Greenhouse operation management and control method, device, equipment, medium and computer program product

Abstract

The invention relates to the field of greenhouse operation management and control, and provides a greenhouse operation management and control method, device and equipment, a medium and a computer program product, and the method comprises the steps: receiving crop visual information and environment information sent by greenhouse monitoring equipment; inputting the crop visual information and the environment information into a greenhouse crop planting knowledge graph to obtain a greenhouse operation management and control macroscopic decision; inputting the greenhouse operation management and control macroscopic decision into a greenhouse model to obtain an operation decision result output by the greenhouse model; the greenhouse model is obtained based on historical greenhouse operation data training. On the basis of the received crop visual information and environment information, the crop visual information and the environment information are analyzed through the established greenhouse crop planting knowledge graph, a greenhouse operation management and control macroscopic decision is obtained, the greenhouse operation management and control macroscopic decision is further combined with a greenhouse model obtained based on historical greenhouse operation data training, and a greenhouse operation management and control result is obtained. Therefore, the accuracy and efficiency of greenhouse operation management and control are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of greenhouse operation control, and in particular to a greenhouse operation control method, device, equipment, medium and computer program product. BACKGROUND

[0002] Current greenhouse operation control still adopts manual mode, and automation and intelligence are low. Due to the difference in professional degree and low efficiency of manual control, the greenhouse crops cannot grow in the suitable environment, which leads to the reduction of the yield of greenhouse crops. The control mode based on preset rules has the problems of low control flexibility and poor control accuracy. Therefore, there is an urgent need for a greenhouse operation control method with high efficiency and accuracy. SUMMARY

[0003] The present application provides a greenhouse operation control method, device, equipment, medium and computer program product to solve the problems of low efficiency and poor accuracy of the existing greenhouse operation control mode.

[0004] The present application provides a greenhouse operation control method, comprising the following steps: receiving crop visual information and environmental information sent by a greenhouse monitoring device; inputting the crop visual information and the environmental information into a greenhouse crop planting knowledge graph to obtain a greenhouse operation control macro decision; inputting the greenhouse operation control macro decision into a greenhouse model to obtain an operation decision result output by the greenhouse model; the greenhouse model is obtained based on historical greenhouse operation data.

[0005] According to the greenhouse operation control method provided by the present application, the greenhouse operation control macro decision is input into the greenhouse model to obtain the operation decision result output by the greenhouse model, and then comprises: determining a current crop state and an optimal crop state based on the crop visual information and the environmental information; inputting a crop deviation into the greenhouse crop planting knowledge graph to obtain a target decision result; the crop deviation is determined based on the current crop state and the optimal crop state; adjusting the initial crop demand based on the target decision result; inputting the adjusted initial crop demand into the greenhouse model to obtain an operation control decision; adjusting the greenhouse operation decision result based on the operation control decision.

[0006] According to the greenhouse operation control method provided by the present application, inputting the adjusted initial crop demand into the greenhouse model to obtain an operation control decision comprises: input the adjusted crop initial demand into the greenhouse model to obtain a predicted yield function, production consumption and operation energy consumption; determine a relationship between control decisions and operation targets based on the predicted yield function, the production consumption and the operation energy consumption; determine operation control decisions based on the relationship between control decisions and operation targets and the adjusted crop initial demand.

[0007] According to the greenhouse operation management and control method provided by the application, the greenhouse model comprises a greenhouse environment model and a decision model; the greenhouse operation management and control macro decision is input into the greenhouse model to obtain operation decision results output by the greenhouse model, and the operation decision results output by the greenhouse model comprise: input the greenhouse operation management and control macro decision into the greenhouse environment model and the decision model to obtain operation control set values output by the greenhouse environment model and greenhouse operation decisions output by the decision model; obtain operation decision results based on the operation control set values and the greenhouse operation decisions.

[0008] According to the greenhouse operation management and control method provided by the application, the greenhouse monitoring device comprises a patrol robot; the greenhouse operation management and control macro decision is input into the greenhouse model to obtain operation decision results output by the greenhouse model, and the method further comprises the following steps: determine a patrol area of the patrol robot; if the greenhouse operation decision corresponding to the patrol area is pollination of crops, determine to-be-pollinated crops and pollination data based on operation control set values corresponding to the patrol area; send operation instructions to the operation robot corresponding to the patrol area based on the to-be-pollinated crops and the pollination data.

[0009] According to the greenhouse operation management and control method provided by the application, the greenhouse operation management and control method further comprises the following steps: obtain planting knowledge texts and historical image data of greenhouse crops; determine knowledge nodes based on knowledge entities and image features; the knowledge entities are obtained by performing entity extraction on the planting knowledge texts; and the image features are obtained by performing feature extraction on the historical image data; construct a greenhouse crop planting knowledge graph based on relationships between the knowledge nodes.

[0010] The application further provides a greenhouse operation management and control device comprising the following modules: an information receiving module, configured to receive crop visual information and environmental information sent by a greenhouse monitoring device; The greenhouse operation control macro decision determination module is configured to input the crop visual information and the environment information into a greenhouse crop planting knowledge graph to obtain a greenhouse operation control macro decision. The operation decision result determination module is configured to input the greenhouse operation control macro decision into a greenhouse model to obtain an operation decision result output by the greenhouse model.

[0011] The present application also provides an electronic device including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the greenhouse operation control method according to any one of the above when executing the computer program.

[0012] The present application also provides a non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the greenhouse operation control method according to any one of the above.

[0013] The present application also provides a computer program product including a computer program, wherein the computer program is executed by a processor to implement the greenhouse operation control method according to any one of the above.

[0014] The greenhouse operation control method, device, equipment, medium and computer program product provided by the present application are based on the crop visual information and the environment information sent by the received greenhouse monitoring equipment, analyze the crop visual information and the environment information through the established greenhouse crop planting knowledge graph, obtain the greenhouse operation control macro decision, further combine the greenhouse operation control macro decision with the greenhouse model trained based on the historical greenhouse operation data, and obtain the greenhouse operation decision result. The greenhouse operation decision is obtained by combining the established greenhouse crop planting knowledge graph with the trained greenhouse model, which improves the accuracy and efficiency of greenhouse operation control. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0016] Figure 1 is one of the flowcharts of the greenhouse operation control method provided by the present application.

[0017] Figure 2 is a construction schematic diagram of the greenhouse crop planting knowledge graph provided by the present application.

[0018] Figure 3A decision schematic diagram of the greenhouse crop planting knowledge graph provided by the present application.

[0019] Figure 4 A second flow schematic diagram of the greenhouse operation management and control method provided by the present application.

[0020] Figure 5 An analysis schematic diagram of the greenhouse crop planting knowledge graph provided by the present application.

[0021] Figure 6 A greenhouse operation management and control schematic diagram provided by the present application.

[0022] Figure 7 A structural schematic diagram of the greenhouse operation management and control device provided by the present application.

[0023] Figure 8 A structural schematic diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0025] The greenhouse operation management and control method, device, equipment, medium and computer program product of the present application will be described below with reference to the drawings. Figures 1-8

[0026] Figure 1 A first flow schematic diagram of the greenhouse operation management and control method provided by the present application, as shown in the figure, the method comprises the following: Figure 1 Step 100, receiving crop visual information and environmental information sent by a greenhouse monitoring device. Specifically, indoor sensors and inspection robots are arranged in the greenhouse to be managed and controlled, and outdoor sensors are arranged outside the greenhouse. The greenhouse monitoring device in the present embodiment comprises the indoor sensors, the outdoor sensors and the inspection robots. The environmental information in the present embodiment comprises indoor environmental information (for example, air temperature, air humidity, carbon dioxide concentration, light intensity and total radiation) acquired by the indoor sensors and outdoor environmental information (for example, atmospheric temperature, air humidity, rainfall, wind speed, wind direction, atmospheric pressure, total radiation and light intensity) acquired by the outdoor sensors. The crop growth picture (i.e. the crop visual information in the present embodiment) is acquired by the visual recognition system of the inspection robot. The crop growth picture can contain information such as leaf number, leaf area, fruit number, fruit size and crop growth period. ​​

[0027] Step 200, input the crop visual information and the environment information into the greenhouse crop planting knowledge graph to obtain a greenhouse operation control macro decision; Specifically, the greenhouse crop is exemplified by tomato. The greenhouse crop grower and expert experience are standardized to establish a standardized tomato planting knowledge graph (i.e. the greenhouse crop planting knowledge graph in this embodiment). As shown in the figure, a large amount of dispersed data in the tomato planting knowledge text and historical image data are linked together to obtain the greenhouse crop planting knowledge graph that turns experience into knowledge. Figure 2 The more interpretable knowledge obtained, the higher the reliability of the knowledge graph. As shown in the figure, the crop visual information and the environment information sent by the greenhouse monitoring device are input into the greenhouse crop planting knowledge graph, and the greenhouse operation control macro decision is obtained by the greenhouse crop planting knowledge graph analysis.

[0028] Figure 3 Step 300, input the greenhouse operation control macro decision into the greenhouse model to obtain the operation decision result output by the greenhouse model; the greenhouse model is obtained by training based on historical greenhouse operation data.

[0029] Specifically, on the basis of the greenhouse operation control macro decision obtained by the greenhouse crop planting knowledge graph analysis, the micro decision of the greenhouse operation control is realized by the greenhouse model.

[0030] Specifically, on the basis of the greenhouse operation control macro decision obtained by the greenhouse crop planting knowledge graph analysis, the micro decision of the greenhouse operation control is realized by the greenhouse model.

[0031] Taking the environment control decision as an example, the environment information sent by the greenhouse monitoring device includes daytime temperature (28℃), humidity (50%) and light intensity; the most suitable growth environment of tomato obtained by knowledge graph matching includes daytime temperature between 22℃ and 26℃, humidity between 60% and 70%, and high temperature and high light are easy to cause sunburn disease; the greenhouse operation control macro decision is to need to control the growth environment of tomato. In the case that the light intensity is higher than the suitable threshold value of tomato, the operation decision result output by the greenhouse model includes opening the greenhouse sunshade net (to reduce the light intensity), starting the wet curtain cooling (to reduce the temperature) and opening the air humidity spray (to increase the humidity).

[0032] Taking the disease and pest control decision as an example, the crop visual information sent by the greenhouse monitoring device is the image of tomato leaf yellow spot and white powdery substance on the back of tomato leaf; the symptom of tomato determined by the knowledge graph matching is white powdery disease easy to occur in high humidity environment, and the corresponding control scheme is to reduce the humidity and spray 25% kresoxim-methyl suspension agent; the greenhouse operation control macro decision is to need to control the disease and pest of tomato. The operation decision result output by the greenhouse model includes strengthening ventilation, reducing the humidity to 50%, and spraying the diseased tomato according to the planned path.

[0033] ​The greenhouse operation control method provided by the embodiment is based on the crop visual information and the environmental information sent by the received greenhouse monitoring device, analyzes the crop visual information and the environmental information through the established greenhouse crop planting knowledge graph, obtains greenhouse operation control macro decision, and further combines the greenhouse operation control macro decision with the greenhouse model trained based on historical greenhouse operation data to obtain a greenhouse operation decision result. The greenhouse operation decision is obtained by combining the established greenhouse crop planting knowledge graph with the trained greenhouse model, thereby improving the accuracy and efficiency of greenhouse operation control.

[0034] Figure 4 is a flowchart of the greenhouse operation control method provided by the embodiment, as shown in Figure 4 , the method can further include: Step 400, determining a current crop state and an optimal crop state based on the crop visual information and the environmental information; Step 500, inputting a crop deviation into a greenhouse crop planting knowledge graph to obtain a target decision result; the crop deviation is determined based on the current crop state and the optimal crop state; Step 600, adjusting a crop initial demand based on the target decision result; Step 700, inputting the adjusted crop initial demand into the greenhouse model to obtain an operation control decision; Step 800, adjusting the greenhouse operation decision result based on the operation control decision.

[0035] Specifically, the crop initial demand in the embodiment is a default setting, and the crop initial demand can be obtained based on the crop visual information and the environmental information in combination with the experience of greenhouse crop planters and expert experience.

[0036] As shown in Figure 5 , based on the crop visual information and the environmental information, the current crop state and the optimal crop state are determined. As shown in formula 1, the crop growth deviation is obtained by comparing the current crop state and the optimal crop state. is a production and consumption function related to climate, energy and raw materials, wherein, is a time-related production and consumption; is a time-related operation energy consumption.

[0037] ; (1) ; (2) The crop initial demand is modified according to the crop bias, and the adjusted crop initial demand is added to the input end of the greenhouse model to make a control decision, and an operation control decision is obtained. On the basis of the operation control decision, the greenhouse operation decision result is adjusted.

[0038] The embodiment adjusts the greenhouse operation decision result through the operation control decision to make a decision within the framework of meeting the crop demand.

[0039] In one embodiment, the greenhouse operation management method provided by the embodiment of the present application can further include: Step 710: inputting the adjusted crop initial demand into the greenhouse model to obtain a predicted yield function, production consumption and operation energy consumption; Step 720: determining the relationship between the control decision and the operation target based on the predicted yield function, the production consumption and the operation energy consumption; Step 730: determining an operation control decision based on the relationship between the control decision and the operation target and the adjusted crop initial demand.

[0040] Specifically, the purpose of the greenhouse operation decision can be to pursue the maximum profit of greenhouse crop planting. In the case that the crop initial demand is determined and the crop is not under stress, reducing consumption is the most effective way to maximize planting profits. Under this decision target of maximizing planting profits, an operation control scheme flowchart as shown in FIG. 3 is formulated. As shown in formula 3 (relationship between control decision and operation target), the optimal solution of the control decision will be given by maximizing the net profit Figure 6 , wherein is the predicted yield function of the crop.

[0041] ; (3) Taking profit maximization as the target may exist a determined optimal operation decision result which is not the best demand of the greenhouse crop. For example, the crop is under stress, and the cost of solving this problem will far exceed the crop yield, therefore, the operation decision is made within the framework of meeting the crop demand, that is, the operation decision result output by the greenhouse model based on the operation control decision is adjusted.

[0042] The embodiment adjusts the greenhouse operation decision result through the operation control decision to make a decision within the framework of meeting the crop demand.

[0043] In one embodiment, the greenhouse operation management method provided by the embodiment of the present application can further include: ​​Step 310, input the greenhouse operation control macro decision into the greenhouse environment model and the decision model, obtain the operation control setting value output by the greenhouse environment model, and the greenhouse operation decision output by the decision model; Step 320, based on the operation control setting value and the greenhouse operation decision, obtain the operation decision result.

[0044] Specifically, the greenhouse model includes a greenhouse environment model and a decision model, and the output of the greenhouse model includes operation control related values and greenhouse operation decisions, that is, what to do and how to do for the greenhouse crops.

[0045] The greenhouse operation control macro decision output by the greenhouse crop planting knowledge graph is input into the decision model to obtain the greenhouse operation decision output by the decision model, that is, the operation control of the greenhouse crops, including operation type and operation mode, etc. The greenhouse operation control macro decision output by the greenhouse crop planting knowledge graph is input into the greenhouse environment model to obtain the operation control setting value output by the greenhouse environment model, that is, the operation specific value of the greenhouse crops, including operation time and operation amount, etc. quantifiable data.

[0046] The operation decision output by the decision model and the operation specific value output by the greenhouse environment model are combined to obtain a comprehensive greenhouse operation decision result.

[0047] In one embodiment, the greenhouse operation control method provided by the embodiment of the present application can further include: Step 900, determine the inspection area of the inspection robot; Step 1000, in the case that the greenhouse operation decision corresponding to the inspection area is crop pollination, determine the to-be-pollinated crops and pollination data based on the operation control setting value corresponding to the inspection area; Step 1100, based on the to-be-pollinated crops and the pollination data, send an operation instruction to the operation robot corresponding to the inspection area.

[0048] Specifically, the main process of the greenhouse operation control method provided by the present application is to dispatch related robots to perform pollination, picking, and inspection, etc. operation control work for greenhouse crops. The operation control includes operation area, operation content, and operation time, etc. Each inspection robot performs inspection in the corresponding inspection area to obtain inspection information.

[0049] In the case that the greenhouse operation decision determined based on the inspection information is crop pollination, the inspection area corresponding to the inspection information is determined, the greenhouse operation decision includes to-be-pollinated tomatoes, pollination types, and pollination amounts, etc. pollination data, and then a pollination operation instruction (determined by the greenhouse operation decision) is sent to the operation robot corresponding to the inspection area.

[0050] Similarly, in the case that the greenhouse operation decision determined based on the inspection information is crop picking, the inspection area corresponding to the inspection information is determined, the greenhouse operation decision includes tomatoes to be picked, picking method and picking position, etc., and then the picking operation instruction determined by the greenhouse operation decision is sent to the operation robot corresponding to the inspection area. In the case that the greenhouse operation decision determined based on the inspection information is crop disease and pest treatment, the inspection area corresponding to the inspection information is determined, the greenhouse operation decision includes tomatoes to be treated, disease and pest treatment method and disease and pest treatment position, etc., and then the disease and pest treatment operation instruction determined by the greenhouse operation decision is sent to the operation robot corresponding to the inspection area.

[0051] The embodiment associates the greenhouse area with the operation decision to obtain accurate operation instructions sent to the corresponding operation robot.

[0052] In one embodiment, the greenhouse operation management method provided by the embodiment of the present application can further include: Step 10, acquiring planting knowledge text and historical image data of greenhouse crops; Step 20, determining knowledge nodes based on knowledge entities and image features; the knowledge entities are obtained by entity extraction on the planting knowledge text; and the image features are obtained by feature extraction on the historical image data; Step 30, constructing a greenhouse crop planting knowledge graph based on the relationships between the knowledge nodes.

[0053] Specifically, the construction process of the greenhouse crop planting knowledge graph mainly includes three parts of data preprocessing, knowledge extraction and knowledge fusion.

[0054] 1. Data preprocessing The collected structured data includes greenhouse environment monitoring data, water and fertilizer irrigation records, operation logs and disease and pest control records. Unstructured data includes tomato growth cycle images (i.e. historical image data), disease and pest symptom image library, agricultural expert manual and literature and planting experience summary documents. Semi-structured data includes agricultural standards and specifications, pesticide use instructions and equipment operation manuals.

[0055] For the processing of historical image data, it includes image cleaning (eliminating images with problems such as blur, overexposure and underexposure), data enhancement (including rotation processing, flip processing and color adjustment, etc. to expand images), labeling processing (labeling tomato growth key features, establishing labeling specifications such as tomato maturity classification and disease spot types) and feature extraction. For the processing of text data, it includes text cleaning, word segmentation processing and entity recognition (constructing an entity dictionary in the field of greenhouse operation).

[0056] 2. Knowledge extraction Knowledge extraction includes entity relation extraction and attribute extraction. Entity relation extraction includes rule-based method and deep learning-based method; the rule-based method is based on the construction of the agricultural field mode matching rule based on the entity dictionary, and the deep learning-based method can extract the relation between entities, detect the key parts of tomatoes, and establish visual relation through a related model. Attribute extraction includes extracting numerical attributes of environmental parameters, time attributes of farming operations, and configuration attributes of device parameters from structured data.

[0057] 3. Knowledge fusion Knowledge fusion includes entity alignment and knowledge completion. The relation between knowledge nodes is obtained through entity alignment and knowledge completion, and the knowledge nodes store the image features or text entities described above. The greenhouse crop planting knowledge graph is constructed through the relation between the knowledge nodes.

[0058] The greenhouse crop planting knowledge graph constructed in the embodiment can knowledgeize the greenhouse operation experience and improve the explainability of the greenhouse operation experience.

[0059] The greenhouse operation management and control device provided by the present application is described below, and the greenhouse operation management and control device described below can be correspondingly referred to the greenhouse operation management and control method described above.

[0060] Please refer to Figure 7 The present application also provides a greenhouse operation management and control device, comprising: An information receiving module 701 is configured to receive crop visual information and environmental information sent by a greenhouse monitoring device; A greenhouse operation management and control macro decision determination module 702 is configured to input the crop visual information and the environmental information into a greenhouse crop planting knowledge graph to obtain a greenhouse operation management and control macro decision; An operation decision result determination module 703 is configured to input the greenhouse operation management and control macro decision into a greenhouse model to obtain an operation decision result output by the greenhouse model; the greenhouse model is obtained by training historical greenhouse operation data.

[0061] Optionally, the greenhouse operation management and control device further comprises: A crop state determination module is configured to determine a current crop state and an optimal crop state based on the crop visual information and the environmental information; A target decision result determination module is configured to input a crop deviation into the greenhouse crop planting knowledge graph to obtain a target decision result; the crop deviation is determined based on the current crop state and the optimal crop state; A crop initial demand adjustment module is configured to adjust a crop initial demand based on the target decision result; The operation control decision determination module inputs the adjusted crop initial demand into the greenhouse model to obtain an operation control decision. The greenhouse operation decision result adjustment module adjusts the greenhouse operation decision result based on the operation control decision.

[0062] Optionally, the operation control decision determination module comprises: The greenhouse model output determination unit inputs the adjusted crop initial demand into the greenhouse model to obtain a predicted yield function, production consumption and operation energy consumption. The relationship determination unit determines the relationship between the control decision and the operation target based on the predicted yield function, the production consumption and the operation energy consumption. The operation control decision determination unit determines the operation control decision based on the relationship between the control decision and the operation target and the adjusted crop initial demand.

[0063] Optionally, the greenhouse model comprises a greenhouse environment model and a decision model; and the operation decision result determination module comprises: The model output unit inputs the greenhouse operation management and control macro decision into the greenhouse environment model and the decision model to obtain an operation control setting value output by the greenhouse environment model and a greenhouse operation decision output by the decision model. The operation decision result determination unit obtains an operation decision result based on the operation control setting value and the greenhouse operation decision.

[0064] Optionally, the greenhouse operation management and control device further comprises: The inspection area determination module determines an inspection area of the inspection robot. The pollination operation determination module determines a to-be-pollinated crop and pollination data based on the operation control setting value corresponding to the inspection area, in a case where the greenhouse operation decision corresponding to the inspection area is crop pollination. The operation instruction sending module sends an operation instruction to an operation robot corresponding to the inspection area based on the to-be-pollinated crop and the pollination data.

[0065] Optionally, the greenhouse operation management and control device further comprises: The acquisition module acquires planting knowledge text and historical image data of a greenhouse crop. The knowledge node determination module determines a knowledge node based on a knowledge entity and an image feature; the knowledge entity is obtained by performing entity extraction on the planting knowledge text; and the image feature is obtained by performing feature extraction on the historical image data. The greenhouse crop planting knowledge graph construction module is configured to construct a greenhouse crop planting knowledge graph based on the relationships between the knowledge nodes.

[0066] Figure 8 An example of an entity structure diagram of an electronic device is shown in FIG. 8. Figure 8 As shown in FIG. 8, the electronic device can include a processor 810, a communications interface 820, a memory 830, and a communications bus 840, wherein the processor 810, the communications interface 820, and the memory 830 communicate with each other through the communications bus 840. The processor 810 can invoke the logic instructions in the memory 830 to execute a greenhouse operation control method, which includes receiving crop visual information and environmental information sent by a greenhouse monitoring device; inputting the crop visual information and the environmental information into a greenhouse crop planting knowledge graph to obtain a greenhouse operation control macro decision; and inputting the greenhouse operation control macro decision into a greenhouse model to obtain an operation decision result output by the greenhouse model; wherein the greenhouse model is trained based on historical greenhouse operation data.

[0067] In addition, the logic instructions in the memory 830 described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium, includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0068] In another aspect, the present application also provides a computer program product comprising a computer program, the computer program being stored in a non-transitory computer-readable storage medium, and the computer program being executable by a processor to enable a computer to perform the greenhouse operation management and control method provided by any of the above methods, the method comprising: receiving crop visual information and environment information sent by a greenhouse monitoring device; inputting the crop visual information and the environment information into a greenhouse crop planting knowledge graph to obtain a greenhouse operation management and control macro decision; and inputting the greenhouse operation management and control macro decision into a greenhouse model to obtain an operation decision result output by the greenhouse model; wherein the greenhouse model is trained based on historical greenhouse operation data.

[0069] In another aspect, the present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, the computer program being executable by a processor to implement the greenhouse operation management and control method provided by any of the above methods, the method comprising: receiving crop visual information and environment information sent by a greenhouse monitoring device; inputting the crop visual information and the environment information into a greenhouse crop planting knowledge graph to obtain a greenhouse operation management and control macro decision; and inputting the greenhouse operation management and control macro decision into a greenhouse model to obtain an operation decision result output by the greenhouse model; wherein the greenhouse model is trained based on historical greenhouse operation data.

[0070] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0071] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by means of software plus necessary general hardware platforms, and of course can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0072] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit the same; and although the present application has been described in detail with reference to the foregoing embodiments, it should be appreciated by those skilled in the art that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features thereof can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for managing greenhouse operations, characterized in that, include: Receive crop visual information and environmental information sent by greenhouse monitoring equipment; The crop visual information and the environmental information are input into a greenhouse crop planting knowledge graph to obtain macro-level decisions for greenhouse operation management. The macro-level decision-making process for greenhouse operation management is input into the greenhouse model to obtain the operation decision-making results output by the greenhouse model; the greenhouse model is trained based on historical greenhouse operation data.

2. The greenhouse operation control method according to claim 1, characterized in that, The step of inputting the macro-level decision on greenhouse operation management into the greenhouse model and obtaining the operation decision results output by the greenhouse model includes: Based on the crop visual information and the environmental information, the current crop state and the optimal crop state are determined; The crop deviation is input into the greenhouse crop planting knowledge graph to obtain the target decision result; the crop deviation is determined based on the current crop state and the optimal crop state. The initial crop requirements are adjusted based on the aforementioned target decision results; The adjusted initial crop requirements are input into the greenhouse model to obtain the operation control decision; Based on the aforementioned operation control decisions, the greenhouse operation decision results are adjusted.

3. The greenhouse operation control method according to claim 2, characterized in that, The process of inputting the adjusted initial crop demand into the greenhouse model to obtain operational control decisions includes: The adjusted initial crop demand is input into the greenhouse model to obtain the predicted yield function, production consumption, and operating energy consumption. Based on the predicted output function, the production consumption, and the operational energy consumption, the relationship between control decisions and operational objectives is determined. Based on the relationship between control decisions and operational objectives, as well as the adjusted initial crop requirements, operational control decisions are determined.

4. The greenhouse operation control method according to claim 1, characterized in that, The greenhouse model includes a greenhouse environment model and a decision-making model; the step of inputting the macro-level decisions on greenhouse operation management into the greenhouse model to obtain the operation decision results output by the greenhouse model includes: The macro-level decision on greenhouse operation management is input into the greenhouse environment model and the decision model to obtain the operation control setpoint output by the greenhouse environment model and the greenhouse operation decision output by the decision model. Based on the operation control settings and the greenhouse operation decision, the operation decision result is obtained.

5. The greenhouse operation control method according to claim 4, characterized in that, The greenhouse monitoring equipment includes an inspection robot; the process of inputting the macro-level decisions for greenhouse operation management into the greenhouse model to obtain the operation decision results output by the greenhouse model further includes: Determine the inspection area of ​​the inspection robot; When the greenhouse operation decision corresponding to the inspection area is crop pollination, the crop to be pollinated and the pollination data are determined based on the operation control setting value corresponding to the inspection area. Based on the crop to be pollinated and the pollination data, a work instruction is sent to the work robot corresponding to the inspection area.

6. The greenhouse operation control method according to claim 1, characterized in that, The greenhouse operation management method also includes: Acquire textual and historical image data on greenhouse crop cultivation; Knowledge nodes are determined based on knowledge entities and image features; the knowledge entities are obtained by entity extraction from the planting knowledge text; the image features are obtained by feature extraction from the historical image data. Based on the relationships between the knowledge nodes, a knowledge graph for greenhouse crop cultivation is constructed.

7. A greenhouse operation control device, characterized in that, include: The information receiving module is used to receive crop visual information and environmental information sent by greenhouse monitoring equipment; The greenhouse operation management macro decision-making module is used to input the crop visual information and the environmental information into the greenhouse crop planting knowledge graph to obtain the greenhouse operation management macro decision. The operation decision result determination module is used to input the macro decision of greenhouse operation management into the greenhouse model and obtain the operation decision result output by the greenhouse model; the greenhouse model is trained based on historical greenhouse operation data.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the greenhouse operation control method as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the greenhouse operation control method as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the greenhouse operation control method as described in any one of claims 1 to 6.

Images