Ginkgo cultivation nutrient solution supplemental irrigation control method and system

Abstract

The application provides a ginkgo cultivation nutrient solution supplementary irrigation control method and system, and determines a target irrigation preference label corresponding to nutrient solution irrigation characteristic data in a plurality of irrigation preference labels based on irrigation frequency of the nutrient solution irrigation characteristic data, the plurality of irrigation preference labels being irrigation preference labels allocated based on irrigation frequency; resolves an irrigation scene characteristic vector from the nutrient solution irrigation characteristic data according to a target irrigation scene characteristic acquisition strategy corresponding to the target irrigation preference label in a plurality of irrigation scene characteristic acquisition strategies; determines a target supplementary irrigation strategy corresponding to the irrigation scene characteristic vector from a mapping vector library corresponding to the target irrigation preference label, and executes a ginkgo cultivation nutrient solution supplementary irrigation task based on the target supplementary irrigation strategy, the mapping vector library representing a mapping relationship between an irrigation characteristic data corresponding irrigation time domain state vector of the target irrigation preference label and the supplementary irrigation strategy, thereby improving the reliability of the supplementary irrigation control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of irrigation control, in particular to a ginkgo cultivation nutrient solution supplemental irrigation control method and system. BACKGROUND

[0002] In order to ensure the normal growth of ginkgo cultivation and obtain high and stable yield, sufficient water must be supplied to ginkgo cultivation. Under natural conditions, the water requirement of ginkgo cultivation cannot be met due to insufficient or uneven distribution of precipitation. Irrigation, i.e. watering the land. The principle of irrigation is that the irrigation amount, irrigation frequency and time should be determined according to the water requirement characteristics of medicinal plants, growth stages, climate and soil conditions, and irrigation should be timely, appropriate and reasonable. Its types mainly include pre-sowing irrigation, seedling accelerating irrigation, growth period irrigation and winter irrigation, etc. How to provide intelligent nutrient solution supplemental irrigation strategy for ginkgo cultivation and improve the reliability of supplemental irrigation control so as to facilitate the improvement of the output benefit of ginkgo cultivation and the reduction of management and operation cost is a technical problem to be solved at present. SUMMARY

[0003] Therefore, the purpose of the present application is to provide a ginkgo cultivation nutrient solution supplemental irrigation control method and system.

[0004] According to a first aspect of the present application, a ginkgo cultivation nutrient solution supplemental irrigation control method is provided, applied to a cloud server, and the method comprises:

[0005] determining a target irrigation preference label corresponding to the nutrient solution irrigation characteristic data in a plurality of irrigation preference labels based on the irrigation frequency of the nutrient solution irrigation characteristic data of the ginkgo cultivation nutrient solution supplemental irrigation task, the plurality of irrigation preference labels being irrigation preference labels distributed based on irrigation frequency;

[0006] parsing an irrigation scene feature vector from the nutrient solution irrigation characteristic data according to a target irrigation scene feature collection strategy corresponding to the target irrigation preference label in a plurality of irrigation scene feature collection strategies;

[0007] determining a target supplemental irrigation strategy corresponding to the irrigation scene feature vector from a mapping vector library corresponding to the target irrigation preference label, the mapping vector library representing the mapping relationship between the irrigation feature data corresponding to the irrigation time domain state vector and the supplemental irrigation strategy of the target irrigation preference label.

[0008] In a possible implementation of the first aspect, determining the target irrigation preference label corresponding to the nutrient solution irrigation characteristic data in a plurality of irrigation preference labels based on the irrigation frequency of the nutrient solution irrigation characteristic data comprises:

[0009] counting the irrigation frequency of the nutrient solution irrigation characteristic data;

[0010] when the irrigation frequency is not greater than a first set frequency, determining that the target irrigation preference label corresponding to the nutrient solution irrigation feature data is a first irrigation preference label;

[0011] when the irrigation frequency is greater than the first set frequency and not greater than a second set frequency, determining that the target irrigation preference label corresponding to the nutrient solution irrigation feature data is a second irrigation preference label; when the irrigation frequency is greater than the second set frequency, determining that the target irrigation preference label corresponding to the nutrient solution irrigation feature data is a third irrigation preference label.

[0012] In a possible implementation of the first aspect, when the target irrigation preference label of the nutrient solution irrigation feature data is the first irrigation preference label, the parsing of the irrigation scene feature vector from the nutrient solution irrigation feature data according to the target irrigation scene feature collection strategy corresponding to the target irrigation preference label in the plurality of irrigation scene feature collection strategies comprises:

[0013] performing irrigation node separation on the nutrient solution irrigation feature data to generate a first separated irrigation node;

[0014] obtaining a first target irrigation control vector corresponding to the first separated irrigation node from a first mapping contact, the first mapping contact being a mapping contact between each first irrigation subarea and an irrigation control vector reflecting the each first irrigation subarea, the first irrigation subarea including the first separated irrigation node;

[0015] outputting the first target irrigation control vector as the irrigation scene feature vector.

[0016] In a possible implementation of the first aspect, before determining the target supplemental irrigation strategy corresponding to the irrigation scene feature vector from the mapping vector library corresponding to the target irrigation preference label, the method further comprises:

[0017] obtaining first sample irrigation feature data;

[0018] performing clustering on the first sample irrigation feature data based on the irrigation preference clustering strategy to generate a plurality of first sample irrigation feature data clusters;

[0019] updating a supplemental irrigation decision network corresponding to each first sample irrigation feature data cluster in the plurality of first sample irrigation feature data clusters according to the each first sample irrigation feature data cluster to generate a plurality of target supplemental irrigation decision networks, the plurality of target supplemental irrigation decision networks representing a mapping contact between an irrigation feature data corresponding to each irrigation preference label and an irrigation time domain state vector and a supplemental irrigation strategy.

[0020] In a possible implementation of the first aspect, the first sample irrigation feature data clusters correspond to a plurality of irrigation preference categories of a supplementary irrigation decision network, updating the supplementary irrigation decision network corresponding to each of the first sample irrigation feature data clusters according to the first sample irrigation feature data clusters comprises:

[0021] updating the supplementary irrigation decision models of the plurality of irrigation preference categories according to the first sample irrigation feature data clusters, to generate a first supplementary irrigation decision network comprising a plurality of target supplementary irrigation decision models representing a mapping relationship between an irrigation feature data corresponding to the irrigation preference category and a supplementary irrigation strategy under the supplementary irrigation decision model of the irrigation preference category;

[0022] obtaining second sample irrigation feature data;

[0023] clustering the second sample irrigation feature data based on the irrigation preference clustering strategy to generate a plurality of second sample irrigation feature data clusters;

[0024] loading each of the second sample irrigation feature data clusters into the first supplementary irrigation decision network corresponding to the second sample irrigation feature data cluster to generate a plurality of supplementary irrigation decision data of each of the first supplementary irrigation decision network, the supplementary irrigation decision data representing a mapping relationship between the second sample irrigation feature data cluster and a support degree of a supplementary irrigation strategy, the support degree of the supplementary irrigation strategy representing a confidence degree of each sample irrigation feature data in the second sample irrigation feature data cluster belonging to each of the supplementary irrigation strategies;

[0025] updating an intermediate supplementary irrigation decision network based on the plurality of supplementary irrigation decision data to generate a second supplementary irrigation decision network, and taking the second supplementary irrigation decision network as the target supplementary irrigation decision network, the intermediate supplementary irrigation decision network representing an attention parameter of the supplementary irrigation decision data of the supplementary irrigation decision model of each irrigation preference category in the plurality of irrigation preference categories to each sample irrigation feature data belonging to each of the supplementary irrigation strategies.

[0026] In a possible implementation of the first aspect, the irrigation preference labels of the sample irrigation feature data included in the first sample irrigation feature data clusters are first irrigation preference labels, and when the irrigation frequency of the irrigation feature data of the first irrigation preference labels is not greater than a first set frequency, updating the supplementary irrigation decision network corresponding to each of the first sample irrigation feature data clusters according to the first sample irrigation feature data clusters comprises:

[0027] performing irrigation node separation on the first sample irrigation feature data cluster to generate a second separated irrigation node;

[0028] obtaining a supplementary irrigation strategy-irrigation node array and an irrigation feature data-supplementary irrigation strategy array of the first sample irrigation feature data cluster;

[0029] obtaining a second target irrigation control vector corresponding to the second separated irrigation node from a second mapping relationship, the second mapping relationship being a mapping relationship between each second irrigation subzone and an irrigation control vector reflecting the each second irrigation subzone, the second irrigation subzone including the second separated irrigation node;

[0030] converting the supplementary irrigation strategy-irrigation node array into a supplementary irrigation strategy-irrigation control vector array based on the second target irrigation control vector corresponding to the second separated irrigation node;

[0031] updating a supplementary irrigation decision network corresponding to the first sample irrigation feature data cluster according to the supplementary irrigation strategy-irrigation control vector array and the irrigation feature data-supplementary irrigation strategy array.

[0032] In a possible implementation of the first aspect, determining the target supplementary irrigation strategy corresponding to the irrigation scene feature vector from the mapping vector library corresponding to the target irrigation preference label comprises:

[0033] loading the irrigation scene feature vector into a third supplementary irrigation decision network corresponding to the target irrigation preference label in the plurality of target supplementary irrigation decision networks to generate supplementary irrigation decision data of the third supplementary irrigation decision network;

[0034] outputting the supplementary irrigation decision data of the third supplementary irrigation decision network as the target supplementary irrigation strategy.

[0035] According to a second aspect of the present application, a cloud server is provided, the cloud server comprising a machine readable storage medium and a processor, the machine readable storage medium storing machine executable instructions, and the processor, when executing the machine executable instructions, implements the ginkgo cultivation nutrient solution supplementary irrigation control method described above.

[0036] According to a third aspect of the present application, a computer readable storage medium is provided, the computer readable storage medium storing computer executable instructions, and when the computer executable instructions are executed, the ginkgo cultivation nutrient solution supplementary irrigation control method described above is implemented.

[0037] According to any of the above aspects, in the present application, the irrigation frequency based on the nutrient solution irrigation feature data is used to determine the target irrigation preference label corresponding to the nutrient solution irrigation feature data in a plurality of irrigation preference labels, and the plurality of irrigation preference labels are irrigation preference labels allocated based on the irrigation frequency; the irrigation scene feature vector is parsed from the nutrient solution irrigation feature data according to the target irrigation scene feature collection strategy corresponding to the target irrigation preference label in a plurality of irrigation scene feature collection strategies; the target supplementary irrigation strategy corresponding to the irrigation scene feature vector is determined from the mapping vector library corresponding to the target irrigation preference label, and the mapping vector library represents the mapping relationship between the irrigation time domain state vector corresponding to the irrigation feature data of the target irrigation preference label and the supplementary irrigation strategy. In other words, the irrigation feature data of different frequencies is suitable for different irrigation scene feature collection strategies and supplementary irrigation decision models, the irrigation feature data is allocated into a plurality of irrigation preference labels based on the irrigation frequency of the irrigation feature data, each irrigation preference label corresponds to an irrigation scene feature collection strategy suitable for the irrigation feature data of the irrigation preference label, and each irrigation preference label also corresponds to a mapping relationship between an irrigation time domain state vector and a supplementary irrigation strategy. When the irrigation feature data is clustered, the target irrigation preference label corresponding to the nutrient solution irrigation feature data is determined based on the irrigation frequency of the nutrient solution irrigation feature data, the irrigation scene feature vector of the nutrient solution irrigation feature data is extracted based on the target irrigation scene feature collection strategy corresponding to the target irrigation preference label, and the target supplementary irrigation strategy of the nutrient solution irrigation feature data is determined based on the target mapping relationship corresponding to the target irrigation preference label. The extraction method and the mapping relationship used when the irrigation feature data is clustered are matched with the irrigation preference category of the irrigation feature data, and the reliability of the supplementary irrigation control is improved. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor.

[0039] Figure 1 The flowchart of the ginkgo cultivation nutrient solution supplementary irrigation control method provided by the embodiments of the present application is shown.

[0040] Figure 2 The component structure schematic diagram of the cloud server for implementing the above-mentioned ginkgo cultivation nutrient solution supplementary irrigation control method provided by the embodiments of the present application is shown. DETAILED DESCRIPTION

[0041] For the purposes of making the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. It should be understood that the drawings in the present application serve merely the purpose of description and illustration, and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in the present application show the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowcharts can not be implemented in sequence, and the steps that have no logical context relationship can be reversed in sequence or implemented simultaneously. In addition, one or more other operations can be added to the flowcharts or one or more operations can be removed from the flowcharts under the guidance of the content of the present application.

[0042] In addition, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art without creative work on the basis of the embodiments of the present application belong to the scope of protection of the present application.

[0043] Figure 1 A flowchart of a ginkgo cultivation nutrient solution supplemental irrigation control method provided by an embodiment of the present application is shown. It should be understood that in other embodiments, the order of some steps of the ginkgo cultivation nutrient solution supplemental irrigation control method of the present embodiment can be shared with each other according to actual needs, or some steps can be omitted or maintained. The ginkgo cultivation nutrient solution supplemental irrigation control method includes the following steps in detail:

[0044] In step S102, a target irrigation preference label corresponding to the nutrient solution irrigation characteristic data in a plurality of irrigation preference labels is determined based on the irrigation frequency of the nutrient solution irrigation characteristic data corresponding to the ginkgo cultivation nutrient solution supplemental irrigation task, and the plurality of irrigation preference labels are irrigation preference labels allocated based on the irrigation frequency.

[0045] In the present embodiment, the ginkgo cultivation nutrient solution supplemental irrigation task is used to represent nutrient solution supplemental irrigation for ginkgo cultivation, and the nutrient solution irrigation characteristic data is used to represent characteristic data related to nutrient solution irrigation, such as environmental state data, historical automatic irrigation operation characteristics, etc. On this basis, a target irrigation preference label corresponding to the nutrient solution irrigation characteristic data in a plurality of irrigation preference labels can be determined based on the irrigation frequency of the nutrient solution irrigation characteristic data, and the target irrigation preference label can be used to represent the corresponding irrigation preference.

[0046] At step S104, the irrigation scene feature vector is parsed from the nutrient solution irrigation feature data according to the target irrigation scene feature collection strategy corresponding to the target irrigation preference label in the plurality of irrigation scene feature collection strategies.

[0047] In the embodiment, the irrigation scene feature vector can represent various environmental state parameters of the irrigation scene, such as humidity parameters, temperature parameters, wind force parameters, etc.

[0048] At step S106, the target supplemental irrigation strategy corresponding to the irrigation scene feature vector is determined from the mapping vector library corresponding to the target irrigation preference label, and the mapping vector library represents the mapping relationship between the irrigation time domain state vector and the supplemental irrigation strategy corresponding to the irrigation feature data of the target irrigation preference label.

[0049] In an alternative embodiment, the irrigation preference clustering strategy can be understood as a strategy for dividing irrigation preference labels based on irrigation frequency, and the irrigation preference clustering strategy can be a clustering strategy set based on the distribution of the irrigation frequency of the stored irrigation feature data.

[0050] In which, the irrigation feature data of different frequencies is suitable for different irrigation scene feature collection strategies and supplemental irrigation decision models, the irrigation feature data is divided into a plurality of irrigation preference labels based on the irrigation frequency of the irrigation feature data, each irrigation preference label corresponds to an irrigation scene feature collection strategy suitable for the irrigation feature data of the irrigation preference label, and each irrigation preference label also corresponds to a mapping relationship between an irrigation time domain state vector and a supplemental irrigation strategy. When clustering the irrigation feature data, the target irrigation preference label corresponding to the nutrient solution irrigation feature data is determined based on the irrigation frequency of the nutrient solution irrigation feature data, the irrigation scene feature vector of the nutrient solution irrigation feature data is extracted based on the target irrigation scene feature collection strategy corresponding to the target irrigation preference label, and the target supplemental irrigation strategy of the nutrient solution irrigation feature data is determined based on the target mapping relationship corresponding to the target irrigation preference label. This can make the extraction method and the mapping relationship used when clustering the irrigation feature data both match the irrigation preference category of the irrigation feature data, thereby improving the reliability of the supplemental irrigation control.

[0051] In an alternative embodiment, determining the target irrigation preference label corresponding to the nutrient solution irrigation feature data in the plurality of irrigation preference labels based on the irrigation frequency of the nutrient solution irrigation feature data comprises:

[0052] S1, statistics the irrigation frequency of the nutrient solution irrigation feature data;

[0053] S2, when the irrigation frequency is not greater than a first set frequency, determining the target irrigation preference label corresponding to the nutrient solution irrigation feature data as a first irrigation preference label;

[0054] Step S1, when the irrigation frequency is greater than the first set frequency and not greater than the second set frequency, determining that the target irrigation preference label corresponding to the nutrient solution irrigation feature data is the second irrigation preference label;

[0055] S4, when the irrigation frequency is greater than the second set frequency, determining that the target irrigation preference label corresponding to the nutrient solution irrigation feature data is the third irrigation preference label.

[0056] In an alternative embodiment, when the target irrigation preference label of the nutrient solution irrigation feature data is the first irrigation preference label, the irrigation scene feature vector is parsed from the nutrient solution irrigation feature data according to the target irrigation scene feature collection strategy corresponding to the target irrigation preference label in the plurality of irrigation scene feature collection strategies, comprising:

[0057] S1, separating the irrigation nodes of the nutrient solution irrigation feature data to generate a first separated irrigation node;

[0058] S2, obtaining the first target irrigation control vector corresponding to the first separated irrigation node from the first mapping contact, the first mapping contact being the mapping contact between each first irrigation partition and the irrigation control vector reflecting each first irrigation partition, and the first irrigation partition including the first separated irrigation node;

[0059] Step S1, outputting the first target irrigation control vector as an irrigation scene feature vector.

[0060] In an alternative embodiment, before determining the target supplementary irrigation strategy corresponding to the irrigation scene feature vector from the mapping vector library corresponding to the target irrigation preference label, further comprising:

[0061] S1, obtaining first sample irrigation feature data;

[0062] S2, clustering the first sample irrigation feature data based on the irrigation preference clustering strategy to generate a plurality of first sample irrigation feature data clusters;

[0063] Step S1, updating the supplementary irrigation decision network corresponding to each first sample irrigation feature data cluster through each first sample irrigation feature data cluster in the plurality of first sample irrigation feature data clusters to generate a plurality of target supplementary irrigation decision networks, the plurality of target supplementary irrigation decision networks representing the mapping contact between the irrigation feature data corresponding to each irrigation preference label and the irrigation time domain state vector and the supplementary irrigation strategy.

[0064] In an alternative embodiment, the same irrigation preference clustering strategy is used for clustering the first sample irrigation feature data as for clustering the irrigation feature data to be clustered, ensuring that the sample irrigation feature data and the irrigation feature data are clustered at the same frequency, thereby making the reliability of determining the supplementary irrigation strategy of the irrigation feature data higher.

[0065] In an alternative embodiment, the mapping relationship between the irrigation time domain state vector and the supplemental irrigation strategy can be determined, but not limited to, by updating the supplemental irrigation decision network using sample irrigation feature data. Each irrigation preference label corresponding to an irrigation frequency division corresponds to a supplemental irrigation decision network. The obtained first sample irrigation feature data is divided into sample irrigation feature data of multiple irrigation preference labels based on irrigation frequency division, and the sample irrigation feature data of each irrigation preference category is used to train the supplemental irrigation decision network corresponding to the irrigation preference category.

[0066] In an alternative embodiment, the first sample irrigation feature data is divided into sample irrigation feature data of three different irrigation preference categories: sample irrigation feature data A corresponding to the first irrigation preference label, sample irrigation feature data B corresponding to the second irrigation preference label, and sample irrigation feature data C corresponding to the third irrigation preference label. The supplemental irrigation decision networks corresponding to the sample irrigation feature data of the three different irrigation preference categories are respectively: supplemental irrigation decision network A, supplemental irrigation decision network B, and supplemental irrigation decision network C. The supplemental irrigation decision network A is updated using the sample irrigation feature data A to generate a target supplemental irrigation decision network A, the supplemental irrigation decision network B is updated using the sample irrigation feature data B to generate a target supplemental irrigation decision network B, and the supplemental irrigation decision network C is updated using the sample irrigation feature data C to generate a target supplemental irrigation decision network C. The target supplemental irrigation decision network A indicates the mapping relationship between the irrigation feature data corresponding to the first irrigation preference label and the irrigation time domain state vector and the supplemental irrigation strategy, and the target supplemental irrigation decision network A can be used to cluster the irrigation feature data with the first irrigation preference label. The target supplemental irrigation decision network B indicates the mapping relationship between the irrigation feature data corresponding to the second irrigation preference label and the irrigation time domain state vector and the supplemental irrigation strategy, and the target supplemental irrigation decision network B can be used to cluster the irrigation feature data with the second irrigation preference label. The target supplemental irrigation decision network C indicates the mapping relationship between the irrigation feature data corresponding to the third irrigation preference label and the irrigation time domain state vector and the supplemental irrigation strategy, and the target supplemental irrigation decision network C can be used to cluster the irrigation feature data with the third irrigation preference label.

[0067] In an alternative embodiment, the supplemental irrigation decision network corresponding to each first sample irrigation feature data cluster can include supplemental irrigation decision models of multiple irrigation preference categories, and the target supplemental irrigation decision network is generated by updating the supplemental irrigation decision network corresponding to each first sample irrigation feature data cluster using each first sample irrigation feature data cluster.

[0068] S1, updating the supplementary irrigation decision model of the plurality of irrigation preference categories by each first sample irrigation feature data cluster to generate a first supplementary irrigation decision network comprising a plurality of target supplementary irrigation decision models, the plurality of target supplementary irrigation decision models representing the mapping relationship between the irrigation feature data corresponding to the irrigation time domain state vector and the supplementary irrigation strategy under the supplementary irrigation decision model of the irrigation preference category;

[0069] S2, obtaining second sample irrigation feature data;

[0070] Step S1, clustering the second sample irrigation feature data based on the irrigation preference clustering strategy to generate a plurality of second sample irrigation feature data clusters;

[0071] S4, inputting each second sample irrigation feature data cluster in the plurality of second sample irrigation feature data clusters into the first supplementary irrigation decision network corresponding to the second sample irrigation feature data cluster to generate a plurality of supplementary irrigation decision data of each first supplementary irrigation decision network, the supplementary irrigation decision data representing the mapping relationship between the second sample irrigation feature data cluster and the support degree of the supplementary irrigation strategy, the support degree of the supplementary irrigation strategy representing the confidence degree of each sample irrigation feature data in the second sample irrigation feature data cluster belonging to each supplementary irrigation strategy in all supplementary irrigation strategies;

[0072] S5, updating the intermediate supplementary irrigation decision network based on the plurality of supplementary irrigation decision data to generate a second supplementary irrigation decision network, and taking the second supplementary irrigation decision network as the target supplementary irrigation decision network, the intermediate supplementary irrigation decision network representing the attention parameter of the supplementary irrigation decision data of each irrigation preference category in the supplementary irrigation decision model of the irrigation preference category in the plurality of irrigation preference category supplementary irrigation decision models to each sample irrigation feature data belonging to each supplementary irrigation strategy in all supplementary irrigation strategies.

[0073] In an alternative embodiment, the same clustering strategy is used when clustering the irrigation feature data to be clustered, the first sample irrigation feature data and the second sample irrigation feature data, ensuring that the clustering frequency based on the irrigation frequency clustering is the same, thereby making the reliability of determining the supplementary irrigation strategy of the irrigation feature data higher.

[0074] In an alternative implementation, the supplementary irrigation decision network corresponding to each first sample irrigation feature data cluster comprises a plurality of supplementary irrigation decision models of irrigation preference categories. The plurality of supplementary irrigation decision models of irrigation preference categories can be Naive Bayes models. In determining the target supplementary irrigation decision network corresponding to each irrigation preference label, the plurality of supplementary irrigation decision models of irrigation preference categories in the supplementary irrigation decision network corresponding to the first sample irrigation feature data cluster can be first updated using the first sample irrigation feature data cluster split from the first sample irrigation feature data, to generate a first supplementary irrigation decision network. Then, second sample irrigation feature data is obtained, and divided based on the same standard to generate a second sample irrigation feature data cluster. The second sample irrigation feature data cluster is loaded into the corresponding first supplementary irrigation decision network to generate corresponding supplementary irrigation decision data. An intermediate supplementary irrigation decision network is updated using the supplementary irrigation decision data corresponding to each irrigation preference label to generate a second supplementary irrigation decision network, which is used as the target supplementary irrigation decision network corresponding to the irrigation preference label. The intermediate supplementary irrigation decision network can represent the attention parameters of the supplementary irrigation decision data of the supplementary irrigation decision models of each irrigation preference category under each irrigation preference label to each sample irrigation feature data belonging to each of the supplementary irrigation strategies. In other words, the intermediate supplementary irrigation decision network indicates the influence degree of the supplementary irrigation decision models of each irrigation preference category to the supplementary irrigation strategies of the irrigation feature data.

[0075] In an alternative embodiment, the total sample irrigation feature data is divided into first sample irrigation feature data and second sample irrigation feature data, the first sample irrigation feature data is divided into sample irrigation feature data of three different irrigation preference categories based on irrigation frequency: sample irrigation feature data A corresponding to the first irrigation preference label, sample irrigation feature data B corresponding to the second irrigation preference label, and sample irrigation feature data C corresponding to the third irrigation preference label, and the sample irrigation feature data of the three different irrigation preference categories correspond to the supplementary irrigation decision network respectively: supplementary irrigation decision network A, supplementary irrigation decision network B, and supplementary irrigation decision network C. Among them, the supplementary irrigation decision network A includes supplementary irrigation decision model A1, supplementary irrigation decision model A2, supplementary irrigation decision model A3, and supplementary irrigation decision model A4; the supplementary irrigation decision network B includes supplementary irrigation decision model B1, supplementary irrigation decision model B2, supplementary irrigation decision model B3, and supplementary irrigation decision model B4; the supplementary irrigation decision network C includes supplementary irrigation decision model C1, supplementary irrigation decision model C2, supplementary irrigation decision model C3, and supplementary irrigation decision model C4; the supplementary irrigation decision model in the supplementary irrigation decision network A is updated using the sample irrigation feature data A to generate the first supplementary irrigation decision network A, the supplementary irrigation decision model in the supplementary irrigation decision network B is updated using the sample irrigation feature data B to generate the first supplementary irrigation decision network B, and the supplementary irrigation decision model in the supplementary irrigation decision network C is updated using the sample irrigation feature data C to generate the first supplementary irrigation decision network C. The second sample irrigation feature data is divided into sample irrigation feature data of three different irrigation preference categories based on irrigation frequency: sample irrigation feature data M corresponding to the first irrigation preference label, sample irrigation feature data N corresponding to the second irrigation preference label, and sample irrigation feature data P corresponding to the third irrigation preference label. Load the sample irrigation feature data M to the first supplementary irrigation decision network A to generate supplementary irrigation decision data A, load the sample irrigation feature data N to the first supplementary irrigation decision network B to generate supplementary irrigation decision data B, and load the sample irrigation feature data P to the first supplementary irrigation decision network C to generate supplementary irrigation decision data C. The intermediate supplementary irrigation decision network A is trained using the supplementary irrigation decision data A to generate the target supplementary irrigation decision network A, the intermediate supplementary irrigation decision network B is trained using the supplementary irrigation decision data B to generate the target supplementary irrigation decision network B, and the intermediate supplementary irrigation decision network C is trained using the supplementary irrigation decision data C to generate the target supplementary irrigation decision network C.

[0076] In an alternative embodiment, the irrigation preference label of the sample irrigation feature data included in the first sample irrigation feature data cluster is a first irrigation preference label, and when the irrigation frequency of the irrigation feature data of the first irrigation preference label is not greater than a first set frequency, the updating of each first sample irrigation feature data cluster by each first sample irrigation feature data cluster updates the corresponding supplementary irrigation decision network of each first sample irrigation feature data cluster to include:

[0077] S1, separating the irrigation nodes of the first sample irrigation feature data cluster to generate second separated irrigation nodes;

[0078] S2, obtaining the supplementary irrigation strategy-irrigation node array and the irrigation feature data-supplementary irrigation strategy array of the first sample irrigation feature data cluster;

[0079] Step S1, obtaining the second target irrigation control vector corresponding to the second separated irrigation node from the second mapping contact, the second mapping contact being a mapping contact between each second irrigation partition and an irrigation control vector reflecting each second irrigation partition, and the second irrigation partition including the second separated irrigation node;

[0080] S4, converting the supplementary irrigation strategy-irrigation node array into a supplementary irrigation strategy-irrigation control vector array based on the second target irrigation control vector corresponding to the second separated irrigation node;

[0081] S5, updating the supplementary irrigation decision network corresponding to the first sample irrigation feature data cluster by the supplementary irrigation strategy-irrigation control vector array and the irrigation feature data-supplementary irrigation strategy array.

[0082] In an alternative embodiment, the supplementary irrigation strategy-irrigation node array can represent the distribution relationship between the supplementary irrigation strategy and the second separated irrigation node, and the irrigation feature data-supplementary irrigation strategy array can represent the distribution relationship between the first sample irrigation feature data cluster and the supplementary irrigation strategy.

[0083] In an alternative embodiment, for irrigation feature data with a low irrigation frequency, the words parsed from the sample irrigation feature data when updating the supplementary irrigation decision network can be converted into irrigation control vectors.

[0084] In an alternative embodiment, determining the target supplementary irrigation strategy corresponding to the irrigation scene feature vector from the mapping vector library corresponding to the target irrigation preference label includes:

[0085] S1, inputting the irrigation scene feature vector into the third supplementary irrigation decision network corresponding to the target irrigation preference label in the plurality of target supplementary irrigation decision networks to generate supplementary irrigation decision data of the third supplementary irrigation decision network;

[0086] S2, output the supplemental irrigation decision data of the third supplemental irrigation decision network as a target supplemental irrigation strategy.

[0087] In an alternative implementation, each target supplemental irrigation decision network represents a mapping relationship between irrigation feature data features of a corresponding irrigation preference label and a supplemental irrigation strategy. Based on the target irrigation preference label corresponding to the nutrient solution irrigation feature data, the target supplemental irrigation decision network corresponding to the target irrigation preference label is determined, that is, the third supplemental irrigation decision network is determined. The irrigation scene feature vector parsed from the nutrient solution irrigation feature data is loaded into the third supplemental irrigation decision network, and the supplemental irrigation decision data of the third supplemental irrigation decision network is generated as the target supplemental irrigation strategy of the nutrient solution irrigation feature data.

[0088] Figure 2 A cloud server 100 that can be used to implement various embodiments described in the present application is schematically shown.

[0089] For one embodiment, Figure 2 The cloud server 100 is shown, which has one or more processors 102, a control module (chipset) 104 coupled to one or more of the processor(s) 102, a memory 106 coupled to the control module 104, a non-volatile memory (NVM) / storage device 108 coupled to the control module 104, one or more input / output devices 110 coupled to the control module 104, and a network interface 112 coupled to the control module 106.

[0090] The processor 102 can include one or more single-core or multi-core processors, which can include any combination of general-purpose processors or special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In some example design ideas, the cloud server 100 can be able to serve as a server device such as a gateway described in the embodiments of the present application.

[0091] In some example design ideas, the cloud server 100 can include one or more computer-readable media (e.g., the memory 106 or the NVM / storage device 108) having instructions 114 and one or more processors 102 integrated with the one or more computer-readable media configured to execute the instructions 114 to implement modules to perform the actions described in the present disclosure.

[0092] For one embodiment, the control module 104 can include any suitable interface controllers to provide any suitable interface to one or more of the processor(s) 102 and / or any suitable device or component in communication with the control module 104.

[0093] The control module 104 can include a memory controller module to provide an interface to the memory 106. The memory controller module can be a hardware module, a software module, and / or a firmware module.

[0094] The memory 106 can be used to load and store data and / or instructions 114, for example, for the cloud server 100. For one embodiment, the memory 106 can include any suitable volatile memory, for example, suitable DRAM. In some example design considerations, the memory 106 can include Double Data Rate Type Four Synchronous Dynamic Random Access Memory (DDR4 SDRAM).

[0095] For one embodiment, the control module 104 can include one or more input / output controllers to provide an interface to the NVM / storage device 108 and the input / output device(s) 110.

[0096] The NVM / storage device 108 can be used to store data and / or instructions 114, for example. The NVM / storage device 108 can include any suitable non-volatile memory (e.g., flash memory) and / or can include any suitable non-volatile storage device(s) (e.g., one or more hard disk drives (HDDs), one or more compact disk (CD) drives, and / or one or more digital versatile disk (DVD) drives).

[0097] The NVM / storage device 108 can include storage resources that are physically part of the device on which the cloud server 100 is installed, or it can be accessed by the device remotely and / or over a network. For example, the NVM / storage device 108 can be accessed by the cloud server 100 via the input / output device(s) 110 over a network.

[0098] The input / output device(s) 110 can provide an interface for the cloud server 100 to communicate with any other suitable device(s), and the input / output device(s) 110 can include communication components, pinyin components, sensor components, etc. The network interface 112 can provide an interface for the cloud server 100 to communicate over one or more networks, and the cloud server 100 can wirelessly communicate with one or more components of a wireless network according to any of one or more wireless network standards and / or protocols, for example, to access a wireless network according to a communication standard, such as 2G, 3G, 4G, 5G, etc., or combinations thereof.

[0099] For one embodiment, one or more of the processor(s) 102 can be loaded with logic of one or more controllers of the control module 104 (e.g., a memory controller module). For one embodiment, one or more of the processor(s) 102 can be loaded with logic of one or more controllers of the control module 104 to form a system on a package (SwP). For one embodiment, one or more of the processor(s) 102 can be integrated on the same die as logic of one or more controllers of the control module 104. For one embodiment, one or more of the processor(s) 102 can be integrated on the same die as logic of one or more controllers of the control module 104 to form a system on a chip (SoC).

[0100] In various embodiments, cloud server 100 can be, but is not limited to, a cloud server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet, a netbook, etc.), and the like. In various embodiments, cloud server 100 can have more or less components, and / or different architectures. For example, in some example design considerations, cloud server 100 includes one or more cameras, a keyboard, a liquid crystal display (LCD) screen (including touch screen displays), non- volatile memory port, multiple antennas, a graphics chip, an application specific integrated circuit (ASIC), and a speaker.

[0101] The above describes the embodiments of the present application in detail, and the principles and implementation modes of the present application are described by applying specific examples. The above description of the embodiments is only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range can be changed, and the above description of the present application should not be understood as a limitation.

Claims

1. A method for controlling the supplemental irrigation of a ginkgo cultivation nutrient solution, characterized by, The method applied to a cloud server comprises: determining a target irrigation preference label corresponding to the nutrient solution irrigation characteristic data in a plurality of irrigation preference labels based on an irrigation frequency of a ginkgo cultivation nutrient solution supplemental irrigation task, the plurality of irrigation preference labels being irrigation preference labels allocated based on irrigation frequency; parsing an irrigation scene feature vector from the nutrient solution irrigation characteristic data according to a target irrigation scene feature collection strategy corresponding to the target irrigation preference label in a plurality of irrigation scene feature collection strategies; determining a target supplemental irrigation strategy corresponding to the irrigation scene feature vector from a mapping vector library corresponding to the target irrigation preference label, and performing a ginkgo cultivation nutrient solution supplemental irrigation task based on the target supplemental irrigation strategy, the mapping vector library representing a mapping relationship between an irrigation feature data corresponding to the target irrigation preference label and a supplemental irrigation strategy; before determining the target supplemental irrigation strategy corresponding to the irrigation scene feature vector from the mapping vector library corresponding to the target irrigation preference label, the method further comprises: obtaining first sample irrigation feature data; clustering the first sample irrigation feature data based on an irrigation preference clustering strategy to generate a plurality of first sample irrigation feature data clusters; updating a plurality of target supplemental irrigation decision networks according to each first sample irrigation feature data cluster in the plurality of first sample irrigation feature data clusters, the plurality of target supplemental irrigation decision networks representing a mapping relationship between an irrigation feature data corresponding to each irrigation preference label and a supplemental irrigation strategy.

2. The ginkgo cultivation nutrient solution supplemental irrigation control method according to claim 1, characterized by, determining the target irrigation preference label corresponding to the nutrient solution irrigation characteristic data in a plurality of irrigation preference labels based on an irrigation frequency of the nutrient solution irrigation characteristic data comprises: counting the irrigation frequency of the nutrient solution irrigation characteristic data; when the irrigation frequency is not greater than a first set frequency, determining the target irrigation preference label corresponding to the nutrient solution irrigation characteristic data as a first irrigation preference label; when the irrigation frequency is greater than the first set frequency and not greater than a second set frequency, determining the target irrigation preference label corresponding to the nutrient solution irrigation characteristic data as a second irrigation preference label; when the irrigation frequency is greater than the second set frequency, determining the target irrigation preference label corresponding to the nutrient solution irrigation characteristic data as a third irrigation preference label.

3. The ginkgo cultivation nutrient solution supplemental irrigation control method according to claim 2, characterized by, when the target irrigation preference label of the nutrient solution irrigation characteristic data is the first irrigation preference label, parsing the irrigation scene feature vector from the nutrient solution irrigation characteristic data according to the target irrigation scene feature collection strategy corresponding to the target irrigation preference label in the plurality of irrigation scene feature collection strategies comprises: performing irrigation node separation on the nutrient solution irrigation characteristic data to generate a first separated irrigation node; obtaining a first target irrigation control vector corresponding to the first separated irrigation node from a first mapping relation, the first mapping relation being a mapping relation between each first irrigation subarea and an irrigation control vector reflecting the each first irrigation subarea, the first irrigation subarea including the first separated irrigation node; outputting the first target irrigation control vector as the irrigation scene feature vector.

4. The ginkgo cultivation nutrient solution supplemental irrigation control method according to claim 1, characterized by, The supplementary irrigation decision network corresponding to each first sample irrigation feature data cluster includes a supplementary irrigation decision model of multiple irrigation preference categories, and updating the supplementary irrigation decision network corresponding to each first sample irrigation feature data cluster according to the each first sample irrigation feature data cluster to generate the target supplementary irrigation decision network includes: updating the supplementary irrigation decision model of the multiple irrigation preference categories according to the each first sample irrigation feature data cluster to generate a first supplementary irrigation decision network including multiple target supplementary irrigation decision models, the multiple target supplementary irrigation decision models representing a mapping relation between an irrigation time domain state vector corresponding to the irrigation feature data and a supplementary irrigation strategy under the supplementary irrigation decision model of the irrigation preference category; obtaining second sample irrigation feature data; clustering the second sample irrigation feature data based on the irrigation preference clustering strategy to generate multiple second sample irrigation feature data clusters; loading each second sample irrigation feature data cluster in the multiple second sample irrigation feature data clusters into the first supplementary irrigation decision network corresponding to the second sample irrigation feature data cluster to generate multiple supplementary irrigation decision data of each first supplementary irrigation decision network, the supplementary irrigation decision data representing a mapping relation between the each second sample irrigation feature data cluster and a support degree of a supplementary irrigation strategy, the support degree of the supplementary irrigation strategy representing a confidence degree of each sample irrigation feature data in the each second sample irrigation feature data cluster belonging to each supplementary irrigation strategy in all supplementary irrigation strategies; updating an intermediate supplementary irrigation decision network based on the multiple supplementary irrigation decision data to generate a second supplementary irrigation decision network, and taking the second supplementary irrigation decision network as the target supplementary irrigation decision network, the intermediate supplementary irrigation decision network representing an attention parameter of the supplementary irrigation decision data of each irrigation preference category in the supplementary irrigation decision model of the irrigation preference category in the multiple irrigation preference category supplementary irrigation decision models to each sample irrigation feature data belonging to each supplementary irrigation strategy in all supplementary irrigation strategies.

5. The ginkgo cultivation nutrient solution supplemental irrigation control method according to claim 1, characterized by, The irrigation preference label of the sample irrigation feature data included in the first sample irrigation feature data cluster is a first irrigation preference label, and when the irrigation frequency of the irrigation feature data of the first irrigation preference label is not greater than a first set frequency, updating the supplementary irrigation decision network corresponding to each first sample irrigation feature data cluster according to the each first sample irrigation feature data cluster includes: generating a second separated irrigation node by separating the first sample irrigation feature data cluster into irrigation nodes; obtaining a supplementary irrigation strategy-irrigation node array and an irrigation feature data-supplementary irrigation strategy array of the first sample irrigation feature data cluster; obtaining a second target irrigation control vector corresponding to the second separate irrigation node from a second mapping relation, the second mapping relation being a mapping relation between each second irrigation subzone and an irrigation control vector reflecting the each second irrigation subzone, the second irrigation subzone including the second separate irrigation node; translating the supplementary irrigation strategy-irrigation node array into a supplementary irrigation strategy-irrigation control vector array based on the second target irrigation control vector corresponding to the second separate irrigation node; updating a supplementary irrigation decision network corresponding to the first sample irrigation feature data cluster according to the supplementary irrigation strategy-irrigation control vector array and the irrigation feature data-supplementary irrigation strategy array.

6. The ginkgo cultivation nutrient solution supplemental irrigation control method according to claim 1, characterized by, determining the target supplementary irrigation strategy corresponding to the irrigation scene feature vector from a mapping vector library corresponding to the target irrigation preference label, the mapping vector library representing a mapping relation between an irrigation feature data corresponding to the target irrigation preference label and an irrigation time domain state vector and a supplementary irrigation strategy; loading the irrigation scene feature vector into a third supplementary irrigation decision network corresponding to the target irrigation preference label in the plurality of target supplementary irrigation decision networks, generating supplementary irrigation decision data of the third supplementary irrigation decision network; outputting the supplementary irrigation decision data of the third supplementary irrigation decision network as the target supplementary irrigation strategy.

7. A computer-readable storage medium, characterized in that, A computer readable storage medium storing machine executable instructions, which when executed by a processor, implement the ginkgo cultivation nutrient solution supplementary irrigation control method of any one of claims 1-6.

8. A cloud server, characterized by, A computer readable storage medium storing machine executable instructions, which when executed by a processor, implement the ginkgo cultivation nutrient solution supplementary irrigation control method of any one of claims 1-6.

9. A ginkgo cultivation nutrient solution supplemental irrigation control system, characterized by, The ginkgo cultivation nutrient solution supplementary irrigation control system comprises a cloud server and an irrigation control device in communication connection with the cloud server, and the cloud server is specifically used for: determining a target irrigation preference label corresponding to the nutrient solution irrigation feature data in a plurality of irrigation preference labels based on an irrigation frequency of the nutrient solution irrigation feature data corresponding to a ginkgo cultivation nutrient solution supplementary irrigation task, the plurality of irrigation preference labels being irrigation preference labels allocated based on irrigation frequency; parsing an irrigation scene feature vector from the nutrient solution irrigation feature data according to a target irrigation scene feature acquisition strategy corresponding to the target irrigation preference label in a plurality of irrigation scene feature acquisition strategies; determining a target supplementary irrigation strategy corresponding to the irrigation scene feature vector from a mapping vector library corresponding to the target irrigation preference label, the mapping vector library representing a mapping relation between an irrigation feature data corresponding to the target irrigation preference label and an irrigation time domain state vector and a supplementary irrigation strategy, and performing a ginkgo cultivation nutrient solution supplementary irrigation task based on the target supplementary irrigation strategy; before determining the target supplementary irrigation strategy corresponding to the irrigation scene feature vector from the mapping vector library corresponding to the target irrigation preference label, the cloud server is further used for: obtaining first sample irrigation feature data; clustering the first sample irrigation feature data based on an irrigation preference clustering strategy, generating a plurality of first sample irrigation feature data clusters; According to each first sample irrigation feature data cluster in the plurality of first sample irrigation feature data clusters, update the corresponding supplementary irrigation decision network of each first sample irrigation feature data cluster, generate a plurality of target supplementary irrigation decision networks, and the plurality of target supplementary irrigation decision networks represent the mapping relationship between the irrigation feature data corresponding to the irrigation preference label and the supplementary irrigation strategy.

Images