Intelligent feed rationing system for breeding industry

By using BP neural network model and optimization algorithm in the feed intelligent distribution system, combined with high-precision dynamic measurement and automatic control, the problem of inaccurate feed distribution in the existing technology is solved, efficient and accurate feed management is achieved, and breeding is improved. industry productivity and sustainability.

CN118972424APending Publication Date: 2024-11-15刘建平
View PDF 0 Cites 9 Cited by

Patent Information

Application Number
CN202410981973.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing feed distribution system relies on experience in determining feeding frequency and quantity, resulting in limited flexibility and accuracy, resulting in malnutrition or feed waste, and lack of precise measurement and control, reducing feed utilization and production efficiency.

Method used

A feed intelligent distribution system for the breeding industry was designed, including data collection, data processing, feed distribution and automatic control modules. The feed demand is predicted through the BP neural network model, combined with sparrow search and firefly algorithm optimization, to achieve high-precision dynamic measurement and Automatic feeding and real-time monitoring of growth and environmental conditions.

Benefits of technology

It improves the accuracy of feed prediction, reduces manpower demand, achieves precise feeding, reduces feed waste and cost, and improves breeding efficiency and sustainability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118972424A_ABST
    Figure CN118972424A_ABST
Patent Text Reader

Abstract

The invention, which belongs to the technical field of agricultural automation and intellectualization, discloses a breeding industry-oriented intelligent feed allocation system comprising: a data acquisition module for acquiring growth data, breeding environment data and feed raw material characteristics of raised animals and historical feed consumption of each animal; the data processing module is used for carrying out data cleaning and standardization processing; the feed allocation module is used for constructing a BP neural network model, optimizing the trained BP neural network model based on a sparrow search algorithm and a firefly algorithm, and predicting the feed demand of fed animals based on the optimized BP neural network model; the automatic control module is used for weighing and metering based on a high-precision dynamic metering technology, and automatic feeding is performed after uniform mixing; and the remote monitoring module is used for monitoring the animal feeding condition and the breeding environment condition in real time. According to the intelligent feed allocation system, accurate control, intelligent management and environment-friendly optimization of animal feed allocation can be achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of agricultural automation and intelligence, and in particular relates to an intelligent feed distribution system for aquaculture. Background Art

[0002] Feed is the core of the breeding industry and directly affects the growth, health and production performance of animals. At present, the feeding frequency and feeding quantity of existing feed distribution systems are mostly determined based on the experience of farmers, which leads to limited flexibility and accuracy in feeding. Feeding too little will lead to malnutrition, while feeding too much will lead to feed waste, increase breeding costs and reduce feed utilization. Some feed distribution systems lack accurate measurement and control systems, resulting in reduced utilization of feed raw materials and waste of feed resources.

[0003] It can be seen that with the advancement of science and technology and the development of animal husbandry, the traditional feed distribution method can no longer meet the multiple needs of modern animal husbandry for improving production efficiency, reducing feed costs and ensuring animal health. Therefore, it is urgent to propose an intelligent feed distribution system for the breeding industry. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes an intelligent feed distribution system for the breeding industry to solve the problems existing in the above prior art.

[0005] To achieve the above-mentioned purpose, the present invention provides an intelligent feed distribution system for the breeding industry, comprising: a data acquisition module, a data processing module, a feed distribution module, an automatic control module and a remote control module connected in sequence;

[0006] The data acquisition module is used to obtain the growth data of the raised animals, the breeding environment data, the feed raw material characteristics and the historical feed consumption of each animal, and upload them to the data processing module in real time;

[0007] The data processing module is used to perform data cleaning and standardization processing on growth data, breeding environment data, feed raw material characteristics and feed consumption;

[0008] The feed rationing module is used to construct a BP neural network model, train the BP neural network model based on standardized growth data, breeding environment data, feed raw material characteristics and historical feed consumption, then optimize the trained BP neural network model based on the sparrow search algorithm and the firefly algorithm, and predict the feed demand of the breeding animals based on the optimized BP neural network model;

[0009] The automatic control module is used to obtain the required weight of each proportion of raw materials based on the predicted feed demand and the preset feed ratio scheme, weigh and measure based on high-precision dynamic metering technology, and then evenly mix all the obtained proportioned raw materials and automatically feed them;

[0010] The remote monitoring module is used to monitor the growth, feeding and breeding environment of animals in real time.

[0011] Optionally, the data acquisition module includes a growth data acquisition unit, an environmental data acquisition unit, a feed raw material acquisition unit and a feed consumption acquisition unit;

[0012] The growth data acquisition unit is used to regularly measure the weight of the animal based on an electronic scale and obtain the body length of the animal based on a laser rangefinder or a camera;

[0013] The environmental data acquisition unit is used to monitor the temperature and humidity in the breeding house based on the temperature and humidity sensor, monitor the light intensity and lighting time in the breeding house based on the light intensity detector, and monitor the gas concentration in the breeding house based on the gas sensor;

[0014] The feed material acquisition unit is used to measure the flow rate and density of the feed based on the mass flow meter, and to measure the number and size of particles in the feed based on the particle counter;

[0015] The feed consumption acquisition unit is used to acquire the historical feed consumption of each animal based on a weighing sensor.

[0016] Optionally, the data processing module includes a data cleaning unit and a data standardization unit;

[0017] The data cleaning unit is used to check and process missing values, abnormal values ​​and duplicate data of growth data, breeding environment data, feed raw material characteristics and historical feed consumption;

[0018] The data standardization unit is used to standardize the growth data and feed consumption based on Z-score normalization, and to standardize the breeding environment data and feed raw material characteristics based on minimum-maximum normalization.

[0019] Optionally, the feed distribution module includes a feed prediction unit and a model optimization unit;

[0020] The feed prediction unit is used to construct a BP neural network model, which takes the standardized growth data, breeding environment data, feed raw material characteristics and historical feed consumption as input, takes the feed requirement as output, trains the BP neural network model using mean square error and gradient descent algorithm, and predicts the feed requirement of the breeding animals based on the trained BP neural network model;

[0021] The model optimization unit is used to optimize the trained BP neural network model based on the sparrow search algorithm and the firefly algorithm.

[0022] Optionally, the model optimization unit includes a position updating subunit and a secondary updating subunit;

[0023] The position updating subunit is used to initialize the parameters of the sparrow search algorithm, take the number of sparrows as the number of BP neural network parameter groups, take the position of the sparrows as the parameter value of the BP neural network in different parameter groups, and obtain the initial position of the sparrows in the population; obtain the mean value of the mean square error between the predicted feed demand and the actual feed demand as the fitness function, and then obtain the fitness values ​​of all sparrows in the population and sort them to obtain the global optimal position; select some sparrows as searchers, some sparrows as guards, and the remaining sparrows as beggars, respectively move the positions of the searchers, guards and beggars, obtain the position of the sparrow population after the move, and then update the global optimal position;

[0024] The secondary update subunit is used to initialize the firefly algorithm parameters. In the firefly population, the firefly with the highest brightness moves to obtain the position update result of the firefly algorithm. Based on the position update result of the firefly algorithm, the position of the sparrow population is further updated in real time to obtain a new BP neural network parameter group.

[0025] Optionally, the automatic control module includes a dynamic metering unit and an automatic control unit;

[0026] The dynamic metering unit is used to obtain the required weight of each ratio of raw materials based on the predicted feed demand and the preset feed ratio scheme, and then weigh and measure each ratio of raw materials based on high-precision dynamic metering technology, and evenly mix all the obtained ratio of raw materials;

[0027] The automatic control unit is used to collect the weight signal of the feed based on the weighing sensor, and then realize automatic feeding by controlling the start and stop of the motor.

[0028] Optionally, the remote monitoring module includes a remote monitoring unit and a remote adjustment unit;

[0029] The remote monitoring unit is used to monitor the growth, feeding and breeding environment of the animals based on the remote terminal device;

[0030] The remote adjustment unit is used to adjust the feed delivery strategy at any time based on real-time monitoring data.

[0031] Optionally, the automatic control module further includes a formula optimization unit, which is used to monitor waste discharge and environmental pollution in the breeding environment, integrate environmentally friendly feed formula technology based on the monitoring results, and automatically adjust the preset feed ratio scheme.

[0032] The present invention also provides an electronic device, comprising: a memory and a processor; the memory is used to store programs; the processor is used to execute the programs to implement various modules of the intelligent feed distribution system for the breeding industry.

[0033] The present invention also provides a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, various modules of the intelligent feed distribution system for the breeding industry are implemented.

[0034] Compared with the prior art, the present invention has the following advantages and technical effects:

[0035] The present invention comprehensively considers multiple factors such as the growth data of the animals, the breeding environment data, the characteristics of the feed raw materials and the historical feed consumption of each animal when predicting the daily feed demand. By integrating these parameters and using the BP neural network model for prediction, the accuracy of the prediction can be improved, providing strong support for the scientific management of the breeding industry.

[0036] The present invention uses the sparrow search algorithm and the firefly algorithm to optimize the BP neural network parameters, thereby improving the prediction accuracy and global search capability of the model.

[0037] The present invention obtains the required weight of each ratio of raw materials based on the predicted feed demand and the preset feed ratio scheme, and then weighs and measures each ratio of raw materials based on high-precision dynamic metering technology, and evenly mixes all the obtained ratio of raw materials and automatically feeds them, which can achieve precise feeding and feed distribution.

[0038] The automated and intelligent feed distribution system proposed by the present invention can reduce manpower requirements and improve breeding efficiency.

[0039] The present invention can monitor the growth, feeding and breeding environment of animals in real time, thereby improving management efficiency.

[0040] The present invention reduces environmental pollution and improves the sustainability of breeding by considering environmentally friendly feed formulation technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0042] Figure 1 The figure is a schematic diagram of the structure of an intelligent feed distribution system for the breeding industry according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0044] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0045] Embodiment 1

[0046] like Figure 1 As shown, in this embodiment, a feed intelligent distribution system for the breeding industry is provided, comprising: a data acquisition module, a data processing module, a feed distribution module, an automatic control module and a remote control module connected in sequence;

[0047] The data acquisition module is used to obtain the growth data of the raised animals, the breeding environment data, the feed raw material characteristics and the historical feed consumption of each animal, and upload them to the data processing module in real time;

[0048] The data processing module is used to perform data cleaning and standardization processing on growth data, breeding environment data, feed raw material characteristics and feed consumption;

[0049] The feed rationing module is used to construct a BP neural network model, train the BP neural network model based on standardized growth data, breeding environment data, feed raw material characteristics and historical feed consumption, then optimize the trained BP neural network model based on the sparrow search algorithm and the firefly algorithm, and predict the feed demand of the breeding animals based on the optimized BP neural network model;

[0050] The automatic control module is used to obtain the required weight of each proportion of raw materials based on the predicted feed demand and the preset feed ratio scheme, weigh and measure based on high-precision dynamic metering technology, and then evenly mix all the obtained proportioned raw materials and automatically feed them;

[0051] The remote monitoring module is used to monitor the growth, feeding and breeding environment of animals in real time.

[0052] It can be implemented that the data acquisition module includes a growth data acquisition unit, an environmental data acquisition unit, a feed raw material acquisition unit and a feed consumption acquisition unit; the growth data acquisition unit is used to regularly measure the weight of animals based on an electronic scale, and to obtain the body length of animals based on a laser rangefinder or a camera; the environmental data acquisition unit is used to monitor the temperature and humidity in the breeding house based on a temperature and humidity sensor, to monitor the light intensity and lighting time in the breeding house based on a light intensity detector, and to monitor the gas concentration in the breeding house based on a gas sensor; the feed raw material acquisition unit is used to measure the flow and density of the feed based on a mass flow meter, and to measure the number and size of particles in the feed based on a particle counter; the feed consumption acquisition unit is used to obtain the historical feed consumption of each animal based on a weighing sensor.

[0053] As a specific example, regarding animal growth data: there are fewer sensors that directly measure animal growth data, but it can be inferred through indirect indicators such as weight and body length. For example, use an electronic scale to regularly measure the weight of the animal, and use a laser rangefinder or camera and image processing technology to measure the body length; mounting method: the electronic scale can be installed under the feeding pen or passage to facilitate the standing measurement of the animal, and the laser rangefinder or camera can be installed at a high place in the feeding house to ensure that the field of view covers the animal's activity area. For weight data, the growth curve can be directly recorded and drawn; for body length data, it needs to be measured regularly and combined with weight data to establish and analyze the growth model.

[0054] As a specific embodiment, a laser rangefinder can be used to collect the body length of an animal. The preparation work includes: determining the starting point and end point of the measurement, and clearing obstacles between the two points to ensure that the measurement path is unobstructed; appropriately setting the laser rangefinder, including adjusting the measurement mode, unit, etc. Operation steps: turn on the laser rangefinder, select the appropriate measurement mode, and set relevant parameters, such as unit, accuracy, etc.; aim the laser rangefinder at the starting point of the animal to ensure that the laser can irradiate the animal's body; press the "start" or "measure" button, and the laser rangefinder will emit a beam of laser to the animal's body; when the laser irradiates the end point of the animal's body, the rangefinder will record this distance and display the length value on the display; move the laser rangefinder away and record the displayed length value, which is the body length of the animal.

[0055] When using a camera for measurement, the specific process includes: selecting a depth camera, using it under the conditions of shielding from sunlight and collecting data at a distance greater than 0.6m and less than 1.2m to ensure imaging accuracy and measurement accuracy. Then use the depth camera to shoot the animal from multiple angles to obtain three-dimensional data of the animal's body surface. Post-process the acquired three-dimensional data, including registration and segmentation, then extract the feature points on the animal's body surface, calculate the animal's body length, and output the calculated animal's body length as a visualization result or numerical data.

[0056] Regarding the breeding environment data: the required sensors include temperature and humidity sensors: monitoring the temperature and humidity in the breeding house; light intensity detectors: monitoring light intensity and lighting time; ammonia sensors and carbon dioxide sensors: monitoring gas concentrations in the breeding house; oxygen sensors: monitoring oxygen content. Mounting method: These sensors should be evenly distributed in the breeding house to ensure that they can accurately reflect the overall environmental conditions. The sensors can be connected to the data acquisition module using wired or wireless methods. For the collected environmental data, pre-processing operations such as filtering and noise reduction are first required to improve the data quality. Then the data can be compressed and feature extracted as needed to facilitate subsequent analysis and modeling.

[0057] Regarding feed raw material characteristics: The sensors required include mass flow meters: measuring the flow rate and density of feed; particle counters: measuring the number and size of particles in feed; temperature sensors: measuring the temperature of feed. Mounting method: These sensors can be installed on feed processing or storage equipment to monitor the characteristics of feed raw materials in real time. For feed raw material characteristic data, data calibration and verification are required to ensure the accuracy of the measurement results. At the same time, feature extraction and data compression can also be performed to facilitate subsequent data analysis and use.

[0058] Regarding historical feed consumption: Feed weight is accurately measured by sensors, and the data is uploaded to the cloud platform. Mounting method: A weighing sensor is installed under the feed tower, and the data is transmitted to the cloud platform in real time through the IoT device. For the historical feed consumption obtained, pre-processing operations such as data cleaning and removal of outliers are required, and then the data can be saved as historical records for subsequent analysis and use.

[0059] It can be implemented that the data processing module includes a data cleaning unit and a data standardization unit; the data cleaning unit is used to check and process missing values, outliers and repeated data of growth data, breeding environment data, feed raw material characteristics and historical feed consumption; the data standardization unit is used to standardize the growth data and feed consumption based on Z-score normalization, and standardize the breeding environment data and feed raw material characteristics based on minimum-maximum normalization.

[0060] As a specific embodiment, cleaning of growth data: checking and processing missing values, outliers and duplicate data; standardization: selecting appropriate standardization methods according to specific needs, such as Z-score normalization, to eliminate dimensional differences between animals of different growth stages or different breeds. Cleaning of breeding environment data: processing missing values ​​and outliers caused by equipment failure or human error; standardization: since environmental data are usually continuous variables, and the dimensions of different environmental parameters may be different, Z-score normalization or minimum-maximum normalization is selected to unify the dimensions. Cleaning of feed raw material characteristics: processing missing values ​​and outliers caused by laboratory equipment failure or human operation errors; standardization: for the laboratory data of feed raw materials, since their dimensions and units may be different, such as protein content expressed as a percentage and moisture content expressed as a decimal, the minimum-maximum normalization is selected to unify to the same scale. Cleaning of historical feed consumption: processing missing values ​​and outliers caused by recording errors or omissions; standardization: it is necessary to compare feed consumption in different time periods or under different feeding conditions, and Z-score normalization is selected to eliminate the impact of time or condition differences.

[0061] It can be implemented that the feed distribution module includes a feed prediction unit and a model optimization unit; the feed prediction unit is used to construct a BP neural network model, taking standardized growth data, breeding environment data, feed raw material characteristics and historical feed consumption as input, and feed demand as output, using mean square error and gradient descent algorithm to train the BP neural network model, and predicting the feed demand of the farmed animals based on the trained BP neural network model; the model optimization unit is used to optimize the trained BP neural network model based on the sparrow search algorithm and the firefly algorithm.

[0062] It can be implemented that the model optimization unit includes a position update subunit and a secondary update subunit; the position update subunit is used to initialize the parameters of the sparrow search algorithm, take the number of sparrows as the number of BP neural network parameter groups, take the position of the sparrows as the parameter value of the BP neural network in different parameter groups, and obtain the initial position of the sparrows in the population; obtain the average value of the mean square error between the predicted feed demand and the actual feed demand as the fitness function, and then obtain the fitness values ​​of all sparrows in the population and sort them to obtain the global optimal position; select some sparrows as searchers, some sparrows as guards, and the remaining sparrows as beggars, move the positions of the searchers, guards and beggars respectively, obtain the position of the sparrow population after the move, and then update the global optimal position; the secondary update subunit is used to initialize the parameters of the firefly algorithm, in the firefly population, the firefly with the highest brightness moves, obtains the position update result of the firefly algorithm, based on the position update result of the firefly algorithm, further updates the position of the sparrow population in real time to obtain a new BP neural network parameter group.

[0063] As a specific embodiment, the feed requirement for the day is predicted based on the bioenergetics model, feed nutrients, stocking conditions of the breeding species, and environmental parameters collected in real time. First, based on the bioenergetics model of the breeding species, the basic energy requirements of the animal at different growth stages and weights, such as maintenance energy, growth energy, etc., are estimated, which usually involves consideration of factors such as the physiological characteristics, growth rate, and weight change of the animal. The nutritional components of the feed, including protein, fat, carbohydrates, etc., are analyzed, and their energy content, such as the energy value per kilogram of feed, is determined. This information will be used to convert the energy requirements of the animal into a specific amount of feed. According to the stocking density, growth stage, and weight range of the breeding species, the basic energy requirements are adjusted. Animals at different growth stages have different requirements for feed, and the stocking density will affect the animal's feeding competition and growth environment, so it needs to be considered comprehensively. The breeding environment parameters are collected in real time. These parameters have an important impact on the physiological state and metabolic rate of the animal, so they need to be taken into account. The above parameters are integrated and an appropriate mathematical model or algorithm is used to predict the feed requirement for the day.

[0064] As an implementable method, the following steps can be followed:

[0065] Basic demand calculation: Calculate the basic energy demand of animals based on the bioenergetics model and the stocking conditions of the farmed species.

[0066] Environmental parameter adjustment: According to the real-time collected environmental parameters, the basic energy requirements are appropriately adjusted. For example, changes in the environment can affect the metabolic rate of animals, which in turn affects feed requirements.

[0067] Feed conversion: The conversion of adjusted energy requirements into specific feed amounts, calculated based on the energy content of the feed.

[0068] Verify and adjust the prediction results: Verify the prediction results by comparing them with the actual breeding situation and make adjustments as needed. Since the growth and feeding behavior of animals are affected by many factors, there may be a certain error in the prediction results. By actually observing and recording the feeding situation and growth status of animals, the prediction model can be continuously optimized and improved to improve the accuracy of the prediction.

[0069] This embodiment adopts BP neural network model to predict the feeding demand, takes the growth data of the raised animals, the breeding environment data, and the feed raw material characteristics as input, and the corresponding feed consumption and demand as output. Select appropriate loss function, such as mean square error, and optimization algorithm, such as gradient descent algorithm, to train the network. Through multiple iterations, the weights and biases of the network are adjusted so that the network can accurately predict the feed demand and the growth of the animals. Finally, the sparrow search algorithm and the firefly algorithm are applied to optimize the BP neural network parameters to improve the prediction accuracy and global search ability of the model.

[0070] As a specific embodiment, the BP neural network sets the number of training times to 1000, the learning rate to 0.01, and the target minimum error to 0.00001; the initial population size of the sparrow search algorithm is 30, the maximum number of iterations is 50, the proportions of searchers and guardians of sparrows are 0.7 and 0.2 respectively, and the safety threshold is 0.6; the population size of the firefly disturbance is 30, the light absorption coefficient is 1, the maximum attraction is 2, and the step factor is 1.

[0071] Among them, the sparrows with the best position are selected as searchers according to the set number, and the moving position of the searcher can be represented by the following function:

[0072]

[0073] in, represents the position of the vth sparrow in the u-dimensional space after iteration t, v∈[1,2,K,n], t represents the current iteration number, t max is a constant, indicating the maximum number of iterations, α is any random number in the interval (0,1]; r d is a random number that follows a normal distribution, K represents a 1×d unit matrix; S a represents the alarm value of the searcher, which is a random number in the interval (0,1); S m The safety value of the searcher is a random number in the interval (0.5,1). They determine the movement range of the sparrow.

[0074] The remaining sparrows act as beggars, and their moving positions can be updated as follows:

[0075]

[0076] Where A represents a 1×d matrix whose elements are random values ​​1 or -1. + =A T (AA T ) -1 ;P wst represents the current global worst position of the searcher, P op Indicates the optimal position it occupies in the current global situation;

[0077] At the same time, some sparrows are randomly selected as guards according to the set number, and the mathematical expression of their moving positions can be obtained as follows:

[0078]

[0079] P best (t) represents the current global optimal value; represents any random parameter that obeys the standard normal distribution with a restricted step size; θ represents a random number in the range [-1,1]; q v The current sparrow's fitness value; q b and q w They represent the optimal and worst fitness values ​​of the current sparrow respectively; γ represents a small constant that must be ensured that the denominator is not zero.

[0080] In the firefly population, brighter fireflies are not attracted by other fireflies and move randomly. The position update formula of the brightest firefly in the population is:

[0081]

[0082] Among them, y fb Indicates the optimal position of the current global firefly individual.

[0083] According to the position update formula of the firefly algorithm mentioned above, the position of the sparrow population is further updated to obtain a new BP neural network parameter group.

[0084] It can be implemented that the automatic control module includes a dynamic metering unit and an automatic control unit; the dynamic metering unit is used to obtain the required weight of each ratio of raw materials based on the predicted feed demand and the preset feed ratio scheme, and then weigh and measure each ratio of raw materials based on high-precision dynamic metering technology, and evenly mix all the obtained ratio of raw materials; the automatic control unit is used to collect the weight signal of the feed based on the weighing sensor, and then realize automatic feeding by controlling the start and stop of the motor to achieve precise feeding.

[0085] As a specific embodiment, the automatic control module is one of the key technologies for achieving precision feeding in modern animal husbandry. It ensures that animals get a balanced nutritional intake by accurately controlling the distribution and delivery of feed, thereby improving breeding efficiency and animal health.

[0086] Among them, the dynamic metering unit is the first step of the automatic control module. Its main task is to determine the required weight of each raw material. This process is based on two key factors: predicted feed demand and preset feed ratio scheme. Predicted feed demand: By analyzing factors such as animal type, age, weight, growth stage and breeding environment, the dynamic metering unit can predict the amount of feed required by the animal in a certain period of time. This prediction can be adjusted based on historical data and real-time monitoring data. Preset feed ratio scheme: According to the nutritional needs of different animals and the characteristics of feed raw materials, professionals will design one or more feed ratio schemes. These schemes will list the ratios of various raw materials in detail to ensure the balance of nutrients in the feed. High-precision dynamic metering technology: After determining the required weight of each raw material, the dynamic metering unit uses high-precision weighing sensors and metering equipment to accurately weigh each raw material. This process requires extremely high accuracy to avoid feed nutritional imbalance caused by raw material weight errors. Even mixing: The weighed raw materials need to be evenly mixed by mixing equipment. This step is crucial because only even mixing can ensure that the nutritional content of each feed meets the preset ratio.

[0087] The automatic control unit is the executive part of the automatic control module, which is responsible for adjusting the feed delivery according to real-time data. Weighing sensor collection: The automatic control unit collects the weight signal of the feed in real time through the weighing sensor installed in the feed delivery system. These signals reflect the actual weight of the current feed and are the basis for the automatic control unit to make decisions. Control the start and stop of the motor: Based on the signals collected by the weighing sensor, the automatic control unit will calculate whether it is necessary to continue to deliver feed. If the feed weight does not reach the preset threshold, the control unit will instruct the motor to continue working and deliver more feed; if it reaches or exceeds the threshold, it will stop the motor to avoid over-delivery. Achieve precise feeding: The automatic control unit can achieve precise feeding by precisely controlling the start and stop of the motor. This means that each portion of feed delivered can accurately meet the nutritional needs of the animal while avoiding waste. Data feedback and adjustment: The automatic control unit will also feed back real-time data to the dynamic metering unit so that the dynamic metering unit can adjust the prediction model and ratio scheme according to the actual situation. This closed-loop control mechanism ensures continuous optimization of the entire feeding process.

[0088] The automatic control module achieves accurate feed metering, uniform mixing and precise delivery through the coordinated work of the dynamic metering unit and the automatic control unit. The application of this technology greatly improves breeding efficiency, reduces feed costs, and ensures the health and production performance of animals.

[0089] It is feasible that the remote monitoring module includes a remote monitoring unit and a remote adjustment unit; the remote monitoring unit is used to monitor the growth, feeding and breeding environment of animals based on a remote terminal device; the remote adjustment unit is used to adjust the feed delivery strategy at any time based on real-time monitoring data.

[0090] It is feasible that the automatic control module further includes a formula optimization unit, which is used to monitor waste discharge and environmental pollution in the breeding environment, integrate environmentally friendly feed formula technology based on the monitoring results, and automatically adjust the preset feed ratio scheme.

[0091] As a specific embodiment, the system can integrate environmentally friendly feed formulation technology to reduce environmental pollution in the breeding process by adopting low-nitrogen, low-phosphorus, low-emission feed raw materials and environmentally friendly additives. The system can monitor the waste discharge and environmental pollution of the farm, and automatically adjust the feed formula and feeding strategy according to the monitoring results to reduce the negative impact on the environment. Through the collaborative work of the above technical components, the intelligent feed distribution system can achieve precise control, intelligent management and environmental optimization of animal feed distribution, and improve breeding efficiency and economic benefits.

[0092] The present invention also provides an electronic device, comprising: a memory and a processor; the memory is used to store programs; the processor is used to execute the programs to implement various modules of the intelligent feed distribution system for the breeding industry.

[0093] The present invention also provides a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, various modules of the intelligent feed distribution system for the breeding industry are implemented.

[0094] The present invention comprehensively considers multiple factors such as the growth data of the animals, the breeding environment data, the characteristics of the feed raw materials and the historical feed consumption of each animal when predicting the daily feed demand. By integrating these parameters and using the BP neural network model for prediction, the accuracy of the prediction can be improved, providing strong support for the scientific management of the breeding industry; the sparrow search algorithm and the firefly algorithm are used to optimize the BP neural network parameters to improve the prediction accuracy and global search capability of the model.

[0095] The automated and intelligent feed distribution system proposed in the present invention can reduce manpower requirements and improve breeding efficiency; improve management efficiency by real-time monitoring of animal growth, feeding conditions and breeding environment conditions; and reduce environmental pollution and improve the sustainability of breeding by considering environmentally friendly feed formulation technology.

[0096] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent feed distribution system for aquaculture, characterized in that: include: A data acquisition module, a data processing module, a feed distribution module, an automatic control module and a remote monitoring module are connected in sequence; The data acquisition module is used to obtain the growth data of the raised animals, the breeding environment data, the feed raw material characteristics and the historical feed consumption of each animal, and upload them to the data processing module in real time; The data processing module is used to perform data cleaning and standardization processing on growth data, breeding environment data, feed raw material characteristics and feed consumption; The feed rationing module is used to construct a BP neural network model, train the BP neural network model based on standardized growth data, breeding environment data, feed raw material characteristics and historical feed consumption, then optimize the trained BP neural network model based on the sparrow search algorithm and the firefly algorithm, and predict the feed demand of the breeding animals based on the optimized BP neural network model; The automatic control module is used to obtain the required weight of each proportion of raw materials based on the predicted feed demand and the preset feed ratio scheme, weigh and measure based on high-precision dynamic metering technology, and then evenly mix all the obtained proportioned raw materials and automatically feed them; The remote monitoring module is used to monitor the growth, feeding and breeding environment of animals in real time.

2. The intelligent feed distribution system for aquaculture according to claim 1 is characterized in that: The data collection module includes a growth data collection unit, an environmental data collection unit, a feed raw material acquisition unit and a feed consumption acquisition unit; The growth data acquisition unit is used to regularly measure the weight of the animal based on an electronic scale and obtain the body length of the animal based on a laser rangefinder or a camera; The environmental data acquisition unit is used to monitor the temperature and humidity in the breeding house based on the temperature and humidity sensor, monitor the light intensity and lighting time in the breeding house based on the light intensity detector, and monitor the gas concentration in the breeding house based on the gas sensor; The feed material acquisition unit is used to measure the flow rate and density of the feed based on the mass flow meter, and to measure the number and size of particles in the feed based on the particle counter; The feed consumption acquisition unit is used to acquire the historical feed consumption of each animal based on a weighing sensor.

3. The intelligent feed distribution system for aquaculture according to claim 1 is characterized in that: The data processing module includes a data cleaning unit and a data standardization unit; The data cleaning unit is used to check and process missing values, abnormal values ​​and duplicate data of growth data, breeding environment data, feed raw material characteristics and historical feed consumption; The data standardization unit is used to standardize the growth data and feed consumption based on Z-score normalization, and to standardize the breeding environment data and feed raw material characteristics based on minimum-maximum normalization.

4. The intelligent feed distribution system for aquaculture according to claim 1 is characterized in that: The feed distribution module includes a feed prediction unit and a model optimization unit; The feed prediction unit is used to construct a BP neural network model, which takes the standardized growth data, breeding environment data, feed raw material characteristics and historical feed consumption as input, takes the feed requirement as output, trains the BP neural network model using mean square error and gradient descent algorithm, and predicts the feed requirement of the breeding animals based on the trained BP neural network model; The model optimization unit is used to optimize the trained BP neural network model based on the sparrow search algorithm and the firefly algorithm.

5. The intelligent feed distribution system for aquaculture according to claim 4 is characterized in that: The model optimization unit includes a position updating subunit and a secondary updating subunit; The position updating subunit is used to initialize the parameters of the sparrow search algorithm, taking the number of sparrows as the number of BP neural network parameter groups, taking the positions of sparrows as the parameter values ​​of the BP neural network in different parameter groups, and obtaining the initial positions of the sparrows in the population; The mean value of the mean square error between the predicted feed demand and the actual feed demand is obtained as the fitness function, and then the fitness values ​​of all sparrows in the population are obtained and sorted to obtain the global optimal position; some sparrows are selected as searchers, some sparrows as guards, and the remaining sparrows as beggars, and the positions of the searchers, guards and beggars are moved respectively to obtain the position of the sparrow population after the movement, and then the global optimal position is updated; The secondary update subunit is used to initialize the firefly algorithm parameters. In the firefly population, the firefly with the highest brightness moves to obtain the position update result of the firefly algorithm. Based on the position update result of the firefly algorithm, the position of the sparrow population is further updated in real time to obtain a new BP neural network parameter group.

6. The intelligent feed distribution system for aquaculture according to claim 1 is characterized in that: The automatic control module includes a dynamic metering unit and an automatic control unit; The dynamic metering unit is used to obtain the required weight of each ratio of raw materials based on the predicted feed demand and the preset feed ratio scheme, and then weigh and measure each ratio of raw materials based on high-precision dynamic metering technology, and evenly mix all the obtained ratio of raw materials; The automatic control unit is used to collect the weight signal of the feed based on the weighing sensor, and then realize automatic feeding by controlling the start and stop of the motor.

7. The intelligent feed distribution system for aquaculture according to claim 1 is characterized in that: The remote monitoring module includes a remote monitoring unit and a remote adjustment unit; The remote monitoring unit is used to monitor the growth, feeding and breeding environment of the animals based on the remote terminal device; The remote adjustment unit is used to adjust the feed delivery strategy at any time based on real-time monitoring data.

8. The intelligent feed distribution system for aquaculture according to claim 6, characterized in that: The automatic control module also includes a formula optimization unit, which is used to monitor waste discharge and environmental pollution in the breeding environment, integrate environmentally friendly feed formula technology based on the monitoring results, and automatically adjust the preset feed ratio scheme.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store programs; the processor is used to execute the programs to implement the various modules of the intelligent feed distribution system for the aquaculture industry as described in any one of claims 1-8.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each module of the intelligent feed distribution system for the aquaculture industry as described in any one of claims 1 to 8 is implemented.

Citation Information

Cited By

  • Intelligent pasture concentrate control management method and system based on Internet of Things

    CN119247911A

  • A method and system for intelligent pasture concentrate control management based on the Internet of Things

    CN119247911B

  • Intelligent automatic feed feeding system based on Internet of Things

    CN119453088A

  • Natural enemy insect mantis feeding management method based on environment simulation

    CN119655228A

  • Automatic feed treatment system and treatment method for livestock breeding

    CN120111070A