An intelligent monitoring and automatic management system for a breeding greenhouse based on an internet of things

Abstract

The application discloses a kind of based on Internet of Things's intelligent monitoring and automatic management system of breeding greenhouse, belong to intelligent agriculture and facility breeding technical field, comprising: data acquisition module, for obtaining the real-time environmental parameter and construction parameter of breeding greenhouse;Digital twin module, for constructing and dynamically updating the digital twin of breeding greenhouse according to the construction parameter and environmental parameter, the digital twin is used to simulate the space-time evolution process of the environment in breeding greenhouse;Control decision module;It can be in advance to predict environmental change trend, fundamentally solves the problem of control lag and large environmental parameter fluctuation in traditional passive response control, can maximize reduce the operating energy consumption of fan, skylight, heater and other equipment under the premise of ensuring high stability of environment, effectively avoid the impact on entity greenhouse environment caused by unreasonable instruction, guarantee the growth safety of breeding animals.

Description

Technical Field

[0001] This invention belongs to the field of intelligent agriculture and facility-based aquaculture technology, specifically relating to an intelligent monitoring and automatic management system for aquaculture greenhouses based on the Internet of Things. Background Technology

[0002] Existing IoT-based monitoring systems for livestock sheds typically use discrete sensors to collect environmental parameters and activate equipment such as fans and skylights when these parameters exceed thresholds, representing a passive response control approach. This method suffers from the following problems: First, the control response is lagging; due to a lack of accurate prediction of the dynamic evolution of environmental conditions, environmental parameters are prone to significant fluctuations and are difficult to maintain stably. Second, it cannot make forward-looking optimization decisions that balance environmental stability and energy conservation. Third, there is a lack of simulation verification before control commands are issued; unreasonable commands may be executed directly, causing stress to the livestock.

[0003] Therefore, there is an urgent need for an intelligent monitoring and automatic management system that can accurately simulate environmental dynamics, predict and optimize control in advance, and perform simulation verification before execution.

[0004] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent monitoring and automatic management system for livestock sheds based on the Internet of Things, which aims to solve the technical problems of existing livestock shed monitoring systems, such as delayed regulation due to passive response control, lack of forward-looking optimization decision-making capabilities, and lack of simulation verification before the execution of control commands.

[0006] To achieve the above objectives, a specific embodiment of the present invention provides the following technical solution: An IoT-based intelligent monitoring and automatic management system for livestock sheds includes: The data acquisition module is used to obtain real-time environmental and structural parameters of the breeding shed; A digital twin module is used to construct and dynamically update a digital twin of the breeding shed based on the construction parameters and environmental parameters. The digital twin is used to simulate the spatiotemporal evolution of the environment inside the breeding shed. The control decision module is used to simulate the environmental state in the future time period based on the digital twin, and generate a control decision scheme with a preset optimization target. The simulation verification module is used to perform simulation verification of the control decision scheme using the digital twin before sending the control decision scheme to the execution device. After the verification is successful, the control command is output.

[0007] In one or more embodiments of the present invention, the digital twin is a multi-physics coupled digital twin, specifically integrating a heat flow field model, a moisture diffusion field model, and an airflow field model to simulate the three-dimensional spatial distribution of temperature, humidity, and airflow velocity inside the breeding shed in real time.

[0008] In one or more embodiments of the present invention, the digital twin module includes a virtual-real synchronization unit, which is used to synchronously input the environmental parameters collected in real time into the edge computing node and the cloud platform, drive both to run the same state update algorithm, and realize the state synchronization between the digital twin and the physical greenhouse by calculating the state change increment between adjacent sampling points.

[0009] In one or more embodiments of the present invention, the synchronization mechanism of the virtual-real synchronization unit is specifically as follows: the edge computing node and the cloud platform run the environmental state update algorithm under the same initialization state. After each sampling, the edge computing node calculates the environmental state change increment between adjacent sampling points and only uploads the change increment to the cloud platform. The cloud platform updates the state of the digital twin based on the increment to reduce the amount of data communication.

[0010] In one or more embodiments of the present invention, the control decision module employs a model predictive control algorithm, taking the current synchronization state of the digital twin as the initial condition, and continuously solves for the optimal control sequence composed of the control quantities of the continuously adjustable execution device, with the optimization objectives being to minimize regulation energy consumption and maximize environmental stability.

[0011] In one or more embodiments of the present invention, the construction process of the multiphysics coupled digital twin includes: using the geometric structure data of the breeding shed, the thermal parameters of the enclosure material and the characteristic curves of the equipment, establishing a heat-humidity-air multiphysics coupled model through the finite element method, and performing online adaptive correction of the model parameters by combining real-time collected environmental data.

[0012] In one or more embodiments of the present invention, if the simulation result exceeds the preset ideal environment range during the simulation verification process, the control decision module triggers the control decision module to perform a constrained secondary optimization adjustment of the regulation decision scheme, and re-invests the adjusted scheme into the digital twin for simulation verification until the simulation result meets the requirements.

[0013] In one or more embodiments of the present invention, it further includes: a multi-scenario simulation module, used to receive custom meteorological scenario parameters, drive the digital twin to perform accelerated simulation simulation, compare and output environmental response results under different control strategies, and generate emergency control plans.

[0014] A method for intelligent monitoring and automatic management of livestock sheds based on the Internet of Things includes the following steps: Obtain real-time environmental and structural parameters of the breeding shed; Based on the construction parameters and environmental parameters, a digital twin of the breeding shed is constructed and dynamically updated. The digital twin is used to simulate the spatiotemporal evolution of the environment inside the breeding shed. Based on the digital twin, the environmental state in the future period is simulated, and a control decision scheme is generated with preset optimization goals; Before the control decision scheme is sent to the execution device, the scheme is simulated and verified using the digital twin. Once the verification is successful, the control command is output.

[0015] In one or more embodiments of the present invention, the step of constructing and dynamically updating the digital twin includes: deploying the same environmental state update algorithm on the edge computing node and the cloud platform, starting with the same initial state; the edge computing node calculates the environmental state change increment in each acquisition cycle, and only sends the increment data to the cloud platform; the cloud platform uses the increment to update the overall state of the digital twin, realizing virtual-real synchronization under low bandwidth conditions.

[0016] Compared with existing technologies, the IoT-based intelligent monitoring and automatic management system for livestock sheds of this invention accurately simulates and pre-simulates the environment inside the shed by constructing a digital twin, enabling it to predict environmental change trends in advance. This fundamentally solves the problems of lagging regulation and large fluctuations in environmental parameters in traditional passive response control. The system introduces a model predictive control algorithm, using the digital twin as the predictive model to continuously optimize and generate smooth control decision schemes. This minimizes the energy consumption of equipment such as fans, skylights, and heaters while ensuring a highly stable environment. A simulation verification step is included, where all control schemes are simulated and verified by the digital twin before execution, effectively preventing unreasonable commands from directly impacting the physical shed environment and ensuring the growth safety of livestock. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the overall architecture of the system of the present invention.

[0019] Figure 2 This is a schematic diagram illustrating the construction process of the multiphysics coupling model of the digital twin in this invention.

[0020] Figure 3 This is a flowchart illustrating the virtual-real synchronization mechanism in this invention.

[0021] Figure 4 This is a schematic diagram of the closed-loop workflow of model predictive control and simulation verification in this invention.

[0022] Figure 5 This is a schematic diagram of the main process steps of the method of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the technical solutions in this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this disclosure.

[0024] like Figures 1 to 5 As shown in Example 1: This example provides an intelligent monitoring and automatic management system for a breeding shed based on the Internet of Things, which is deployed in a standardized egg-laying hen breeding shed; the system includes a data acquisition module, a digital twin module, a control decision module, and a simulation verification module.

[0025] The data acquisition module consists of a sensor network and data acquisition terminals deployed within the breeding shed. The sensor network includes a distributed array of temperature sensors, a humidity sensor array, a light sensor, a CO2 concentration sensor, and wind speed and direction sensors arranged in a three-dimensional spatial grid. These sensors are installed at different heights and planar locations within the breeding shed to acquire three-dimensional spatial distribution data of environmental parameters within the shed. Simultaneously, operating parameter acquisition units are set up at each actuator to acquire real-time equipment status data such as fan speed percentage, skylight opening angle, and heater power percentage. The data acquisition module also stores the structural parameters of the breeding shed, including geometric dimensions, thermal parameters of the enclosure structure materials, and characteristic curves of each actuator.

[0026] like Figure 2As shown, the digital twin module receives real-time environmental and structural parameters from the data acquisition module to construct a multi-physics coupled digital twin of the breeding shed. The construction process of this digital twin is as follows: a three-dimensional geometric model is established based on the geometric structure data of the breeding shed; based on the three-dimensional geometric model, a heat flow field model, a moisture diffusion field model, and an airflow field model are established respectively; the three physical field models are coupled and solved using the finite element method to form a multi-physics coupled model capable of simulating the three-dimensional spatiotemporal evolution of the temperature field, humidity field, and airflow velocity field inside the shed; during operation, the digital twin module continuously receives real-time environmental data and performs online adaptive correction of key parameters in the multi-physics coupled model to ensure that the simulation accuracy of the digital twin remains consistent with that of the physical shed.

[0027] like Figure 3 As shown, the digital twin module also includes a virtual-real synchronization unit to achieve high-precision synchronization between the digital twin and the physical greenhouse. The specific working mechanism of the virtual-real synchronization unit is as follows: identical environmental state update algorithms are deployed on the edge computing nodes at the greenhouse site and on the remote cloud digital twin platform. This algorithm is based on the thermal balance equation, the humidity balance equation, and the momentum equation, and can recursively deduce the greenhouse state distribution at the next moment based on the input environmental parameters. When the system starts, the edge computing nodes and the cloud platform start the algorithm with the same initial state. In each sampling period, the edge computing nodes calculate the temperature field change increment, humidity field change increment, and airflow field change increment between adjacent sampling points based on the latest collected environmental data, and package the change increment into a lightweight data packet and upload it to the cloud platform. After receiving the incremental data, the cloud platform superimposes it onto the current state field of the digital twin to achieve step-by-step update of the digital twin state. By transmitting only the change increment instead of the entire field data, the data communication volume of the system is greatly reduced, and virtual-real synchronization with second-level accuracy can be achieved under conventional 4G / 5G network conditions.

[0028] The control decision module generates control decision schemes based on the synchronized digital twin. In this embodiment, the control decision module employs a model predictive control algorithm, and the specific workflow is as follows: Figure 4As shown. In each control cycle, the control decision module uses the current synchronization state of the digital twin as the initial condition and a multiphysics coupling model as the prediction model to predict the environmental response of the breeding shed under the combined effects of changes in external meteorological conditions and internal heat and humidity loads over the next N control cycles. Based on this prediction, the module performs rolling optimization using the following multi-objective optimization function: J = α·E + β·S, where E is the regulation energy consumption index, representing the total energy consumption estimate of equipment such as fans, skylights, and heaters within the prediction domain; S is the environmental stability index, representing the integral value of the deviation of the temperature and humidity inside the shed from the ideal physical setting zone for breeding within the prediction domain; α and β are weighting coefficients, which can be adaptively adjusted according to the different growth stages of the breeding organisms to match the differentiated priority requirements for environmental stability and energy saving at different stages.

[0029] By solving the above optimization function, the control decision module rolls out the optimal control sequence consisting of continuously adjustable control quantities. The control quantities are stepless continuous percentage or angle signals, specifically including the variable frequency fan speed percentage signal, the proportional regulation heater power percentage signal, and the electric sunroof rotation angle signal. Unlike traditional on / off control, stepless continuous control quantities can achieve smooth regulation of the breeding shed environment, avoiding energy waste and environmental impact caused by frequent equipment start-ups and shutdowns. After the optimization solution is completed, the control decision module, according to the rolling optimization principle of MPC, only outputs the control command of the first control cycle in the optimal control sequence to the simulation verification module.

[0030] After receiving the control decision scheme (i.e., the first control command of the optimal control sequence) output by the control decision module, the simulation verification module does not directly send it to the physical execution equipment. Instead, it first inputs the control command into the digital twin and performs a "hypothetical execution" simulation on the digital twin. Starting from the current synchronization state, the digital twin simulates the response of the greenhouse environment within one control cycle after the execution of the control command, and outputs the spatiotemporal distribution simulation results of parameters such as temperature, humidity, and airflow velocity within the greenhouse. The simulation verification module compares the simulation results with the preset ideal environmental zone (the optimal temperature and humidity range set according to the species and growth stage of the farmed animals): if the simulation results fall within the ideal environmental zone, the verification is successful, and the control command is sent to the physical execution equipment such as the variable frequency fan, electric skylight, and heater; if the simulation results exceed the ideal environmental zone, the simulation verification module triggers the control decision module to perform a constrained secondary optimization adjustment of the control scheme—resolving the optimal control quantity within the temperature and humidity constraints necessary to ensure the safety of the farmed animals, and then putting the adjusted scheme back into the digital twin simulation verification. The closed-loop iterative process of "optimization-simulation-verification-adjustment" continues until the simulation results meet the environmental stability requirements, at which point the confirmed control commands are output to the physical devices for execution. This mechanism effectively builds a safety gate for the execution of control commands, fundamentally avoiding stress on aquaculture caused by unreasonable control commands due to model deviations, instantaneous sensor disturbances, or control parameter mismatches.

[0031] Furthermore, the system in this embodiment is also equipped with a multi-scenario simulation module. This module provides a human-computer interaction interface for users, allowing them to customize input parameters for extreme weather scenarios, such as a cold wave scenario with an external temperature of -15°C for 6 hours, or a heat wave scenario with an external temperature of 40°C for 8 hours. After receiving the scenario parameters, the multi-scenario simulation module drives the digital twin to simulate the extreme scenarios in an accelerated manner, i.e., the simulation time step is larger than the actual time step, for example, using a 1-minute simulation to represent a 1-hour real-time acceleration ratio. The module compares and outputs the time response curves and spatial distribution maps of the greenhouse environmental parameters under different control strategies (such as full-power heating strategy and intermittent heating and insulation strategy), and automatically ranks and scores the strategies based on preset environmental stability criteria (such as the percentage of time that the temperature and humidity in the main activity area of ​​the greenhouse are within a set zone), providing recommended emergency control plans and providing quantitative decision-making basis for managers to respond to extreme weather events.

[0032] like Figure 5 As shown in Example 2: This example provides a method for intelligent monitoring and automatic management of livestock sheds based on the Internet of Things. This method can be applied to the system of Example 1, or it can be implemented independently on a computing platform with equivalent functions. The method includes the following steps: S1. Data Acquisition Steps: Through the sensor network and data acquisition terminals deployed in the breeding shed, multi-dimensional environmental data, equipment status data, and structural parameters of the breeding shed are acquired in real time.

[0033] S2. Digital Twin Construction and Synchronization Steps: Using the acquired construction parameters and environmental data, a multi-physics coupled digital twin integrating a thermal flow field model, a moisture diffusion field model, and an airflow field model is constructed. The real-time collected environmental data is simultaneously input into the environmental state update algorithm running on the edge computing node and the cloud platform. The edge computing node calculates the state change increment between adjacent sampling points in each collection cycle and only sends the increment data to the cloud platform. The cloud platform uses the increment data to update the overall state of the digital twin, achieving low-bandwidth, high-precision synchronization between the digital twin and the physical greenhouse.

[0034] S3. Model Predictive Control Decision Steps: Using the synchronized digital twin as the prediction model and the current state of the digital twin as the initial condition, the model predictive control algorithm is adopted. Within the given prediction domain, the optimization objectives are to minimize regulation energy consumption and maximize environmental stability. The optimal control sequence containing continuously adjustable control quantities is generated by rolling solution.

[0035] S4. Simulation Verification and Execution Steps: Simulate and execute the first control command of the generated optimal control sequence on the digital twin to verify its control effect on the indoor environment. If the simulation result is within the preset ideal environment zone, the verification is successful, and the control command is sent to physical execution devices such as variable frequency fans, electric skylights, and heaters. If the simulation result exceeds the ideal environment zone, return to step S3 and perform secondary optimization and adjustment under environmental safety constraints until the simulation verification is successful and then the execution is carried out.

[0036] Example 3: This example uses a pig breeding shed as an application scenario to further illustrate the technical effects of the present invention.

[0037] The pig farming shed has a building area of ​​approximately 2,000 square meters and is equipped with 12 variable frequency fans, 8 sets of electric skylights, and 4 sets of proportionally adjustable gas heaters. Before the implementation of the plan, traditional threshold switching control was used. When the temperature inside the shed was higher than 28°C, the fans started at full speed and when it was lower than 22°C, the heaters started at full power. The system triggered the equipment switching action once every 15 minutes on average, with a daily energy consumption of approximately 480 kWh and a daily fluctuation range of ±4°C in ambient temperature.

[0038] After deploying the system of this invention, the digital twin synchronizes its state at a 1-second cycle, and the model predictive control performs rolling optimization with a 60-second control cycle and a 10-minute prediction domain. After 30 days of continuous operation and comparison, the results show that the system of this invention achieves stepless smooth control of fans, skylights, and heaters, reducing the frequency of equipment start-up and shutdown to about 2 times per hour; the daily fluctuation range of ambient temperature is controlled within ±1℃, significantly reducing the stress response of pigs caused by environmental fluctuations; the average daily energy consumption is reduced to about 310kWh, with an energy saving rate of 35.4%. At the same time, the system's simulation verification module successfully intercepted 3 erroneous control commands caused by instantaneous abnormal data from sensors during operation, effectively avoiding potential environmental accidents.

[0039] Example 4: This example uses a large-scale broiler chicken farm as an example to demonstrate the application effect of the multi-scenario simulation module. The farm's sheds are located in northern China, where cold waves are frequent in winter. Operators input an extreme cold wave scenario of "external temperature -20℃" into the multi-scenario simulation module. The system automatically accelerated the simulation and extrapolated the temperature changes inside the sheds over 6 hours. The simulation compared two strategies: "full-power continuous heating" and "intermittent heating combined with minimum ventilation." The results showed that while the full-power heating strategy could maintain the temperature, the total gas consumption reached 85 cubic meters over 6 hours. The intermittent heating strategy combined with minimum ventilation resulted in a total gas consumption of only 52 cubic meters over 6 hours, and the temperature in the core breeding area remained consistently within the target range of 20℃-24℃. Based on this, the farm developed a winter cold wave emergency plan, reducing heating energy costs during cold waves by approximately 38% in actual operation.

[0040] It will be apparent to those skilled in the art that this disclosure is not limited to the details of the exemplary embodiments described above, and that this disclosure can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of this disclosure is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this disclosure. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0041] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A smart monitoring and automatic management system for livestock sheds based on the Internet of Things, characterized in that, include: The data acquisition module is used to obtain real-time environmental and structural parameters of the breeding shed; A digital twin module is used to construct and dynamically update a digital twin of the breeding shed based on the construction parameters and environmental parameters. The digital twin is used to simulate the spatiotemporal evolution of the environment inside the breeding shed. The control decision module is used to simulate the environmental state in the future time period based on the digital twin, and generate a control decision scheme with a preset optimization target. The simulation verification module is used to perform simulation verification of the control decision scheme using the digital twin before sending the control decision scheme to the execution device. After the verification is successful, the control command is output.

2. The system according to claim 1, characterized in that, The digital twin is a multi-physics coupled digital twin, which specifically integrates a heat flow field model, a moisture diffusion field model, and an airflow field model to simulate the three-dimensional spatial distribution of temperature, humidity, and airflow velocity inside the breeding shed in real time.

3. The system according to claim 1, characterized in that, The digital twin module includes a virtual-real synchronization unit, which is used to synchronously input the environmental parameters collected in real time into the edge computing node and the cloud platform, drive both to run the same state update algorithm, and realize the state synchronization between the digital twin and the physical greenhouse by calculating the state change increment between adjacent sampling points.

4. The system according to claim 3, characterized in that, The synchronization mechanism of the virtual-real synchronization unit is as follows: the edge computing node and the cloud platform run the environmental state update algorithm under the same initialization state. After each sampling, the edge computing node calculates the environmental state change increment between adjacent sampling points and only uploads the change increment to the cloud platform. The cloud platform updates the state of the digital twin based on the increment to reduce the amount of data communication.

5. The system according to claim 1, characterized in that, The control decision module employs a model predictive control algorithm, taking the current synchronization state of the digital twin as the initial condition, and continuously solves for the optimal control sequence composed of the control quantities of the continuously adjustable execution device, with the optimization objectives being to minimize regulation energy consumption and maximize environmental stability.

6. The system according to claim 2, characterized in that, The construction process of the multiphysics coupled digital twin includes: using the geometric structure data of the breeding shed, the thermal parameters of the enclosure material and the characteristic curves of the equipment, establishing a heat-humidity-air multiphysics coupled model through the finite element method, and combining the real-time collected environmental data to perform online adaptive correction of the model parameters.

7. The system according to claim 1, characterized in that, If the simulation result exceeds the preset ideal environment range during the simulation verification process, the control decision module will trigger the control decision module to perform a constrained secondary optimization adjustment of the regulation decision scheme, and then re-invest the adjusted scheme into the digital twin for simulation verification until the simulation result meets the requirements.

8. The system according to claim 1, characterized in that, Also includes: The multi-scenario simulation module is used to receive custom meteorological scenario parameters, drive the digital twin to perform accelerated simulation simulation, compare and output the environmental response results under different control strategies, and generate emergency control plans.

9. A method for intelligent monitoring and automatic management of livestock sheds based on the Internet of Things, characterized in that, Includes the following steps: Obtain real-time environmental and structural parameters of the breeding shed; Based on the construction parameters and environmental parameters, a digital twin of the breeding shed is constructed and dynamically updated. The digital twin is used to simulate the spatiotemporal evolution of the environment inside the breeding shed. Based on the digital twin, the environmental conditions in the future period are simulated, and a control decision scheme is generated with preset optimization goals; Before the control decision scheme is sent to the execution device, the scheme is simulated and verified using the digital twin. Once the verification is successful, the control command is output.

10. The method according to claim 9, characterized in that, The steps for constructing and dynamically updating the digital twin include: deploying the same environment state update algorithm on the edge computing node and the cloud platform, starting with the same initial state; the edge computing node calculates the incremental change in environment state in each collection cycle, and only sends the incremental data to the cloud platform; the cloud platform uses the incremental change to update the overall state of the digital twin, achieving virtual-real synchronization under low bandwidth conditions.

Images