Intelligent culture management system and method for aquaculture

Abstract

The invention discloses an intelligent culture management system and method for aquaculture. The method comprises the steps that a detection module detects environment parameters, culture pond parameters, feed feeding parameters and the survival rate of juvenile shrimps; the correlation analysis module analyzes the environment parameters, the breeding pond parameters, the feed feeding parameters and the survival rate of juvenile shrimps to obtain a plurality of juvenile shrimp breeding key parameters; the training module carries out training to obtain a juvenile shrimp survival rate prediction model; the function establishing unit establishes a fitness function according to the survival rate of the juvenile shrimps and the predicted survival rate of the juvenile shrimps; the optimization unit introduces a sparrow search algorithm, calculates a global optimal sparrow and a fitness function value after each iteration by using an improved investigation early warning formula, and solves the global optimal sparrow when the fitness function value is minimum; and the cultivation adjustment module dynamically adjusts parameters of each cultivation pond, feeding parameters of each feed and environment parameters in the cultivation shed according to the optimal cultivation parameters. The survival rate of juvenile shrimps is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of juvenile shrimp farming, and in particular to an intelligent farming management system and method for aquaculture. Background Art

[0002] In the shrimp farming process of aquaculture, shrimps need to go through three stages: shrimp seedlings, young shrimps, and adult shrimps. Shrimp seedlings are cultivated in the nursery ponds in the nursery and then diluted out of the ponds. The physique of shrimp seedlings is young and tender, and their feeding ability and disease resistance are poor. Therefore, the cultivation of young shrimps in the process of cultivating shrimp seedlings into young shrimps requires strict control of the cultivation environment to ensure that the young shrimps maintain a high survival rate.

[0003] Young shrimps are generally cultivated in artificially built outdoor cultivation pools, which are greatly disturbed by external environmental factors. When the external environment suddenly changes to be unfavorable for the growth of seedlings, the survival rate of seedlings will plummet. In the prior art, the environment in the cultivation pool is generally adjusted manually based on the cultivation experience of farmers. This method requires high artificial cultivation experience and is slow to respond, resulting in untimely remediation and the death of a large number of seedlings. Therefore, the traditional method of cultivating young shrimps is highly dependent on the external weather environment and cannot achieve a high survival rate of young shrimps. Summary of the invention

[0004] In view of the deficiencies in the prior art, the object of the present invention is to provide an intelligent breeding management system for aquaculture, which is used to significantly improve the survival rate of young shrimps.

[0005] To achieve the above object, the present invention provides the following technical solution: a smart aquaculture management system for aquaculture, comprising:

[0006] A detection module is used to detect environmental parameters in the cultivation shed where the juvenile shrimp cultivation pond is located at a number of historical moments, detect cultivation pond parameters in the juvenile shrimp cultivation pond, detect feed feeding parameters in the juvenile shrimp cultivation pond, and detect the survival rate of the juvenile shrimp in the juvenile shrimp cultivation pond, wherein the historical moments are associated with the number of days for the juvenile shrimp to grow and the optimal survival rate of the juvenile shrimp;

[0007] A correlation analysis module, connected to the detection module, is used to perform correlation analysis on the environmental parameters, the culture pool parameters, the feed feeding parameters and the survival rate of the juvenile shrimp, and obtain several key parameters for juvenile shrimp cultivation, wherein the key parameters for juvenile shrimp cultivation include oxygen concentration in the culture pool, temperature in the culture pool, pH value in the culture pool, ambient air pressure, light intensity, feeding frequency and single feeding amount;

[0008] A training module, connected to the correlation analysis module, used to take the key parameters for juvenile shrimp cultivation at each historical moment as input parameters, and the juvenile shrimp survival rate at the corresponding historical moment as output parameters, to train a juvenile shrimp survival rate prediction model, wherein the juvenile shrimp survival rate prediction model is used to predict the juvenile shrimp predicted survival rate based on the input key parameters for juvenile shrimp cultivation;

[0009] The parameter optimization module is connected to the training module and includes:

[0010] A function establishment unit, used for establishing a fitness function according to the survival rate of the juvenile shrimp and the predicted survival rate of the juvenile shrimp, wherein the fitness function is used for outputting a fitness function value;

[0011] An optimization unit, connected to the function establishment unit, is used to introduce a preset sparrow search algorithm, and regards the key parameters for shrimp cultivation as a sparrow individual, and uses the improved detection and early warning formula to calculate the global optimal sparrow position and the corresponding global optimal fitness function value after each iteration, and finally obtains the global optimal sparrow position corresponding to the global optimal fitness function value at this time after the iterative update is completed, and the shrimp cultivation key parameters contained in the global optimal sparrow position are the optimal cultivation parameters, and the optimal training parameters are associated with the number of shrimp growth days corresponding to the historical moment;

[0012] A cultivation adjustment module is connected to the parameter optimization module, and is used to obtain the number of days for the juvenile shrimp to grow at the current moment, and dynamically adjust the cultivation pond parameters, the feed feeding parameters and the environmental parameters in the cultivation shed in the juvenile shrimp cultivation pond at the current moment according to the optimal cultivation parameters at the same historical moment as the number of days for the juvenile shrimp to grow at the current moment.

[0013] Preferably, the environmental parameters include ambient temperature, ambient humidity, ambient oxygen concentration, air flow rate, light intensity and ambient air pressure; the cultivation tank parameters include cultivation tank temperature, cultivation tank pH value, disinfectant concentration, cultivation tank oxygen concentration and algae concentration; the feed feeding parameters include feeding frequency, single feeding amount and feed components.

[0014] Preferably, the correlation analysis module includes:

[0015] A coefficient analysis unit is used to respectively calculate the correlation coefficients between the ambient temperature, ambient humidity, ambient oxygen concentration, air flow rate, light intensity, ambient air pressure, culture tank temperature, culture tank pH value, disinfectant concentration, culture tank oxygen concentration, algae concentration, feeding frequency, single feeding amount and feed components and the survival rate of the juvenile shrimp;

[0016] A screening unit is connected to the coefficient analysis unit, and is used to sort the calculated correlation coefficients from high to low, and use the environmental parameters, the cultivation pond parameters and the feed feeding parameters whose correlation coefficients are higher than a preset coefficient threshold as the key parameters for the cultivation of young shrimp.

[0017] Preferably, the optimization unit generates a first step length improvement parameter and a second step length improvement parameter respectively according to the algae concentration, ambient temperature, ambient humidity, ambient oxygen concentration, and air flow rate whose correlation coefficient is not higher than the coefficient threshold, and generates an improved detection and warning formula according to the first step length improvement parameter and the second step length improvement parameter.

[0018] Preferably, the improved detection warning formula is configured as follows:

[0019]

[0020] in, Used to represent the global optimal sparrow position, Used to indicate the global worst sparrow position, f i It is used to represent the fitness function value corresponding to the sparrow individual, f g It is used to represent the global optimal fitness function value, f w It is used to represent the global worst fitness function value, β is used to represent the initial first step length control parameter, β is a normally distributed random number with a mean of 0 and a variance of 1, K is used to represent the initial second step length control parameter, K is a random number between [-1,1], φ is used to represent the first step length improvement parameter, ω is used to represent the second step length improvement parameter, It is used to represent the average value of the algae concentration in the past preset time period. It is used to indicate the average value of the ambient temperature in the past preset time period. It is used to represent the average value of the ambient humidity in the past preset time period. It is used to represent the average value of the ambient oxygen concentration in the past preset time period. It is used to represent the average value of the air flow rate in the past preset time period, and σ1, σ2, σ3, σ4, and σ5 are used to represent the preset first configuration coefficient, second configuration coefficient, third configuration coefficient, fourth configuration coefficient, and fifth configuration coefficient, respectively.

[0021] Preferably, the calculation process of the average value of the algae concentration in the past preset time period is:

[0022]

[0023] in, Used to indicate the entire preset time period in the past, Tx Used to indicate a sub-time period within a preset time period in the past. cx It is used to represent the algae concentration in a sub-time period, A tx It is used to indicate the ambient temperature corresponding to a sub-time period. hx It is used to indicate the ambient humidity in a sub-time period. ocx It is used to indicate the ambient oxygen concentration in a sub-time period, A frx Used to indicate the air flow rate in a sub-time period.

[0024] Preferably, the optimization unit is further used to perform parameter correction on the first step length improvement parameter and the second step length improvement parameter according to the disinfectant concentration and the feed component, and the first step length improvement parameter and the second step length improvement parameter after correction are configured as follows:

[0025]

[0026]

[0027] Wherein, ω' is used to represent the first step length improvement parameter after correction, φ' is used to represent the second step length improvement parameter after correction, τ1 and τ2 are used to represent the preset first correction coefficient and the second correction coefficient respectively, D fc Used to indicate the disinfectant concentration, D f0 Used to indicate the preset standard concentration, f e Used to indicate the feed ingredients, f e0 Used to indicate preset standard components.

[0028] Preferably, the juvenile shrimp survival rate prediction model is obtained based on XGBoost model training.

[0029] Preferably, the fitness function is:

[0030] f i =R0-R(C po ,C pt ,C ph ,A ap ,L i ,F f ,S fa )

[0031] Among them, R0 is used to represent the optimal survival rate of the juvenile shrimp, R(C po ,C pt ,C ph ,A ap ,L i ,F f ,S fa) is used to represent the predicted survival rate of the juvenile shrimp, C po Used to indicate the oxygen concentration in the cultivation pool, C pt Used to indicate the temperature of the cultivation tank, C ph Used to indicate the pH value of the cultivation pool, A ap Used to indicate the ambient air pressure, L i Used to represent the light intensity, F f Used to indicate the feeding frequency, S fa Used to indicate the single feeding amount.

[0032] A smart breeding management method for aquaculture, applied to the above-mentioned smart breeding management system for aquaculture, comprising:

[0033] Step S1, the detection module detects the environmental parameters in the cultivation shed where the juvenile shrimp cultivation pond is located at a number of historical moments, detects the cultivation pond parameters in the juvenile shrimp cultivation pond, detects the feed feeding parameters in the juvenile shrimp cultivation pond, and detects the survival rate of the juvenile shrimp in the juvenile shrimp cultivation pond, and the historical moments are associated with the number of days the juvenile shrimp grows;

[0034] Step S2, the correlation analysis module performs correlation analysis on each of the environmental parameters, each of the cultivation pool parameters, each of the feed feeding parameters and the survival rate of the juvenile shrimp to obtain a number of key parameters for juvenile shrimp cultivation, wherein the key parameters for juvenile shrimp cultivation include oxygen concentration in the cultivation pool, temperature in the cultivation pool, pH value in the cultivation pool, ambient air pressure, light intensity, feeding frequency and single feeding amount;

[0035] Step S3, the training module uses the key parameters for juvenile shrimp cultivation at each historical moment as input parameters, and uses the juvenile shrimp survival rate at the corresponding historical moment as output parameters, and trains to obtain a juvenile shrimp survival rate prediction model, wherein the juvenile shrimp survival rate prediction model is used to predict the juvenile shrimp predicted survival rate based on the input key parameters for juvenile shrimp cultivation;

[0036] Step S4, the function establishment unit establishes a fitness function according to the optimal survival rate of the young shrimp and the predicted survival rate of the young shrimp at the same historical moment, and the fitness function is used to output the fitness function value; the optimization unit introduces a preset sparrow search algorithm, and takes the key parameters for young shrimp cultivation as a sparrow individual, and uses the improved detection and early warning formula to calculate the global optimal sparrow position and the corresponding global optimal fitness function value after each iteration, and finally obtains the global optimal sparrow position corresponding to the global optimal fitness function value at this time after the final iterative update is completed, and the key parameters for young shrimp cultivation contained in the global optimal sparrow position are optimal cultivation parameters, and the optimal training parameters are associated with the number of days for young shrimp growth corresponding to the historical moment;

[0037] Step S5, the cultivation adjustment module obtains the number of days for the juvenile shrimp to grow at the current moment, and dynamically adjusts the cultivation pond parameters, the feed feeding parameters and the environmental parameters in the cultivation shed in the juvenile shrimp cultivation pond at the current moment according to the optimal cultivation parameters at the same historical moment as the number of days for the juvenile shrimp to grow at the current moment.

[0038] Beneficial effects of the present invention:

[0039] The present invention first detects and obtains environmental parameters, cultivation pond parameters, feed feeding parameters and juvenile shrimp survival rate at various historical moments, then performs correlation analysis on various environmental parameters, various cultivation pond parameters, various feed feeding parameters and juvenile shrimp survival rate, analyzes and screens to obtain a number of key parameters for juvenile shrimp cultivation, then trains according to the key parameters for juvenile shrimp cultivation and the juvenile shrimp survival rate at various historical moments to obtain a juvenile shrimp survival rate prediction model, then uses the juvenile shrimp survival rate prediction model to predict and output the predicted survival rate of juvenile shrimp and the juvenile shrimp survival rate at the same moment to establish a fitness function value, then uses the key parameters for juvenile shrimp cultivation as sparrow individuals, and uses an improved sparrow search algorithm to calculate the global optimal sparrow position to obtain the optimal cultivation parameters, and finally dynamically adjusts various cultivation pond parameters, various feed feeding parameters and various environmental parameters in the cultivation shed in the juvenile shrimp cultivation pond at the current moment according to the optimal cultivation parameters, so that the juvenile shrimp survival rate at the current moment is infinitely close to the optimal survival rate of the juvenile shrimp at the historical moment, so that the survival rate of the juvenile shrimp in the cultivation pond can always be maintained near the optimal survival rate of the juvenile shrimp during the dynamic adjustment process, thereby significantly improving the survival rate of the juvenile shrimp. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a structural schematic diagram of the intelligent breeding management system for aquaculture in the present invention;

[0041] Figure 2 It is a schematic diagram of the structure of the correlation analysis module in the present invention;

[0042] Figure 3 It is a flow chart of the steps of the intelligent breeding management method for aquaculture in the present invention.

[0043] Figure numerals: 1. Detection module; 2. Correlation analysis module; 21. Coefficient analysis unit; 22. Screening unit; 3. Training module; 4. Parameter optimization module; 41. Function establishment unit; 42. Optimization unit; 7. Cultivation and adjustment module. DETAILED DESCRIPTION

[0044] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. The same parts are represented by the same reference numerals. It should be noted that the words "front", "rear", "left", "right", "upper" and "lower" used in the following description refer to directions in the accompanying drawings, and the words "bottom surface" and "top surface", "inner" and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0045] Example 1, reference Figure 1 , which is the first embodiment of the present invention, provides a smart breeding management system for aquaculture, which can significantly improve the survival rate of young shrimps, including:

[0046] Detection module 1 is used to detect environmental parameters in the cultivation shed where the juvenile shrimp cultivation pond is located at a number of historical moments, detect cultivation pond parameters in the juvenile shrimp cultivation pond, detect feed feeding parameters in the juvenile shrimp cultivation pond, and detect the survival rate of the juvenile shrimp in the juvenile shrimp cultivation pond, wherein the historical moments are associated with the number of days for the juvenile shrimp to grow and the optimal survival rate of the juvenile shrimp;

[0047] The correlation analysis module 2 is connected to the detection module 1 and is used to perform correlation analysis on various environmental parameters, various cultivation pool parameters, various feed feeding parameters and the survival rate of young shrimps to obtain several key parameters for young shrimp cultivation. The key parameters for young shrimp cultivation include the oxygen concentration in the cultivation pool, the temperature in the cultivation pool, the pH value in the cultivation pool, the ambient air pressure, the light intensity, the feeding frequency and the single feeding amount;

[0048] The training module 3 is connected to the correlation analysis module 2 and is used to use the key parameters of juvenile shrimp cultivation at each historical moment as input parameters and the juvenile shrimp survival rate at the corresponding historical moment as output parameters to train a juvenile shrimp survival rate prediction model, which is used to predict the juvenile shrimp predicted survival rate based on the input key parameters of juvenile shrimp cultivation;

[0049] The parameter optimization module 4 is connected to the training module 3 and includes:

[0050] A function establishment unit 41 is used to establish a fitness function according to the optimal survival rate of young shrimp and the predicted survival rate of young shrimp, and the fitness function is used to output a fitness function value;

[0051] The optimization unit 42 and the connection function establishment unit 41 are used to introduce a preset sparrow search algorithm, and take the key parameters of shrimp breeding as a sparrow individual, and use the improved detection and early warning formula to calculate the global optimal sparrow position and the corresponding global optimal fitness function value after each iteration, and finally obtain the global optimal sparrow position corresponding to the global optimal fitness function value at this time after the iteration update is completed. The key parameters of shrimp breeding included in the global optimal sparrow position are the optimal breeding parameters, and the optimal training parameters are associated with the number of shrimp growth days corresponding to the historical moment;

[0052] The cultivation adjustment module 7 is connected to the parameter optimization module 4, which is used to obtain the number of days for the juvenile shrimp to grow at the current moment, and dynamically adjust the cultivation pond parameters, feed feeding parameters and environmental parameters in the cultivation shed in the juvenile shrimp cultivation pond at the current moment according to the optimal cultivation parameters at the same historical moment as the number of days for the juvenile shrimp to grow at the current moment.

[0053] Preferably, the environmental parameters include ambient temperature, ambient humidity, ambient oxygen concentration, air flow rate, light intensity and ambient air pressure; the cultivation tank parameters include cultivation tank temperature, cultivation tank pH value, disinfectant concentration, cultivation tank oxygen concentration and algae concentration; the feed feeding parameters include feeding frequency, single feeding amount and feed composition.

[0054] Specifically, in this embodiment, the detection module 1 may include a number of temperature sensors, humidity sensors, oxygen concentration sensors, wind speed sensors, illumination sensors, air pressure sensors, pH testers, disinfectant detectors, algae analyzers, feeding frequency counters, feeding amount weighing meters and feed component detection equipment, wherein each temperature sensor is used to detect the ambient temperature and the temperature of the cultivation pool, each humidity sensor is used to detect the ambient humidity, each oxygen concentration sensor is used to detect the ambient oxygen concentration and the oxygen concentration of the cultivation pool, the wind speed sensor is used to detect the air flow rate, the illumination sensor is used to detect the light intensity, the air pressure sensor is used to detect the ambient air pressure, the pH tester is used to detect the pH value of the cultivation pool, the disinfectant detector is used to detect the disinfectant concentration, the algae analyzer is used to detect the algae concentration, and the feeding frequency counter is used to detect the feeding frequency in a unit time, and the unit time can be 6 hours. The feeding amount weighing meter is used to detect the single feeding amount, and the feed component detection equipment is used to detect the feed component. The correlation analysis module 2 may be an analysis and calculation program in the intelligent processing terminal, which is used to calculate the correlation coefficients between each environmental parameter, each cultivation pool parameter, each feed feeding parameter and the survival rate of the young shrimp, respectively obtain the correlation coefficients between each environmental parameter, each cultivation pool parameter, each feed feeding parameter and the survival rate of the young shrimp, and obtain the best training parameters by screening. The training module 3 may be a model training program configured in the intelligent processing terminal, which is used to use the key parameters of young shrimp cultivation at each historical moment as input parameters, and the survival rate of the young shrimp at the corresponding historical moment as output parameters, and train to obtain a prediction model for the survival rate of the young shrimp. The parameter optimization module 4 may be a parameter optimization program configured on the intelligent control terminal, which is used to use the improved sparrow search algorithm to find the best cultivation parameters contained in the global optimal sparrow position. Finally, the number of days for the growth of young shrimps at the current moment is matched with the number of days for the growth of young shrimps at the historical moment, and the optimal cultivation parameters at the historical moment are used to dynamically adjust the parameters of each cultivation pool, each feed feeding parameter in the young shrimp cultivation pool at the current moment, and the environmental parameters, each cultivation pool parameters, and each feed feeding parameter in the cultivation shed, so that the internal environment of the young shrimp cultivation pool and the cultivation shed is infinitely close to the environment when the young shrimps have the highest optimal survival rate, so that the survival rate of the young shrimps is significantly improved. The cultivation adjustment module 7 includes a number of environmental adjustment devices, including a humidifier, a heating device, a blower, a light emitting device, a feeding device, etc., to adjust the environmental parameters, each cultivation pool parameters, and each feed feeding parameter in the young shrimp cultivation pool and the cultivation shed.

[0055] Working principle of embodiment 1:

[0056] The present invention first detects and obtains environmental parameters, cultivation pond parameters, feed feeding parameters and juvenile shrimp survival rate at various historical moments, then performs correlation analysis on various environmental parameters, various cultivation pond parameters, various feed feeding parameters and juvenile shrimp survival rate, analyzes and screens to obtain a number of key parameters for juvenile shrimp cultivation, then trains according to the key parameters for juvenile shrimp cultivation and the juvenile shrimp survival rate at various historical moments to obtain a juvenile shrimp survival rate prediction model, then uses the juvenile shrimp survival rate prediction model to predict and output the predicted survival rate of juvenile shrimp and the juvenile shrimp survival rate at the same moment to establish a fitness function value, then uses the key parameters for juvenile shrimp cultivation as sparrow individuals, and uses an improved sparrow search algorithm to calculate the global optimal sparrow position to obtain the optimal cultivation parameters, and finally dynamically adjusts various cultivation pond parameters, various feed feeding parameters and various environmental parameters in the cultivation shed in the juvenile shrimp cultivation pond at the current moment according to the optimal cultivation parameters, so that the juvenile shrimp survival rate at the current moment is infinitely close to the optimal survival rate of the juvenile shrimp at the historical moment, so that the survival rate of the juvenile shrimp in the cultivation pond can always be maintained near the optimal survival rate of the juvenile shrimp during the dynamic adjustment process, thereby significantly improving the survival rate of the juvenile shrimp.

[0057] Preferably, Figure 2 As shown, the relevant analysis module 2 includes:

[0058] The coefficient analysis unit 21 is used to respectively calculate the correlation coefficients between the ambient temperature, ambient humidity, ambient oxygen concentration, air velocity, light intensity, ambient air pressure, culture tank temperature, culture tank pH value, disinfectant concentration, culture tank oxygen concentration, algae concentration, feeding frequency, single feeding amount and feed components and the survival rate of young shrimp;

[0059] The screening unit 22 is connected to the coefficient analysis unit 21, and is used to sort the calculated correlation coefficients from high to low, and use the environmental parameters, culture pond parameters and feed feeding parameters with correlation coefficients higher than a preset coefficient threshold as key parameters for juvenile shrimp cultivation.

[0060] Specifically, in this embodiment, the coefficient threshold can be 0.6. By screening and retaining the oxygen concentration in the breeding pond, the temperature in the breeding pond, the pH value in the breeding pond, the ambient air pressure, the light intensity, the feeding frequency and the single feeding amount whose correlation coefficients with the survival rate of the juvenile shrimps are greater than 0.6 as the key parameters for juvenile shrimp breeding, the screening of data with strong correlation coefficients is achieved.

[0061] Preferably, the optimization unit 42 generates the first step length improvement parameters and the second step length improvement parameters respectively according to the algae concentration, ambient temperature, ambient humidity, ambient oxygen concentration, and air flow rate whose correlation coefficients are not higher than the coefficient threshold, and generates an improved detection and warning formula according to the first step length improvement parameters and the second step length improvement parameters.

[0062] Specifically, in this embodiment, the optimization unit 42 uses the algae concentration, ambient temperature, ambient humidity, ambient oxygen concentration, and air flow rate, which are not key parameters for juvenile shrimp cultivation, to calculate the first step length improvement parameters and the second step length improvement parameters. This not only avoids the waste of effective data, but also optimizes and improves the sparrow search algorithm, thereby enhancing the optimization ability of the sparrow search algorithm.

[0063] Preferably, the improved detection warning formula is configured as follows:

[0064]

[0065] in, Used to indicate the global optimal sparrow position, Used to indicate the global worst sparrow position, f i Used to represent the fitness function value corresponding to the sparrow individual, f g Used to represent the global optimal fitness function value, f w It is used to represent the global worst fitness function value, β is used to represent the initial first step length control parameter, β is a normally distributed random number with a mean of 0 and a variance of 1, K is used to represent the initial second step length control parameter, K is a random number between [-1,1], φ is used to represent the first step length improvement parameter, ω is used to represent the second step length improvement parameter, It is used to indicate the average value of algae concentration in the past preset time period. It is used to indicate the average value of the ambient temperature in the past preset time period. It is used to indicate the average value of ambient humidity in the past preset time period. It is used to indicate the average value of ambient oxygen concentration in the past preset time period. It is used to represent the average value of the air flow rate in the past preset time period, and σ1, σ2, σ3, σ4, and σ5 are used to represent the preset first configuration coefficient, second configuration coefficient, third configuration coefficient, fourth configuration coefficient, and fifth configuration coefficient, respectively.

[0066] Specifically, in the present embodiment, in the sparrow search algorithm, the finder's first step length control parameter β and the second step length control parameter K play an important role in balancing the global search capability and the local development capability. However, since β and K are both random numbers, they cannot satisfy the exploration of the sparrow search algorithm in the solution space, causing the sparrow search algorithm to easily fall into the local optimum. Therefore, the present embodiment optimizes the first step length control parameter β and the second step length control parameter K, and uses algae concentration, ambient temperature, ambient humidity, ambient oxygen concentration, and air flow rate to optimize the calculation of the first step length control parameter β and the second step length control parameter K, so that the first step length control parameter β and the second step length control parameter K finally calculated balance the global search capability, while also avoiding falling into the local optimum, thereby achieving the optimization and improvement of the sparrow search algorithm.

[0067] Preferably, the calculation process of the average value of the algae concentration in the past preset time period is:

[0068]

[0069] in, Used to indicate the entire preset time period in the past, T x Used to indicate a sub-time period within a preset time period in the past. cx Used to indicate the algae concentration in a sub-time period, A tx Used to indicate the ambient temperature corresponding to a sub-time period, A hx It is used to indicate the ambient humidity within a certain sub-time period. ocx Used to indicate the ambient oxygen concentration in a sub-time period, A frx Used to indicate the air velocity within a sub-time period.

[0070] Specifically, in this embodiment, the length of the entire preset time period may be 60 minutes, and the length of each sub-time period may be 60 seconds. It represents the average value of algae concentration in the past 60 minutes. Indicates the average value of the ambient temperature in the past 60 minutes. Indicates the average value of ambient humidity in the past 60 minutes. Indicates the average value of ambient oxygen concentration in the past 60 minutes. Indicates the average air velocity over the past 60 minutes.

[0071] Preferably, the optimization unit 42 is further used to perform parameter correction on the first step length improvement parameter and the second step length improvement parameter according to the disinfectant concentration and the feed composition, and the first step length improvement parameter and the second step length improvement parameter after correction are configured as follows:

[0072]

[0073] Among them, ω' is used to represent the first step length improvement parameter after correction, φ' is used to represent the second step length improvement parameter after correction, τ1 and τ2 are used to represent the preset first correction coefficient and the second correction coefficient respectively, D fc Used to indicate disinfectant concentration, D f0 Used to indicate the preset standard concentration, f e Used to indicate feed ingredients, f e0 Used to indicate preset standard components.

[0074] Specifically, in this embodiment, the disinfectant concentration and feed components will also have a certain impact on the survival rate of young shrimp. Using the disinfectant concentration and feed components to correct the first step improvement parameters and the second step improvement parameters can further optimize the optimized sparrow search algorithm, further balance the global search capability, and avoid falling into local optimality.

[0075] Preferably, the juvenile shrimp survival rate prediction model is obtained based on XGBoost model training.

[0076] Preferably, the fitness function is:

[0077] f i =R0-R(C po ,C pt ,C ph ,A ap ,L i ,F f ,S fa )

[0078] Among them, R0 is used to represent the optimal survival rate of young shrimp, R(C po ,C pt ,C ph ,A ap ,L i ,F f ,S fa ) is used to represent the predicted survival rate of juvenile shrimp, C po Used to indicate the oxygen concentration in the culture tank, C pt Used to indicate the temperature of the cultivation tank, C ph Used to indicate the pH value of the cultivation pool, A ap Used to indicate ambient air pressure, L i Used to indicate light intensity, F f Used to indicate feeding frequency, S fa Used to indicate the amount of food fed at a single time.

[0079] A smart breeding management method for aquaculture, applied to the above-mentioned smart breeding management system for aquaculture, such as Figure 3 As shown, including:

[0080] Step S1, the detection module 1 detects the environmental parameters in the cultivation shed where the juvenile shrimp cultivation pond is located at several historical moments, detects the cultivation pond parameters in the juvenile shrimp cultivation pond, detects the feed feeding parameters in the juvenile shrimp cultivation pond, and detects the survival rate of the juvenile shrimp in the juvenile shrimp cultivation pond, and the historical moments are associated with the growth days of the juvenile shrimp and the optimal survival rate of the juvenile shrimp;

[0081] Step S2, the correlation analysis module 2 performs correlation analysis on various environmental parameters, various culture pond parameters, various feed feeding parameters and juvenile shrimp survival rate to obtain several key parameters for juvenile shrimp cultivation, the key parameters for juvenile shrimp cultivation include the oxygen concentration in the culture pond, the temperature in the culture pond, the pH value in the culture pond, the ambient air pressure, the light intensity, the feeding frequency and the single feeding amount;

[0082] Step S3, the training module 3 uses the key parameters of juvenile shrimp cultivation at each historical moment as input parameters, and uses the juvenile shrimp survival rate at the corresponding historical moment as output parameters, and trains to obtain a juvenile shrimp survival rate prediction model, which is used to predict the juvenile shrimp predicted survival rate based on the input juvenile shrimp cultivation key parameters;

[0083] Step S4, the function establishment unit 41 establishes a fitness function according to the optimal survival rate of the young shrimp and the predicted survival rate of the young shrimp, and the fitness function is used to output the fitness function value; the optimization unit 42 introduces a preset sparrow search algorithm, and takes the key parameters of young shrimp cultivation as a sparrow individual, and uses the improved detection and early warning formula to calculate the global optimal sparrow position and the corresponding global optimal fitness function value after each iteration, and finally obtains the global optimal sparrow position corresponding to the global optimal fitness function value at this time after the final iteration update is completed, and the key parameters of young shrimp cultivation contained in the global optimal sparrow position are the optimal cultivation parameters, and the optimal training parameters are associated with the number of days of growth of young shrimp corresponding to the historical moment;

[0084] Step S5, the cultivation adjustment module 7 obtains the number of days for the juvenile shrimp to grow at the current moment, and dynamically adjusts the cultivation pond parameters, feed feeding parameters and environmental parameters in the juvenile shrimp cultivation pond at the current moment according to the optimal cultivation parameters at the same historical moment as the number of days for the juvenile shrimp to grow at the current moment.

[0085] The above are only preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A smart breeding management system for aquaculture, characterized in that: include: A detection module (1) is used to detect environmental parameters in a cultivation shed where the juvenile shrimp cultivation pond is located, detect cultivation pond parameters in the juvenile shrimp cultivation pond, detect feed feeding parameters in the juvenile shrimp cultivation pond, and detect the survival rate of juvenile shrimp in the juvenile shrimp cultivation pond at a number of historical moments, wherein the historical moments are associated with the number of days for the juvenile shrimp to grow and the optimal survival rate of the juvenile shrimp; A correlation analysis module (2) is connected to the detection module (1) and is used to perform correlation analysis on the environmental parameters, the culture pond parameters, the feed feeding parameters and the survival rate of the juvenile shrimp to obtain a number of key parameters for juvenile shrimp cultivation, wherein the key parameters for juvenile shrimp cultivation include oxygen concentration in the culture pond, temperature in the culture pond, pH value in the culture pond, ambient air pressure, light intensity, feeding frequency and single feeding amount; A training module (3) is connected to the correlation analysis module (2) and is used to use the key parameters for juvenile shrimp cultivation at each historical moment as input parameters and the juvenile shrimp survival rate at the corresponding historical moment as output parameters to train a juvenile shrimp survival rate prediction model, wherein the juvenile shrimp survival rate prediction model is used to predict the juvenile shrimp predicted survival rate based on the input key parameters for juvenile shrimp cultivation; The parameter optimization module (4) is connected to the training module (3) and comprises: A function establishment unit (41) is used to establish a fitness function according to the optimal survival rate of the juvenile shrimp and the predicted survival rate of the juvenile shrimp at the same historical moment, wherein the fitness function is used to output a fitness function value; An optimization unit (42) is connected to the function establishment unit (41) and is used to introduce a preset sparrow search algorithm, and take the key parameters for young shrimp cultivation as a sparrow individual, and use the improved detection and early warning formula to calculate the global optimal sparrow position and the corresponding global optimal fitness function value after each iteration, and finally obtain the global optimal sparrow position corresponding to the global optimal fitness function value at this time after the iteration update is completed, and the key parameters for young shrimp cultivation contained in the global optimal sparrow position are optimal cultivation parameters, and the optimal training parameters are associated with the number of days for young shrimp growth corresponding to the historical moment; A cultivation adjustment module (7) is connected to the parameter optimization module (4) and is used to obtain the number of days for the juvenile shrimp to grow at the current moment, and dynamically adjust the cultivation pond parameters, feed feeding parameters and environmental parameters in the cultivation shed in the juvenile shrimp cultivation pond at the current moment according to the optimal cultivation parameters at the same historical moment as the number of days for the juvenile shrimp to grow at the current moment.

2. The intelligent aquaculture management system for aquaculture according to claim 1, characterized in that: The environmental parameters include ambient temperature, ambient humidity, ambient oxygen concentration, air flow rate, light intensity and ambient air pressure; the cultivation pool parameters include cultivation pool temperature, cultivation pool pH value, disinfectant concentration, cultivation pool oxygen concentration and algae concentration; the feed feeding parameters include feeding frequency, single feeding amount and feed components.

3. The intelligent aquaculture management system for aquaculture according to claim 2, characterized in that: The correlation analysis module (2) comprises: A coefficient analysis unit (21) is used to respectively calculate the correlation coefficients between the environmental temperature, environmental humidity, environmental oxygen concentration, air flow rate, light intensity, environmental pressure, culture tank temperature, culture tank pH value, disinfectant concentration, culture tank oxygen concentration, algae concentration, feeding frequency, single feeding amount and feed components and the survival rate of the juvenile shrimp; A screening unit (22) is connected to the coefficient analysis unit (21) and is used to sort the calculated correlation coefficients from high to low, and use the environmental parameters, the cultivation pond parameters and the feed feeding parameters whose correlation coefficients are higher than a preset coefficient threshold as the key parameters for juvenile shrimp cultivation.

4. The intelligent aquaculture management system for aquaculture according to claim 3, characterized in that: The optimization unit (42) generates a first step length improvement parameter and a second step length improvement parameter respectively according to the algae concentration, ambient temperature, ambient humidity, ambient oxygen concentration, and air flow rate whose correlation coefficient is not higher than the coefficient threshold, and generates an improved detection and early warning formula according to the first step length improvement parameter and the second step length improvement parameter.

5. The intelligent aquaculture management system for aquaculture according to claim 4, characterized in that: The improved detection and early warning formula is configured as follows: in, Used to represent the global optimal sparrow position, Used to indicate the global worst sparrow position, f i It is used to represent the fitness function value corresponding to the sparrow individual, f g It is used to represent the global optimal fitness function value, f w It is used to represent the global worst fitness function value, β is used to represent the initial first step length control parameter, β is a normally distributed random number with a mean of 0 and a variance of 1, K is used to represent the initial second step length control parameter, K is a random number between [-1,1], φ is used to represent the first step length improvement parameter, ω is used to represent the second step length improvement parameter, It is used to represent the average value of the algae concentration in the past preset time period. It is used to indicate the average value of the ambient temperature in the past preset time period. It is used to represent the average value of the ambient humidity in the past preset time period. It is used to represent the average value of the ambient oxygen concentration in the past preset time period. It is used to represent the average value of the air flow rate in the past preset time period, and σ1, σ2, σ3, σ4, and σ5 are used to represent the preset first configuration coefficient, second configuration coefficient, third configuration coefficient, fourth configuration coefficient, and fifth configuration coefficient, respectively.

6. The intelligent aquaculture management system for aquaculture according to claim 5, characterized in that: The calculation process of the average value of the algae concentration in the past preset time period is: in, Used to indicate the entire preset time period in the past, T x Used to indicate a sub-time period within a preset time period in the past. cx It is used to represent the algae concentration in a sub-time period, A tx It is used to indicate the ambient temperature corresponding to a sub-time period. hx It is used to indicate the ambient humidity in a sub-time period. ocx It is used to indicate the ambient oxygen concentration in a sub-time period, A frx Used to indicate the air flow rate in a sub-time period.

7. The intelligent aquaculture management system for aquaculture according to claim 5, characterized in that: The optimization unit (42) is further used to perform parameter correction on the first step length improvement parameter and the second step length improvement parameter according to the disinfectant concentration and the feed composition, and the first step length improvement parameter and the second step length improvement parameter after correction are configured as follows: Wherein, ω' is used to represent the first step length improvement parameter after correction, φ' is used to represent the second step length improvement parameter after correction, τ1 and τ2 are used to represent the preset first correction coefficient and the second correction coefficient respectively, D fc Used to indicate the disinfectant concentration, D f0 Used to indicate the preset standard concentration, f e Used to indicate the feed ingredients, f e0 Used to indicate preset standard components.

8. The intelligent aquaculture management system for aquaculture according to claim 1, characterized in that: The juvenile shrimp survival rate prediction model is obtained based on XGBoost model training.

9. The intelligent aquaculture management system for aquaculture according to claim 1, characterized in that: The fitness function is: f i =|R0-R(C po ,C pt ,C ph ,A ap ,L i ,F f ,S fa )| Among them, R0 is used to represent the optimal survival rate of the juvenile shrimp, R(C po ,C pt ,C ph ,A ap ,L i ,F f ,S fa ) is used to represent the predicted survival rate of the juvenile shrimp, C po Used to indicate the oxygen concentration in the cultivation pool, C pt Used to indicate the temperature of the cultivation tank, C ph Used to indicate the pH value of the cultivation pool, A ap Used to indicate the ambient air pressure, L i Used to represent the light intensity, F f Used to indicate the feeding frequency, S fa Used to indicate the single feeding amount.

10. A smart farming management method for aquaculture, applied to the smart farming management system for aquaculture according to any one of claims 1 to 9, characterized in that: include: Step S1, the detection module 1 detects the environmental parameters in the cultivation shed where the juvenile shrimp cultivation pond is located at several historical moments, detects the cultivation pond parameters in the juvenile shrimp cultivation pond, detects the feed feeding parameters in the juvenile shrimp cultivation pond, and detects the survival rate of the juvenile shrimp in the juvenile shrimp cultivation pond, and the historical moments are associated with the growth days of the juvenile shrimp and the optimal survival rate of the juvenile shrimp; Step S2, the correlation analysis module 2 performs correlation analysis on each of the environmental parameters, each of the cultivation pool parameters, each of the feed feeding parameters and the survival rate of the juvenile shrimp to obtain a number of key parameters for juvenile shrimp cultivation, wherein the key parameters for juvenile shrimp cultivation include the oxygen concentration in the cultivation pool, the temperature in the cultivation pool, the pH value in the cultivation pool, the ambient air pressure, the light intensity, the feeding frequency and the single feeding amount; Step S3, the training module 3 uses the key parameters for juvenile shrimp cultivation at each historical moment as input parameters, and uses the juvenile shrimp survival rate at the corresponding historical moment as output parameters, and trains to obtain a juvenile shrimp survival rate prediction model, wherein the juvenile shrimp survival rate prediction model is used to predict the juvenile shrimp predicted survival rate based on the input key parameters for juvenile shrimp cultivation; Step S4, the function establishment unit 41 establishes a fitness function according to the optimal survival rate of the young shrimps and the predicted survival rate of the young shrimps at the same historical moment, and the fitness function is used to output a fitness function value; The optimization unit 42 introduces a preset sparrow search algorithm, and takes the key parameters for shrimp breeding as a sparrow individual, and uses the improved detection and early warning formula to calculate the global optimal sparrow position and the corresponding global optimal fitness function value after each iteration, and finally obtains the global optimal sparrow position corresponding to the global optimal fitness function value at this time after the iteration update is completed, and the key parameters for shrimp breeding contained in the global optimal sparrow position are the optimal breeding parameters, and the optimal training parameters are associated with the number of shrimp growth days corresponding to the historical moment; Step S5, the cultivation adjustment module 7 obtains the number of days for the juvenile shrimp to grow at the current moment, and dynamically adjusts the cultivation pond parameters, the feed feeding parameters and the environmental parameters in the cultivation shed in the juvenile shrimp cultivation pond at the current moment according to the optimal cultivation parameters at the same historical moment as the number of days for the juvenile shrimp to grow at the current moment.

Images