A method for monitoring and controlling agricultural environment

Abstract

The invention relates to an agricultural environment monitoring and control method, which relates to the field of environmental control, and comprises: interacting with a target agricultural scene, acquiring scene feature information of the target agricultural scene, and analyzing and determining an environmental control area of ​​the target agricultural scene; according to the regional function in the environmental control area, interactively acquiring an environmental control record set through agricultural big data, and determining a control index set of the environmental control area; based on the control index set, deploying sensors in the target agricultural scene, and activating the deployed acquired sensor network to perform real-time environmental monitoring and acquire real-time environmental data; constructing a control efficiency mapping model based on the environmental control record set, and making a control decision on the control index set according to the control efficiency mapping model and the environmental control record set; converting a control decision result into an environmental control instruction set, and issuing the environmental control instruction set to a corresponding environmental control terminal to execute environmental control.

Description

Technical Field

[0001] The invention relates to the field of environmental control, and in particular to an agricultural environmental monitoring and control method. Background Art

[0002] In the field of smart agriculture, through environmental monitoring and control of agricultural scenes, key parameters such as temperature, humidity, light and soil conditions can be dynamically adjusted to meet the growth needs of crops.

[0003] Existing agricultural environmental monitoring and control technologies usually rely on a single sensor or a simple monitoring network. They collect environmental data and perform basic processing to generate control instructions to adjust the agricultural environment. The environmental monitoring area is divided in a single way, which makes it difficult to meet the multi-area and multi-functional needs in complex agricultural scenarios. It can only achieve simple reactive control based on real-time data, relies on a single variable, and it is difficult to establish a control efficiency mapping relationship between multiple factors, which leads to limitations in the control effect. Summary of the invention

[0004] The present invention aims to solve the technical problems of extensive control, insufficient pertinence of control decisions and lack of dynamic response capability in the prior art by providing an agricultural environment monitoring and control method.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] The present invention provides an agricultural environment monitoring and control method, comprising:

[0007] Interact with the target agricultural scene, obtain the scene feature information of the target agricultural scene, and analyze and determine the environmental control area of ​​the target agricultural scene.

[0008] According to the regional function in the environmental control area, the interactive agricultural big data obtains the environmental control record set and determines the control indicator set of the environmental control area.

[0009] Based on the control indicator set, sensors are deployed in the target agricultural scene, and the deployed sensor network is activated to perform real-time environmental monitoring and obtain real-time environmental data.

[0010] A control effectiveness mapping model based on the environmental control record set is constructed, and a control decision of the control indicator set is made according to the control effectiveness mapping model and the environmental control record set.

[0011] The real-time environmental data is compared with the control decision result to determine an environmental control instruction set, and the environmental control instruction set is sent to the corresponding environmental control terminal to execute environmental control.

[0012] The beneficial effects of the present invention are as follows: by interacting with the target agricultural scene, the scene feature information is acquired and parsed to determine the environmental control area of ​​the target agricultural scene; according to the functional characteristics of the environmental control area, the agricultural big data platform is connected, the environmental control record set is extracted, and the control index set of the area is determined; based on the control index set, sensor deployment is implemented in the target agricultural scene, the sensor network is activated to carry out real-time environmental monitoring, and real-time environmental data is collected; a control efficiency mapping model is constructed using the environmental control record set, and the decision of the control index set is optimized by combining the record set with the model results; the real-time environmental data is compared with the control decision result, the environmental control instruction set is determined, and it is sent to the corresponding environmental control terminal to complete the environmental control task, thereby achieving the technical effect of improving control accuracy, control decision targeting, and enhancing response capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A schematic flow chart of an agricultural environment monitoring and control method provided by the present invention;

[0014] Figure 2 A schematic diagram of a flow chart for obtaining the control decision result in an agricultural environment monitoring and control method provided by the present invention. DETAILED DESCRIPTION

[0015] 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0016] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0017] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0018] Embodiment 1, as Figure 1 As shown, an embodiment of the present invention provides an agricultural environment monitoring and control method.

[0019] Interact with the target agricultural scene, obtain the scene feature information of the target agricultural scene, and analyze and determine the environmental control area of ​​the target agricultural scene.

[0020] Specifically, first determine the environmental area that needs to be controlled for the target agricultural scene based on the scene characteristics of the target agricultural scene. Different agricultural scenes correspond to different areas that need to be controlled, and correspond to different control requirements and control characteristics, such as different temperature, humidity or light conditions.

[0021] Specifically, different control areas are delineated according to the spatial layout of the scene (such as greenhouses, farmlands, agricultural production lines, etc.) through sensor distribution or environmental control requirements, which will help to subsequently select specific control targets for each environmental control area. For example, some areas may need to maintain higher humidity to support planting, while other areas require lower humidity to avoid pests and diseases.

[0022] According to the regional function in the environmental control area, the interactive agricultural big data obtains an environmental control record set and determines a control indicator set of the environmental control area.

[0023] Specifically, the system interacts with the agricultural big data platform or the agricultural Internet of Things system through an API interface or a data warehouse to obtain relevant environmental control records and output them as an environmental control record set, which includes historical operation data of environmental control equipment. For example, the environmental control records include sensor data (such as temperature, humidity, light, soil moisture, CO2 concentration, etc.), data of automated control equipment (such as irrigation systems, air conditioning systems, heating equipment, etc.), and control operation logs (equipment start, stop, adjustment and other operations).

[0024] Specifically, according to the functions of each control area in the agricultural scene, the control indicators corresponding to each area are analyzed and determined. In other words, the control indicators of environmental control areas with different functions depend on the characteristics of the agricultural environment, crop needs and regional environmental regulation goals. For example, the greenhouse planting area needs to control temperature, humidity, CO2 concentration, light, etc.; the seedling area needs to control temperature, humidity, irrigation amount, etc.; the irrigation area controls soil moisture, irrigation amount, etc.

[0025] Through the above steps, the functional requirements of each environmental control area are analyzed to determine the corresponding control indicators for real-time monitoring and adjustment, so as to achieve precise environmental regulation and real-time feedback, thereby ensuring that environmental conditions meet the requirements for crop growth.

[0026] In some embodiments, according to the regional function in the environmental control area, the interactive agricultural big data obtains the environmental control record set, including:

[0027] Based on the regional function, a first call constraint is configured; based on the scene scale information of the target agricultural scene, a second call constraint is configured; according to the first call constraint and the second call constraint, the corresponding environmental control records are extracted from the interactive agricultural big data and output as the environmental control record set.

[0028] Specifically, first, clarify the functions of different areas in the target agricultural scenario, and configure corresponding calling constraints based on the functions of each environmental control area; regional functions usually correspond to different environmental control requirements. In other words, each function may require different types of environmental data and equipment control; exemplarily, regional functions include but are not limited to greenhouse areas, seedling areas, irrigation areas, livestock and poultry breeding areas, and storage / warehousing areas.

[0029] Specifically, according to the scale information of the target agricultural scene, a suitable second call constraint is configured to ensure that the acquired environmental control record set is consistent with or similar to the scale of the target agricultural scene; exemplarily, the scale of the agricultural scene involves site area, number of equipment, planting or breeding scale, etc.

[0030] Furthermore, according to the first call constraint and the second call constraint, the corresponding environmental control records are extracted from the agricultural big data platform to generate an environmental control record set, wherein the first call constraint clarifies the type of data to be extracted, in other words, different functions require different combinations of control indicators; the second call constraint ensures that the extracted data matches the actual scenario.

[0031] By combining regional functions and scene scales, configuring call constraints and interacting with agricultural big data, the environmental control record set of the target scene can be accurately extracted to meet the personalized needs of different agricultural scenes and provide data support for subsequent environmental control and optimization.

[0032] In some embodiments, determining the control indicator set of the environmental control area includes:

[0033] Statistically analyzing the occurrence frequencies of all control indicators in the environmental control record set, and serializing the control indicators based on the occurrence frequencies to obtain a control indicator sequence; selecting the first N control indicators from the control indicator sequence as the control indicator set, where N is a positive integer greater than 1.

[0034] Specifically, first, all the control indicators involved need to be extracted from the environmental control record set. Each environmental control record contains different control indicators, such as temperature, humidity, soil moisture, CO2 concentration, etc. Then, the frequency of all control indicators is counted to calculate the number of times each control indicator appears in all environmental control records.

[0035] Specifically, the control indicators are then sorted by frequency of occurrence to obtain an ordered control indicator sequence, and the first N control indicators are selected from the control indicator sequence to form a final control indicator set, where N is a positive integer greater than 1, depending on the specific scenario requirements. The higher the control requirements of the target agricultural scenario, the more indicator dimensions that need to be controlled, and the larger the corresponding N value.

[0036] Through the above steps, a representative control indicator set can be effectively extracted from the environmental control record set. In other words, indicators with higher frequencies are selected into the control indicator set. Through this process, it can be ensured that the most critical control indicators in the environmental control area receive priority attention, thereby providing data support for subsequent environmental control strategies.

[0037] Based on the control indicator set, sensors are deployed in the target agricultural scene, and the deployed sensor network is activated to perform real-time environmental monitoring and obtain real-time environmental data.

[0038] Specifically, according to each control indicator in the control indicator set, based on the required monitoring parameter type and accuracy requirements, appropriate sensors are selected for deployment to ensure that the required environmental data can be accurately monitored.

[0039] Optionally, the sensor deployment location is determined according to the actual situation of the target agricultural scenario. The deployment factors involve: site area. In larger areas, sensors need to be evenly distributed to ensure that the monitoring data is representative; microclimate differences. For areas with obvious microclimate differences, sensors should be deployed in places with obvious environmental characteristics; and sensor deployment coverage to ensure that the monitoring range of each sensor can cover the key parts of the target control area.

[0040] Specifically, after deploying the sensors, a sensor network is built to achieve real-time monitoring of environmental data. Sensor connection and data transmission are achieved through wired communication (such as Ethernet) or wireless sensor networks (such as LoRa, Zigbee, Wi-Fi, etc.); after the sensors are deployed and activated, the target agricultural scene will conduct real-time environmental monitoring and obtain real-time environmental data provided by the sensors. These data will reflect the environmental changes related to the control indicators in the target scene.

[0041] A control effectiveness mapping model based on the environmental control record set is constructed, and a control decision of the control indicator set is made according to the control effectiveness mapping model and the environmental control record set.

[0042] In some embodiments, a control effectiveness mapping model based on the environmental control record set is constructed, including: parsing the environmental control record set to obtain multiple control indicator data groups and corresponding environmental status labels; taking the control requirements of the environmental control area as a constraint, binary-dividing the multiple control indicator data groups and the corresponding environmental status labels to obtain a standard control sample set and an additional control sample set; based on the standard control sample set and the additional control sample set, constructing and supervised training the control effectiveness mapping model, wherein the multiple control indicator data groups are input independent variables, and the corresponding environmental status labels are output dependent variables.

[0043] Specifically, first, the environmental control record set needs to be parsed to extract the required information, including: control index data group, that is, the control index data contained in each record, such as temperature, humidity, light intensity, etc., which are input variables for environmental control. Environmental status labels are labels marked according to the actual effect of environmental control, such as qualified / unqualified or specific environmental status levels, representing whether the environment has achieved the expected control target.

[0044] Specifically, after parsing the record set, the control indicator data group and the environmental status label are divided into two parts based on the control requirements of the environmental control area, and a standard control sample set and an additional control sample set are obtained, wherein the control requirement constraint is to set the standard of control effect according to the actual needs of the target agricultural scenario or the control area, such as crop yield, disease rate, etc.; the standard control sample set contains samples that meet the set environmental control requirements, while the samples in the additional control sample set fail to fully meet the set environmental control requirements, that is, the environmental status label indicates that the environment does not meet the standards.

[0045] Furthermore, according to the extracted control indicator data and environmental condition labels, a suitable machine learning or regression analysis model is selected to construct a control effectiveness mapping model, exemplarily including a decision tree, a random forest, a support vector machine (SVM), a neural network, etc.

[0046] Specifically, the standard control sample set and the additional control sample set will be used as training data for the model. The model training is carried out through supervised learning, in which the standard control sample set will provide control effects that meet the standards as positive samples, and the additional control sample set will provide control effects that do not meet the standards as negative samples, the control indicator data group will be used as input features (independent variables), and the environmental condition label will be used as output (dependent variable).

[0047] Through the iterative supervised training process, it is ensured that the acquired control effectiveness mapping model can accurately capture the relationship between the input control indicators and the environmental condition labels. The model is cross-validated during the training process to evaluate the prediction accuracy of the model. The evaluation criteria include accuracy, precision, recall rate, F1 value and other indicators.

[0048] By constructing a control effectiveness mapping model based on the environmental control record set, it is possible to perform real-time analysis and prediction of environmental control data to achieve more accurate agricultural environmental control and improve agricultural production efficiency and resource utilization.

[0049] In some embodiments, making a control decision for the control indicator set according to the control effectiveness mapping model and the environmental control record set includes:

[0050] Based on the standard control sample set, the control indicator decision space of the environmental control area is defined; random extraction is performed in the standard control sample set, and the extraction results are mapped to initial nodes in the control indicator decision space to generate an initial node set; the initial node set is fused and optimized in combination with an optimization algorithm to obtain the control decision result, wherein the control effectiveness mapping model is an optimization cost function of fusion optimization.

[0051] Specifically, according to the numerical distribution of control indicators of multiple groups of samples in the standard control sample set, the possible value range of each control indicator is determined, and a multidimensional decision space is constructed; for example, if temperature and humidity are control indicators, the control range of temperature may be from 20℃ to 30℃, and the control range of humidity may be from 50% to 80%, then the control indicator decision space is the Cartesian product of these two ranges.

[0052] Specifically, based on the data in the standard control sample set, multiple samples are randomly extracted, and these samples are mapped to the initial nodes in the control indicator decision space. First, several sample data are randomly selected from the standard control sample set, and the corresponding control indicator data group is extracted. Then, each sample is mapped to the corresponding initial node in the control indicator decision space. Each initial node represents a possible control strategy, and the multiple nodes obtained can be regarded as potential control strategies.

[0053] Specifically, an optimization algorithm is used to fuse and optimize the initial node set to obtain the optimal control decision result. The optimization algorithm uses the control effectiveness mapping model as the optimization cost function to perform global multi-objective optimization. Optional optimization algorithms include genetic algorithm, particle swarm optimization, simulated annealing algorithm, gradient descent method, etc.

[0054] Specifically, during the fusion process, the optimization algorithm is combined to continuously search and iterate in the control index decision space, and multiple initial nodes are fused into better nodes to find the optimal control decision. After completing the fusion optimization of the initial node set, a control decision result will be obtained, that is, an optimal control strategy to guide the adjustment of agricultural environmental control equipment.

[0055] In some embodiments, Figure 2 As shown, the fusion optimization of the initial node set is performed in combination with the optimization algorithm to obtain the control decision result, wherein the control effectiveness mapping model is an optimization cost function of the fusion optimization, including:

[0056] Based on a random number generator, the initial optimization parameters of the initial node set are defined, wherein the initial optimization parameters include an optimization step and an optimization direction; based on the initial optimization parameters, the initial node set is iterated to generate an updated node set, the cost value of the updated node set is calculated by the optimization cost function, and the optimization direction is adjusted according to the cost value; if there is a unique common intersection point in the optimization directions of M nodes in the updated node set, the M nodes are merged to the common intersection point, and a new node is generated and added to the updated node set, wherein M is a positive integer greater than or equal to 2; when the number of nodes in the updated node set is less than or equal to a preset node number limit or the number of iterations meets a preset number, the node with the best corresponding cost value in the updated node set is output as the control decision result.

[0057] Specifically, first, the optimization parameters are defined for the initial node set through a random number generator to guide the direction and step size of the optimization process, where the initial optimization parameters include: optimization step size, which controls the update amplitude of the node during each optimization. A larger step size may help to quickly jump out of the local optimum, but may also miss the global optimum. A smaller step size is helpful for fine-tuning, but the calculation time may be longer; optimization direction, which is used to determine the optimization direction of the node at each iteration; based on the random number generator, the initial optimization parameters can be set to random values, which provides an initial exploration space for the optimization process.

[0058] Specifically, based on the defined initial optimization parameters, the initial node set is iterated, the cost value of each updated node set is calculated according to the optimization cost function (i.e., the control effectiveness mapping model), and an updated node set is generated; in each iteration, the optimization direction of the node is adjusted according to the position of the current node and the optimization step size. Exemplarily, the optimization direction is determined according to the cost function value of the current node, that is, the node is updated in the direction that reduces the cost value.

[0059] Furthermore, during the optimization process, if the optimization directions of multiple nodes tend to a common intersection, it means that these nodes have found a potential optimal solution during the optimization process. At this time, these nodes can be merged into one node, which can reduce the number of nodes and improve computing efficiency.

[0060] Optionally, for each group of nodes that converge to the same direction, their mean or weighted average is calculated and added as a new node to the update node set.

[0061] Specifically, when the number of nodes in the update node set is less than or equal to a preset limit or when the number of iterations reaches a preset number, the optimization is stopped to ensure that the optimization process is completed within a reasonable time; the limit is used to avoid too many nodes slowing down the calculation speed, while also ensuring a certain search space; when the stopping condition is met, the node with the best cost value in the update node set (that is, the node with the best control efficiency) is selected as the final control strategy, and the control indicator combination corresponding to the node is the optimal control decision.

[0062] Through the optimization cost function and optimization algorithm based on the control effectiveness mapping model, the initial node set is iteratively optimized, the nodes are fused and merged, and finally the optimal control decision result is output, which effectively transforms the complex control decision problem into a multidimensional optimization problem, thereby providing efficient and intelligent control strategies for the agricultural environment to ensure that environmental conditions meet the optimal agricultural production needs.

[0063] The real-time environmental data is compared with the control decision result to determine an environmental control instruction set, and the environmental control instruction set is sent to the corresponding environmental control terminal to execute environmental control.

[0064] Specifically, the real-time environmental data is first compared with the control decision results obtained through the optimization model, that is, the key environmental parameters in the real-time environmental data and the control decision results are compared, and the difference value (for example, temperature deviation, humidity deviation, etc.) is calculated. If the environmental data is significantly different from the target control value, a control instruction is generated to adjust the environment.

[0065] Optionally, a threshold range is set to determine whether the deviation between the actual data and the decision result exceeds a preset tolerance range. If it exceeds the range, control measures need to be triggered to generate an environmental control instruction set.

[0066] Furthermore, the generated environmental control instruction set is sent to the corresponding environmental control terminal, and the environmental control terminal performs corresponding operations according to the received instructions, wherein each control terminal executes the equipment's start, stop, adjustment of working parameters, etc.

[0067] Optionally, environmental data can be obtained regularly or in real time and compared with the target value to ensure that the control effect of the equipment meets expectations. If the control result is found to be inconsistent with expectations, such as the temperature still deviates from the target value, a new control instruction can be generated for adjustment.

[0068] By comparing real-time environmental data and control decision results, an environmental control instruction set is generated and sent to the environmental control terminal to achieve automated and intelligent regulation of the agricultural environment, thereby helping agricultural production to respond quickly to environmental changes and ensure that environmental conditions meet preset standards, thereby improving agricultural production efficiency and quality.

[0069] In some implementations, the method further includes:

[0070] Based on the additional control sample set, a taboo space of the environmental control area is defined; the taboo space is introduced into a fusion optimization process to perform fusion optimization of the initial node set.

[0071] Optionally, a taboo space is defined in the fusion optimization process. The taboo space is a constraint area defined based on an additional control sample set. In this area, the control strategy is determined to be inappropriate or not allowed. Through the taboo space, some unreasonable or invalid optimization solutions can be excluded, preventing the algorithm from exploring solutions that do not meet actual needs, thereby improving optimization efficiency.

[0072] Specifically, during the optimization process, each node generated by each iteration will be checked to see if it is in the taboo space. If a solution is in the taboo space, it will be excluded and will not participate in the subsequent optimization process. The introduction of the taboo space reduces the generation of invalid solutions, allowing the optimization algorithm to converge to the effective decision space faster, thereby improving the optimization efficiency.

[0073] In some implementations, determining the control indicator set of the environmental control area further includes:

[0074] Analyze the environmental control record set to extract multiple control indicator data groups and corresponding environmental status labels; apply principal component analysis to the multiple control indicator data groups and corresponding environmental status labels to obtain a control indicator list; and match the corresponding first N control indicators in the control indicator list according to the environmental control terminal of the target agricultural scene as the control indicator set, where N is a positive integer greater than 1.

[0075] Specifically, in order to screen and optimize the control indicator set more scientifically, the control indicator data group and environmental condition labels are reduced in dimension through the principal component analysis (PCA) method, so as to extract the control indicators that have the greatest impact on the target agricultural scenario.

[0076] Specifically, after extracting the control indicator list through PCA, combined with the environmental control terminal of the target agricultural scene, the top N most relevant control indicators are selected from the control indicator list as the target control indicator set. In other words, the top N control indicators that can be controlled by the environmental control terminal of the target agricultural scene are selected from the control indicator list to ensure the control adaptability of subsequent control decision results to the target agricultural scene.

[0077] The agricultural environment monitoring and control method provided by the embodiment of the present invention has at least the following technical effects:

[0078] By interacting with the target agricultural scene, the scene feature information is obtained and parsed to determine the environmental control area of ​​the target agricultural scene; according to the functional characteristics of the environmental control area, the agricultural big data platform is connected, the environmental control record set is extracted, and the control indicator set of the area is determined; based on the control indicator set, sensor deployment is implemented in the target agricultural scene, and the sensor network is activated to carry out real-time environmental monitoring and collect real-time environmental data; a control effectiveness mapping model is constructed using the environmental control record set, and the decision of the control indicator set is optimized by combining the record set with the model results; the real-time environmental data is compared with the control decision results, the environmental control instruction set is determined, and it is sent to the corresponding environmental control terminal to complete the environmental control task, thereby achieving the technical effect of improving control accuracy, control decision-making targeting, and enhancing response capabilities.

[0079] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0081] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0083] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0084] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A method for monitoring and controlling an agricultural environment, characterized in that: The method comprises: Interact with the target agricultural scene, obtain the scene feature information of the target agricultural scene, and analyze and determine the environmental control area of ​​the target agricultural scene; According to the regional function in the environmental control area, the interactive agricultural big data obtains the environmental control record set, and determines the control indicator set of the environmental control area; Based on the control indicator set, sensors are deployed in the target agricultural scene, and the deployed sensor network is activated to perform real-time environmental monitoring and obtain real-time environmental data; Constructing a control effectiveness mapping model based on the environmental control record set, and making a control decision for the control indicator set according to the control effectiveness mapping model and the environmental control record set; Comparing the real-time environmental data with the control decision result, determining an environmental control instruction set, and sending the environmental control instruction set to a corresponding environmental control terminal to execute environmental control; Wherein, according to the regional function in the environmental control area, the interactive agricultural big data obtains the environmental control record set, including: Based on the regional function, configuring a first call constraint; Based on the scene scale information of the target agricultural scene, configure the second call constraint; According to the first call constraint and the second call constraint, the corresponding environmental control record is extracted from the interactive agricultural big data, and the output is the environmental control record set; Wherein, determining the control indicator set of the environmental control area includes: Statistically analyzing the occurrence frequencies of all control indicators in the environmental control record set, and serializing the control indicators based on the occurrence frequencies to obtain a control indicator sequence; Selecting the first N control indicators from the control indicator sequence as the control indicator set, where N is a positive integer greater than 1; Wherein, determining the control indicator set of the environmental control area also includes: Analyze the environmental control record set to extract multiple control indicator data groups and corresponding environmental status labels; Applying principal component analysis to the plurality of control indicator data groups and corresponding environmental condition labels to obtain a control indicator list; According to the environmental control terminal of the target agricultural scene, the corresponding first N control indicators are matched in the control indicator list as the control indicator set, where N is a positive integer greater than 1.

2. The agricultural environment monitoring and control method according to claim 1, characterized in that: Constructing a control effectiveness mapping model based on the environmental control record set, including: Parsing the environmental control record set to obtain multiple control indicator data groups and corresponding environmental status labels; Taking the control requirements of the environmental control area as constraints, a plurality of the control indicator data groups and the corresponding environmental condition labels are divided into two groups to obtain a standard control sample set and an additional control sample set; Based on the standard control sample set and the additional control sample set, the control effectiveness mapping model is constructed and supervised for training, wherein the plurality of control indicator data sets are input independent variables and the corresponding environmental condition labels are output dependent variables.

3. The agricultural environment monitoring and control method according to claim 2, characterized in that: According to the control effectiveness mapping model and the environmental control record set, a control decision of the control indicator set is made, including: Based on the standard control sample set, defining a control index decision space of the environmental control area; Performing random extraction from the standard control sample set, and mapping the extraction results to initial nodes in the control indicator decision space to generate an initial node set; The initial node set is fused and optimized in combination with an optimization algorithm to obtain the control decision result, wherein the control effectiveness mapping model is an optimization cost function of fusion optimization.

4. The agricultural environment monitoring and control method according to claim 3, wherein the initial node set is fused and optimized in combination with an optimization algorithm to obtain the control decision result, wherein: The control effectiveness mapping model is an optimization cost function of fusion optimization, including: Based on a random number generator, defining initial optimization parameters of the initial node set, wherein the initial optimization parameters include an optimization step size and an optimization direction; Iterating the initial node set based on the initial optimization parameters to generate an updated node set, calculating a cost value of the updated node set by using the optimization cost function, and adjusting the optimization direction according to the cost value; If there is a unique common intersection point between the optimization directions of M nodes in the update node set, then the M nodes are correspondingly merged to the common intersection point, and a new node is generated and added to the update node set, where M is a positive integer greater than or equal to 2; When the number of nodes in the update node set is less than or equal to the preset node number limit or the number of iterations meets the preset number, the node with the best corresponding cost value in the update node set is output as the control decision result.

5. The agricultural environment monitoring and control method according to claim 3, characterized in that: The method further comprises: Based on the additional control sample set, defining a taboo space of the environmental control area; The taboo space is introduced into the fusion optimization process to perform fusion optimization of the initial node set.

Images