Agricultural production virtual construction method based on digital twinning

Abstract

The invention belongs to the technical field of modern agriculture, and particularly relates to an agricultural production virtual construction method based on digital twinning, which comprises the following steps: step 1, a data acquisition unit which acquires various data d = {d1, d2... dn} in an agricultural production environment by using a sensor network, and n is a positive integer; 2, a data processing unit, wherein the data preprocessing unit is used for preprocessing the data acquired by the data acquisition unit and preprocessing an original data set; 3, a digital twinning virtual model unit: constructing a digital twinning virtual model of agricultural production based on the preprocessed data; the real-time monitoring and feedback unit is used for monitoring the agricultural production process in real time based on a sensor network, inputting collected new data Dreal (t) into a virtual model, calculating a virtual predicted value Mpred (t), and performing feedback adjustment by comparing an actual value with the predicted value; and 5, an optimization unit: performing optimization adjustment on the agricultural production process according to the model adjustment amount.

Description

Technical Field

[0001] The present invention relates to the field of modern agricultural technology, and in particular to a method for virtual construction of agricultural production based on digital twins. Background Art

[0002] Existing agricultural production simulation methods often lack sufficient accuracy and flexibility, and are unable to meet the diversified needs of modern agricultural production. With the rapid development of information technology, the application of digital twin technology in agricultural production has gradually deepened. Digital twins make full use of physical models, sensors, operation history and other data, integrate multi-disciplinary, multi-physical quantity, multi-scale, and multi-probability simulation processes, and complete mapping in virtual space, thereby reflecting the entire life cycle of the corresponding physical equipment.

[0003] However, current digital twin applications mainly focus on data collection and basic model construction, lacking comprehensive virtual construction and optimization methods for complex agricultural production systems. At the same time, existing technologies cannot accurately simulate changes in farmland status, and do not have the ability to continuously iterate and accurately optimize simulation algorithms, which not only causes a waste of resources, but also reduces work efficiency. Summary of the invention

[0004] To solve the above problems, the present invention provides the following technical solutions:

[0005] The following steps are involved:

[0006] Step 1: Data collection unit, which uses a sensor network to collect various types of data in the agricultural production environment d = {d1, d2..., d n}, where n is a positive integer;

[0007] Step 2: a data processing unit, wherein the data preprocessing unit preprocesses the original data set on the data collected by the data collection unit;

[0008] Step 3, digital twin virtual model unit: construct a digital twin virtual model of agricultural production based on the preprocessed data;

[0009] Step 4, real-time monitoring and feedback unit: Real-time monitoring of agricultural production process based on sensor network, and the new data collected D real (t) Input into the virtual model and calculate the virtual predicted value M pred (t), make feedback adjustments by comparing actual values ​​with predicted values;

[0010] Step 5, optimization unit: optimize and adjust the agricultural production process according to the model adjustment amount;

[0011] Step 6, repeat steps 1 to 5 until the maximum number of cycles is reached, and output the final optimization model;

[0012] Step 7, the early warning unit includes an early warning light, and the optimized production parameters are compared with the set standard threshold value to make a judgment and early warning to continue the cycle or output the optimized production parameters.

[0013] Furthermore, the data acquisition unit includes a storage, a humidity sensor, a temperature sensor and a light intensity sensor, and the collected data includes soil moisture, temperature and light intensity, etc. The various types of data collected by using sensors and drones to collect environmental data in real time are defined as d={d1, d2..., d n}, where n is a positive integer.

[0014] Furthermore, the data preprocessing unit includes a processor, which is used to perform preprocessing analysis on the data acquisition unit, and the preprocessed data is recorded as D = {d i1 , d i2 ......,d in}, where i represents the time series of the data. The preprocessed data can be represented as a multidimensional time series matrix D:

[0015]

[0016] Furthermore, based on the preprocessed data, a digital twin virtual model of agricultural production is constructed, and the virtual model is described by the following complex formula:

[0017]

[0018] Where M(t) represents the virtual model that changes over time, α ij is the model parameter, φ ij is a nonlinear function, D ij (t) is the data in the i-th row and j-th column, and t is the time.

[0019] Furthermore, the φ ij The specific expression for nonlinear function is as follows:

[0020]

[0021] Where σ is the activation function, ω k is the weight, b is the bias, t is the time, D ij (t) is the data in the i-th row and j-th column.

[0022] Furthermore, the virtual model is specifically as follows:

[0023]

[0024] Where σ is the activation function, ω k is the weight, b is the bias, M(t) represents the virtual model that changes over time, α ij is the model parameter, φ ij is a nonlinear function, D ij (t) is the data in the i-th row and j-th column.

[0025] Furthermore, a real-time monitoring and feedback unit is provided, wherein the real-time monitoring and feedback unit monitors the agricultural production process in real time based on a sensor network, and the collected new data D real (t) Input into the virtual model and calculate the virtual predicted value M pred (t);

[0026] Furthermore, through real-time monitoring and feedback units, actual values ​​and predicted values ​​are compared and feedback adjustments are made.

[0027] ΔM(t)=M real (t)-M pred (t)

[0028] Among them, ΔM(t) represents the model adjustment amount that changes with time, M real (t) represents the actual measured value, M pred (t) represents the model prediction value, and t is time.

[0029] Furthermore, the agricultural production process is optimized and adjusted according to the model adjustment amount ΔM(t), and the optimization process is based on a complex iterative formula:

[0030]

[0031] Among them, P opt (t) represents the optimized production parameters, t is time, P current (t) represents the current production parameters, β k represents the adjustment coefficient, λ is the attenuation factor, k is the time step, ΔM(tk) is the model adjustment amount, and the model adjustment amount at time tk represents the difference between the actual measured value and the model predicted value. At the same time, P is set 阈值 (t) is the standard parameter value;

[0032] When t = H time, stop the cycle optimization and output P opt (t) results.

[0033] Furthermore, the warning unit includes a warning light, t = H time, stop cycle optimization, output P opt (t) The result P opt (t) and the threshold value P of the setting standard 阈值 (t) for comparison, that is, P阈值 (t) and P opt (t) contrast, i.e. P 阈值 (t) <P opt (t) Stop the loop and output P opt (t), otherwise repeat steps 1 to 5 to continue optimizing P opt (t).

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. In the present invention, the deployed sensor nodes and drones are first used to collect environmental data in real time, such as temperature, humidity and light intensity, so as to realize comprehensive monitoring of agricultural production. The accuracy of data collection and preprocessing ensures the reliability and real-time nature of the information, providing a solid data foundation for subsequent analysis and decision-making;

[0036] 2. In the present invention, by constructing a digital twin model, the agricultural environment and production process in the real world are converted into a virtual model, which can accurately simulate the changes in the state of farmland. This process includes dynamic updating and simulation of the environmental state, ensuring high accuracy and high fidelity of the simulation results;

[0037] 3. In the present invention, algorithm optimization tools are applied to intelligently optimize agricultural production strategies. In each simulation cycle, model parameters are continuously updated, simulation is performed, optimization effects are evaluated, and strategies are adjusted through cyclic steps to achieve gradual improvement and optimization of production strategies. This method not only improves resource utilization efficiency, but also promotes intelligent management of the production process.

[0038] 4. In the present invention, through accurate model simulation and optimization decision-making, the operational efficiency of agricultural production is significantly improved, resource waste and environmental burden are reduced, and at the same time, through intelligent adjustment and optimization measures, the sustainable development of agricultural production is promoted, ensuring the minimum impact of the production process on the environment;

[0039] 5. The present invention provides an efficient, precise and intelligent agricultural production solution through digital twin technology, which greatly enhances the management capabilities of agricultural production and brings significant benefits and sustainable development opportunities to the agricultural industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0041] Figure 1 This is a schematic diagram of the steps of a method for virtual construction of agricultural production based on digital twins of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] like Figure 1 As shown, a virtual construction method for agricultural production based on digital twins includes the following steps:

[0044] The following steps are involved:

[0045] Step 1: Data collection unit, the data collection unit uses the sensor network to collect various types of data in the agricultural production environment d = {d1, d2..., d n}, where n is a positive integer;

[0046] Step 2: The data processing unit and the data preprocessing unit preprocess the original data set on the data collected by the data collection unit;

[0047] Step 3, digital twin virtual model unit: construct a digital twin virtual model of agricultural production based on the preprocessed data;

[0048] Step 4, real-time monitoring and feedback unit: Real-time monitoring of agricultural production process based on sensor network, and the new data collected D real (t) Input into the virtual model and calculate the virtual predicted value M pred (t), make feedback adjustments by comparing actual values ​​with predicted values;

[0049] Step 5, optimization unit: optimize and adjust the agricultural production process according to the model adjustment amount;

[0050] Step 6, repeat steps 1 to 5 until the maximum number of cycles is reached, and output the final optimization model;

[0051] Step 7, the early warning unit includes an early warning light, and the optimized production parameters are compared with the set standard threshold value to make a judgment and early warning to continue the cycle or output the optimized production parameters.

[0052] The data acquisition unit includes storage, humidity sensor, temperature sensor and light intensity sensor. The collected data include soil moisture, temperature and light intensity. The various data collected by sensors and drones are defined as d = {d1, d2..., d n}, where n is a positive integer.

[0053] Using humidity sensors, temperature sensors, and light intensity sensors in conjunction with drones, environmental data is collected in real time and the collected data is stored through a data collection unit including a storage device;

[0054] The data preprocessing unit includes a processor, which is used to perform preprocessing analysis on the data acquisition unit. The preprocessed data is recorded as D = {d i1 , d i2 ......,d in}, where i represents the time series of the data. The preprocessed data can be represented as a multidimensional time series matrix D:

[0055]

[0056] The processor of the data preprocessing unit is used to preprocess and analyze the data collected by the data collection unit, and store the analyzed data in a storage device;

[0057] Based on the preprocessed data, a digital twin virtual model of agricultural production is constructed. The virtual model is described by the following complex formula:

[0058]

[0059] Where M(t) represents the virtual model that changes over time, α ij is the model parameter, φ ij is a nonlinear function, D ij (t) is the data in the i-th row and j-th column, and t is the time.

[0060] φ ij The specific expression for nonlinear function is as follows:

[0061]

[0062] Where σ is the activation function, ω k is the weight, b is the bias, t is the time, D ij (t) is the data in the i-th row and j-th column.

[0063] The virtual model is as follows:

[0064]

[0065] Where σ is the activation function, ω k is the weight, b is the bias, M(t) represents the virtual model that changes over time, α ij is the model parameter, φ ij is a nonlinear function, D ij (t) is the data in the i-th row and j-th column.

[0066] Real-time monitoring and feedback unit, which monitors the agricultural production process in real time based on the sensor network, collects new data D real (t) Input into the virtual model and calculate the virtual predicted value M pred (t);

[0067] Through real-time monitoring and feedback units, actual values ​​are compared with predicted values, and feedback adjustments are made.

[0068] ΔM(t)=M real (t)-M pred (t)

[0069] Among them, ΔM(t) represents the model adjustment amount that changes with time, M real (t) represents the actual measured value, M pred (t) represents the model prediction value, and t is time.

[0070] The agricultural production process is optimized according to the model adjustment value ΔM(t). The optimization process is based on a complex iterative formula:

[0071]

[0072] Among them, P opt (t) represents the optimized production parameters, t is time, P current (t) represents the current production parameters, β k represents the adjustment coefficient, k is the time step, and the time distance represents the step from the current time t to the past. As k increases from 0 to n, it represents all adjustments from the current moment to the past n time units. ΔM(tk) is the model adjustment, which is the model adjustment at time tk, indicating the difference between the actual measurement value and the model prediction value. At the same time, P is set 阈值 (t) is the standard parameter value;

[0073] e -λk is the decay factor, a factor that decays exponentially with increasing time distance k, controlling the impact of past adjustments on the current optimization process, λ is the decay coefficient, which determines the decay rate, and the decay factor ensures that recent adjustments have a greater impact, while earlier adjustments have a smaller impact;

[0074] β k Represents the adjustment coefficient, which is used to balance the contribution of the model adjustment amount at each time point in the total adjustment item. The adjustment coefficient can be set according to the specific application to ensure the stability and accuracy of the optimization process;

[0075] P current(t) represents the current production parameters, which are the production parameters actually used at time t. These parameters reflect the settings and status of the current agricultural production process;

[0076] P opt (t) represents the optimized production parameters, which are the production parameters adjusted by optimization at time t. These parameters will be used to guide actual agricultural production operations to improve production efficiency and resource utilization;

[0077] Parameter optimization is performed by accumulating the model adjustment amount ΔM within a certain period of time (from t to tn), and each adjustment amount is attenuated according to the time distance. The most recent adjustment amount has a larger weight, and the earlier adjustment amount has a smaller weight.

[0078] When t = H time, stop the cycle optimization and output P opt (t) the result;

[0079] The warning unit includes a warning light, t = H time, stop cycle optimization, output P opt (t) The result P opt (t) and the threshold value P of the setting standard 阈值 (t) for comparison, that is, P 阈值 (t) and P opt (t) contrast, i.e. P 阈值 (t) <P opt (t) Stop the loop and output P opt (t), otherwise repeat steps 1 to 5 to continue optimizing P opt (t);

[0080] When P 阈值 (t) <P opt (t) indicates that the requirements for model optimization have been met, and the optimization is stopped at this time, and the output P opt (t)Numerical value.

[0081] The working principle of the present invention is as follows: a data acquisition unit is used, and the data acquisition unit uses a sensor network to collect various types of data d={d1, d2..., d n}, where n is a positive integer, and then the data preprocessing unit preprocesses the data collected by the data acquisition unit. The preprocessed data is recorded as D = {d i1 , d i2 ......,d in}, where i represents the time series of the data. After preprocessing the data, a digital twin virtual model of agricultural production is constructed. The virtual model is described by the following complex formula:

[0082]

[0083] Where σ is the activation function, ω k is the weight, b is the bias, M(t) represents the virtual model that changes over time, α ij is the model parameter, φ ij is a nonlinear function, D ij (t) is the data in row i and column j;

[0084] At the same time, the real-time monitoring and feedback unit monitors the agricultural production process in real time based on the sensor network, and the new data collected is real (t) Input into the virtual model and calculate the virtual predicted value M pred (t); through real-time monitoring and feedback units, compare the actual value and the predicted value, and make feedback adjustments ΔM(t)=M real (t)-M pred (t)

[0085] The agricultural production process is optimized according to the model adjustment value ΔM(t). The optimization process is based on a complex iterative formula:

[0086]

[0087] Among them, P opt (t) represents the optimized production parameters, t is time, P current (t) represents the current production parameters, β k represents the adjustment coefficient, λ is the attenuation factor, k is the time step, ΔM(tk) is the model adjustment amount, and the model adjustment amount at time tk represents the difference between the actual measured value and the model predicted value. At the same time, P is set 阈值 (t) is the standard parameter value;

[0088] When t = H time, stop the cycle optimization and output P opt (t) result, t = H time, stop loop optimization, output P opt (t) The result P opt (t) and the threshold value P of the setting standard 阈值 (t) for comparison, that is, P 阈值 (t) and P opt (t) contrast, i.e. P 阈值 (t) <P opt (t) Stop the loop and output P opt (t), otherwise repeat steps 1 to 5 to continue optimizing P opt (t).

[0089] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for virtual construction of agricultural production based on digital twins, characterized in that: The following steps are involved: Step 1: Data collection unit, which uses a sensor network to collect various types of data in the agricultural production environment d = {d1, d2..., d n }, where n is a positive integer; Step 2: a data processing unit, wherein the data preprocessing unit preprocesses the original data set on the data collected by the data collection unit; Step 3, digital twin virtual model unit: construct a digital twin virtual model of agricultural production based on the preprocessed data; Step 4, real-time monitoring and feedback unit: Real-time monitoring of agricultural production process based on sensor network, and the new data collected D real (t) Input into the virtual model and calculate the virtual predicted value M pred (t), by comparing the actual value with the predicted value, feedback adjustment is performed; Step 5, optimization unit: optimize and adjust the agricultural production process according to the model adjustment amount; Step 6, repeat steps 1 to 5 until the maximum number of cycles is reached, and output the final optimization model; Step 7, the early warning unit includes an early warning light, and the optimized production parameters are compared with the set standard threshold value to make a judgment and early warning to continue the cycle or output the optimized production parameters.

2. The method for virtual construction of agricultural production based on digital twin according to claim 1, characterized in that: The data acquisition unit includes a storage, a humidity sensor, a temperature sensor and a light intensity sensor. The collected data include soil moisture, temperature and light intensity, etc. The various types of data collected by using sensors and drones to collect environmental data in real time are defined as d={d1, d2..., d n }, where n is a positive integer.

3. The method for virtual construction of agricultural production based on digital twin according to claim 2, characterized in that: The data preprocessing unit includes a processor, which is used to perform preprocessing analysis on the data acquisition unit. The preprocessed data is recorded as D={d i1 , d i2 ......,d in }, where i represents the time series of the data. The preprocessed data can be represented as a multidimensional time series matrix D:

4. The method for virtual construction of agricultural production based on digital twin according to claim 3 is characterized in that: Based on the preprocessed data, a digital twin virtual model of agricultural production is constructed, and the virtual model is described by the following complex formula: Where M(t) represents the virtual model that changes over time, α ij is the model parameter, φ ij is a nonlinear function, D ij (t) is the data in the i-th row and j-th column, and t is the time.

5. The method for virtual construction of agricultural production based on digital twin according to claim 4, characterized in that: The φ ij The specific expression for nonlinear function is as follows: Where σ is the activation function, ω k is the weight, b is the bias, t is the time, D ij (t) is the data in the i-th row and j-th column.

6. The method for virtual construction of agricultural production based on digital twin according to claim 5, characterized in that: The virtual model is specifically as follows: Where σ is the activation function, ω k is the weight, b is the bias, M(t) represents the virtual model that changes over time, α ij is the model parameter, φ ij is a nonlinear function, D ij (t) is the data in the i-th row and j-th column.

7. The method for virtual construction of agricultural production based on digital twin according to claim 6, characterized in that: A real-time monitoring and feedback unit is used to monitor the agricultural production process in real time based on a sensor network. The new data collected is real (t) Input into the virtual model and calculate the virtual predicted value M pred (t).

8. The method for virtual construction of agricultural production based on digital twin according to claim 7, characterized in that: Through real-time monitoring and feedback units, actual values ​​are compared with predicted values, and feedback adjustments are made. ΔM(t)=M real (t)-M pred (t) Among them, ΔM(t) represents the model adjustment amount that changes with time, M real (t) represents the actual measured value, M pred (t) represents the model prediction value, and t is time.

9. The method for virtual construction of agricultural production based on digital twin according to claim 8, characterized in that: The agricultural production process is optimized and adjusted according to the model adjustment amount ΔM(t), and the optimization process is based on a complex iterative formula: Among them, P opt (t) represents the optimized production parameters, t is time, P current (t) represents the current production parameters, β k represents the adjustment coefficient, e -λk is the attenuation factor, k is the time step, ΔM(tk) is the model adjustment amount, and the model adjustment amount at time tk represents the difference between the actual measured value and the model predicted value. At the same time, P is set 阈值 (t) is the standard parameter value; When t = H time, stop the cycle optimization and output P opt (t) results.

10. The method for virtual construction of agricultural production based on digital twin according to claim 9, characterized in that: The warning unit includes a warning light, t = H time, stop cycle optimization, output P opt (t) The result P opt (t) and the threshold value P of the setting standard 阈值 (t) for comparison, that is, P 阈值 (t) and P opt (t) contrast, i.e. P 阈值 (t) <P opt (t) Stop the loop and output P opt (t), otherwise repeat steps 1 to 5 to continue optimizing P opt (t).

Images