Optimization method, device and computer-readable storage medium for highland barley irrigation and fertilization

Abstract

This application discloses a method, device, and computer-readable storage medium for optimizing highland barley fertilization. The method comprises: collecting environmental information of a highland barley planting area within a preset period; determining probability values ​​for various highland barley yield intervals based on the environmental information using a preset yield prediction model; generating corresponding correlation information based on the environmental information and the probability values, wherein the correlation information indicates a correlation between the environmental information and highland barley yield; and optimizing the ratio of water and fertilizer components used in highland barley fertilization based on the correlation information using a preset ratio model.

Description

Technical Field

[0001] The present application relates to the technical field of crop cultivation, and in particular to an optimization method, device and computer-readable storage medium for highland barley irrigation and fertilization. Background Art

[0002] Highland barley, a member of the genus Hordeum in the Poaceae family, is an ancient cultivated crop. In my country, it is one of the most important specialty crops in Tibet, cultivated in highland areas at altitudes between 2,400 and 4,500 meters. It provides the staple food for 50% of the Tibetan population. Highland barley boasts a nutritional profile characterized by "three highs and two lows" (high protein, high soluble fiber, and high vitamins, and low fat and sugar), along with a rich mineral content, making it a premium cereal crop. Furthermore, highland barley straw serves as high-quality forage, and its grains are a key raw material for the brewing and food processing industries. In recent years, as the value of highland barley has been gradually discovered, its products have gained widespread consumer favor, and demand for the crop has increased annually. Currently, water and fertilizer management technologies for high-quality, high-yield, and environmentally friendly cultivation have been established for crops such as rice, wheat, and corn, tailored to their specific growing regions. Although highland barley has a long history of cultivation, its production has been limited by both natural and human factors. There are still problems such as the lack of scientific application of chemical fertilizers and the failure to apply fertilizers reasonably according to climatic conditions and environmental factors. This not only increases production costs, but also results in low barley yield, poor quality, and low fertilizer utilization rate, which seriously restricts barley production.

[0003] With respect to the technical problems in the above-mentioned prior art, such as the incompatibility between climatic conditions and environmental factors and fertilizers, which leads to high production costs, low barley yields and low fertilizer utilization rates, no effective solution has been proposed so far. Summary of the Invention

[0004] The embodiments of the present application provide a method, device and computer-readable storage medium for optimizing highland barley irrigation and fertilization, so as to at least solve the technical problems existing in the prior art in that factors such as climatic conditions and environment are not coordinated with fertilizers, resulting in high production costs, low highland barley yields and low fertilizer utilization.

[0005] According to one aspect of an embodiment of the present application, a method for optimizing highland barley irrigation and fertilization is provided, comprising: collecting environmental information of a highland barley planting area within a preset period; determining the probability value of each highland barley yield interval based on the environmental information through a preset yield prediction model; generating corresponding association information based on the environmental information and the probability value, wherein the association information is used to indicate the association relationship between the environmental information and the highland barley yield; and optimizing the component ratios of water and fertilizer components used for highland barley irrigation and fertilization based on the association information through a preset ratio model, wherein the operation of determining the probability value of each highland barley yield interval based on the environmental information through the preset yield prediction model comprises: generating corresponding first feature information based on the environmental information through an RNN model of the yield prediction model; generating corresponding second feature information based on the first feature information through the first fully connected layer of the yield prediction model; and determining the probability value of each highland barley yield interval based on the second feature information through the first classifier of the yield prediction model.

[0006] According to another aspect of an embodiment of the present application, an optimization device for highland barley irrigation and fertilization is also provided, including: an information collection module for collecting environmental information of a highland barley planting area within a preset period; a probability determination module for determining the probability value of each highland barley yield interval based on the environmental information through a preset yield prediction model; an information generation module for generating corresponding association information based on the environmental information and the probability value, wherein the association information is used to indicate the association relationship between the environmental information and the highland barley yield; and a proportion optimization module for optimizing the component ratio of water and fertilizer components used for highland barley irrigation and fertilization based on the association information through a preset ratio model, wherein the probability determination module includes: a first generation submodule for generating corresponding first feature information based on the environmental information through the RNN model of the yield prediction model; a second generation submodule for generating corresponding second feature information based on the first feature information through the first fully connected layer of the yield prediction model; and a first determination submodule for determining the probability value of each highland barley yield interval based on the second feature information through the first classifier of the yield prediction model.

[0007] According to another aspect of an embodiment of the present application, a device for optimizing highland barley irrigation and fertilization is also provided, comprising: a processor; and a memory connected to the processor, for providing the processor with instructions for processing the following processing steps: collecting environmental information of the highland barley planting area within a preset period; determining the probability value of each highland barley yield interval based on the environmental information through a preset yield prediction model; generating corresponding association information based on the environmental information and the probability value, wherein the association information is used to indicate the association relationship between the environmental information and the highland barley yield; and optimizing the component ratio of water and fertilizer components used for highland barley irrigation and fertilization based on the association information through a preset ratio model, wherein the operation of determining the probability value of each highland barley yield interval based on the environmental information through the preset yield prediction model includes: generating corresponding first feature information based on the environmental information through the RNN model of the yield prediction model; generating corresponding second feature information based on the first feature information through the first fully connected layer of the yield prediction model; and determining the probability value of each highland barley yield interval based on the second feature information through the first classifier of the yield prediction model.

[0008] According to another aspect of an embodiment of the present application, a computer system is further provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0009] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0010] In an embodiment of the present application, a computing device uses environmental information of a barley planting area suitable for a period of barley planting as an influencing factor affecting barley yield, and uses a yield prediction model to predict the probability value of each barley yield in the barley planting area based on the environmental information. Thus, the technical solution determines the influence relationship of environmental information on barley yield, thereby accurately predicting barley yield. Furthermore, the technical solution optimizes the component ratio of water and fertilizer based on the generated associated information by associating environmental information with the corresponding barley yield probability value, thereby avoiding the problems of increased production costs, low barley yield, poor quality, and low fertilizer utilization rate caused by the failure to reasonably apply fertilizer according to factors such as climatic conditions and the environment. Thus, a reasonable component ratio is determined under the influence of environmental and yield factors, so that the environmental information, yield, and fertilizer are more coordinated, and the technical effect of stronger adaptability is achieved. This solves the technical problem in the prior art that factors such as climatic conditions and the environment are not coordinated with fertilizers, resulting in high production costs, low barley yield, and low fertilizer utilization rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0012] Figure 1 is a hardware structure block diagram of a computing device for implementing the method according to embodiment 1 of the present application;

[0013] Figure 2 1 is a flow chart of the optimization method for highland barley fertilization according to the first aspect of Example 1 of the present application;

[0014] Figure 3 1 is a schematic diagram of the overall process of the optimization method for highland barley fertilization according to Example 1 of the present application;

[0015] Figure 4 is a schematic diagram of the yield prediction model according to Example 1 of the present application;

[0016] Figure 5 This is a flow chart of associating environmental information with probability values ​​according to Example 1 of the present application;

[0017] Figure 6 is a schematic diagram of a proportioning model according to Example 1 of the present application;

[0018] Figure 7 1 is a sequential flow diagram of the optimization method for highland barley fertilization according to Example 1 of the present application;

[0019] Figure 8 is a schematic diagram of the optimized device for highland barley fertilization according to Example 2 of the present application; and

[0020] Figure 9 This is a schematic diagram of the optimized device for highland barley irrigation and fertilization according to Example 3 of the present application. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the technical solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of this application.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following interpretations:

[0024] Fertigation: It is the process of dissolving fertilizer in irrigation water and simultaneously irrigating and fertilizing through the irrigation system.

[0025] Example 1

[0026] According to this embodiment, a method embodiment of an optimization method for highland barley irrigation and fertilization is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0027] The method embodiment provided in this embodiment can be executed in a mobile terminal, a computer terminal, a server or a similar computing device. Figure 1 The hardware structure block diagram of a computing device for implementing an optimization method for highland barley irrigation and fertilization is shown. Figure 1 As shown, the computing device may include one or more processors (the processor may include but is not limited to a microprocessor MCU or a programmable logic device FPGA, etc.), a memory for storing data, and a transmission device for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0028] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry." The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single, independent processing module, or may be incorporated in whole or in part into any of the other components of the computing device. As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0029] The memory can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the optimization method for highland barley fertilization in the embodiment of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the optimization method for highland barley fertilization of the above-mentioned application. The memory may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computing device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0030] The transmission device is used to receive or send data via a network. Specific examples of the aforementioned network may include a wireless network provided by a communications provider of the computing device. In one embodiment, the transmission device includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission device may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0031] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computing device.

[0032] It should be noted that, in some optional embodiments, the above Figure 1 The computing device shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. Figure 1 This is merely one example of a particular embodiment and is intended to illustrate the types of components that may be present in the computing device described above.

[0033] Under the above operating environment, according to the first aspect of this embodiment, a method for optimizing highland barley fertilization is provided, which comprises: Figure 1 The computing device implementation shown in . Figure 2 A schematic diagram showing the process of the method is shown in FIG. Figure 2 As shown, the method includes:

[0034] S202: Collecting environmental information of the highland barley planting area within a preset period;

[0035] S204: Determine the probability value of each highland barley yield range based on the environmental information using a preset yield prediction model;

[0036] S206: generating corresponding association information according to the environmental information and the probability value, wherein the association information is used to indicate the association relationship between the environmental information and the highland barley yield; and

[0037] S208: Optimizing the proportions of water and fertilizer components used for highland barley irrigation and fertilization according to the associated information using a preset matching model.

[0038] Specifically, the computing device predicts the environmental information of a certain highland barley planting area within a preset period in the future (for example, from March to September 2025) based on historical environmental data through a preset environmental information prediction model. The environmental information prediction model can be any meteorological prediction model in the prior art and is not specifically limited here. The environmental information includes: monthly average air pressure, monthly average temperature, monthly average water vapor pressure, monthly precipitation, monthly sunshine hours, relative humidity and wind speed, etc. m information. The environmental information [TA] = [A1, A2, ..., A n ]. Where n represents the number of months in the preset period, and A i represents the environmental feature vector of the i-th month, where i = 1 to n. For example, n = 7 (i.e., there are 7 months from March to September).

[0039] Where A1=[a 1,1 ,a 2,1 ,...,a m,1 ] T ;

[0040] A2=[a 1,2 ,a 2,2 ,...,a m,2 ] T ;

[0041] …

[0042] A n =[a 1,n ,a 2,n ,...,a m,n ] T .

[0043] where a j,i Indicates the j-th environmental information in the i-th month within a preset period, where j=1~m.

[0044] Furthermore, the computing device is pre-set with a yield prediction model for predicting highland barley yield. Figure 3 As shown, the computing device inputs the environmental information [TA] into the yield prediction model, and the yield prediction model predicts the barley yield under the influence of the environment based on the environmental information [TA]. The yield prediction model then outputs the probability value P = [p1, p2, p3, p4] of the barley yield interval corresponding to the barley yield. T The highland barley yield ranges include: 10,000-15,000 tons, 16,000-20,000 tons, 21,000-25,000 tons, and 26,000-30,000 tons. The yield prediction model thus outputs a probability value p1 corresponding to the highland barley yield range of 10,000-15,000 tons, a probability value p2 corresponding to the highland barley yield range of 16,000-20,000 tons, a probability value p3 corresponding to the highland barley yield range of 21,000-25,000 tons, and a probability value p4 corresponding to the highland barley yield range of 26,000-30,000 tons.

[0045] Furthermore, the computing device converts the environmental information [TA]=[A1, A2, ..., A n ] and probability value P = [p1,p2,p3,p4] T Perform association to generate corresponding association information C = [c1, c2, ..., c L ] T The correlation information is used to indicate the correlation between environmental information and highland barley yield.

[0046] Furthermore, the computing device is also pre-set with a ratio model for determining the ratio between the various components in the water fertilizer. Thus, the computing device converts the associated information C = [c1, c2, ..., c L ] T Input matching model, and use matching model to match the associated information C=[c1,c2,...,c L ] T Processing is performed so that the proportion model outputs the proportion of water and fertilizer components Z = [z1, z2, z3, z4] T The water-fertilizer components include nitrogen fertilizer, phosphate fertilizer, potash fertilizer, and water. The proportioning model outputs a nitrogen fertilizer component ratio value z1, a phosphate fertilizer component ratio value z2, a potash fertilizer component ratio value z3, and a water component ratio value z4. Furthermore, z1+z2+z3+z4=1.

[0047] Therefore, the computing device optimizes and calculates the water and fertilizer component ratios of water and fertilizer used for highland barley irrigation and fertilization based on environmental factors and corresponding outputs, and then mixes the corresponding water and fertilizer components according to the obtained component ratio, so that the water and fertilizer obtained by the mixture are applied to the highland barley planting area through irrigation.

[0048] As described in the background art, highland barley, a plant of the genus Hordeum in the family Poaceae, is an ancient cultivated crop. In my country, it is one of the most important specialty crops in the Tibetan region, cultivated in plateau areas at altitudes of 2,400 to 4,500 meters. It serves as the staple food for 50% of the Tibetan population. Highland barley boasts a nutritional profile characterized by "three highs and two lows" (high protein, high soluble fiber, high vitamins, and low fat and sugar), and is rich in mineral elements, making it a top-quality cereal crop. Furthermore, highland barley stalks can be used as high-quality fodder, and its grains are an important raw material for the brewing and food processing industries. In recent years, as the value of highland barley has gradually been discovered, highland barley products have become increasingly popular among consumers, and demand for highland barley has also increased year by year. Currently, water and fertilizer management technology models for high-quality, high-yield, green, and efficient cultivation of crops such as rice, wheat, and corn have been established across different growing regions. Although highland barley has a long history of cultivation, its production has lagged behind due to limitations imposed by natural conditions and human factors. Problems persist, such as the lack of scientific application of chemical fertilizers and the failure to rationally apply fertilizers based on climatic and environmental factors. This not only increases production costs but also results in low yields, poor quality, and low fertilizer utilization, severely restricting highland barley production.

[0049] In view of the technical problems described above, through the technical solutions of the embodiments of the present application, the computing device uses the environmental information of the barley planting area of ​​the period suitable for planting barley as an influencing factor affecting the barley yield, and predicts the probability value of each barley yield in the barley planting area according to the environmental information through the yield prediction model. Thus, the technical solution determines the influence relationship of environmental information on barley yield, thereby accurately predicting the yield of barley. And the technical solution optimizes the component ratio of water and fertilizer according to the generated associated information by associating the environmental information and the corresponding barley yield probability value, thereby avoiding the problems of increased production costs, low barley yield, poor quality, low fertilizer utilization rate, etc. caused by the inability to carry out reasonable fertilization according to factors such as climatic conditions and environment. Thus, it is achieved that under the influencing factors of environment and yield, a reasonable component ratio is determined, so that environmental information, yield and fertilizer are more coordinated and more adaptable. This solves the technical problems in the prior art of high production costs, low barley yields and low fertilizer utilization rates caused by factors such as climate conditions and environment not being in harmony with fertilizers.

[0050] Optionally, the operation of determining the probability value of each highland barley yield interval based on environmental information through a preset yield prediction model includes: generating corresponding first feature information based on environmental information through the RNN model of the yield prediction model; generating corresponding second feature information based on the first feature information through the first fully connected layer of the yield prediction model; and determining the probability value of each highland barley yield interval based on the second feature information through the first classifier of the yield prediction model.

[0051] Specifically, refer to Figure 4 As shown, the pre-trained yield prediction model includes an RNN model, a fully connected layer (i.e., the first fully connected layer) and a classifier (i.e., the first classifier).

[0052] The computing device inputs the environmental information [TA] into the yield prediction model, and the yield prediction model analyzes the environmental information [TA] through the RNN model, thereby outputting corresponding feature information (i.e., first feature information). The computing device then inputs the first feature information into the fully connected layer (i.e., the first fully connected layer) of the yield prediction model, extracts features from the first feature information through the first fully connected layer, and outputs corresponding feature information (i.e., second feature information). The computing device then inputs the second feature information into the classifier (i.e., the first classifier) ​​of the yield prediction model. The first classifier can be, for example, a softmax classifier. Thus, the yield prediction model classifies the second feature information through the first classifier and outputs the probability value P = [p1, p2, p3, p4] of the highland barley yield interval corresponding to the highland barley yield. T .

[0053] Barley growth is continuously influenced by multiple environmental factors, which change over time, forming complex time series data. The RNN model, with its unique memory units, is able to capture and leverage the long-term dependencies within these time series, thereby more accurately reflecting the impact of environmental changes on barley growth. This enables the model to take into account the cumulative effect of historical environmental information when predicting barley yield, improving forecast accuracy.

[0054] Optionally, the operation of generating corresponding association information based on environmental information and probability values ​​includes: determining multiple environmental feature vectors corresponding to the environmental information; fusing the multiple environmental feature vectors to generate third feature information; and splicing the third feature information and the probability value to generate association information.

[0055] Specifically, refer to Figure 5 As shown, the computing device determines a plurality of environmental feature vectors A1 to A2 corresponding to the environmental information. n .

[0056] Where A1=[a 1,1 ,a2,1 ,...,a m,1 ] T ;

[0057] A2=[a 1,2 ,a 2,2 ,...,a m,2 ] T ;

[0058] …

[0059] A n =[a 1,n ,a 2,n ,...,a m,n ] T .

[0060] Furthermore, the computing device calculates A1 to A n The average value of each feature in b j , thereby fusing multiple environmental feature vectors to generate the third feature information B = [b1, b2, ..., b m ] T . The third feature information B is generated as follows: m ] T The calculation formula is:

[0061]

[0062] Furthermore, the computing device converts the third feature information B=[b1, b2, ..., b m ] T And the probability value P = [p1, p2, p3, p4] T Perform splicing processing to generate corresponding association information C = [c1, c2, ..., c L ] T .Right now:

[0063]

[0064] Where L = m + 4.

[0065] Therefore, this technical solution splices the third characteristic information with the probability value of the yield prediction, and the generated correlation information not only contains a detailed description of the environmental characteristics, but also incorporates the barley yield prediction value, so that each environmental indicator can be directly mapped to the possible yield result, greatly facilitating the intuitive understanding and analysis of the correlation between environmental factors and yield.

[0066] Optionally, the operation of optimizing the component ratio of water and fertilizer components used for highland barley irrigation and fertilization according to the association information through a preset matching model includes: generating fourth feature information according to the association information through the second fully connected layer of the matching model; and optimizing the component ratio of water and fertilizer components according to the fourth feature information through the second classifier of the matching model.

[0067] Specifically, refer to Figure 6 As shown, the pre-trained matching model includes a fully connected layer (ie, the second fully connected layer) and a classifier (ie, the second classifier).

[0068] The computing device inputs the correlation information C into the matching model, which then extracts features from the correlation information through a fully connected layer (i.e., the second fully connected layer) and outputs corresponding feature information (i.e., the fourth feature information). The computing device then inputs the fourth feature information output by the second fully connected layer into the classifier of the matching model (i.e., the second classifier). The second classifier can be, for example, a softmax classifier. The matching model then classifies the fourth feature information through the second classifier and outputs the component ratio Z = [z1, z2, z3, z4] of the water and fertilizer components of the water and fertilizer. T .

[0069] Therefore, this technical solution performs deep feature extraction on the associated information through the second fully connected layer of the ratio model, and then converts the fourth feature information into clear water and fertilizer component optimization suggestions through the second classifier, thereby clearly displaying the water and fertilizer ratio plan.

[0070] Optionally, the steps for training the yield prediction model are as follows: collecting sample environment information and the corresponding first sample yield in the sample period; and training the yield prediction model according to the sample environment information and the corresponding first sample yield by using a gradient descent algorithm.

[0071] Specifically, before the yield prediction model is put into use, the computing device trains the yield prediction model, wherein the training steps are as follows:

[0072] The computing device collects sample environment information and corresponding first sample output in a sample period (e.g., March to September 2014 to March to September 2023) to construct a first sample set. The first sample set is shown in Table 1:

[0073] Table 1

[0074] Sample serial number Sample environment information First sample yield 1 <![CDATA[[STA 1 ]]]> <![CDATA[SP 1 ]]> 2 <![CDATA[[STA 2 ]]]> <![CDATA[SP 2 ]]> ... ... ... h <![CDATA[[STA h ]]]> <![CDATA[SP h ]]>

[0075] The sample environment information [STA f ]=[SA1 f ,SA2 f ,...,SA nf ]. Where n represents the number of months of each sample in the sample period, and SA i f The sample environment feature vector of the i-th month of the f-th sample is represented by i=1-n and f=1-h. For example, n=7 (i.e., 7 months from March to September) and h=10 (10 years from 2014 to 2023).

[0076] Among them SA1 f =[sa f 1,1 ,sa f 2,1 ,...,sa f m,1 ] T ;

[0077] SA2 f =[sa f 1,2 ,sa f 2,2 ,...,sa f m,2 ] T ;

[0078] …

[0079] SA n f =[sa f 1,n ,sa f 2,n ,...,sa f m,n ] T .

[0080] And the first sample yield SP f =[sp1 f ,sp2 f ,sp3 f ,sp4 f ] T . Among them sp1 f represents the sample probability value of the fth sample corresponding to the highland barley yield range of 10,000 to 15,000 tons; sp2 f represents the sample probability value of the fth sample corresponding to the highland barley yield range of 16,000 to 20,000 tons; sp3 f The sample probability value corresponding to the barley yield range of 21,000 to 25,000 tons for the f-th sample; sp4 f It represents the sample probability value of the f-th sample corresponding to the highland barley production range of 26,000 to 30,000 tons.

[0081] Furthermore, the computing device trains the yield prediction model according to the sample environment information and the corresponding first sample yield through a gradient descent algorithm, determines the parameters in the yield prediction model, and obtains a yield prediction model that can be put into use.

[0082] Therefore, this technical solution uses a gradient descent algorithm to train the yield prediction model. By iteratively adjusting the model parameters, the difference between the predicted results and the actual output is minimized, thereby continuously improving the model's generalization ability and prediction accuracy.

[0083] Optionally, the steps for training the ratio model are as follows: collecting sample environmental information and sample component ratios of corresponding water and fertilizer components in the sample period; inputting the sample environmental information into the trained yield prediction model, and outputting the second sample yield according to the sample environmental information through the yield prediction model; and training the ratio model according to the second sample yield, the sample environmental information and the sample component ratio through the gradient descent algorithm.

[0084] Specifically, before the matching model is put into use, the computing device trains the matching model, wherein the training steps are as follows:

[0085] The computing device collects sample environmental information during the sample period (e.g., March to September 2014 to March to September 2023) and the sample component ratios of the water and fertilizer components used at that time to construct a second sample set. The second sample set is shown in Table 2:

[0086] Table 2

[0087] Sample serial number Sample environment information Sample component ratio 1 <![CDATA[[STA 1 ]]]> <![CDATA[SZ 1 ]]> 2 <![CDATA[[STA 2 ]]]> <![CDATA[SZ 2 ]]> ... ... ... h <![CDATA[[STA h ]]]> <![CDATA[SZ h ]]>

[0088] The sample component ratio SZ f =[sz1 f ,sz2 f ,sz3 f ,sz4 f ] T . Among them sz1 f Indicates the sample component ratio value of nitrogen fertilizer in the fth sample, sz2 f Indicates the sample component ratio of the phosphate fertilizer of the fth sample, sz3 f Indicates the sample component ratio value of the potassium fertilizer of the fth sample, sz4 f Represents the sample component ratio value of water in the fth sample.

[0089] Furthermore, the computing device sends the sample environment information [STA f ] is input into the trained yield prediction model, so that the yield prediction model can predict the sample environment information [STA f ] is processed and the corresponding second sample output YP is output f=[yp1 f ,yp2 f ,yp3 f ,yp4 f ] T . Among them yp1 f represents the sample probability value of the highland barley yield range of 10,000 to 15,000 tons corresponding to the f-th sample environmental information; yp2 f represents the sample probability value of the highland barley yield range of 16,000 to 20,000 tons corresponding to the f-th sample environmental information; yp3 f The sample probability value of the highland barley yield range of 21,000 to 25,000 tons corresponding to the f-th sample environmental information; yp4 f It represents the sample probability value of the highland barley yield range of 26,000 to 30,000 tons corresponding to the f-th sample environmental information.

[0090] Furthermore, the computing device sends the sample environment information [STA f ] and the second sample yield YP f Perform association to generate corresponding sample association information SC f The computing device then uses the gradient descent algorithm to calculate the sample association information SC f and sample component ratio SZ f , train the matching model, determine the parameters in the matching model, and obtain a matching model that can be put into use.

[0091] Therefore, this technical solution can obtain accurate second sample yield and sample correlation information through the trained yield prediction model. In the subsequent training of the matching model, the parameters of the matching model can be adjusted through the accurate second sample yield and sample correlation information, thereby improving the applicability between the yield prediction model and the matching model.

[0092] In summary, reference Figure 7 As shown, the sequential steps of the optimization method for highland barley fertilization are as follows:

[0093] S701: The computing device collects environmental information of the highland barley planting area within a preset period;

[0094] S702: The computing device generates corresponding first feature information according to the environmental information using the RNN model of the yield prediction model;

[0095] S703: The computing device generates corresponding second feature information based on the first feature information through the first fully connected layer of the yield prediction model;

[0096] S704: The computing device determines the probability value of each highland barley yield interval based on the second feature information using the first classifier of the yield prediction model;

[0097] S705: The computing device determines a plurality of environmental feature vectors corresponding to the environmental information;

[0098] S706: The computing device fuses the multiple environmental feature vectors to generate third feature information;

[0099] S707: The computing device performs splicing processing on the third feature information and the probability value to generate correlation information;

[0100] S708: The computing device generates fourth feature information according to the association information through the second fully connected layer of the matching model;

[0101] S709: The computing device optimizes the component ratio of the water and fertilizer components according to the fourth feature information through the second classifier of the matching model.

[0102] In addition, according to a second aspect of this embodiment, a computer system is provided, including a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps of the above method.

[0103] In addition, according to a third aspect of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0104] According to this embodiment, the computing device uses environmental information of a barley-growing region suitable for a barley-growing period as an influencing factor affecting barley yield, and uses a yield prediction model to predict the probability values ​​of barley yields for each barley-growing region based on this environmental information. This technical solution thus determines the relationship between environmental information and barley yield, thereby accurately predicting barley yield. Furthermore, this technical solution associates environmental information with corresponding barley yield probability values, thereby optimizing the component ratios of water and fertilizer based on the generated association information through a ratioing model. This avoids problems such as increased production costs, low barley yield, poor quality, and low fertilizer utilization efficiency caused by the failure to rationally apply fertilizers based on climatic conditions and environmental factors. This achieves the technical effect of determining a reasonable component ratio based on the influencing factors of environment and yield, achieving greater coordination between environmental information, yield, and fertilizer, and achieving greater adaptability. This solves the technical problem in the prior art of high production costs, low barley yield, and low fertilizer utilization efficiency caused by the incompatibility between factors such as climatic conditions and environment and fertilizer.

[0105] It should be noted that for the aforementioned method embodiments, for simplicity of description, they are all expressed as a series of action combinations. However, those skilled in the art should be aware that the present invention is not limited by the order of the actions described, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0106] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0107] Example 2

[0108] Figure 8 The apparatus 800 for optimizing highland barley fertilization according to the first aspect of this embodiment is shown. The apparatus 800 corresponds to the method according to the first aspect of embodiment 1. Figure 8 As shown, the device 800 includes: an information collection module 810, which is used to collect environmental information of the barley planting area within a preset period; a probability determination module 820, which is used to determine the probability value of each barley yield interval according to the environmental information through a preset yield prediction model; an information generation module 830, which is used to generate corresponding correlation information according to the environmental information and the probability value, wherein the correlation information is used to indicate the correlation relationship between the environmental information and the barley yield; and a ratio optimization module 840, which is used to optimize the component ratio of water and fertilizer components used for barley irrigation and fertilization according to the correlation information through a preset ratio model.

[0109] Optionally, the probability determination module 820 includes: a first generation submodule, used to generate corresponding first feature information based on environmental information through the RNN model of the yield prediction model; a second generation submodule, used to generate corresponding second feature information based on the first feature information through the first fully connected layer of the yield prediction model; and a first determination submodule, used to determine the probability value of each barley yield interval based on the second feature information through the first classifier of the yield prediction model.

[0110] Optionally, the information generation module 830 includes: a second determination submodule, used to determine multiple environmental feature vectors corresponding to the environmental information; a third generation submodule, used to fuse the multiple environmental feature vectors to generate third feature information; and a fourth generation submodule, used to splice the third feature information and the probability value to generate associated information.

[0111] Optionally, the ratio optimization module 840 includes: a fifth generation submodule, which is used to generate fourth feature information based on the association information through the second fully connected layer of the ratio model; and a sixth generation submodule, which is used to optimize the component ratio of the water and fertilizer components based on the fourth feature information through the second classifier of the ratio model.

[0112] Optionally, the first training module is used to train the yield prediction model in the following steps: a first training submodule is used to collect sample environment information and the corresponding first sample yield in the sample period; and a second training submodule is used to train the yield prediction model based on the sample environment information and the corresponding first sample yield through a gradient descent algorithm.

[0113] Optionally, the second training module is used to train the ratio model in the following steps: a third training submodule is used to collect sample environmental information and sample component ratios of corresponding water and fertilizer components in a sample period; a fourth training submodule is used to input the sample environmental information into a trained yield prediction model, and output a second sample yield based on the sample environmental information through the yield prediction model; and a fifth training submodule is used to train the ratio model based on the second sample yield, sample environmental information and sample component ratio through a gradient descent algorithm.

[0114] According to this embodiment, the computing device uses environmental information of a barley-growing region suitable for a barley-growing period as an influencing factor affecting barley yield, and uses a yield prediction model to predict the probability values ​​of barley yields for each barley-growing region based on this environmental information. This technical solution thus determines the relationship between environmental information and barley yield, thereby accurately predicting barley yield. Furthermore, this technical solution associates environmental information with corresponding barley yield probability values, thereby optimizing the component ratios of water and fertilizer based on the generated association information through a ratioing model. This avoids problems such as increased production costs, low barley yield, poor quality, and low fertilizer utilization efficiency caused by the failure to rationally apply fertilizers based on climatic conditions and environmental factors. This achieves the technical effect of determining a reasonable component ratio based on the influencing factors of environment and yield, achieving greater coordination between environmental information, yield, and fertilizer, and achieving greater adaptability. This solves the technical problem in the prior art of high production costs, low barley yield, and low fertilizer utilization efficiency caused by the incompatibility between factors such as climatic conditions and environment and fertilizer.

[0115] Example 3

[0116] Figure 9 The apparatus 900 for optimizing highland barley fertilization according to the first aspect of this embodiment is shown. The apparatus 900 corresponds to the method according to the first aspect of embodiment 1. Figure 9 As shown, the device 900 includes: a processor 910; and a memory 920, which is connected to the processor 910 and is used to provide the processor 910 with instructions for processing the following processing steps: collecting environmental information of the barley planting area within a preset period; determining the probability value of each barley yield interval based on the environmental information through a preset yield prediction model; generating corresponding correlation information based on the environmental information and the probability value, wherein the correlation information is used to indicate the correlation relationship between the environmental information and the barley yield; and optimizing the component ratio of water and fertilizer components used for barley irrigation and fertilization based on the correlation information through a preset ratio model.

[0117] Optionally, the operation of determining the probability value of each highland barley yield interval based on environmental information through a preset yield prediction model includes: generating corresponding first feature information based on environmental information through the RNN model of the yield prediction model; generating corresponding second feature information based on the first feature information through the first fully connected layer of the yield prediction model; and determining the probability value of each highland barley yield interval based on the second feature information through the first classifier of the yield prediction model.

[0118] Optionally, the operation of generating corresponding association information based on environmental information and probability values ​​includes: determining multiple environmental feature vectors corresponding to the environmental information; fusing the multiple environmental feature vectors to generate third feature information; and splicing the third feature information and the probability value to generate association information.

[0119] Optionally, the operation of optimizing the component ratio of water and fertilizer components used for highland barley irrigation and fertilization according to the association information through a preset matching model includes: generating fourth feature information according to the association information through the second fully connected layer of the matching model; and optimizing the component ratio of water and fertilizer components according to the fourth feature information through the second classifier of the matching model.

[0120] Optionally, the steps for training the yield prediction model are as follows: collecting sample environment information and the corresponding first sample yield in the sample period; and training the yield prediction model according to the sample environment information and the corresponding first sample yield by using a gradient descent algorithm.

[0121] Optionally, the steps for training the ratio model are as follows: collecting sample environmental information and sample component ratios of corresponding water and fertilizer components in the sample period; inputting the sample environmental information into the trained yield prediction model, and outputting the second sample yield according to the sample environmental information through the yield prediction model; and training the ratio model according to the second sample yield, the sample environmental information and the sample component ratio through the gradient descent algorithm.

[0122] According to this embodiment, the computing device uses environmental information of a barley-growing region suitable for a barley-growing period as an influencing factor affecting barley yield, and uses a yield prediction model to predict the probability values ​​of barley yields for each barley-growing region based on this environmental information. This technical solution thus determines the relationship between environmental information and barley yield, thereby accurately predicting barley yield. Furthermore, this technical solution associates environmental information with corresponding barley yield probability values, thereby optimizing the component ratios of water and fertilizer based on the generated association information through a ratioing model. This avoids problems such as increased production costs, low barley yield, poor quality, and low fertilizer utilization efficiency caused by the failure to rationally apply fertilizers based on climatic conditions and environmental factors. This achieves the technical effect of determining a reasonable component ratio based on the influencing factors of environment and yield, achieving greater coordination between environmental information, yield, and fertilizer, and achieving greater adaptability. This solves the technical problem in the prior art of high production costs, low barley yield, and low fertilizer utilization efficiency caused by the incompatibility between factors such as climatic conditions and environment and fertilizer.

[0123] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0124] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0126] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0127] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0128] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc. Various media that can store program codes.

[0129] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for optimizing highland barley fertilization, characterized in that: include: The environmental information prediction model is used to predict the environmental information of the highland barley planting area in the future preset period based on historical environmental information [TA] = [A1, A2, ..., A n ], where n represents the number of months in the preset period, A i represents the environmental feature vector of the i-th month, where i=1-n, and wherein the environmental information includes the monthly average air pressure, monthly average temperature, monthly average water vapor pressure, monthly precipitation, monthly sunshine hours, relative humidity, and wind speed; Determining the probability value of each highland barley yield interval based on the environmental information through a preset yield prediction model; generating corresponding association information according to the environmental information and the probability value, wherein the association information is used to indicate an association relationship between the environmental information and highland barley yield; as well as The component ratio of water and fertilizer components used for highland barley irrigation and fertilization is optimized according to the associated information through a preset matching model, wherein the water and fertilizer components include nitrogen fertilizer, phosphorus fertilizer, potash fertilizer and water, wherein The operation of determining the probability value of each highland barley yield interval according to the environmental information through a preset yield prediction model includes: generating corresponding first feature information according to the environmental information through an RNN model of the yield prediction model; generating corresponding second feature information according to the first feature information through a first fully connected layer of the yield prediction model; and determining the probability value of each highland barley yield interval according to the second feature information through a first classifier of the yield prediction model, and The operation of generating corresponding association information according to the environmental information and the probability value includes: Determine a plurality of environmental feature vectors A1 to A2 corresponding to the environmental information n : A1=[a 1,1 ,a 2,1 ,...,a m,1 ] T ; A2=[a 1,2 ,a 2,2 ,...,a m,2 ] T ; …… A n =[a 1,n ,a 2,n ,...,a m,n ] T , Among them A i represents the environmental feature vector of the i-th month, i = 1 to n, and n represents the number of months in the preset cycle; Calculate the multiple environmental feature vectors A1~A n The average value of each feature in b j , thereby fusing the multiple environmental feature vectors to generate the third feature information B = [b1, b2, ..., b m ] T , wherein the third feature information B is generated as [b1, b2, ..., b m ] T The calculation formula is: where a j,i represents the jth environmental information of the i-th month within a preset period, where j=1-m; and The third feature information B=[b1, b2, ..., b m ] T And the probability value P = [p1, p2, p3, p4] T Perform splicing processing to generate the associated information C = [c1, c2, ..., c L ] T : Where L = m + 4; p1, p2, p3 and p4 represent the probability values ​​of different highland barley yield intervals, and The operation of optimizing the component ratio of water and fertilizer components used for highland barley irrigation and fertilization according to the association information through a preset matching model includes: generating fourth feature information according to the association information through a second fully connected layer of the matching model; and optimizing the component ratio of the water and fertilizer components according to the fourth feature information through a second classifier of the matching model, wherein the second classifier is used to output the optimized component ratio of the water and fertilizer components according to the fourth feature information, and The steps of training the yield prediction model are as follows: collecting sample environment information and the corresponding first sample yield in a sample period; and training the yield prediction model according to the sample environment information and the corresponding first sample yield by a gradient descent algorithm, and The steps of training the matching model are as follows: collecting the sample environmental information and the sample component ratios of the corresponding water and fertilizer components in the sample period; inputting the sample environmental information into the trained yield prediction model, and outputting a second sample yield according to the sample environmental information through the yield prediction model; and training the matching model according to the second sample yield, the sample environmental information and the sample component ratio through the gradient descent algorithm, wherein: Training the matching model according to the second sample yield, the sample environment information, and the sample component ratio using the gradient descent algorithm includes: Associating the sample environment information with the second sample output to obtain sample association information; The matching model is trained according to the sample association information and the sample component ratio through the gradient descent algorithm.

2. An optimization device for highland barley irrigation and fertilization, characterized in that: include: The information collection module is used to predict the environmental information of the barley planting area in the future preset period based on the historical environmental information through the environmental information prediction model [TA] = [A1, A2, ..., A n ], where n represents the number of months in the preset period, A i represents the environmental feature vector of the i-th month, where i=1-n, and wherein the environmental information includes the monthly average air pressure, monthly average temperature, monthly average water vapor pressure, monthly precipitation, monthly sunshine hours, relative humidity, and wind speed; A probability determination module is used to determine the probability value of each highland barley yield interval according to the environmental information through a preset yield prediction model; An information generation module, configured to generate corresponding association information based on the environmental information and the probability value, wherein the association information is used to indicate an association relationship between the environmental information and highland barley yield; as well as A ratio optimization module is used to optimize the component ratio of water and fertilizer components used for highland barley irrigation and fertilization according to the associated information through a preset ratio model, wherein the water and fertilizer components include nitrogen fertilizer, phosphorus fertilizer, potash fertilizer and water, wherein The probability determination module includes: a first generation submodule for generating corresponding first feature information according to the environmental information through the RNN model of the yield prediction model; a second generation submodule for generating corresponding second feature information according to the first feature information through the first fully connected layer of the yield prediction model; and a first determination submodule for determining the probability value of each highland barley yield interval according to the second feature information through the first classifier of the yield prediction model, and The information generation module includes: Determine a plurality of environmental feature vectors A1 to A2 corresponding to the environmental information n : A1=[a 1,1 ,a 2,1 ,...,a m,1 ] T ; A2=[a 1,2 ,a 2,2 ,...,a m,2 ] T ; …… A n =[a 1,n ,a 2,n ,...,a m,n ] T , Among them A i represents the environmental feature vector of the i-th month, i = 1 to n, and n represents the number of months in the preset cycle; Calculate the multiple environmental feature vectors A1~A n The average value of each feature in b j , thereby fusing the multiple environmental feature vectors to generate the third feature information B = [b1, b2, ..., b m ] T , wherein the third feature information B is generated as [b1, b2, ..., b m ] T The calculation formula is: where a j,i represents the jth environmental information of the i-th month within a preset period, where j=1-m; and The third feature information B=[b1, b2, ..., b m ] T And the probability value P = [p1, p2, p3, p4] T Perform splicing processing to generate the associated information C = [c1, c2, ..., c L ] T : Where L = m + 4; p1, p2, p3 and p4 represent the probability values ​​of different highland barley yield intervals, and The operation of optimizing the component ratio of water and fertilizer components used for highland barley irrigation and fertilization according to the association information through a preset matching model includes: generating fourth feature information according to the association information through a second fully connected layer of the matching model; and optimizing the component ratio of the water and fertilizer components according to the fourth feature information through a second classifier of the matching model, wherein the second classifier is used to output the optimized component ratio of the water and fertilizer components according to the fourth feature information, and The steps of training the yield prediction model are as follows: collecting sample environment information and the corresponding first sample yield in a sample period; and training the yield prediction model according to the sample environment information and the corresponding first sample yield by a gradient descent algorithm, and The steps of training the matching model are as follows: collecting the sample environmental information and the sample component ratios of the corresponding water and fertilizer components in the sample period; inputting the sample environmental information into the trained yield prediction model, and outputting a second sample yield according to the sample environmental information through the yield prediction model; and training the matching model according to the second sample yield, the sample environmental information and the sample component ratio through the gradient descent algorithm, wherein: The matching model is trained according to the second sample yield, the sample environmental information and the sample component ratio by the gradient descent algorithm, including: associating the sample environmental information with the second sample yield to obtain sample association information; and training the matching model according to the sample association information and the sample component ratio by the gradient descent algorithm.

3. An optimization device for highland barley irrigation and fertilization, characterized in that: include: processor; as well as A memory, connected to the processor, configured to provide the processor with instructions for processing the following processing steps: The environmental information prediction model is used to predict the environmental information of the highland barley planting area in the future preset period based on historical environmental information [TA] = [A1, A2, ..., A n ], where n represents the number of months in the preset period, A i represents the environmental feature vector of the i-th month, where i=1-n, and wherein the environmental information includes the monthly average air pressure, monthly average temperature, monthly average water vapor pressure, monthly precipitation, monthly sunshine hours, relative humidity, and wind speed; Determining the probability value of each highland barley yield interval based on the environmental information through a preset yield prediction model; generating corresponding association information according to the environmental information and the probability value, wherein the association information is used to indicate an association relationship between the environmental information and highland barley yield; as well as The component ratio of water and fertilizer components used for highland barley irrigation and fertilization is optimized according to the associated information through a preset matching model, wherein the water and fertilizer components include nitrogen fertilizer, phosphorus fertilizer, potash fertilizer and water, wherein The operation of determining the probability value of each highland barley yield interval according to the environmental information through a preset yield prediction model includes: generating corresponding first feature information according to the environmental information through an RNN model of the yield prediction model; generating corresponding second feature information according to the first feature information through a first fully connected layer of the yield prediction model; and determining the probability value of each highland barley yield interval according to the second feature information through a first classifier of the yield prediction model, and The operation of generating corresponding association information according to the environmental information and the probability value includes: Determine a plurality of environmental feature vectors A1 to A2 corresponding to the environmental information n : A1=[a 1,1 ,a 2,1 ,...,a m,1 ] T ; A2=[a 1,2 ,a 2,2 ,...,a m,2 ] T ; …… A n =[a 1,n ,a 2,n ,...,a m,n ] T , Among them A i represents the environmental feature vector of the i-th month, i = 1 to n, and n represents the number of months in the preset cycle; Calculate the multiple environmental feature vectors A1~A n The average value of each feature in b j , thereby fusing the multiple environmental feature vectors to generate the third feature information B = [b1, b2, ..., b m ] T , wherein the third feature information B is generated as [b1, b2, ..., b m ] T The calculation formula is: where a j,i represents the jth environmental information of the i-th month within a preset period, where j=1-m; and The third feature information B=[b1, b2, ..., b m ] T And the probability value P = [p1, p2, p3, p4] T Perform splicing processing to generate the associated information C = [c1, c2, ..., c L ] T : Where L = m + 4; p1, p2, p3 and p4 represent the probability values ​​of different highland barley yield intervals, and The operation of optimizing the component ratio of water and fertilizer components used for highland barley irrigation and fertilization according to the association information through a preset matching model includes: generating fourth feature information according to the association information through a second fully connected layer of the matching model; and optimizing the component ratio of the water and fertilizer components according to the fourth feature information through a second classifier of the matching model, wherein the second classifier is used to output the optimized component ratio of the water and fertilizer components according to the fourth feature information, and The steps of training the yield prediction model are as follows: collecting sample environment information and the corresponding first sample yield in a sample period; and training the yield prediction model according to the sample environment information and the corresponding first sample yield by a gradient descent algorithm, and The steps of training the matching model are as follows: collecting the sample environmental information and the sample component ratios of the corresponding water and fertilizer components in the sample period; inputting the sample environmental information into the trained yield prediction model, and outputting a second sample yield according to the sample environmental information through the yield prediction model; and training the matching model according to the second sample yield, the sample environmental information and the sample component ratio through the gradient descent algorithm, wherein: The matching model is trained according to the second sample yield, the sample environmental information and the sample component ratio by the gradient descent algorithm, including: associating the sample environmental information with the second sample yield to obtain sample association information; and training the matching model according to the sample association information and the sample component ratio by the gradient descent algorithm.

4. A computer system comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method according to claim 1.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to claim 1 are implemented.

Images