Water and fertilizer integrated intelligent irrigation method and device and electronic equipment

Abstract

The invention relates to the technical field of water and fertilizer intelligent control, in particular to a water and fertilizer integrated intelligent irrigation method and device and electronic equipment.The method comprises the steps that the crop growth stage of each cultivation layer is obtained, and a crop transpiration rate model of each cultivation layer is established according to crop environment data and the crop growth stage of each cultivation layer; outputting a real-time water stress index of each cultivation layer; performing feature classification on the historical cultivation data according to the historical crop growth data to establish a growth-water and fertilizer mapping relation table; collecting crop growth data of each cultivation layer in the monitoring period, extracting crop characteristics of each cultivation layer according to the crop growth data of each cultivation layer and the crop growth stage, and then setting a characteristic weight of each cultivation layer; and generating a water and fertilizer configuration scheme of each cultivation layer based on the real-time water stress index and the feature weight of each cultivation layer. The intelligent control efficiency of water and fertilizer irrigation is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water and fertilizer intelligent control, and particularly relates to a water and fertilizer integrated intelligent irrigation method and device and electronic equipment. BACKGROUND

[0002] In the northwest arid region, greenhouse planting often faces harsh environmental challenges such as high evaporation, strong sunlight and large diurnal temperature difference. The traditional irrigation method is difficult to accurately match the actual water and fertilizer demand of crops, which easily causes problems such as waste of water resources, low fertilizer utilization rate and crop growth stress. Especially in the multi-layer vertical cultivation system, the microclimate of each cultivation layer is significantly different, which further aggravates the uneven supply of water and fertilizer, affecting the overall yield and quality.

[0003] The existing water and fertilizer integrated technology is mostly based on fixed period or single environmental parameter for control, and lacks comprehensive consideration of crop growth stage, real-time physiological state and interlayer mutual interference. At the same time, historical cultivation experience cannot be effectively converted into reusable intelligent decision basis, resulting in weak system self-adaptation ability and insufficient control precision. Therefore, an intelligent irrigation method capable of integrating multi-source data, dynamically responding to crop demand and adapting to multi-layer structure is needed to improve resource utilization efficiency and planting benefits. SUMMARY

[0004] The present application relates to the technical field of water and fertilizer intelligent control, and particularly relates to a water and fertilizer integrated intelligent irrigation method and device and electronic equipment.

[0005] To achieve the above-mentioned purpose, according to one aspect of the present application, a water and fertilizer integrated intelligent irrigation method is provided, comprising: Obtaining the crop growth stage of each cultivation layer, and establishing the crop transpiration rate model of each cultivation layer according to the crop environmental data and the crop growth stage of each cultivation layer, and then outputting the real-time water stress index of each cultivation layer; Classifying the historical cultivation data according to the historical crop growth data to establish a growth-water and fertilizer mapping relationship table; Collecting the crop growth data of each cultivation layer in the monitoring period, and extracting the crop characteristics of each cultivation layer according to the crop growth data and the crop growth stage of each cultivation layer, and then setting the feature weight of each cultivation layer; Generating the water and fertilizer configuration scheme of each cultivation layer based on the real-time water stress index and the feature weight of each cultivation layer.

[0006] Optionally, the crop growth history data is obtained, and the average value of the crop transpiration rate of different crop growth stages is calculated, and the calculation result is taken as the reference transpiration rate, denoted as ZT(i), i=1,2,3,4,5; According to the crop environmental data of each cultivation layer, an environmental interference coefficient HG(j) is constructed; Determine the real-time transpiration rate SZ(j) of each cultivation layer according to the crop growth stage of each cultivation layer and the environmental interference coefficient, and set SZ(j)=ZT(i,j)×HG(j); wherein ZT(i,j) represents the reference transpiration rate of the jth cultivation layer.

[0007] Optionally, the transpiration state of each cultivation layer is determined by setting a transpiration rate threshold, and then the real-time water stress index of each cultivation layer is constructed: when SZ(j) is less than the transpiration rate threshold, the real-time water stress index is set to 1; when SZ(j) is greater than or equal to the transpiration rate threshold, the real-time water stress index is set to exp{[SZ(j)-transpiration rate threshold] / transpiration rate threshold}.

[0008] Optionally, the interlayer interference coefficient β(j,j+1) is constructed according to the interlayer environmental data between each cultivation layer, and set β(j,j+1)=[a1×(△CO-YC) / YC+a2×(△T-YT) / YT] / H; wherein △CO represents the interlayer carbon dioxide concentration difference, △T represents the interlayer temperature difference, YC and YT respectively represent the carbon dioxide concentration difference threshold and the interlayer temperature difference threshold, a1 and a2 respectively represent the interlayer carbon dioxide difference weight and the interlayer temperature weight, and a1+a2=1; Take |β(j,j+1)-β(j-1,j)| as the interlayer stress coefficient of the jth cultivation layer, set the interlayer interference constant, and compare it with the interlayer stress coefficient of the jth cultivation layer: when |β(j,j+1)-β(j-1,j)| is greater than or equal to the interlayer interference constant, the real-time water stress index of the jth cultivation layer is updated to F(j)×ln{e+[|β(j,j+1)-β(j-1,j)|-interlayer interference constant] / interlayer interference constant}; otherwise, no update is performed; wherein e is the natural logarithm, and F(j) is the real-time water stress index of the jth cultivation layer.

[0009] Optionally, the fruit quality index PZ of the historical mature period is calculated according to the historical crop growth data, and set PZ={Σ[a(k)-A(k)] / A(k)} / K; wherein a(k) represents the kth quality index of the fruit of the historical mature period, A(k) represents the quality standard value of the kth quality index, and K is the number of quality indexes of the fruit; According to the fruit quality index and the historical batch yield, the historical cultivation data is classified according to characteristics: when PZ is less than the quality judgment constant or CL is less than the minimum yield, the historical cultivation data corresponding to the historical crop growth data is stored as a failure configuration component data; When PZ is greater than or equal to the quality judgment constant×η and CL is less than the minimum yield×η, the historical cultivation data corresponding to the historical crop growth data is stored as quality priority configuration component data; When PZ is less than the quality judgment constant x η and CL is greater than or equal to the minimum yield x η, the historical cultivation data corresponding to the historical crop growth data is stored as yield priority configuration component data; When PZ is greater than or equal to the quality judgment constant x η and CL is greater than or equal to the minimum yield x η, the historical cultivation data corresponding to the historical crop growth data is stored as standard configuration component data; Wherein, η is a preset offset constant.

[0010] Optionally, the historical cultivation data classified by features is subjected to similar feature extraction to obtain feature configurations of each feature classification, and the process is as follows: The coefficient of variation CV(s, d) of the dth configuration component in the historical cultivation data under the sth feature classification is calculated, and the calculation result is compared with the preset variation abnormality number. When CV(s, d) is greater than or equal to the preset variation abnormality number, it is determined that the dth configuration component does not belong to the feature configuration component under the sth feature classification; otherwise, it is determined that the dth configuration component belongs to the feature configuration component under the sth feature classification; The average value of the feature configuration components in the historical cultivation data under the sth feature classification is calculated, and the average value set of the feature configuration components is taken as the feature configuration of the sth feature classification; The plant growth feature of the sth feature classification is calculated; The key-value pair of the plant growth feature and the feature configuration of the feature classification is taken as a group of data of the growth-water fertilizer mapping relationship table, and the growth-water fertilizer mapping relationship table is constructed therefrom.

[0011] Optionally, when the crop growth stage of the cultivation layer is the fruit setting period, the cosine similarity between the crop growth bias feature vector β1(j) and the plant growth feature of each feature classification is calculated, and the calculation result is taken as the feature weight set of the cultivation layer; When the crop growth stage of the cultivation layer is not the fruit setting period, the cosine similarity between the crop growth defect feature vector β2(j) and the plant growth feature of each feature classification is calculated, and the calculation result is taken as the feature weight set of the cultivation layer; The feature weights of each cultivation layer are combined and denoted as S(j).

[0012] Optionally, the average value of the real-time water stress index of each cultivation layer in the monitoring period is calculated, and the calculation result is denoted as μ(j); The selection sequence C(j) of each cultivation layer is constructed from the feature weight set of each cultivation layer and μ(j); When max{C(j)} is not equal to s1(j)×μ(j)×b1, the feature configuration corresponding to the feature classification of max{C(j)} is used as the irrigation scheme for the next monitoring period and output to the user; s1(j) is the weight of the j-th cultivation layer irrigated with the first feature configuration, and b1 is the influence coefficient of water stress on the first feature classification. Conversely, an alarm should be triggered for the user.

[0013] According to another aspect of this application, an integrated water and fertilizer intelligent irrigation device is provided, comprising: The water stress analysis unit is used to obtain the crop growth stage of each cultivation layer, and establish a crop transpiration rate model for each cultivation layer based on the crop environmental data and crop growth stage of each cultivation layer, and then output the real-time water stress index of each cultivation layer. The mapping relationship construction unit is used to classify historical cultivation data by feature based on historical crop growth data in order to establish a growth-water and fertilizer mapping relationship table. The feature weight construction unit is used to collect crop growth data of each cultivation layer during the monitoring period, extract crop features of each cultivation layer based on the crop growth data and crop growth stage, match the crop features with the growth-water-fertilizer mapping table, and then set the feature weight of each cultivation layer. The irrigation scheme generation unit is used to generate water and fertilizer configuration schemes for each cultivation layer based on the real-time water stress index and feature weights of each cultivation layer.

[0014] According to another aspect of this application, an electronic device is provided, comprising: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the integrated water and fertilizer intelligent irrigation method.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: The integrated water and fertilizer intelligent irrigation solution provided by this invention achieves precise, dynamic, and stratified water and fertilizer regulation for low-growing crops such as greenhouse tomatoes by integrating multi-layer cultivation environment perception, crop transpiration model, inter-layer interference correction, historical data mapping, and real-time growth characteristic matching. This solution effectively overcomes the problems of slow response, resource waste, and uneven inter-layer growth in traditional irrigation methods, significantly improving water and fertilizer utilization efficiency, crop quality stability, and system intelligence. It is particularly suitable for the complex greenhouse environment of high evaporation and strong sunlight in the arid Northwest region, and has good application prospects and promotion value. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the process of the integrated water and fertilizer intelligent irrigation method in this embodiment.

[0018] Figure 2 This is a flowchart illustrating the crop transpiration model construction method in this embodiment.

[0019] Figure 3 This is a flowchart illustrating the feature weight construction method in this embodiment.

[0020] Figure 4 This is a schematic diagram of the integrated water and fertilizer intelligent irrigation device provided in this embodiment.

[0021] Figure 5 This is a schematic diagram of the structure of the electronic device provided in this embodiment. Detailed Implementation

[0022] To more clearly illustrate the present invention, the following description, in conjunction with preferred embodiments and accompanying drawings, further explains the invention. Similar components in the drawings are indicated by the same reference numerals. Those skilled in the art should understand that the specific description below is illustrative rather than restrictive and should not be construed as limiting the scope of protection of the present invention.

[0023] It should be noted that although the terms first, second, third, etc., may be used in the embodiments of this application for description, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, without departing from the scope of the embodiments of this application, first can also be referred to as second, and similarly, second can also be referred to as first.

[0024] The acquisition, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.

[0025] Specifically, the integrated water and fertilizer intelligent irrigation method described in this application is applied to a multi-layer three-dimensional water and fertilizer cultivation system in the arid Northwest region characterized by high evaporation, strong sunlight, and large diurnal temperature differences. The multi-layer three-dimensional water and fertilizer cultivation system described in this application is set up in a greenhouse planting setting for indoor cultivation of low-growing crops. For example, the low-growing crop in this application is tomato. To apply this to the above-mentioned scenarios, it is necessary to comprehensively analyze the impact of crop growth between different cultivation layers, the effects of environmental evaporation and temperature differences on crops, and the real-time absorption status of crops by water and fertilizer, so as to adjust the water and fertilizer irrigation process in real time and realize intelligent irrigation of multi-layer indoor cultivated crops based on water and fertilizer integration.

[0026] All data used in this application can be collected through historical logs and smart sensors.

[0027] To apply to the above-mentioned application scenarios, this application provides a smart irrigation method integrating water and fertilizer, the process of which can be found in the following diagram. Figure 1 As shown, it includes: Step S101: Obtain the crop growth stage of each cultivation layer, and establish a crop transpiration rate model for each cultivation layer based on the crop environment data and crop growth stage of each cultivation layer, and then output the real-time water stress index of each cultivation layer.

[0028] Specifically, by collecting real-time crop environment data and water and fertilizer irrigation data from each cultivation layer, a refined perception of the microenvironment in the multi-layered vertical cultivation system was achieved. This step provides a timely and accurate data foundation for subsequent model building and decision-making, effectively overcoming the problems of regulatory lag and resource waste caused by traditional irrigation relying on manual experience or single-point monitoring.

[0029] Please see Figure 2 The diagram shown is a flowchart illustrating the crop transpiration model construction method provided in this application, including: Step S201: Obtain the crop growth stage of each cultivation layer; the crop growth stage in this application is specifically the seedling stage, growth stage, transition stage, fruit setting stage and maturity stage. The process of obtaining the crop growth stage can be achieved by image recognition or by manual judgment based on the identification features. The specific process is a publicly known technical means familiar to those skilled in the art, and will not be described in detail in this application.

[0030] Step S202: Based on the crop environment data and crop growth stage of each cultivation layer, establish a crop transpiration rate model for each cultivation layer, and output the real-time water stress index of each cultivation layer based on the crop transpiration model.

[0031] Specifically, in step S202, the process of establishing the crop transpiration rate model for each cultivation layer is as follows: Obtain historical crop growth data and calculate the average crop transpiration rate at different crop growth stages. Use the calculation result as the baseline transpiration rate, denoted as ZT(i), i=1,2,3,4,5, where i is the growth stage number. An environmental disturbance coefficient HG(j) was constructed based on crop environmental data for each cultivation layer. HG(j) was set as c1×I(j) / Iref+c2×[T(j)-Tmin] / [Tmax-Tmin]; where I(j) is the light intensity of the j-th layer, Iref is the reference light intensity, T(j) is the air temperature of the j-th layer, Tmin and Tmax are the lower and upper limits of the suitable growth temperature range, respectively, c1 and c2 are the light stress weight and temperature stress weight, respectively, and c1+c2=1; The real-time transpiration rate SZ(j) of each cultivation layer is determined based on the crop growth stage and environmental disturbance coefficient of each cultivation layer, and SZ(j) is set as ZT(i,j)×HG(j); where ZT(i,j) represents the baseline transpiration rate of the j-th cultivation layer.

[0032] It is understood that in this application, ZT(1), ZT(2), ZT(3), ZT(4), and ZT(5) refer to the average transpiration rates of crops during the seedling stage, growth stage, transition stage, fruit setting stage, and maturity stage, respectively.

[0033] Specifically, in step S202, the transpiration state of each cultivation layer is determined by setting a transpiration rate threshold, thereby constructing a real-time water stress index for each cultivation layer: when SZ(j) is less than the transpiration rate threshold, the real-time water stress index is set to 1; when SZ(j) is greater than or equal to the transpiration rate threshold, the real-time water stress index is set to exp{[SZ(j)-transpiration rate threshold] / transpiration rate threshold}.

[0034] Specifically, the transpiration rate threshold mentioned in this application is 0.03 mm / h.

[0035] Please continue reading. Figure 2 As shown, the method for constructing the crop transpiration rate model further includes: Step S203: Construct an interlayer interference coefficient based on the interlayer environmental data between each cultivation layer, and then update the real-time water stress index based on the interlayer interference coefficient; the interlayer environmental data between each cultivation layer includes interlayer height, interlayer carbon dioxide concentration difference, and interlayer temperature difference.

[0036] Specifically, in step S203, the process of updating the real-time water stress index is as follows: Based on the interlayer environmental data between each cultivation layer, an interlayer interference coefficient β(j,j+1) is constructed. β(j,j+1) is set as β(j,j+1) = [a1×(△CO-YC) / YC+a2×(△T-YT) / YT] / H; where △CO represents the interlayer carbon dioxide concentration difference, △T represents the interlayer temperature difference, YC and YT represent the carbon dioxide concentration difference threshold and the interlayer temperature difference threshold, respectively; a1 and a2 represent the interlayer carbon dioxide difference weight and the interlayer temperature weight, respectively, and a1+a2=1; H is the interlayer height between the j-th cultivation layer and the (j+1)-th cultivation layer. |β(j,j+1)-β(j-1,j)| is used as the interlayer stress coefficient of the j-th cultivation layer, and an interlayer interference constant is set. The interlayer interference constant is compared with the interlayer stress coefficient of the j-th cultivation layer: when |β(j,j+1)-β(j-1,j)| is greater than or equal to the interlayer interference constant, the real-time water stress index of the j-th cultivation layer is updated to F(j)×ln{e+[|β(j,j+1)-β(j-1,j)|-interlayer interference constant] / interlayer interference constant}; otherwise, no update is performed; where e is the natural logarithm and F(j) is the real-time water stress index of the j-th cultivation layer.

[0037] Specifically, the interlayer interference constant described in this application is 0.3m. -1 .

[0038] Please continue reading. Figure 1 As shown, the integrated water and fertilizer intelligent irrigation method further includes: Step S102: Based on historical crop growth data, historical cultivation data is classified by feature to establish a growth-water-fertilizer mapping table. The historical crop growth data consists of historical crop growth data records for each cultivation layer, including historical fruit quality indicators, historical plant growth parameters, and quality standard values ​​of the quality indicators. The historical fruit quality indicators include, but are not limited to, water content, lycopene content, and vitamin C. The historical plant growth parameters include, but are not limited to, plant height, stem diameter, number of leaves, leaf area, number of branches, number of fruit clusters, chlorophyll content, and net photosynthetic rate. The historical cultivation data consists of historical water and fertilizer irrigation data for each cultivation layer. The feature classification results in this application include: failed configuration component data, quality-priority configuration component data, yield-priority configuration component data, and standard configuration component data, which respectively refer to the first feature classification, the second feature classification, the third feature classification, and the fourth feature classification. The quality standard values ​​of the quality indicators are obtained through user interaction input.

[0039] Specifically, by combining crop growth stages with environmental data to establish a transpiration rate model and output a water stress index, this method can dynamically reflect the actual water requirements of crops. Compared to a fixed irrigation cycle model, this method significantly improves the sensitivity and scientific rigor of water response, avoids over-irrigation or water stress, and ensures healthy crop growth.

[0040] Specifically, in step S102, the process of feature classification of historical cultivation data is as follows: Based on historical crop growth data, the fruit quality index PZ at historical maturity is calculated, and PZ is set as PZ={Σ[a(k)-A(k)] / A(k)} / K; where a(k) represents the k-th quality index of the fruit at historical maturity, A(k) represents the quality standard value of the k-th quality index, and K is the number of quality indexes of the fruit. Based on the fruit quality index and historical batch yield, historical cultivation data are classified by characteristics: when PZ is less than the quality judgment constant or CL is less than the minimum yield, the historical cultivation data corresponding to the historical crop growth data is stored as the failed configuration component data. When PZ is greater than or equal to the quality judgment constant × η and CL is less than the minimum yield × η, the historical cultivation data corresponding to the historical crop growth data is stored as the quality priority configuration component data. When PZ is less than the quality judgment constant × η and CL is greater than or equal to the minimum yield × η, the historical cultivation data corresponding to the historical crop growth data is stored as the yield priority configuration component data. When PZ is greater than or equal to the quality judgment constant × η and CL is greater than or equal to the minimum yield × η, the historical cultivation data corresponding to the historical crop growth data is stored as the standard configuration component data. Where η is a preset offset constant.

[0041] Specifically, the preset offset constant mentioned in this application is 0.2; at the same time, the quality judgment constant mentioned in this application is 0.3, and the minimum yield is set by the user based on the cultivation cost, and is not specifically limited in this application.

[0042] Specifically, in step S103, the process of establishing the growth-water-fertilizer mapping table is as follows: Similar features are extracted from the historical cultivation data after feature classification to obtain the feature configuration for each feature classification. The process is as follows: Calculate the coefficient of variation (CV(s,d) of the d-th configuration component within the historical cultivation data under the s-th feature classification, and compare the calculation result with the preset constant of variation. If CV(s,d) is greater than or equal to the preset constant of variation, it is determined that the d-th configuration component does not belong to the feature configuration component under the s-th feature classification; otherwise, it is determined that the d-th configuration component belongs to the feature configuration component under the s-th feature classification. Calculate the average value of the feature configuration components within the historical cultivation data under the s-th feature category, and use the set of average values ​​of the feature configuration components as the feature configuration of the s-th feature category; Calculate the plant growth characteristics for the s-th feature classification; The key-value pairs configured with plant growth characteristics and feature classification are used as a set of data for the growth-water-fertilizer mapping table, and the growth-water-fertilizer mapping table is constructed based on this.

[0043] It is worth noting that the process of "calculating the plant growth characteristics of the s-th feature category" in this application is the same as the process of calculating the "feature configuration of the s-th feature category", which is obtained by calculation based on historical plant growth parameters, and will not be described in detail in this application; at the same time, the value of the preset variation constant mentioned in this application is 0.15.

[0044] Please continue reading. Figure 1 As shown, the integrated water and fertilizer intelligent irrigation method further includes: Step S103: Collect crop growth data for each cultivation layer during the monitoring period, and extract crop characteristics for each cultivation layer based on the crop growth data and crop growth stage. Then, match the crop characteristics with the growth-water-fertilizer mapping table, and set the feature weights for each cultivation layer. The crop growth data is the average value of various crop growth data within the cultivation layer. The crop growth data includes the same parameter types as the historical crop growth data, including fruit quality indicators and plant growth parameters, which will not be elaborated here.

[0045] Specifically, by constructing a growth-water-fertilizer mapping table using historical data, complex agronomic experience is transformed into quantifiable and reusable configuration strategies. This step enables a shift from "trial-and-error" management to "data-driven" decision-making, significantly improving the adaptability of water and fertilizer programs and the predictability of cultivation results.

[0046] Specifically, this application does not impose a specific limit on the value of the monitoring period duration. Those skilled in the art can set it freely, as long as the value requirement of the monitoring period duration is met. In this application, 2 days is used as the duration of the monitoring period.

[0047] Please see Figure 3 The diagram shown is a flowchart illustrating the feature weight construction method provided in this application, including: Step S301: Extract crop characteristics of each cultivation layer based on crop growth data and crop growth stage of each cultivation layer.

[0048] Specifically, in step S301, the process of extracting crop features for each cultivation layer is as follows: When the crop growth stage in the cultivation layer is the fruit setting period, a crop growth bias feature vector β1(j) is constructed based on the crop growth data: the fruit quality indicators are arranged in the manner of plant growth characteristics, and each parameter in the crop growth data is used as a sub-vector of the crop growth bias feature vector. When the crop growth stage in the cultivation layer is not the fruit-setting period, a crop growth defect feature vector β2(j) is constructed based on the crop growth data: the plant growth parameters are arranged in the manner of plant growth characteristics, and each parameter in the crop growth data is used as a sub-vector of the crop growth defect feature vector.

[0049] Please continue reading. Figure 3 As shown, the feature weight construction method further includes: Step S302: Set the feature weights of each cultivation layer based on the crop characteristics of each cultivation layer.

[0050] Specifically, in step S302, the process of setting the feature weights of each cultivation layer is as follows: When the crop growth stage of the cultivation layer is the fruit setting period, calculate the cosine similarity between the crop growth bias feature vector β1(j) and the plant growth features of each feature classification, and use the calculation result as the feature weight set of the cultivation layer. When the crop growth stage in the cultivation layer is not the fruit-setting period, calculate the cosine similarity between the crop growth defect feature vector β2(j) and the plant growth features of each feature classification, and use the calculation result as the feature weight set of the cultivation layer. The feature weights of each cultivation layer are combined and denoted as S(j).

[0051] The feature weight set S(i) = [s1(j), s2(j), s3(j), s4(j)] described in this application, where s1(j), s2(j), s3(j), and s4(j) are the weights of irrigation for the j-th cultivation layer using the first feature classification, the second feature classification, the third feature classification, and the fourth feature classification, respectively.

[0052] Specifically, in this application, when calculating cosine similarity, the feature components that are the same in the plant growth characteristics of each feature category and the plant growth characteristics of each feature category are combined into feature vectors that are the same in the plant growth characteristics of each feature category and the plant growth characteristics of each feature category, and then the cosine similarity is calculated.

[0053] Please continue reading. Figure 1 As shown, the integrated water and fertilizer intelligent irrigation method further includes: Step S104: Generate water and fertilizer configuration schemes for each cultivation layer based on the real-time water stress index and feature weights of each cultivation layer.

[0054] Specifically, by matching current crop characteristics with a mapping table and setting feature weights, irrigation strategies can be dynamically adjusted according to the actual crop growth. This mechanism takes into account the priority of different growth objectives (such as quality or yield), enhancing the system's adaptability and flexibility in complex cultivation scenarios. It integrates the water stress index and feature weights to generate personalized water and fertilizer configuration schemes, achieving intelligent decision-making that integrates "environment-physiology-objective." This step not only optimizes water and fertilizer use efficiency but also proactively warns of abnormal conditions, significantly improving the stability, sustainability, and economic benefits of greenhouse tomato production.

[0055] Specifically, in step S104, the process of generating the water and fertilizer configuration scheme for each cultivation layer is as follows: Calculate the average real-time water stress index of each cultivation layer during the monitoring period, and record the result as μ(j); The selection sequence C(j) for each cultivation layer is constructed using the feature weight set of each cultivation layer and μ(j). C(j) is set as [s1(j)×μ(j)×b1,s2(j)×μ(j)×b2,s3(j)×μ(j)×b3,s4(j)×μ(j)×b4]; where b1, b2, b3, and b4 are the influence coefficients of water stress on the classification of each feature, and b1+b2+b3+b4=1. When max{C(j)} is not equal to s1(j)×μ(j)×b1, the feature configuration corresponding to the feature classification of max{C(j)} is used as the irrigation scheme for the next monitoring cycle and output to the user. Conversely, an alarm should be triggered for the user.

[0056] Specifically, in this application, b1=0.2, b2=0.2, b3=0.4, and b4=0.2.

[0057] Please see Figure 4 The diagram shown is a structural schematic of the integrated water and fertilizer intelligent irrigation device provided in this application, including: The water stress analysis unit is used to obtain the crop growth stage of each cultivation layer, and establish a crop transpiration rate model for each cultivation layer based on the crop environmental data and crop growth stage of each cultivation layer, and then output the real-time water stress index of each cultivation layer. The mapping relationship construction unit is used to classify historical cultivation data by feature based on historical crop growth data in order to establish a growth-water and fertilizer mapping relationship table. The feature weight construction unit is used to collect crop growth data of each cultivation layer during the monitoring period, extract crop features of each cultivation layer based on the crop growth data and crop growth stage, match the crop features with the growth-water-fertilizer mapping table, and then set the feature weight of each cultivation layer. The irrigation scheme generation unit is used to generate water and fertilizer configuration schemes for each cultivation layer based on the real-time water stress index and feature weights of each cultivation layer.

[0058] The integrated water and fertilizer intelligent irrigation device provided in the embodiments of this application can execute the integrated water and fertilizer intelligent irrigation method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects of the method.

[0059] From a hardware perspective, to realize the functionality of the integrated water and fertilizer intelligent irrigation method in a computer, this application also provides an electronic device; please refer to [link to relevant documentation]. Figure 5 As shown, it is a schematic diagram of the structure of the electronic device described in this application, including: The system comprises a processor 1, a memory 2, a communication interface 3, and a bus 4; wherein the processor 1 and the memory 2, and the memory 2 and the communication interface 3, transmit data via the bus 4; the processor is used to process data in the memory and generate commands, the memory is used to store data, the communication interface is used to receive and send data, and the bus is used to realize data transmission between the processor, the memory, and the communication interface.

[0060] In this embodiment, the integrated water and fertilizer intelligent irrigation method can be implemented as a runnable computer program. When the computer program is loaded into the processor or memory and processed by the processor via the bus, one or more steps of the integrated water and fertilizer intelligent irrigation can be executed.

[0061] This embodiment also provides a computer-readable storage medium for storing the computer-executable instructions. The computer-readable storage medium is a tangible physical storage medium that can store the computer program and various types of data used in the program. The physical storage medium includes, but is not limited to, existing physical storage media or combinations thereof, such as random access memory, read-only memory, optical disk, and hard disk.

[0062] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A smart irrigation method integrating water and fertilizer, characterized in that, include: The crop growth stage of each cultivation layer is obtained, and a crop transpiration rate model for each cultivation layer is established based on the crop environmental data and crop growth stage of each cultivation layer, thereby outputting the real-time water stress index of each cultivation layer. Historical crop growth data are classified by characteristics to establish a growth-water-fertilizer mapping table; Crop growth data of each cultivation layer is collected during the monitoring period, and crop characteristics of each cultivation layer are extracted based on the crop growth data and crop growth stage, and then feature weights of each cultivation layer are set. Water and fertilizer configuration schemes for each cultivation layer are generated based on the real-time water stress index and feature weights of each cultivation layer.

2. The integrated water and fertilizer intelligent irrigation method according to claim 1, characterized in that, Obtain historical crop growth data and calculate the average crop transpiration rate at different crop growth stages. Use the calculation result as the baseline transpiration rate, denoted as ZT(i), i=1,2,3,4,5. An environmental disturbance coefficient HG(j) was constructed based on crop environmental data for each cultivation layer. The real-time transpiration rate SZ(j) of each cultivation layer is determined based on the crop growth stage and environmental disturbance coefficient of each cultivation layer, and SZ(j) is set as ZT(i,j)×HG(j); where ZT(i,j) represents the baseline transpiration rate of the j-th cultivation layer.

3. The integrated water and fertilizer intelligent irrigation method according to claim 2, characterized in that, The transpiration state of each cultivation layer is determined by setting a transpiration rate threshold, and then a real-time water stress index for each cultivation layer is constructed: when SZ(j) is less than the transpiration rate threshold, the real-time water stress index is set to 1; when SZ(j) is greater than or equal to the transpiration rate threshold, the real-time water stress index is set to exp{[SZ(j)-transpiration rate threshold] / transpiration rate threshold}.

4. The integrated water and fertilizer intelligent irrigation method according to claim 3, characterized in that, Based on the interlayer environmental data between each cultivation layer, an interlayer interference coefficient β(j,j+1) is constructed. β(j,j+1) is set as β(j,j+1) = [a1×(△CO-YC) / YC+a2×(△T-YT) / YT] / H; where △CO represents the interlayer carbon dioxide concentration difference, △T represents the interlayer temperature difference, YC and YT represent the carbon dioxide concentration difference threshold and the interlayer temperature difference threshold, respectively, and a1 and a2 represent the interlayer carbon dioxide difference weight and the interlayer temperature weight, respectively, with a1+a2=1. |β(j,j+1)-β(j-1,j)| is used as the interlayer stress coefficient of the j-th cultivation layer, and an interlayer interference constant is set. The interlayer interference constant is compared with the interlayer stress coefficient of the j-th cultivation layer: when |β(j,j+1)-β(j-1,j)| is greater than or equal to the interlayer interference constant, the real-time water stress index of the j-th cultivation layer is updated to F(j)×ln{e+[|β(j,j+1)-β(j-1,j)|-interlayer interference constant] / interlayer interference constant}; otherwise, no update is performed; where e is the natural logarithm and F(j) is the real-time water stress index of the j-th cultivation layer.

5. The integrated water and fertilizer intelligent irrigation method according to claim 4, characterized in that, Based on historical crop growth data, the fruit quality index PZ at historical maturity is calculated, and PZ is set as PZ={Σ[a(k)-A(k)] / A(k)} / K; where a(k) represents the k-th quality index of the fruit at historical maturity, A(k) represents the quality standard value of the k-th quality index, and K is the number of quality indexes of the fruit. Based on the fruit quality index and historical batch yield, historical cultivation data are classified by characteristics: when PZ is less than the quality judgment constant or CL is less than the minimum yield, the historical cultivation data corresponding to the historical crop growth data is stored as the failed configuration component data. When PZ is greater than or equal to the quality judgment constant × η and CL is less than the minimum yield × η, the historical cultivation data corresponding to the historical crop growth data is stored as the quality priority configuration component data. When PZ is less than the quality judgment constant × η and CL is greater than or equal to the minimum yield × η, the historical cultivation data corresponding to the historical crop growth data is stored as the yield priority configuration component data. When PZ is greater than or equal to the quality judgment constant × η and CL is greater than or equal to the minimum yield × η, the historical cultivation data corresponding to the historical crop growth data is stored as the standard configuration component data. Where η is a preset offset constant.

6. The integrated water and fertilizer intelligent irrigation method according to claim 5, characterized in that, Similar features are extracted from the historical cultivation data after feature classification to obtain the feature configuration for each feature classification. The process is as follows: Calculate the coefficient of variation (CV(s,d) of the d-th configuration component within the historical cultivation data under the s-th feature classification, and compare the calculation result with the preset constant of variation. If CV(s,d) is greater than or equal to the preset constant of variation, it is determined that the d-th configuration component does not belong to the feature configuration component under the s-th feature classification; otherwise, it is determined that the d-th configuration component belongs to the feature configuration component under the s-th feature classification. Calculate the average value of the feature configuration components within the historical cultivation data under the s-th feature category, and use the set of average values ​​of the feature configuration components as the feature configuration of the s-th feature category; Calculate the plant growth characteristics for the s-th feature classification; The key-value pairs configured with plant growth characteristics and feature classification are used as a set of data for the growth-water-fertilizer mapping table, and the growth-water-fertilizer mapping table is constructed based on this.

7. The integrated water and fertilizer intelligent irrigation method according to claim 6, characterized in that, When the crop growth stage of the cultivation layer is the fruit setting period, calculate the cosine similarity between the crop growth bias feature vector β1(j) and the plant growth features of each feature classification, and use the calculation result as the feature weight set of the cultivation layer. When the crop growth stage in the cultivation layer is not the fruit-setting period, calculate the cosine similarity between the crop growth defect feature vector β2(j) and the plant growth features of each feature classification, and use the calculation result as the feature weight set of the cultivation layer. The feature weights of each cultivation layer are combined and denoted as S(j).

8. The integrated water and fertilizer intelligent irrigation method according to claim 7, characterized in that, Calculate the average real-time water stress index of each cultivation layer during the monitoring period, and record the result as μ(j); Construct the selection sequence C(j) for each cultivation layer using the feature weight set of each cultivation layer and μ(j); When max{C(j)} is not equal to s1(j)×μ(j)×b1, the feature configuration corresponding to the feature classification of max{C(j)} is used as the irrigation scheme for the next monitoring period and output to the user; s1(j) is the weight of the j-th cultivation layer irrigated with the first feature configuration, and b1 is the influence coefficient of water stress on the first feature classification. Conversely, an alarm should be triggered for the user.

9. A water and fertilizer integrated intelligent irrigation device, applied to the water and fertilizer integrated intelligent irrigation method as described in any one of claims 1-8, characterized in that, include: The water stress analysis unit is used to obtain the crop growth stage of each cultivation layer, and establish a crop transpiration rate model for each cultivation layer based on the crop environmental data and crop growth stage of each cultivation layer, and then output the real-time water stress index of each cultivation layer. The mapping relationship construction unit is used to classify historical cultivation data by feature based on historical crop growth data in order to establish a growth-water and fertilizer mapping relationship table. The feature weight construction unit is used to collect crop growth data of each cultivation layer during the monitoring period, extract crop features of each cultivation layer based on the crop growth data and crop growth stage, match the crop features with the growth-water-fertilizer mapping table, and then set the feature weight of each cultivation layer. The irrigation scheme generation unit is used to generate water and fertilizer configuration schemes for each cultivation layer based on the real-time water stress index and feature weights of each cultivation layer.

10. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the integrated water and fertilizer intelligent irrigation method according to any one of claims 1-8.

Images