11,
The first feed amount may have a positive feed weight or a negative feed weight according to the feeding desire of the fish, measured tank data from the plurality of sensors, and expert knowledge, and the correction calculator is the first feed An intelligent feeding system for deriving the second feed amount by calculating the amount and the feed weight included in the feed supply schedule data. 13. The method of claim 12,
The second feed amount may be equal to, greater than, or less than the weight of the feed included in the feeding schedule. According to claim 1,
The feed supply device is connected to a manager terminal, and the manager terminal receives the feed amount data received from the feed amount management system and uses an MQTT protocol to control the feed supply device. 15. The method of claim 14, wherein
The feeding device is an intelligent feeding system for accumulating a second amount of feed in the feed box by using the opening and closing operation of the slide and then dropping it and supplying it to the water tank using a high-pressure blower. 15. The method of claim 14, wherein
and the feeding device includes a slide for moving the feed container in which the second feed amount is accumulated to a feed pipe connected to a water tank to which the second feed amount is to be fed. According to claim 1,
The tank environment data includes at least one of temperature data in the tank, dissolved oxygen data, and fish weight data.

Images