Digital intelligent agricultural planting and breeding combined system

Abstract

The invention relates to the technical field of intelligent agricultural planting and breeding, and discloses a digital intelligent agricultural planting and breeding combined system which comprises a multi-source sensing module used for collecting environment data, biological sign data and equipment operation data through a planting area sensor, a breeding area sensor and a circular processing equipment sensor; the data verification module is used for dynamically verifying the collected data based on a preset agronomic mechanism model; the hierarchical digital twinning module is used for carrying out collaborative decision making in a three-level architecture of the edge computing node, the local server and the cloud platform and generating an optimization instruction; a virtual-real decision engine module; and a closed loop execution network module. According to the invention, a lightweight twinborn model is deployed through the edge layer to realize millisecond-level equipment control and guarantee timely response of extreme working conditions, the field area layer coordinates planting and breeding material circulation through a dynamic optimization algorithm, and the cloud layer integrates climate data to generate a carbon sink strategy of block chain evidence storage. And the real-time decision-making precision, the resource utilization efficiency and the carbon economic value are improved through three-level cooperation.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent agricultural breeding technology, in particular to a digital intelligent agricultural breeding combination system. BACKGROUND

[0002] In the field of intelligent agricultural technology, the digital collaborative management of the breeding combination system is the core link to realize the efficient circulation of agricultural resources and the precise prevention and control of production risks, especially suitable for multi-industry integration scenarios such as livestock breeding, crop planting and organic fertilizer production. The multi-source data fusion accuracy and dynamic decision real-time of the system directly affect the utilization rate of fecal resource, biological safety prevention and control effect and carbon sink asset development efficiency, and put forward strict requirements on data collection reliability, decision chain response speed and execution closed loop accuracy.

[0003] Among the current commonly used technical means in the industry, the decision execution system relies on a central server for unified scheduling, and the edge node only bears the function of data forwarding, resulting in a delay in device control response, which cannot meet the millisecond-level emergency scenarios such as sudden increase in ammonia concentration and sharp drop in dissolved oxygen, and the carbon sink measurement is based on fixed conversion coefficient manual record processing, lacking full-link trusted traceability support.

[0004] The above prior art scheme at least has the following technical problems: the centralized decision architecture leads to a lag in response to extreme working conditions, the static resource model causes a decrease in nitrogen utilization rate, and the carbon sink assets are difficult to market due to low measurement reliability. SUMMARY

[0005] In order to make up for the above shortcomings, the present application provides a digital intelligent agricultural breeding combination system, which aims to improve the problems of lag in response to extreme working conditions under centralized decision architecture, low nitrogen utilization rate caused by static resource model, and difficulty in market circulation of carbon sink assets due to lack of measurement reliability.

[0006] In a first aspect, the present application provides the following technical scheme: a digital intelligent agricultural breeding combination system, comprising: A multi-source perception module for collecting environmental data, biological sign data and device operation data through planting area sensors, breeding area sensors and circulating treatment equipment sensors; A data verification module for dynamically verifying the collected data based on a pre-set agricultural mechanism model; A hierarchical digital twin module for collaborative decision-making in a three-level architecture of edge computing nodes, local servers and cloud platforms, and generating optimized instructions; A virtual-real decision engine module for sand table pre-performance and backtracking retry of the decision instructions, and outputting verified instructions; A closed-loop execution network module for converting the verified instructions into device control signals and collecting execution effect data feedback to the multi-source perception module.

[0007] By adopting the technical scheme, the multi-source heterogeneous sensor network is deployed to synchronously collect planting environment, breeding signs and circulating equipment data, a global digital base of agriculture is constructed, dynamic compensation of environmental interference and spatio-temporal consistency verification are performed based on an agricultural mechanism model, a trusted data set with a confidence label is output, millisecond-level equipment control, resource dynamic optimization and blockchain carbon sink strategy generation are realized relying on a three-level twin architecture of edge, field and cloud, a virtual sand table deduction and decay weight retry mechanism is introduced to intercept high-risk instructions, finally, a protocol adaptive conversion execution signal is output, and closed-loop feedback is performed to the perception layer based on a deviation rate, and production safety is strengthened, resource efficiency is improved and carbon economy is increased in value.

[0008] Preferably, the multi-source perception module comprises: a planting environment perception unit, which collects soil moisture, crop growth and pest characteristics data through soil temperature and humidity sensors, light intensity sensors, leaf humidity sensors and multispectral imagers; a breeding sign perception unit, which collects biological physiological indicators and activity anomaly data through livestock body temperature monitoring ear tags, respiration rate sensors, aquaculture dissolved oxygen sensors and behavior analysis cameras; a circulating equipment monitoring unit, which collects resource circulation equipment operating state parameters through biogas flow meters, manure solid-liquid separator current sensors, water treatment pH probes and organic fertilizer fermentation temperature probes.

[0009] Preferably, the data verification module comprises: an environmental coupling compensation unit for constructing an environmental interference matrix , wherein is the influence coefficient of the th environmental factor on the th sensor; a multi-source data collaborative correction is performed; wherein is the original value of the th sensor, is the change amount of the environmental factor; a spatio-temporal consistency unit for establishing a spatio-temporal distribution model by historical data flow to detect abnormal deviation points; when multiple sensors monitor the same physical quantity, a fusion value is calculated according to a confidence weight ; a confidence encapsulation unit for calculating an overall confidence ; when or , a multi-source conflict label is embedded, and a uniform spatio-temporal stamp is added to output a data packet.

[0010] Preferably, the hierarchical digital twin module comprises: An edge response unit deployed at an edge computing node, configured to call a device-level twin model, including a livestock and poultry environment regulation model and an irrigation response model, to generate device control instructions according to real-time data; A field area coordination unit deployed at a local server, for running a breeding-coupling twin sub-model and optimizing resource allocation paths through a material flow balance equation, specifically calculating the amount of manure returned to the field; A strategy generation unit deployed at a cloud platform, for generating a global strategy by integrating market, climate and policy data, calculating carbon sink assets based on a specific carbon conversion rate, and generating a blockchain-verified carbon sink report.

[0011] Preferably, the livestock and poultry environment regulation model in the edge response unit is a PID controller, with a proportional coefficient , and an integral coefficient ; The material flow balance equation in the field area coordination unit is ; Wherein, is the dynamic utilization efficiency is the monthly average temperature; The specific carbon conversion rate in the strategy generation unit is , and a unique associated carbon sink report is generated through blockchain.

[0012] Preferably, the virtual-real decision engine module comprises: A pre-play deduction unit that loads the optimization instructions generated by the hierarchical digital twin module into a virtual twin body; A call history disaster model, including a disease transmission path and an extreme weather impact, simulates the consequences of instruction execution, and the pre-play duration is dynamically set to ; A backtracking retry unit that triggers strategy backtracking when the pre-play shows economic loss > 15% or resource waste > 20%, performs multiple rounds of retry with a decay weight, and the decision weight of the first backtracking , the maximum number of retries ; A conflict arbitration unit that receives multi-source conflict markers embedded in the data verification module, freezes the instructions and requests human intervention when the conflict markers trigger or the number of pre-play failures > 2, and outputs the final verified instructions to the closed-loop execution network module.

[0013] Preferably, the closed-loop execution network module comprises: A protocol conversion unit that converts the verified instructions into executable signals for industrial devices, and dynamically loads driving plug-ins based on different agricultural breeding devices; An execution feedback unit collects execution state data in real time, and calculates an execution deviation rate , wherein is a device feedback value, is an instruction expected value; A data closed loop unit generates a calibration instruction, updates a twin model parameter, and structures feedback data including a time and space stamp and a device ID when , and returns the data to a multi-source perception module.

[0014] In a second aspect, the present application provides the following technical solution, a digital smart agricultural planting and breeding combination method, the method comprising: S1, multi-source data acquisition, through a sensor cluster of a planting area, a breeding area and a circulating treatment device, synchronously collecting environmental data, biological sign data and device operation data; S2, dynamic data verification, based on a pre-set agricultural mechanism model, performing cross-domain verification on the collected data, and outputting a trusted data set with a confidence label; S3, hierarchical collaborative decision-making, running a digital twin in a three-level architecture of an edge node, a local server and a cloud platform, and generating planting and breeding resource optimization instructions; S4, instruction pre-play verification, sand table deduction and backtracking retry are performed on the optimization instructions, and a risk-controllable verified instruction is outputted; S5, closed loop execution feedback, converting the verified instruction into a device control signal, collecting execution effect data and feeding back to step S1.

[0015] In a third aspect, the present application provides the following technical solution, a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to realize the digital smart agricultural planting and breeding combination method described above. In a fourth aspect, the present application provides the following technical solution, a readable storage medium, wherein the readable storage medium stores a computer program, and the computer program is executed by a processor to realize the digital smart agricultural planting and breeding combination method described above.

[0016] The present application has the following beneficial effects: 1、In the present application, a lightweight twin model is deployed in an edge layer to realize millisecond-level device control, guaranteeing timely response in extreme working conditions, a field area layer coordinates planting and breeding material circulation through a dynamic optimization algorithm, and a cloud layer integrates climate data to generate a carbon sink strategy stored in a blockchain, and three-level collaboration improves real-time decision-making accuracy, resource utilization efficiency and carbon economic value.

[0017] 2、In the application, through planting, breeding, and synchronous collection of full-factor data by multi-source sensors of circulation equipment, an environmental interference matrix is constructed based on an agronomic mechanism model to eliminate cross-domain influence, and through spatiotemporal consistency fusion and credibility packaging, check data with conflict markers are output, thereby realizing high-trust data supply in complex agricultural scenes.

[0018] 3、In the application, the historical disaster model verification instruction risk is loaded through sand table deduction, the strategy is optimized by using a decay weight backtracking mechanism, high-risk operations are frozen in combination with conflict markers, industrial signals are dynamically converted in the execution stage, and a deviation rate is calculated, a twin model is triggered for online calibration, structured feedback data drive dynamic revision of the perception end, and a decision-making, execution, and calibration closed loop is realized. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The figure is a structural diagram of the digital smart agricultural planting and breeding combination system provided by the application. Figure 2 The figure is a flowchart of the digital smart agricultural planting and breeding combination method provided by the application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0021] Embodiment one Reference Figure 1 In the first embodiment of the application, the application provides a digital smart agricultural planting and breeding combination system, which comprises: A multi-source perception module is used to collect environmental data, biological sign data, and equipment operation data through planting area sensors, breeding area sensors, and circulation processing equipment sensors. A data check module is used to dynamically check the collected data based on a pre-set agronomic mechanism model. A hierarchical digital twin module is used to generate optimization instructions through collaborative decision-making in a three-level architecture of edge computing nodes, local servers, and cloud platforms. A virtual-real decision engine module is used to perform sand table pre-performance and backtracking retry on the decision instructions, and output verified instructions. A closed-loop execution network module is used to convert the verified instructions into equipment control signals, and collect execution effect data feedback to the multi-source perception module.

[0022] Specifically, the multi-source perception module uses three types of sensor networks deployed in planting areas, breeding areas, and recycling processing equipment to simultaneously collect environmental parameters (temperature, humidity, light, and gases), biological signs (body temperature, respiration, and activity), and equipment operation data. This builds a comprehensive agricultural data base covering the entire supply chain, providing the system with real-time, multi-dimensional, and heterogeneous raw perception information of the physical world, and achieving comprehensive digital mapping of agricultural production factors. The data verification module dynamically corrects the raw data collected by the multi-source perception module based on a preset agronomic mechanism model, eliminates environmental interference, and converts sensor signals into reliable data sets that conform to the laws of agricultural science, achieving reliable input from raw physical quantities to agronomic decision-making. The hierarchical digital twin module deploys linked twins across a three-tiered architecture: edge, field, and cloud. This integrates agronomic rules with real-time data to generate collaborative optimization instructions covering equipment control, farming resource allocation, and carbon sequestration asset development, enabling an intelligent transition from local response to system-level decision-making. The virtual-reality decision engine module conducts virtual sandbox simulations and multiple rounds of backtracking and retrying on the optimization instructions generated by the hierarchical digital twin module, intercepting high-risk erroneous decisions and outputting reliable instructions that have been verified by the risk. This enables the safe transformation from theoretical decisions to executable instructions, providing a high-confidence action plan guarantee for agricultural operations. The closed-loop execution network module converts the verified instructions output by the virtual-reality decision engine into drive signals recognizable by industrial control equipment. At the same time, it collects the actuator response status and physical environment change data in real time to form a closed-loop control of instructions, execution and feedback, and realize the dynamic calibration cycle of agricultural control strategies from digital space to physical manipulation.

[0023] The multi-source perception module includes: The planting environment perception unit collects data on soil moisture, crop growth, and pest and disease characteristics through soil temperature and humidity sensors, light intensity sensors, leaf wetness sensors, and multispectral imagers; The livestock and poultry vital signs sensing unit collects data on biophysiological indicators and abnormal activities through livestock and poultry temperature monitoring ear tags, respiratory rate sensors, aquatic dissolved oxygen sensors, and behavior analysis cameras; The circulation equipment monitoring unit collects the operating status parameters of the resource circulation equipment through the biogas flow meter, the manure solid-liquid separator current sensor, the water treatment pH probe and the organic fertilizer fermentation temperature probe.

[0024] Specifically, the planting environment perception unit monitors the root layer water dynamics in real time through the soil temperature and humidity sensor, captures the photosynthetically active radiation flux through the light intensity sensor, warns of the breeding environment of fungal diseases through the leaf humidity sensor, analyzes the crop chlorophyll content and stress characteristics through the multispectral imager, constructs the holographic image of crop growth, realizes the stereoscopic monitoring from soil moisture to canopy physiology, and provides data foundation for precision irrigation and disease control.

[0025] The breeding sign perception unit uses the livestock ear tag to continuously track the core body temperature fluctuation, the respiratory frequency sensor to diagnose respiratory disease signs, the dissolved oxygen sensor to ensure the water breathing safety threshold of aquatic organisms, and the behavior analysis camera to capture abnormal activities (such as pecking anus and fighting) through posture recognition algorithm, realizing the digital health management from individual physiology to group behavior. The circulating equipment monitoring unit quantifies the manure gas production efficiency through the biogas flow meter, detects the bearing wear and overload risk through the current sensor of the solid-liquid separator, dynamically controls the neutralization reaction process through the water treatment pH probe, and ensures the activity threshold of thermophilic bacteria through the organic fertilizer fermentation temperature probe, realizing the process closed-loop optimization from energy conversion to fertilizer generation.

[0026] The data verification module includes: The environment coupling compensation unit is used to build an environment interference matrix , wherein, is the influence coefficient of the first environmental factor on the first sensor; perform multi-source data collaborative correction ; , wherein, is the original value of the first sensor, is the change amount of the environmental factor; The space-time consistency unit establishes a space-time distribution model through historical data flow to detect abnormal deviation points; When multiple sensors monitor the same physical quantity, the fusion value is calculated according to the confidence weight ; The credibility packaging unit calculates the overall confidence ; When or , embed multi-source conflict markers and add a unified space-time stamp output data packet.

[0027] Specifically, the environmental coupling compensation unit constructs an environmental interference matrix to quantify the cross-system influencing factors (such as the corrosion deviation of ammonia gas in livestock and poultry houses on EC sensors), eliminates cross interference between devices through a dynamic compensation formula, restores multi-source sensing data to physical true values, and provides clean data foundation conforming to agricultural mechanisms for decision-making. The spatio-temporal consistency unit constructs a spatio-temporal probability distribution model based on historical data flow (such as the gradual change law of soil moisture content with rainfall events), detects abnormal deviation points and automatically reduces the weight of failed sensors, and when multiple sensors monitor the same target, it fuses the optimal estimate value according to the confidence weight, solves the problem of accidental error of single-point data, and guarantees the continuity and reliability of data in time dimension and space grid. The credibility packaging unit calculates the overall confidence and identifies conflicting data combinations (such as sudden drop of dissolved oxygen but normal fish activity), embeds conflict markers and spatio-temporal stamps for low credibility (ρ<0.7) or significantly deviated data, and outputs structured data packets with quality labels, providing risk interception anchor points for subsequent decision engine.

[0028] The hierarchical digital twin module includes: The edge response unit is deployed on the edge computing node and is configured to call the device-level twin model, including the livestock and poultry environment regulation model and the irrigation response model, to generate device control instructions according to real-time data. The field area coordination unit is deployed on the local server to run the crop-livestock coupled twin sub-model and optimize resource allocation paths through material flow balance equations, specifically calculating the amount of manure returned to the field. The strategy generation unit is deployed on the cloud platform to generate global strategies by integrating market, climate and policy data, calculate carbon sink assets based on specific carbon conversion rates, and generate blockchain-verified carbon sink reports.

[0029] Specifically, the edge response unit deploys lightweight twin models (such as livestock and poultry ventilation PID controllers) on the edge computing node, generates control instructions based on real-time sensing data (ammonia concentration / soil moisture), and realizes precise regulation of livestock shed temperature and humidity and dynamic response of irrigation valve opening, ensuring the safe operation of devices in extreme conditions (such as automatic oxygenation in hypoxia). The field area coordination unit analyzes the material flow relationship (such as poultry manure nitrogen and phosphorus-crop fertilizer requirement) through the crop-livestock coupled twin body, dynamically optimizes the resource allocation path with the minimum supply-demand balance equation, synchronously integrates weather forecast and device state data, realizes precise calculation of manure returned to the field and feed feeding ratio, and improves resource recycling rate. The strategy generation unit integrates futures prices, carbon sink policies and climate models to generate quarterly production plans and carbon asset development schemes, calculates the carbon sink value of manure treatment based on carbon conversion rate (β=0.28), and generates spatio-temporally bound tamper-proof reports through blockchain, making agricultural waste resourceization produce tradable carbon credit assets.

[0030] The livestock and poultry environment regulation model in the edge response unit is a PID controller, the proportional coefficient , and the integral coefficient ; The material flow balance equation in the field area coordination unit ; , wherein is the dynamic utilization efficiency is the monthly average temperature The specific carbon conversion rate in the strategy generation unit , and a unique associated carbon sink report is generated through the blockchain.

[0031] Specifically, the PID controller of the edge response unit realizes rapid response to environmental mutations (such as a sudden increase in ammonia concentration) through the proportional coefficient , eliminates steady-state control bias through the integral coefficient , ensures millisecond-level accurate adjustment of the livestock and poultry house environment parameters, and guarantees the continuous stability of the biological growth environment. The material flow balance of the field area coordination unit drives the cross-season adaptive allocation of resources with dynamic efficiency, optimizes the amount of manure returned to the field in combination with the minimum supply and demand principle, avoids resource mismatching problems in the crop-livestock system, and realizes efficient recycling of nitrogen and phosphorus nutrients. The carbon sink assetization in the strategy unit is based on the experiment-locked carbon conversion rate to calculate the carbon sink value of manure treatment, and a time and space bound tamper-proof report (hash value fusion of processing amount, timestamp and geographic coordinates) is generated through the blockchain to create verifiable agricultural carbon sink income.

[0032] The virtual-real decision engine module includes: The pre-play deduction unit loads the optimization instructions generated by the hierarchical digital twin module to the virtual twin body. The historical disaster model is called, including the epidemic disease transmission path and the influence of extreme weather, the consequences after the execution of the instructions are simulated, and the pre-play duration is dynamically set to ; The backtracking retry unit triggers strategy backtracking when the pre-play shows economic loss > 15% or resource waste > 20%, adopts a decay weight for multiple rounds of retry, and the decision weight of the first backtracking , the maximum number of retries ; The conflict arbitration unit receives the multi-source conflict markers embedded by the data verification module, freezes the instructions and requests manual intervention when the conflict markers trigger or the number of pre-play failures > 2, and outputs the final verification instructions to the closed-loop execution network module.

[0033] Specifically, the pre-rehearsal deduction unit loads the optimization instructions into the virtual twin, calls the historical disaster model (such as the spread dynamics of avian influenza, the impact of rainstorm floods), simulates the consequences of execution, dynamically sets the pre-rehearsal duration (intelligently adjusted based on historical loss cycles), quantifies potential economic losses and resource waste risks, visualizes the consequences of decision-making, and intercepts high-misjudgment instructions in advance; The backtracking and retry unit triggers strategy backtracking when the pre-rehearsal loss exceeds the threshold (economic loss > 15% or resource waste > 20%), uses an exponential decay weight for multiple rounds of instruction optimization, limits the maximum number of retries (N = 3) to avoid invalid calculations, preserves the core logic of the original strategy while gradually approaching the optimal solution through a weight decay mechanism, and balances decision-making efficiency and reliability; The conflict arbitration unit analyzes the conflict markers embedded in the data verification module in real time, freezes the instruction stream when the markers trigger or the pre-rehearsal fails consecutively, automatically pushes an alarm to the artificial decision-making end for intervention, and only outputs verified instructions that pass arbitration, ensuring the safety of executed instructions in extreme scenarios.

[0034] The closed-loop execution network module includes: The protocol conversion unit converts the verified instructions into executable signals for industrial devices and dynamically loads driver plug-ins based on different agricultural breeding devices; The execution feedback unit collects execution state data in real time and calculates the execution deviation rate , where is the device feedback value, is the expected value of the instruction; The data closed-loop unit generates calibration instructions, updates twin model parameters, and structures feedback data when , which includes time and space stamps and device IDs, and is returned to the multi-source perception module.

[0035] Specifically, the protocol conversion unit dynamically converts the verified instructions into executable signals (such as PWM waves) for industrial control devices, loads special driver plug-ins (Modbus / CANopen protocol libraries) for different agricultural devices (feeding machines, oxygenation pumps, irrigation valves), eliminates communication between heterogeneous devices, and ensures the accurate delivery of control signals in complex agricultural scenarios; The execution feedback unit collects actuator response data (such as valve opening and motor speed) in real time, quantifies execution reliability through the deviation rate formula, identifies mechanical failures (when sudden increase > 20%) or signal distortion (when continuously > 10%); The data closed-loop unit automatically generates twin model calibration instructions when > 10% or consecutively for 3 times > 5%, dynamically updates control parameters, and simultaneously structures feedback data (including values, environmental interference factors) are sent back to the perception module to drive the dynamic revision of the perception end benchmark value.

[0036] Example 2: Reference Figure 2 In a second embodiment of the present invention, the present invention provides a digital smart agricultural planting and breeding integration method, the method comprising: S1. Multi-source data collection: Through the sensor clusters in the planting area, breeding area and recycling treatment equipment, environmental data, biological sign data and equipment operation data are synchronously collected; S2, dynamic data verification, performs cross-domain verification on collected data based on preset agronomic mechanism models, and outputs a trusted dataset with confidence labels; S3, hierarchical collaborative decision-making, runs digital twins in a three-level architecture of edge nodes, local servers, and cloud platforms to generate instructions for optimizing planting and breeding resources; S4: Instruction rehearsal verification: sandbox simulation and backtracking retry are performed on the optimized instructions to output verified instructions with controllable risks; S5: Closed-loop execution feedback, converting the verification instruction into a device control signal, collecting execution effect data and feeding it back to step S1.

[0037] Specifically, through the multi-source data collection step, the environmental parameters of the planting area (soil moisture / crop growth), the biological signs of the breeding area (body temperature / respiration rate) and the working conditions of the circulation equipment (current / flow) are obtained synchronously. Through the cross-domain compensation and credibility packaging output of dynamic data verification, the Labeled data packets; in hierarchical collaborative decision-making, edge nodes generate device instructions through PID controllers, local servers optimize resource allocation according to material flow balance equations, cloud platforms integrate market climate data to generate carbon sequestration strategies, instruction preview verification loads historical disaster models for sandbox simulations, and when losses are greater than 15%, attenuation weight backtracking is triggered. Conflict markers or failures exceed the limit and instructions are frozen. Afterwards, closed-loop execution feedback converts instructions into industrial signals and calculates the deviation rate. ,when When the error rate is greater than 10%, the twin model parameters are updated, and the structured feedback data is transmitted back to the collection end to calibrate the closed loop and achieve precise control and continuous self-optimization of agricultural system resources.

[0038] Example 3 The third embodiment of the present invention is based on the same inventive concept. The present invention proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the digital smart agricultural breeding and farming integration method of the above embodiment.

[0039] Example 4 The fourth embodiment of the present application is based on the same inventive concept, and the present application provides a computer, which comprises a processor and a memory; the processor and the memory are in communication with each other; the memory is used for storing instructions; and the processor is used for executing the instructions in the memory to execute the digitalized wisdom agricultural planting and breeding combination method of the above-mentioned embodiments.

[0040] It should be understood that parts of the present application can be realized in hardware, software, firmware, or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be realized by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if realized in hardware, and as in another embodiment, any one or a combination of the following technologies known in the art can be used: discrete logic circuit with logic gate circuit for implementing logic functions on data signals, application specific integrated circuit with suitable combination logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA), etc.

[0041] Finally, it should be noted that: the above only describes the preferred embodiments of the present application and is not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. The digital smart agricultural planting and breeding system is characterized by: include: Multi-source sensing module, used to collect environmental data, biological sign data and equipment operation data through planting area sensors, breeding area sensors and recycling equipment sensors; Data verification module, which dynamically verifies the collected data based on the preset agronomic mechanism model; Hierarchical digital twin module, which makes collaborative decisions and generates optimization instructions in a three-level architecture of edge computing nodes, local servers, and cloud platforms; The virtual-reality decision engine module is used to perform sandbox rehearsals and backtrack retries on decision instructions, and output verified instructions; The closed-loop execution network module converts the verified instructions into device control signals, collects execution effect data and feeds it back to the multi-source perception module.

2. The digital smart agricultural planting and breeding system according to claim 1 is characterized in that: The multi-source perception module includes: The planting environment perception unit collects data on soil moisture, crop growth, and pest and disease characteristics through soil temperature and humidity sensors, light intensity sensors, leaf wetness sensors, and multispectral imagers; The livestock and poultry vital signs sensing unit collects data on biophysiological indicators and abnormal activities through livestock and poultry temperature monitoring ear tags, respiratory rate sensors, aquatic dissolved oxygen sensors, and behavior analysis cameras; The circulation equipment monitoring unit collects the operating status parameters of the resource circulation equipment through the biogas flow meter, the manure solid-liquid separator current sensor, the water treatment pH probe and the organic fertilizer fermentation temperature probe.

3. The digital smart agricultural planting and breeding system according to claim 1 is characterized in that: The data verification module includes: Environmental coupling compensation unit, used to construct environmental interference matrix ,in, For the Environmental factors on the The influence coefficient of the sensor type; Perform collaborative correction of multi-source data ; in, For the Class sensor raw value, is the change of environmental factors; Spatiotemporal consistency unit, which builds a spatiotemporal distribution model through historical data streams , detect abnormal deviation points; When multiple sensors monitor the same physical quantity, the confidence weight Calculate fusion value ; Credibility encapsulation unit, calculates overall confidence ; when or When ,multi-source conflict markers are embedded, and a unified time and space stamp is added to the output data packet.

4. The digital smart agricultural planting and breeding system according to claim 1 is characterized in that: The hierarchical digital twin module includes: The edge response unit is deployed on the edge computing node and is configured to call the device-level twin model, including the livestock and poultry environment control model and the irrigation response model, to generate device control instructions based on real-time data; The field coordination unit, deployed on the local server, runs the crop-livestock coupling twin model and optimizes resource allocation paths through the material flow balance equation, specifically calculating the amount of manure and sewage returned to the fields. The strategy generation unit, deployed on the cloud platform, generates global strategies for integrating market, climate and policy data, calculates carbon sink assets based on specific carbon conversion rates, and generates blockchain-verified carbon sink reports.

5. The digital smart agricultural planting and breeding system according to claim 4 is characterized in that: The livestock and poultry environment control model in the edge response unit is a PID controller, and its proportional coefficient , integral coefficient ; The material flow balance equation in the field collaborative unit ; in, Dynamic utilization efficiency is the monthly mean temperature; The strategy generates a specific carbon conversion rate in the unit and generate a uniquely linked carbon sequestration report through blockchain.

6. The digital smart agricultural planting and breeding system according to claim 1 is characterized in that: The virtual-real decision engine module includes: The rehearsal and deduction unit loads the optimization instructions generated by the hierarchical digital twin module into the virtual twin; Call the historical disaster model, including the disease transmission path and extreme weather impact, to simulate the consequences of command execution. The rehearsal duration is dynamically set to ; Backtracking retry unit, when the preview shows that the economic loss is greater than 15% or the resource waste is greater than 20%, the strategy backtracking is triggered and multiple rounds of retries are performed using the decaying weight. The decision weight of the backtracking , maximum number of retries ; The conflict arbitration unit receives the multi-source conflict markers embedded in the data verification module. When the conflict marker is triggered or the number of rehearsal failures is greater than 2, it freezes the instructions and requests manual intervention, and outputs the final verification instructions to the closed-loop execution network module.

7. The digital smart agricultural planting and breeding system according to claim 1 is characterized in that: The closed-loop execution network module includes: The protocol conversion unit converts the verified instructions into executable signals for industrial equipment and dynamically loads driver plug-ins based on different agricultural breeding equipment; Execution feedback unit, real-time collection of execution status data, calculation of execution deviation rate ,in is the device feedback value, is the expected value of the instruction; Data closed loop unit, when , generate calibration instructions, update twin model parameters, and structure feedback data, including time and space stamps and device ID, and send it back to the multi-source perception module.

8. A digital smart agricultural planting and breeding method, characterized in that: The digital smart agricultural planting and breeding system according to any one of claims 1 to 7, wherein the method comprises: S1. Multi-source data collection: Through the sensor clusters in the planting area, breeding area and recycling treatment equipment, environmental data, biological sign data and equipment operation data are synchronously collected; S2, dynamic data verification, performs cross-domain verification on collected data based on preset agronomic mechanism models, and outputs a trusted dataset with confidence labels; S3, hierarchical collaborative decision-making, runs digital twins in a three-level architecture of edge nodes, local servers, and cloud platforms to generate instructions for optimizing planting and breeding resources; S4: Instruction rehearsal verification: sandbox simulation and backtracking retry are performed on the optimized instructions to output verified instructions with controllable risks; S5: Closed-loop execution feedback, converting the verification instruction into a device control signal, collecting execution effect data and feeding it back to step S1.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the digital smart agricultural planting and breeding integration method as described in claim 8 is implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by the processor, the digital smart agricultural planting and breeding integration method as claimed in claim 8 is implemented.

Images