CN111436386A - A method and system for aquaculture of swimming fish based on feeding intensity measurement - Google Patents

Abstract

The invention discloses a breeding method and system for swimming-type cultured fish based on feeding intensity measurement, and relates to the field of fish culture. This solution is proposed in view of the problem that breeding intelligence is not precise enough in the prior art. The breeding method mainly includes the steps of acquiring images and binarizing, calculating the feeding intensity, and controlling the feeding according to the feeding intensity. The cultivation system mainly realizes the control realization of the cultivation method through the host computer. The advantage is that it is possible to quantitatively describe the feeding intensity of fish and use the feeding intensity as an important reference standard for feed feeding. The feeding amount and feeding time of different breeding batches are no longer dependent on empirical formulas, and can be corrected based on changes in feeding intensity. With the growth of fish and changes in the number of mantissas in the breeding process, corresponding adjustments can be made in time, and the intelligence of breeding production has been significantly improved.

Description

A method and system for aquaculture of swimming fish based on feeding intensity measurement

technical field

The invention relates to a method and system for culturing fish, in particular to a method and system for culturing swimming fish based on feeding intensity measurement.

Background technique

With the advancement of science and technology, the breeding industry has gradually developed from extensive artificial breeding to refined breeding. The machine system can automatically feed according to the manually set time and feeding amount, or perform operations such as pumping oxygen at the manually preset time. However, they are only limited to manual presets, and the system cannot automatically determine the status of the fish and perform corresponding operations.

The manual preset must be set accurately, and the ideal feeding amount and timing of the specified breeding batch can be obtained after multiple trials, and it cannot respond to changes in various parameters of the weather. For example, with the change of cloudy weather, air pressure, temperature, etc., the feeding desire of fish will be significantly different. Relying on the system preset program to complete, it is easy to cause excessive or insufficient feed, or feeding in the absence of oxygen, etc., which will seriously lead to the death of the entire batch of fish.

Traditional breeding methods are difficult to achieve accurate feeding of feed. Excessive supply of feed will increase the amount of residual bait and excrement in the breeding environment, which may easily lead to the outbreak of fish diseases and affect the growth of fish. Insufficient supply of feed will lead to slow growth of fish and affect breeding efficiency. Feed cost is the main consumption in the whole breeding process. How to achieve accurate feeding of feed is an urgent problem to be solved in the production process.

With the development of computer technology and visual technology, it is a research hotspot at home and abroad to use visual technology to judge the hunger degree of fish, and then assist the breeding system to achieve accurate feeding. The application of visual technology in fish farming at home and abroad is based on strict experimental conditions, such as high requirements for water clarity, small water monitoring range, high program calculation pressure, and a wide variety of collected parameters. The inventor previously disclosed a three-dimensional fish activity monitoring system, CN109389623A, but did not disclose the behavior analysis of fish activity. A feeding system based on feed information after feeding is also disclosed, CN109757419A, but the objects collected and analyzed are not based on the fish itself, and fish analysis must be performed with the help of other reference materials. Therefore, how to obtain a reasonable and easy-to-calculate fish physiological parameter, and then use the fish physiological parameter as an automated breeding application is an urgent problem to be solved by those skilled in the art.

SUMMARY OF THE INVENTION

In order to solve the above-mentioned problems in the prior art, the present invention aims to provide a method and system for culturing swimming-type farmed fish based on the measurement of feeding intensity.

A method for culturing swimming-type aquaculture fish based on the measurement of feeding intensity according to the present invention comprises the following steps:

Use a somatosensory camera to obtain a depth map with a resolution of m*n; take the depth value as a comparison parameter, and perform binarization f(x,y) processing on each pixel in the depth map: the depth value must be valid if the depth value is located above and below the water surface. The pixels within the interval are recorded as 1, otherwise they are recorded as 0;

获取，其中k为深度图的获取时刻；Monitor the feeding intensity y(k) of the fish, the feeding intensity y(k) uses the formula

Get, where k is the acquisition moment of the depth map;

When the feeding intensity y(k) remains below the threshold for a certain period of time and reaches the preset feeding time, feed feeding;

Continue to monitor the feeding intensity. When the feeding intensity decreases with time, it rises again to the set proportion of the feeding intensity at the feeding time, and a new round of reduced feeding is performed; this step is repeated until the feeding intensity decreases with time and cannot rise to this level. The set ratio of the feeding intensity at the time of the second feeding, and the feeding of the feed is stopped.

Preferably, when the feeding intensity is maintained above the threshold for a certain period of time, the derivation formula is used to monitor the fluctuation rate of the feeding intensity: when the absolute value of the fluctuation rate y'(k) is greater than the preset value, the fish school is judged to be normal Feeding state; when the absolute value of the fluctuation rate is less than the preset value, it is judged that the fish is in an oxygen-deficient state, and the circulating pump is controlled to start.

Preferably, after the circulation pump is started, the feeding intensity drops to a threshold value and is maintained for a certain period of time, and the circulation pump is controlled to be turned off.

Preferably, when it is not within the preset feeding time, the feeding intensity is higher than the threshold, and the absolute value of the fluctuation rate is greater than the preset value, it is determined that the time is the physiological feeding time of the corresponding cultured fish; the physiological feeding time is used. Correct the preset feeding time.

Preferably, the effective interval is 500mm below the somatosensory camera to 60mm underwater; and the somatosensory camera is not less than 500mm above the water surface.

Preferably, the certain period of time is 10 minutes.

Preferably, when the feeding intensity exceeds the threshold, the time, feeding intensity and corresponding depth value are recorded.

Preferably, the set ratio is 90%.

A culturing system for swimming-type culturing fish based on measurement of feeding intensity, comprising a culturing pond, a somatosensory camera arranged above the culturing pond, a feeding device for feeding feed, and a host computer; the host computer applies the breeding method Control the somatosensory camera and feeding device to work.

Preferably, the breeding system further comprises a circulating pump, which is connected to the breeding pond through a pipeline and is electrically connected to the upper computer; the upper computer detects when the feeding intensity is maintained above a threshold for a certain period of time When the derivation formula is used to monitor the fluctuation rate of feeding intensity: when the absolute value of the fluctuation rate y`(k) is greater than the preset value, it is judged that the fish is in a normal feeding state; when the absolute value of the fluctuation rate is less than the preset value , judge that the fish is in anoxic state, and control the circulation pump to start; after the circulation pump is started, the feeding intensity drops to the threshold value and maintains it for a certain period of time, and the circulation pump is controlled to close; when the feeding intensity is not within the preset feeding time, the feeding intensity And if the absolute value of the fluctuation rate is greater than the preset value, it is determined that the time corresponds to the physiological feeding time of the cultured fish group; the preset feeding time is corrected by using the physiological feeding time.

The method and system for culturing swimming fish based on feeding intensity measurement of the invention has the advantages that the feeding intensity of fish can be quantitatively described, and the feeding intensity can be used as an important reference standard for feed feeding. The feeding amount and feeding time of different breeding batches are no longer dependent on empirical formulas, and can be corrected based on changes in feeding intensity. With the growth of fish and changes in the number of mantissas in the breeding process, corresponding adjustments can be made in time, and the intelligence of breeding production has been significantly improved.

Further, by cleverly using the feeding intensity and the corresponding fluctuation rate as the judgment basis, the hypoxia phenomenon of the fish group was detected early, which greatly reduced the occurrence of death accidents. Finally, the feeding intensity can be used to obtain the physiological characteristics of different fish, so as to understand the physiological feeding time as soon as possible. Regardless of the fish that forage at night or during the day, a common breeding system can be used, and automatic corrections are made based on the physiological feeding time.

The breeding method and the corresponding breeding system have a wide range of applications, simple data operation, and very low hardware requirements. Somatosensory cameras mainly use infrared light time-of-flight for sampling, so the entire breeding process does not depend on light conditions and does not affect the physiology of fish.

Description of drawings

FIG. 1 is a schematic diagram of the structure of the breeding system for swimming fish cultured fish based on the measurement of feeding intensity according to the present invention.

Fig. 2 is a time chart of feeding intensity after feeding.

FIG. 3 is a time chart of the fluctuation change rate of the feeding intensity corresponding to the time shown in FIG. 2 .

Figure 4 is a time chart of feeding intensity in fish hypoxia.

FIG. 5 is a time chart of the fluctuation change rate of feeding intensity corresponding to time shown in FIG. 4 .

Fig. 6 is a time chart of feeding intensity after multiple feedings.

Reference numerals: 1-culture pond, 2-somatosensory camera, 3-host computer, 4-feeding device, 5-filter tank, 6-circulation pump. The corresponding time of t1 is 10:29:21, the corresponding time of t2 is 10:29:31, the corresponding time of t3 is 10:29:41, the corresponding time of t4 is 10:29:51, and the corresponding time of t5 is 10:30:01. The corresponding time of t6 is 05:18:00, the corresponding time of t7 is 05:18:50, the corresponding time of t8 is 05:19:40, the corresponding time of t9 is 05:20:30, and the corresponding time of t10 is 05:21:20. Region a is the data set of the first feeding, region b is the data set of the first foraging state, region c is the data set of the second feeding, and region d is the data set of the second foraging state.

Detailed ways

How to quantify and program the physiological characteristics of fish is an important technical problem to improve intelligent breeding. The invention quantitatively describes the feeding intensity of one of the physiological characteristics of fish, and utilizes the descriptable physiological characteristics for breeding management application. During the feeding process, swimming fish will surface to grab food, then dive, and then surface to grab food and then dive in a continuous cycle, until the degree of satiety is high, which will gradually reduce the frequency and desire of grabbing food. The behavior of swimming fish rushing to eat is the technical basis of the present invention, and the fish feeding intensity is judged by the somatosensory camera based on the information of the fish appearing and appearing on the water surface.

The somatosensory camera is used to obtain the depth map. The somatosensory camera belongs to the mature prior art. The present invention selects one of them for use, and adopts the KINECT 2.0 depth camera produced by Microsoft Corporation, but does not exclude the application of other depth images with infrared functions. The technical solution of the camera.

In this embodiment, the resolution of the depth map is set to 424*512, that is, m is set to 424, n is set to 512, and the coordinate value ranges of the pixels in the depth map are x∈[0,424-1], y∈[0,512- 1]. The somatosensory camera is set at a position of 850mm above the water surface, and the effective range for shooting is set from 500mm to 910mm. Among them, 500mm corresponds to the minimum shooting distance of the somatosensory camera, and 910mm corresponds to the distance of 60mm underwater. The water surface specifically refers to the water surface of the culture pond.

The fish adopts common carp in the southern market as the sample collection object, and the carp fry is provided by Yulin Xinjian Planting and Breeding Co., Ltd., the fry is 6-10cm in length, and the color is gray-brown. In the actual breeding process, the body size and body color of the cultured fish may not be completely consistent. In order to show the actual breeding scene and to verify the reliability of the system in complex situations, this example is used in the gray-brown carp population. 10% koi with different colors were added together. The length of the koi is about 15cm, which is slightly larger than the size of other carp fry. Since they all belong to the Cyprinidae family, the co-culture group did not show robbing. Food fighting behavior. The culturing method and system of the present invention are suitable for various single-species culturing or mixed culturing. In this example, the co-cultivation group was cultivated in the cultivation pond for 30 days, which was fully adapted to the current cultivation environment. The hardware configuration used is a Lenovo brand machine, and the hardware configuration is higher than the basic requirements required by KINECT 2.0. The operating system is WINDOWS 10, and the CPU adopts a 64-bit processor. The software processing platform used is Microsoft's C#, and the database background uses ACCESS to store experimental data. The cultivation for 30 days to adapt to the current cultivation environment is only for the purpose of simplifying the extraction and analysis of the data chart of the technical solution in this embodiment. Due to the self-correction function of this program, it can be automatically operated and monitored at the beginning of the breeding and production, and the monitoring and analysis of feeding intensity can be carried out without prior breeding in actual production.

As shown in FIG. 1 , a breeding system for swimming type cultured fish based on the measurement of feeding intensity according to the present invention includes a culture pond, a somatosensory camera arranged above the culture pond, a feeding device for feeding feed, a circulating pump and an upper machine. The circulating pump is connected to the culture tank through a pipeline. The upper computer is respectively electrically connected with the somatosensory camera, the feeding device and the circulating pump. The somatosensory camera is used to shoot the depth map and transmit it to the host computer. The feeding device is used to feed the feed according to the control of the host computer. The circulating pump is used to start or stop according to the control of the host computer. The upper computer is used to realize the operation of each computer program in the breeding method, and send corresponding control signals to the circulating pump and/or the feeding device according to the operation result.

完成，公式中的500和910单位均为mm。除了提取水下距离之外，还提取水上距离，目的在于使图像更准确还原实际情况，因为鱼类抢食的时候会有部分区域高于水面。After obtaining the depth map, use the depth value as a comparison parameter, and perform binarization f(x,y) processing on each pixel in the depth map: the pixel whose depth value is located within a certain valid interval above and below the water surface is recorded as 1, otherwise it is recorded as 1. is 0. Binarization by the formula

Done, both 500 and 910 in the formula are in mm. In addition to extracting the underwater distance, the water distance is also extracted, the purpose is to make the image more accurately restore the actual situation, because some areas will be higher than the water surface when the fish are rushing to eat.

计算得到。其意义为将深度在有效区间内的所有像素点，以逐行逐列扫描的方式计算总和，以该总和为摄食强度的具体数值。当鱼类进入有效区间内，深度图中的鱼类或池壁、饲料对应区域的所有像素点会被赋值为1，反之没有鱼类或饲料的区域对应像素点会被赋值为0。鱼类在非缺氧或没有抢食欲望的时候会潜游在水下60mm的位置，此时y(k)的值会保持在比较稳定的区间，如图2中y(k)在t2之前维持在23000±200的范围内。没有投食的情况下依然有一定摄食强度，是因为体感摄像机拍摄到养殖池池壁，而微小波动变化是水波造成的。在t2时刻投食后一两秒内，鱼类出现明显的抢食行为，在t3前后5秒的时刻分别出现了高峰，y(k)接近34000。随着饲料的消耗，鱼类抢食现象减弱，t4时刻开始逐渐下降。但鱼类此时并不饱食，因此还有鱼类重新上浮觅食，因此t5时刻后的摄食强度略有上升。图2很好地展示出鱼类抢食行为和摄食强度的对应关系，也表明了摄食强度的设置能直接反馈出鱼类的真实行为。The feeding intensity obtained at time k is recorded as y(k), which is calculated by the formula

Calculated. Its meaning is to calculate the sum of all the pixels whose depth is within the valid interval in a scanning manner of row by column, and use the sum as the specific value of feeding intensity. When the fish enters the valid range, all the pixels in the area corresponding to the fish or the pool wall and feed in the depth map will be assigned a value of 1, otherwise, the corresponding pixels of the area without fish or feed will be assigned a value of 0. When the fish are not hypoxic or have no desire to grab food, they will swim at a position of 60mm underwater. At this time, the value of y(k) will remain in a relatively stable range. In Figure 2, y(k) is before t2. maintained within the range of 23000±200. In the case of no feeding, there is still a certain feeding intensity, because the somatosensory camera captures the wall of the breeding pond, and the small fluctuations are caused by water waves. Within a second or two after feeding at t2, the fish showed obvious predation behavior, and peaked at 5 seconds before and after t3, and y(k) was close to 34000. With the consumption of feed, the phenomenon of fish grabbing weakened, and it began to decline gradually at time t4. However, the fish were not full at this time, so some fish re-floated for food, so the feeding intensity after t5 increased slightly. Figure 2 shows the corresponding relationship between fish prey behavior and feeding intensity, and also shows that the setting of feeding intensity can directly feedback the real behavior of fish.

At a certain moment, the feeding intensity y(k) showed a high value, which may be caused by the lack of oxygen for fish to float to the surface. Hypoxia is particularly prone to occur at night, and human involvement during this time is significantly lower than during the day. If hypoxia is not accurately monitored, it can easily lead to widespread death. When the fish is hypoxic, as shown in Figure 4, the feeding intensity y(k) is significantly higher than 23000 in the non-preemptive state, and maintained in the range of 28000±2000. Therefore, there is still room for improvement in the analysis of whether or not to grab food based on feeding intensity alone, although hypoxia does not necessarily occur.

对摄食强度的时刻表曲线进行求导。分别得到图2的对应波动变化率时刻表，如图3所示。得到图4对应的波动变化率时刻表，如图5所示。从图3中可见，当鱼类在t2后一两秒出现抢食时，波动变化率y'(k)最高点逼近5000，最低点逼近-4000。而鱼类缺氧时，波动变化率如图5所示，仅维持在28000正负不到2000的区间。本实施例中将缺氧判断的预设值设置为30000，即波动变化率的绝对值小于30000就判断为缺氧。上位机发现鱼类缺氧后，会控制循环泵启动，使得养殖池内的氧分得到补充。同时监控到鱼类获取足够氧分再次下潜水底时，控制循环泵关闭节省能量。In order to more accurately analyze fish behavior through numerical values, the present invention also introduces a method of taking the derivative of the time-to-moment change of feeding intensity, so as to further screen whether the fish are in a predation or anoxic state. In this embodiment, time k is discretized data, so the discretized derivative formula is used

Derivation of the timetable curve of feeding intensity. The corresponding volatility change rate timetables in Figure 2 are obtained respectively, as shown in Figure 3. The timetable of volatility rate of change corresponding to Figure 4 is obtained, as shown in Figure 5. It can be seen from Figure 3 that when the fish scramble for food one or two seconds after t2, the highest point of the fluctuation rate y'(k) is close to 5000, and the lowest point is close to -4000. When the fish is hypoxic, the fluctuation rate is shown in Figure 5, which is only maintained in the range of 28,000 plus or minus less than 2,000. In this embodiment, the preset value of the determination of hypoxia is set to 30000, that is, the absolute value of the fluctuation rate of change is less than 30000, and it is determined to be hypoxia. When the host computer finds that the fish is deficient, it will control the circulation pump to start, so that the oxygen in the breeding pond can be replenished. At the same time, when it is monitored that the fish has obtained enough oxygen to go down to the bottom again, the circulating pump is controlled to turn off to save energy.

It shows that the breeding method of the present invention can analyze and apply fish feeding behavior by feeding intensity, and can analyze and apply feeding or hypoxia after adding fluctuation rate.

The main control computer monitors the feeding intensity of the fish in real time, when the feeding intensity is maintained below the threshold for a certain period of time. In this example, the specified time is set to 10 minutes, and the threshold is set to 24000, indicating that the fish is not in an oxygen-deficient state. When the preset feeding time is reached, the feeding device is controlled to feed. After the feeding intensity decreases with time, it rises again to the set proportion of the feeding intensity at the feeding time, and a new round of reduced feeding is performed; this step is repeated until the feeding intensity decreases with time and cannot rise to the feeding intensity at the feeding time. the set ratio, stop feeding. The set ratio is set to 90% in this embodiment. The parameters such as the setting ratio, monitoring time and threshold value can be adjusted by those skilled in the art according to common knowledge according to different conditions of different fish, individual size or co-cultivation group, without the need for creative work.

Due to the existence of the inherent pool wall, in the absence of predation and hypoxia, when all the fish dive to the bottom, the value of y(k) will still float around 23000. Therefore, the ratio judgment in the circulating feeding should be understood as the amount increased on the basis of 23000 as the judgment basis. At the same time, in order to make the illustration clearer and simplify the expression, the values between 23000 and 28500 are deleted in Figure 6. After the first feeding, the feeding intensity y(k) increased from 23,000 to 34,500. With the passage of time, the feed decreased and the snatch intensity decreased significantly. However, at this time, the fish did not enter a state of satiation, and would re-migrate to the water surface to feed, with a peak of 34,000 in area b. (34000-23000) ÷ (34500-23000) × 100% = 95.65%, which meets the requirements of secondary reduction feeding. After the second feeding, the feeding intensity y(k) reached a peak value of 33,000. With the passage of time, the fish migrated to the water surface to feed, and a peak value of 31,400 appeared in the area d. (33000-23000)÷(31400-23000)×100%=84%, which does not meet the requirement of feeding again.

During the breeding cycle, the host computer found that the co-cultivation group described in the example had a strong appetite for food from 8:00 to 10:00 in the morning. That is, this period of time is not within the feeding time, the feeding intensity is higher than the threshold value, and the absolute value of the fluctuation rate is greater than the preset value, it is determined that this time is the physiological feeding time of the corresponding cultured fish group. Adjust the first feeding time to start at 8 am.

When the feeding intensity exceeds the threshold, the host computer will also store the time, feeding intensity and corresponding depth value in the database in the background in real time. The database uses Microsoft's ACCESS platform, ACCESS creates relational data tables, and the C# program writes the data exceeding the set threshold into the table one by one, and the data can be retrieved at any time for analysis and statistics. When analyzing, export the data table to EXCLE format, and use various statistical functions of EXCEL to mine rich feeding information, and then draw the corresponding data map to realize the analysis of fish behavior.

For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all these changes and deformations should fall within the protection scope of the claims of the present invention.

Claims (10)

1. A swimming type cultured fish culture method based on ingestion intensity measurement comprises the following steps:

acquiring a depth map with the resolution of m x n by using a somatosensory camera; taking the depth value as a contrast parameter, and carrying out binarization f (x, y) processing on each pixel in the depth map: marking the pixel with the depth value positioned in a certain effective interval above and below the water surface as 1, otherwise, marking as 0;

the method is characterized by further comprising the following steps:

monitoring the feeding intensity y (k) of the fish herd by using a formula

Obtaining, wherein k is the obtaining time of the depth map;

when the ingestion intensity y (k) is kept below a threshold value within a certain time and reaches the preset feeding time, feeding the feed;

continuously monitoring the feeding intensity, and increasing the feeding intensity to the set proportion of the feeding intensity at the feeding moment again after the feeding intensity is reduced along with the time, and performing a new round of reduced feeding; the step is circulated until the feeding intensity can not rise to the set proportion of the feeding intensity at the feeding moment after the feeding intensity is reduced along with the time, and the feed feeding is stopped.

2. The swimming type cultured fish farming method based on feeding intensity measurement according to claim 1, wherein when the feeding intensity is maintained above a threshold value for a certain period of time, the fluctuation rate of the feeding intensity is monitored using a derivation formula: when the absolute value of the fluctuation change rate y' (k) is larger than a preset value, judging that the fish shoal is in a normal feeding state; and when the absolute value of the fluctuation change rate is smaller than a preset value, judging that the fish school is in an anoxic state, and controlling the circulating pump to start.

3. The cultivation method of swimming type cultured fish based on feeding intensity measurement as claimed in claim 2, characterized in that after the circulation pump is started, the feeding intensity is reduced to a threshold value and maintained for a certain time, and the circulation pump is controlled to be turned off.

4. The cultivation method of swimming type cultivated fish based on feeding intensity measurement according to claim 2, characterized in that when the feeding intensity is not within the preset feeding time, the feeding intensity is higher than the threshold value, and the absolute value of the fluctuation change rate is larger than the preset value, the time is judged as the physiological feeding time of the corresponding cultivated fish; and correcting the preset feeding time by using the physiological feeding time.

5. The cultivation method of the swimming type farmed fish based on the feeding intensity measurement as claimed in claim 1, characterized in that the effective interval is 500mm to 60mm under water below the somatosensory camera; and the somatosensory camera is arranged above the water surface by not less than 500 mm.

6. The swimming farmed fish of claim 1, the certain period of time being 10 minutes, based on feeding intensity measurements.

7. The swimming farmed fish based on feeding intensity measurement as claimed in claim 1, characterized in that when the feeding intensity exceeds a threshold value, the time, the feeding intensity and the corresponding depth value are recorded.

8. The swimming farmed fish of claim 1, the farming method based on feeding intensity measurement, wherein the set ratio is 90%.

9. A swimming type cultured fish culture system based on ingestion intensity measurement comprises a culture pond, a somatosensory camera arranged above the culture pond, a feeding device used for feeding feed and an upper computer; the cultivation method is characterized in that the upper computer controls the motion sensing camera and the feeding device to work by applying the cultivation method as claimed in claim 1, 5, 6, 7 or 8.

10. The swimming type cultured fish feeding system based on feeding intensity measurement according to claim 9, which comprises a circulating pump, wherein the circulating pump is connected with the culture pond through a pipeline and is electrically connected with the upper computer; when the upper computer detects that the ingestion intensity is maintained above a threshold value within a certain time, the fluctuation change rate of the ingestion intensity is monitored by using a derivation formula: when the absolute value of the fluctuation change rate y' (k) is larger than a preset value, judging that the fish shoal is in a normal feeding state; when the absolute value of the fluctuation change rate is smaller than a preset value, judging that the fish school is in an anoxic state, and controlling a circulating pump to start; after the circulating pump is started, the ingestion intensity is reduced to a threshold value and is maintained for a certain time, and the circulating pump is controlled to be closed; when the feeding intensity is not within the preset feeding time, the feeding intensity is higher than the threshold value, and the absolute value of the fluctuation change rate is larger than the preset value, the time is judged to be the physiological feeding time corresponding to the cultured fish shoal; and correcting the preset feeding time by using the physiological feeding time.

Images