A method for intelligently regulating fish growth

Abstract

The present invention discloses a method for intelligently regulating fish growth. The method uses a multi-sensor wireless water quality monitoring module to monitor the physical and chemical index data of water quality in a pond and transmits it to a cloud platform. The acquired data is put into a support vector machine model to establish a prediction model for a specific growth rate. Then, a corresponding control instruction set is issued based on the prediction result of the specific growth rate of farmed fish, so that the execution control module controls the corresponding controlled device to execute according to the received control instruction set, thereby realizing the automation and intelligence of pond farming. The present invention establishes a mathematical model of the correlation between fish growth and physical and chemical factors such as temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity. Then, the execution mechanism at different stages is controlled based on the fish growth control model, and the control result is evaluated and fed back to form a closed-loop control system, thereby ensuring the stability of the water environment, reducing resource consumption, promoting fish growth, shortening the growth cycle, and greatly improving the farming efficiency.

Description

Technical Field

[0001] The invention relates to the technical field of aquaculture, and in particular to a method for intelligently regulating fish growth. Background Art

[0002] Aquaculture is greatly affected by environmental changes. It is necessary to collect corresponding environmental information and dynamic information on aquaculture status in a timely manner to achieve environmental prediction and take appropriate measures to ensure the aquaculture environment. However, producers currently lack real-time information collection technology and means, and most of them rely on manual collection, which has high collection costs, poor data accuracy, and difficult maintenance. The analysis of sensor data is still mainly based on single data processing. There is no fusion analysis of water quality physical and chemical indicators, and no correlation analysis and modeling with biological growth, which makes water quality data still extensive threshold control and cannot be coordinated with multiple variables according to specific aquaculture organisms. Therefore, there are deficiencies in the degree of intelligence of aquaculture, the degree of informatization of aquaculture management, intelligent equipment, operating energy consumption, water resource recycling, green and efficient, etc. It is urgent to carry out technical equipment improvement, optimization integration and model demonstration to achieve precise energy saving and intelligent control. Summary of the invention

[0003] In order to overcome the above-mentioned shortcomings and deficiencies of the prior art, the purpose of the present invention is to provide an intelligent regulation method for fish growth, which realizes the precise dynamic regulation of the healthy growth of fish, ensures the stability of the water environment, reduces resource consumption, and improves the economic benefits of aquaculture.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] A method for intelligent regulation of fish growth is based on an intelligent regulation system for fish growth, wherein the intelligent regulation system for fish growth comprises a communication and data wireless transmission module, a cloud processing platform, an execution regulation module and at least one multi-sensor wireless water quality monitoring module; the execution regulation module is connected to the cloud processing platform via the communication and data wireless transmission module; the multi-sensor wireless water quality monitoring module is connected to the cloud processing platform via the communication and data wireless transmission module;

[0006] The multi-sensor wireless water quality monitoring module includes a data processing module, an analog-to-digital conversion module, a microprocessor, an image acquisition module, a water temperature sensor, a dissolved oxygen sensor, an ammonia nitrogen sensor, a nitrite nitrogen sensor, a pH sensor and a salinity sensor;

[0007] The image acquisition module, water temperature sensor, dissolved oxygen sensor, ammonia nitrogen sensor, nitrite nitrogen sensor, pH sensor and salinity sensor are respectively connected to the data processing module;

[0008] The data processing module, analog-to-digital conversion module, and microprocessor are connected in sequence;

[0009] The execution control module includes at least one pond control module; the pond control module includes a controller, a relay group and a controlled device, and the controlled device includes a temperature adjustment unit, a dissolved oxygen adjustment unit, an ammonia nitrogen adjustment unit, a nitrite nitrogen adjustment unit, a pH adjustment unit, and a salinity adjustment unit;

[0010] The temperature regulating unit, dissolved oxygen regulating unit, ammonia nitrogen regulating unit, nitrite nitrogen regulating unit, pH regulating unit and salinity regulating unit are respectively connected to the controller via a relay group;

[0011] The fish growth intelligent regulation method comprises the following steps:

[0012] The S1 multi-sensor wireless water quality monitoring module sends the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity in the pond at n historical collection times, and normalizes them by dimension; n is greater than or equal to 1;

[0013] S2: the data processing module performs noise reduction processing on the collected sensor data to obtain an analog signal of the processed monitoring data;

[0014] S3: The analog-to-digital conversion module converts the analog signals output by each sensor into digital signals, and transmits the digital signals to the microprocessor;

[0015] S4 The microprocessor receives water temperature information sent by the water temperature sensor, dissolved oxygen information sent by the dissolved oxygen sensor, ammonia nitrogen information sent by the ammonia nitrogen sensor, nitrite nitrogen information sent by the nitrite nitrogen sensor, pH value sent by the pH sensor, salinity information sent by the salinity sensor, and fish image information sent by the image acquisition module, and sends the received temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, salinity, and fish image information to the cloud processing platform;

[0016] S5: The cloud processing platform receives the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, salinity and fish image information in the pond at n historical collection moments sent by the multi-sensor wireless water quality monitoring module, compares the size of the fish in the received image information with the set multiple standard fish image information one by one, screens out the standard fish image information that matches the size of the fish in the received pond image information, and calculates the average mass of the fish in each image according to the weight of the fish in the standard fish image information;

[0017] S6 The cloud processing platform obtains the specific growth rate SGR of the fish body according to the image information data;

[0018] The cloud processing platform S7 uses the least squares support vector machine to establish a prediction model for the specific growth rate SGR, and the model equation is specifically:

[0019] Y SGR =C0+aX1+bX2+cX3+dX4+eX5+fX6

[0020] Among them, Y SGR is the specific growth rate of fish; X1, X2, X3, X4, X5, and X6 correspond to temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity, respectively; C0, a, b, c, d, e, and f are all constants;

[0021] When SGR value is ≤10%, C0=-0.1947, a=0.0016, b=0.0076, c=0.38, d=0.01, e=0.021, f=0.21;

[0022] When 10%<SGR value≤30%, C0=-0.3285, a=0.0028, b=0.0091, c=0.27, d=0.0076, e=0.036, f=0.25;

[0023] When SGR value>30%, C0=-0.1563, a=0.0021, b=0.013, c=0.22, d=0.0034, e=0.029, f=0.36;

[0024] S8 The cloud processing platform predicts the specific growth rate of the current farmed fish based on the prediction model and the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity of the current farmed fish in the pond at the current moment, and then issues a corresponding control instruction set so that the execution control module controls the corresponding controlled device to execute according to the received control instruction set.

[0025] Preferably, the fish growth intelligent regulation method further comprises the following steps:

[0026] When the multi-sensor wireless water quality monitoring module detects that the temperature in the aquaculture pond exceeds 30°C or is lower than 8°C, the dissolved oxygen content is lower than 3mg / L or higher than 8mg / L, the pH is lower than 5.5 or higher than 9, and the salinity is higher than 5%, the cloud processing platform sends a warning signal to the external client and the user through the communication and data wireless transmission module.

[0027] Preferably, the prediction root mean square error RMSEP of the prediction model is less than 0.05, and the prediction determination coefficient R 2 >0.95.

[0028] Preferably, the cloud processing platform in step S6 obtains the specific growth rate SGR of the fish body according to the image information data, specifically:

[0029] in, represents the average mass of fish at the end of the measurement period, represents the average mass of fish at the beginning of the measurement period; t represents the number of days in a measurement period.

[0030] Preferably, the data processing module in step S2 performs noise reduction processing on the collected sensor data, specifically:

[0031] The noise reduction random filter processing model is used to reduce the interference of noise on the collected data through multidimensional regression algorithm.

[0032] Preferably, the average mass of the fish in each image is calculated in step S5, specifically:

[0033] The weight of fish in the image information collected by the multi-sensor wireless water quality monitoring module is quantified to obtain the weight set Mn(m1,m2,...,mk) of the fish in the pond, where k is the number of fish in each image. is the average mass of the fish in each image.

[0034] Preferably, the communication and data wireless transmission module performs wireless bidirectional transmission via 4G / NBIOT.

[0035] Preferably, the fish growth intelligent regulation system further includes an external client, and the external client is connected to the cloud processing platform via a wired or wireless network.

[0036] Preferably, the pond control module also includes a monitoring display screen.

[0037] Preferably, the analog-to-digital conversion module in step S3 converts the analog signals output by each sensor into digital signals, and transmits the digital signals to the microprocessor, specifically:

[0038] The analog signals output by each sensor are converted into 485 digital signals, and the 485 digital signals are transmitted to the microprocessor.

[0039] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0040] (1) The present invention provides an intelligent fish growth control method, which detects the environment in the pond and collects image information in the pond, processes the detected environmental parameters and image information, and establishes a mathematical model of the correlation between fish growth and physical and chemical factors such as temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity. Then, the fish growth control model is used to control the actuators at different stages, and the control results are evaluated and fed back to form a closed-loop control system, which ensures the stability of the water environment, reduces resource consumption, promotes fish growth, shortens the growth cycle, and greatly improves the breeding efficiency.

[0041] (2) After verifying the correlation between ammonia nitrogen and nitrite nitrogen and specific growth rate, the present invention introduces ammonia nitrogen and nitrite nitrogen as direct indicators of water quality, further improving the regulation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 The figure is a schematic diagram of the intelligent control system for fish growth of an embodiment of the present invention. In the figure, the components indicated by the reference numerals are as follows: 1. Data processing module, 2. Analog-to-digital conversion module, 3. Microprocessor, 4. Image acquisition module, 5. Water temperature sensor, 6. Dissolved oxygen sensor, 7. Ammonia nitrogen sensor, 8. Nitrite nitrogen sensor, 9. pH sensor, 10. Salinity sensor, 11. Power module, 12. Communication and data wireless transmission module, 13. Cloud processing platform, 14. Execution control module, 15. Temperature control unit, 16. Dissolved oxygen control unit, 17. Ammonia nitrogen control unit, 18. Nitrite nitrogen control unit, 19. pH control unit, 20. Salinity control unit, 21. Peripheral client.

[0043] Figure 2 This is the effect of ammonia nitrogen content on the specific growth rate in the examples of the present invention.

[0044] Figure 3 This is the effect of nitrite nitrogen content on specific growth rate in the examples of the present invention. DETAILED DESCRIPTION

[0045] The present invention will be further described in detail below in conjunction with the examples, but the embodiments of the present invention are not limited thereto.

[0046] like Figure 1As shown, an intelligent fish growth control system based on multi-sensor measurement of the present invention includes a plurality of multi-sensor wireless water quality monitoring modules, a power module 11, a communication and data wireless transmission module 12, a cloud processing platform 13 and an execution control module 14, wherein the power module 11 is used to power each module in the system, the multi-sensor wireless water quality monitoring module is connected to the cloud processing platform 13 via the communication and data wireless transmission module 12, and the execution control module 14 is connected to the cloud processing platform 13 via the communication and data wireless transmission module 12.

[0047] The multi-sensor wireless water quality monitoring module is arranged at different positions of the pond, and is used to monitor the physical and chemical index data of water quality in the pond and output monitoring signals, including a data processing module 1, an analog-to-digital conversion module 2, a microprocessor 3, an image acquisition module 4, a water temperature sensor 5, a dissolved oxygen sensor 6, an ammonia nitrogen sensor 7, a nitrite nitrogen sensor 8, a pH sensor 9 and a salinity sensor 10.

[0048] The data processing module 1 is used to process the collected sensor data using a noise reduction random filtering processing model and a multidimensional regression algorithm to further reduce the interference of noise on the collected data, so as to obtain a high-precision analog signal of the monitoring data; the analog-to-digital conversion module 2 directly performs a 485 interface design, converts the analog signal output by each sensor into a 485 digital signal, and transmits the 485 digital signal to the microprocessor.

[0049] The microprocessor 3 is used to receive water temperature information sent by the water temperature sensor 5, dissolved oxygen information sent by the dissolved oxygen sensor 6, ammonia nitrogen information sent by the ammonia nitrogen sensor 7, nitrite nitrogen information sent by the nitrite nitrogen sensor 8, pH value sent by the pH sensor 9, salinity information sent by the salinity sensor 10 and grass carp image information sent by the image acquisition module 4, and send the received temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, salinity and grass carp fish body image information to the cloud processing platform 13.

[0050] The image acquisition module 4 and multiple sensors can collect the growth image information, temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH and salinity of grass carp in the pond in real time. Among them, ammonia nitrogen and nitrite nitrogen are introduced as direct indicators of water quality, and their toxicity is closely related to the pH value and temperature of the aquaculture pond.

[0051] The communication and data wireless transmission module 12 performs wireless bidirectional transmission via 4G / NBIOT. On the one hand, it receives the water quality model control requirements of the cloud processing platform, and on the other hand, it transmits the control status and feedback signal of the controlled equipment to the cloud processing platform to realize wireless intelligent control.

[0052] The execution control module 14 includes a controller, a relay group, a monitoring display screen, and a controlled device. The controlled device includes: a temperature adjustment unit 15, a dissolved oxygen adjustment unit 16, an ammonia nitrogen adjustment unit 17, a nitrite nitrogen adjustment unit 18, a pH adjustment unit 19, and a salinity adjustment unit 20. The temperature adjustment unit 15, the dissolved oxygen adjustment unit 16, the ammonia nitrogen adjustment unit 17, the nitrite nitrogen adjustment unit 18, the pH adjustment unit 19, and the salinity adjustment unit 20 are connected to the relay group. The controller controls the temperature adjustment unit 15, the dissolved oxygen adjustment unit 16, the ammonia nitrogen adjustment unit 17, the nitrite nitrogen adjustment unit 18, the pH adjustment unit 19, and the salinity adjustment unit 20 to achieve precise adjustment of the aquaculture pond.

[0053] The fish growth intelligent control method of this embodiment comprises the following steps:

[0054] The S1 multi-sensor wireless water quality monitoring module sends the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity in the pond at n historical collection moments and normalizes them by dimension.

[0055] S2 The analog-to-digital conversion module converts the analog signals output by each sensor into 485 digital signals, and transmits the 485 digital signals to the microprocessor.

[0056] S3 The microprocessor is used to receive water temperature information sent by the water temperature sensor, dissolved oxygen information sent by the dissolved oxygen sensor, ammonia nitrogen information sent by the ammonia nitrogen sensor, nitrite nitrogen information sent by the nitrite nitrogen sensor, pH value sent by the pH sensor, salinity information sent by the salinity sensor and grass carp image information sent by the image acquisition module, and sends the received temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, salinity and grass carp body image information to the cloud processing platform.

[0057] S4 The cloud processing platform receives the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, salinity and grass carp body image information in the pond sent by the multi-sensor wireless water quality monitoring module at n historical collection moments, compares the size of the grass carp in the received image information with the set multiple standard grass carp image information one by one, and selects the standard grass carp image information that matches the size of the grass carp in the received pond image information. According to the weight of the grass carp in the standard grass carp image information, the weight of the grass carp in the image information collected by the multi-sensor wireless water quality monitoring module is quantified to obtain the weight set Mn (m1, m2, ..., mk) of the grass carp in the pond, where k is the number of grass carp in each image. is the average mass of grass carp in each image.

[0058] S5: the cloud processing platform obtains the specific growth rate (SGR) of the grass carp body according to the image information data. in, represents the average mass of grass carp at the end of the measurement period, represents the average mass of grass carp at the beginning of the measurement period; t represents the number of days in a measurement period. Specifically, the number of days in the measurement period is 7 days.

[0059] S6 The cloud processing platform uses the least squares support vector machine (LS-SVM) to establish a prediction model for specific growth rate (SGR), and the model equation is specifically:

[0060] Y SGR =C0+aX1+bX2+cX3+dX4+eX5+fX6

[0061] Among them, Y SGR is the specific growth rate of grass carp; X1, X2, X3, X4, X5, and X6 are the corresponding temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity respectively; C0, a, b, c, d, e, and f are all constants;

[0062] When SGR value is ≤10%, C0=-0.1947, a=0.0016, b=0.0076, c=0.38, d=0.01, e=0.021, f=0.21;

[0063] When 10%<SGR value≤30%, C0=-0.3285, a=0.0028, b=0.0091, c=0.27, d=0.0076, e=0.036, f=0.25;

[0064] When the SGR value is greater than 30%, C0=-0.1563, a=0.0021, b=0.013, c=0.22, d=0.0034, e=0.029, f=0.36.

[0065] Furthermore, the prediction root mean square error RMSEP of the support vector machine model is less than 0.05, and the prediction determination coefficient R 2 Specifically, the prediction model accuracy established in this embodiment is the prediction root mean square error RMSEP = 1.8%, and the prediction determination coefficient R 2 =0.968.

[0066] The cloud processing platform described in S7 predicts the specific growth rate prediction result of farmed fish based on the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity of the fish currently farmed in the pond at the current moment based on the support vector machine model, and then issues a corresponding control instruction set based on the specific growth rate prediction result of farmed fish, so that the execution control module controls the corresponding controlled device to execute according to the received control instruction set, thereby realizing the automation and intelligence of pond farming. Specifically, the corresponding relationship between the preset specific growth rate prediction result of farmed grass carp and the control amount of the controlled device is set according to the grass carp in the pond.

[0067] At the same time, the cloud processing platform obtains real-time monitoring data of water quality physical and chemical indicators, and detects that the temperature of the aquaculture pond is too high over 30°C or too low under 8°C, the dissolved oxygen content is lower than 3mg / L or higher than 8mg / L, the pH is lower than 5.5 or higher than 9, and the salinity is higher than 5%. It will send a warning signal to the external client and the user through the communication and data wireless transmission module. Specifically, a warning signal will be issued if the ammonia nitrogen content in the grass carp aquaculture pond is lower than 1.0mg / L and the nitrite nitrogen content is higher than 0.15mg / L.

[0068] The execution and control module described in S8 controls the pond according to the optimization calculation results of the cloud processing platform or the parameters set independently by the peripheral client. The target customers include pond managers, pond security personnel, pond maintenance personnel, breeding workers and other relevant pond breeding personnel. The peripheral client includes PC, notebook, tablet and other peripherals. The peripheral client is connected to the end processing platform via a wired or wireless network.

[0069] The present invention provides an intelligent fish growth control system based on multi-sensor measurement, which detects the environment in the pond and collects image information in the pond, processes the detected environmental parameters and image information, establishes a mathematical model of the correlation between fish growth and physical and chemical factors such as temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, salinity, etc., and then controls the actuators at different stages according to the fish growth control model, evaluates and feeds back the control results, forms a closed-loop control system, ensures the stability of the water environment, reduces resource consumption, promotes fish growth, shortens the growth cycle, and greatly improves the breeding efficiency.

[0070] This example experiments on the effects of ammonia nitrogen and nitrite nitrogen on specific growth rate (large yellow croaker and grass carp), the results are as follows Figure 2 and 3 As shown, it can be seen that there is a correlation between ammonia nitrogen and nitrite nitrogen in fish specific growth rate (large yellow croaker and grass carp), and different types of fish have different tolerance to ammonia nitrogen and nitrite nitrogen in water, and the concentration range of ammonia nitrogen and nitrite nitrogen in water needs to be strictly controlled.

[0071] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for intelligently regulating fish growth, characterized in that: Based on the intelligent control system for fish growth, the intelligent control system for fish growth includes a communication and data wireless transmission module, a cloud processing platform, an execution control module and at least one multi-sensor wireless water quality monitoring module; the execution control module is connected to the cloud processing platform through the communication and data wireless transmission module; the multi-sensor wireless water quality monitoring module is connected to the cloud processing platform through the communication and data wireless transmission module; The multi-sensor wireless water quality monitoring module includes a data processing module, an analog-to-digital conversion module, a microprocessor, an image acquisition module, a water temperature sensor, a dissolved oxygen sensor, an ammonia nitrogen sensor, a nitrite nitrogen sensor, a pH sensor and a salinity sensor; The image acquisition module, water temperature sensor, dissolved oxygen sensor, ammonia nitrogen sensor, nitrite nitrogen sensor, pH sensor and salinity sensor are respectively connected to the data processing module; The data processing module, analog-to-digital conversion module, and microprocessor are connected in sequence; The execution control module includes at least one pond control module; the pond control module includes a controller, a relay group and a controlled device, and the controlled device includes a temperature adjustment unit, a dissolved oxygen adjustment unit, an ammonia nitrogen adjustment unit, a nitrite nitrogen adjustment unit, a pH adjustment unit, and a salinity adjustment unit; The temperature regulating unit, dissolved oxygen regulating unit, ammonia nitrogen regulating unit, nitrite nitrogen regulating unit, pH regulating unit and salinity regulating unit are respectively connected to the controller via a relay group; The fish growth intelligent regulation method comprises the following steps: The S1 multi-sensor wireless water quality monitoring module sends the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity in the pond at n historical collection times, and normalizes them by dimension; n is greater than or equal to 1; S2: the data processing module performs noise reduction processing on the collected sensor data to obtain an analog signal of the processed monitoring data; S3: The analog-to-digital conversion module converts the analog signals output by each sensor into digital signals, and transmits the digital signals to the microprocessor; S4 The microprocessor receives water temperature information sent by the water temperature sensor, dissolved oxygen information sent by the dissolved oxygen sensor, ammonia nitrogen information sent by the ammonia nitrogen sensor, nitrite nitrogen information sent by the nitrite nitrogen sensor, pH value sent by the pH sensor, salinity information sent by the salinity sensor, and fish image information sent by the image acquisition module, and sends the received temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, salinity, and fish image information to the cloud processing platform; S5: The cloud processing platform receives the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, salinity and fish image information in the pond at n historical collection moments sent by the multi-sensor wireless water quality monitoring module, compares the size of the fish in the received image information with the set multiple standard fish image information one by one, screens out the standard fish image information that matches the size of the fish in the received pond image information, and calculates the average mass of the fish in each image according to the weight of the fish in the standard fish image information; S6 The cloud processing platform obtains the specific growth rate SGR of the fish body according to the image information data; The cloud processing platform S7 uses the least squares support vector machine to establish a prediction model for the specific growth rate SGR, and the model equation is specifically: Y SGR =C0+aX1+bX2+cX3+dX4+eX5+fX6 Among them, Y SGR is the specific growth rate of fish; X1, X2, X3, X4, X5, and X6 correspond to temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity, respectively; C0, a, b, c, d, e, and f are all constants; When SGR value is ≤10%, C0=-0.1947, a=0.0016, b=0.0076, c=0.38, d=0.01, e=0.021, f=0.21; When 10%<SGR value≤30%, C0=-0.3285, a=0.0028, b=0.0091, c=0.27, d=0.0076, e=0.036, f=0.25; When SGR value>30%, C0=-0.1563, a=0.0021, b=0.013, c=0.22, d=0.0034, e=0.029, f=0.36; S8 The cloud processing platform predicts the specific growth rate of the current farmed fish based on the prediction model and the temperature, dissolved oxygen, ammonia nitrogen, nitrite nitrogen, pH, and salinity of the current farmed fish in the pond at the current moment, and then issues a corresponding control instruction set so that the execution control module controls the corresponding controlled device to execute according to the received control instruction set.

2. The intelligent fish growth control method according to claim 1, characterized in that: The following steps are also included: When the multi-sensor wireless water quality monitoring module detects that the temperature in the aquaculture pond exceeds 30°C or is lower than 8°C, the dissolved oxygen content is lower than 3mg / L or higher than 8mg / L, the pH is lower than 5.5 or higher than 9, and the salinity is higher than 5%, the cloud processing platform sends a warning signal to the external client and the user through the communication and data wireless transmission module.

3. The intelligent fish growth control method according to claim 1, characterized in that: The prediction root mean square error RMSEP of the prediction model is less than 0.05, and the prediction determination coefficient R 2 >0.

95.

4. The intelligent fish growth control method according to claim 1, characterized in that: In step S6, the cloud processing platform obtains the specific growth rate SGR of the fish body according to the image information data, which is specifically: in, represents the average mass of fish at the end of the measurement period, represents the average mass of fish at the beginning of the measurement period; t represents the number of days in a measurement period.

5. The intelligent fish growth control method according to claim 1, characterized in that: The data processing module in step S2 performs noise reduction processing on the collected sensor data, specifically: The noise reduction random filtering processing model is used to reduce the interference of noise on the collected data through multidimensional regression algorithm.

6. The intelligent fish growth control method according to claim 1, characterized in that: Step S5 calculates the average mass of the fish in each image, specifically: The weight of fish in the image information collected by the multi-sensor wireless water quality monitoring module is quantified to obtain the weight set Mn(m1,m2,...,mk) of the fish in the pond, where k is the number of fish in each image. is the average mass of the fish in each image.

7. The intelligent fish growth control method according to claim 1, characterized in that: The communication and data wireless transmission module performs wireless bidirectional transmission via 4G / NBIOT.

8. The intelligent fish growth control method according to claim 1, characterized in that: The fish growth intelligent regulation system also includes an external client, which is connected to the cloud processing platform via a wired or wireless network.

9. The intelligent fish growth control method according to claim 1, characterized in that: The pond control module also includes a monitoring display screen.

10. The intelligent fish growth control method according to claim 1, characterized in that: Step S3: The analog-to-digital conversion module converts the analog signals output by each sensor into digital signals and transmits the digital signals to the microprocessor, specifically: The analog signals output by each sensor are converted into 485 digital signals, and the 485 digital signals are transmitted to the microprocessor.

Images