Precise fertilization and irrigation method and system based on corn and soybean ontology information

Abstract

The invention discloses a precise fertilization and irrigation method and system based on corn and soybean ontology information, and the method comprises the following steps: building a leaf nitrogen content estimation model and a moisture estimation model based on spectral information according to the growth big data of two crops in the whole growth period, and installing an innovatively developed spectral probe sensor, an environment monitoring station is installed in the field, and the predicted nitrogen content and moisture content are obtained according to canopy spectrum information, collected by spectrum equipment, of available corn and soybean leaves. According to the precise fertilization and irrigation method and system based on the corn and soybean ontology information, by establishing the leaf nitrogen content estimation model of the corn and the soybean, the nitrogen content of the corn and the soybean in different growth stages can be precisely calculated, the nitrogen condition of crops can be accurately obtained according to actually measured spectral data, and the accuracy of fertilization and irrigation is improved. Therefore, on-demand fertilization is achieved, nutrient insufficiency or excess possibly occurring in traditional fertilization is avoided, the utilization rate of the fertilizer is increased, and waste of the fertilizer is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent water-fertilizer integration, and in particular to a method and system for precise fertilization and irrigation based on corn and soybean ontology information. Background Art

[0002] Corn and soybeans are important crops for both food and feed in my country, and are vital to national food security. In recent years, with limited arable land resources, how to effectively increase corn and soybean production without increasing the planting area to ensure my country's food security has become an important issue that needs to be solved. The corn and soybean strip composite planting model can achieve efficient resource utilization, increase land output, and increase corn and soybean production. It is one of the effective ways to increase soybean production and expand soybean area.

[0003] In the corn-soybean strip-composite planting model, the two crops have very different requirements for water and fertilizer. Corn requires a lot of water and nitrogen fertilizer during its growth, while soybeans are nitrogen-fixing crops that require less nitrogen fertilizer. In production, people manage the fields roughly and apply excessive amounts of water and fertilizer, resulting in uneven fertilization and fertilizer waste, and soybeans grow too fast and fall over, resulting in reduced yields. Although some farmers have adopted integrated water and fertilizer fertilization technology, whether it is corn-soybean strip-composite planting or monoculture of corn or soybeans, they rely more on subjective experience to judge whether there is a lack of water and fertilizer. The water and fertilizer management is extensive, which easily leads to low utilization of water and fertilizer resources.

[0004] Crop ontology information can directly reflect the nutrient and water status of crops. Previous researchers have conducted some theoretical research on monitoring and predicting the moisture and nitrogen content of crop leaves, but have not applied it to actual production. In particular, there is a lack of crop ontology information models and systems for precision fertilization and irrigation that integrate water-fertilizer integration technology and Internet of Things technology. It is urgent to invent a precision fertilization and irrigation method and system based on corn and soybean ontology information. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method and system for precise fertilization and irrigation based on corn and soybean ontology information, which aims to solve the problems raised in the above-mentioned background technology.

[0006] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0007] A precise fertilization and irrigation method based on corn and soybean ontology information comprises the following steps:

[0008] S1: Based on the big data of the full growth period of the two crops, a leaf nitrogen content estimation model and a water estimation model based on spectral information were established, and the feasibility of the model was verified on the data set of independent samples;

[0009] S2: Based on the sensitive spectral band information contained in the model constructed in step S1, an innovatively developed spectral probe sensor is installed;

[0010] S3: Install an environmental monitoring station in the field, which includes temperature and humidity, wind speed and wind direction sensors and a sensor that can sense external irradiance, and import the collected spectral reflectance data into the intelligent water-fertilizer integrated control platform;

[0011] S4: Substituting the available canopy spectral information of corn and soybean leaves collected by the spectral equipment into the model in step S1 to obtain the predicted nitrogen content and water content;

[0012] S5: The minimum thresholds of nitrogen content and water content corresponding to nutrient and water deficit are set in advance in the intelligent water-fertilizer integrated control platform. When the predicted value calculated by the model is lower than the set minimum threshold for a period of time, the system automatically makes a judgment through the built-in program, opens the dual-channel solenoid valve, and then the water and fertilizer enter the dripper belt through tube A and are finally replenished to the crops. Here, a fixed drip irrigation time is set. After the time is up, the system automatically terminates and closes the valve at the same time. After 24 hours, the spectral information of the crop canopy is collected again and substituted into the model to calculate the current nutrient and water content. If it is still lower than the set minimum threshold, the above steps are repeated. If it is higher than the set minimum threshold, the water-fertilizer integrated drip irrigation task is terminated.

[0013] Furthermore, in step S1, the nitrogen content model of corn and soybean is:

[0014] Corn: Y(corn leaf nitrogen content) = 0.1988*(RVI(696,531))2+18.756*RVI(696,531)+124.31, R2 = 0.868, R2 of independent sample test = 0.915, RMSE = 5.022;

[0015] Soybean: Y (nitrogen content of soybean leaves) = -0.0387 (RVI (723, 531))2 + 6.318 * RVI (723, 531) + 0.7848, R2 = 0.813, R2 of independent sample data set test = 0.852, RMSE = 3.962.

[0016] Furthermore, in step S1, the general model of moisture content of corn and soybean leaves is:

[0017] Y=-31.439*(NDVI(1498,1551))2+8.6383*NDVI(1498,1551)+0.2405, R2=0.565;

[0018] The independent sample test has R2=0.344 and RMSE=0.153.

[0019] Furthermore, in step S2, the spectrum probe sensor has spectrum information of five independent sensitive bands (531nm, 696nm, 723nm, 1498nm, 1551nm).

[0020] Furthermore, in step S3, if the light intensity is lower than 90,000 Lux, it is determined to be cloudy / overcast, and the collected spectral sensor data is automatically filtered out; if the light intensity is higher than 90,000 Lux, it is determined to be sunny.

[0021] A precise fertilization and irrigation system based on corn and soybean ontology information comprises a water source, a constant pressure controller, a debris filter 1, a debris filter 2, a fertilizer storage tank, a water-fertilizer integrated control platform module and an environmental monitoring station module.

[0022] The beneficial effects of the present invention are:

[0023] The invention discloses a precise fertilization and irrigation method and system based on the ontological information of corn and soybeans. By establishing a leaf nitrogen content estimation model for corn and soybeans, the nitrogen content of corn and soybeans at different growth stages can be accurately calculated, and the nitrogen status of crops can be accurately obtained according to the actually measured spectral data, thereby realizing fertilization on demand, avoiding nutrient deficiency or excess that may occur in traditional fertilization, improving fertilizer utilization, and reducing fertilizer waste.

[0024] Using the universal model of corn and soybean leaf moisture content, the moisture status of crops can be monitored in real time. The system automatically controls irrigation based on the model calculation results and the set moisture threshold to ensure that crops can obtain the right amount of moisture at different growth stages, thus avoiding the inhibition of crop growth caused by drought and preventing problems such as water waste and soil compaction caused by over-irrigation.

[0025] This method integrates an intelligent water and fertilizer integrated control platform, from spectral data collection, nutrient and moisture content calculation, to automatic water and fertilizer supply, the entire process does not require much human intervention. Once the threshold and related parameters are set, the system will automatically operate according to the actual needs of the crop, saving a lot of manpower and time costs and improving agricultural production efficiency.

[0026] The environmental monitoring station in the system can monitor environmental factors such as temperature and humidity, wind speed, wind direction and light intensity in real time. When the light intensity is lower than 90,000 Lux, the spectral sensor data is automatically filtered to avoid data errors that may occur in poor lighting conditions such as cloudy / overcast days, so that the system can better adapt to different environmental conditions and ensure the stability of data collection and fertilization and irrigation decisions.

[0027] Precision fertilization and irrigation can create a suitable growth environment for corn and soybeans, so that the crops can obtain sufficient nutrients and water throughout the growth period, which helps the development of crop roots, plant growth and photosynthesis, enhances crop resistance, reduces the occurrence of diseases and pests, and improves the quality and safety of agricultural products.

[0028] By accurately meeting the water and fertilizer needs of corn and soybeans, the growth potential of crops is fully utilized. In conjunction with other cultivation measures such as reasonable density planting, the yield and total output of corn and soybeans can be effectively increased, increasing farmers' economic benefits.

[0029] Quasi-irrigation avoids the waste of water resources caused by flooding, and performs drip irrigation according to the actual water demand of crops, thereby improving the efficiency of water resource utilization, helping to ease the tense situation of agricultural water use and achieving sustainable use of water resources.

[0030] Precision fertilization reduces excessive application of fertilizers and reduces the risk of non-point source pollution caused by fertilizers being lost into water bodies with rainwater. It also reduces problems such as soil compaction and soil pollution caused by unreasonable fertilization, which is beneficial to protecting the agricultural ecological environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the precise fertilization and irrigation method and system based on corn and soybean ontology information of the present invention;

[0032] Figure 2 It is a schematic diagram of the core components of the present invention;

[0033] Figure 3 This is a schematic diagram of the operation process of the intelligent water-fertilizer integrated control platform of the present invention;

[0034] Figure 4 Schematic diagram of the environmental monitoring station of the present invention. DETAILED DESCRIPTION

[0035] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings, wherein the same components are represented by the same reference numerals.

[0036] It should be noted that the words "front", "rear", "left", "right", "up" and "down" used in the following description refer to directions in the drawings, and the words "inside" and "outside" refer to directions toward or away from the geometric center of a specific component, respectively.

[0037] In order to make the contents of the present invention more clearly understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0038] A method and system for precise fertilization and irrigation based on corn and soybean ontology information, comprising the following steps:

[0039] 1. Based on the big data of the growth of the two crops throughout their entire growth period, a leaf nitrogen content estimation model and a water estimation model based on spectral information were established, and the feasibility of the model was verified on an independent sample data set to prove the feasibility of the model.

[0040] 2. Based on the sensitive spectral band information contained in the model constructed in step 1, an innovatively developed spectral probe sensor is installed. This probe has spectral information of five independent sensitive bands (531nm, 696nm, 723nm, 1498nm, 1551nm). There is no need to install a probe that can perceive full spectral information, and the equipment cost is low.

[0041] 3. Install an environmental monitoring station in the field. It contains conventional sensors such as temperature and humidity, wind speed, wind direction, and a sensor that can sense external irradiance to determine whether it is sunny or cloudy. If the light intensity is lower than 90,000 Lux, it is judged as cloudy, and the collected spectral sensor data is automatically filtered out. If the light intensity is higher than 90,000 Lux, it is judged as sunny, and the collected spectral reflectance data is imported into the intelligent water-fertilizer integrated control platform.

[0042] 4. Based on the available canopy spectral information of corn and soybean leaves collected by the spectral equipment, substitute it into the model in step 1 to obtain the predicted nitrogen content and moisture content.

[0043] 5. Set the minimum thresholds of nitrogen content and water content corresponding to nutrient and water deficit in advance in the intelligent water-fertilizer integrated control platform (Table 1 and Table 2). When the predicted value calculated by the model is lower than the set minimum threshold for a period of time, the system automatically makes a judgment through the built-in program, opens the dual-channel solenoid valve, and then the water and fertilizer enter the dripper belt through tube A, and finally replenishes the crops. Here, set a fixed drip irrigation time (for example, 90 minutes), and the system automatically terminates after the time is up, and closes the valve at the same time. After 24 hours, repeat the spectral information of the crop canopy and substitute it into the model to calculate the current nutrient and water content. If it is still lower than the set minimum threshold, continue to repeat the above steps. If it is higher than the set minimum threshold, the water-fertilizer integrated drip irrigation task is terminated.

[0044] 6. In the present invention, the differences in the demand for fertilizer and water and the demand period of corn and soybean are taken into consideration. In addition, the growth period of corn and soybean is in spring and summer, and they are prone to high temperature and drought weather. Compared with lack of fertilizer, water shortage is more likely to occur, and the sensitive bands of fertilizer model and water are quite different. Therefore, a separate pipeline that does not flow through the fertilizer storage barrel is set up. The pipeline is directly connected to the intelligent water-fertilizer integrated control platform. When water is needed but fertilizer is not needed, the system responds and opens one side of the dual-channel solenoid valve to allow water to flow through the pipeline alone and finally replenish the crops. By setting up such an independent pipeline, water is replenished without causing excess fertilizer.

[0045] The present invention constructs a spectral model suitable for rapid diagnosis of nitrogen and water content in corn and soybean leaves throughout their growth period, and verifies it with an independent sample set, demonstrating the feasibility of the model.

[0046] The specific nitrogen content models for corn and soybeans are:

[0047] Corn: Y (nitrogen content of corn leaves) = 0.1988*(RVI(696,531))2+18.756*RVI(696,531)+124.31, R2 = 0.868. R2 of independent sample test = 0.915, RMSE = 5.022.

[0048] Soybean: Y (nitrogen content of soybean leaves) = -0.0387 (RVI (723, 531))2 + 6.318 * RVI (723, 531) + 0.7848. R2 = 0.813, R2 of independent sample data set test = 0.852, RMSE = 3.962. The construction and test of the model show that it has high accuracy and can be applied to actual production.

[0049] The specific general model of corn and soybean leaf moisture content is Y = -31.439*(NDVI(1498,1551))2+8.6383*NDVI(1498,1551)+0.2405, R2 = 0.565. The independent sample test R2 = 0.344, RMSE = 0.153. This model calculates the real-time percentage of leaf moisture.

[0050] In the present invention, the above model is introduced into the intelligent integrated water and fertilizer control platform. By placing a developed 5-band spectral sensor (531nm, 696nm, 723nm, 1498nm, 1551nm) above the corn and soybean canopy, the spectral reflectance of the crop canopy is collected, and the spectral information is transmitted to the platform and substituted into the model for operation, the parameter information of nutrients and water content of corn and soybean can be quickly obtained.

[0051] like Figures 1 to 4 As shown, the present invention includes a water source, a constant pressure controller, a debris filter 1, a debris filter 2, a fertilizer storage tank, a water-fertilizer integrated control platform module, and an environmental monitoring station module. The present invention is described in detail below in conjunction with the accompanying drawings.

[0052] In the present invention, river water or well water is extracted through a constant pressure controller, and then debris such as silt and weeds that may clog the pipeline are filtered out through two debris filters.

[0053] The water then flows into the core component of the present invention, such as Figure 2 As shown, A. Single water flow pipe. B. Water and fertilizer pipe. C. Fertilizer storage tank: for storing liquid fertilizer. D. Dual-channel solenoid valve, which connects two pipes, and the opening and closing of the valve are controlled by the intelligent water and fertilizer integrated control platform. E. The intelligent water and fertilizer integrated platform is the core of this system. In its built-in system, the two core models of this system are stored: corn and soybean fertilizer demand model. Corn and soybean water demand model. When the results calculated by the model show that the plant is short of water, the platform controls the switch of the dual-channel solenoid valve so that only the water flow of the single water flow pipe can pass through. When the results calculated by the model show that the plant is short of nitrogen, the solenoid valve is controlled to allow only the water flow of the water and fertilizer pipe to pass through. Here, the water and fertilizer integrated control platform controls the switch direction of the solenoid valve to divert the water source and flow to pipe A and pipe B respectively. When only nutrients need to be supplemented and both nutrients and water need to be supplemented, the solenoid valve is controlled to allow the water flow to flow through pipe A. When only water needs to be supplemented, the solenoid valve is controlled to allow the water flow to flow through pipe B.

[0054] The program workflow of the intelligent water and fertilizer integrated control platform is as follows: Figure 3 After the program starts, open the valve to allow water to fill the fertilizer storage tank and automatically stir to fully mix the fertilizer and water.

[0055] After the program is started, the spectral probe in the environmental monitoring station begins to record the spectral reflectance information of the crop canopy in real time and transmits it to the system water and fertilizer integrated control platform. Figure 4 As shown, G. The spectral probe with a rainproof cover can measure the spectral reflectance of the plant canopy. H. White board. Used for reflectance calibration of the probe. It can be removed and properly stored when not in use. I. Conventional environmental monitoring sensors. Including temperature and humidity, air pressure, wind speed and direction, solar irradiance sensors, wireless transmission equipment, solar panels, etc. J. Retractable and detachable spectral fixing frame. K. The retractable and detachable fixing frame is a schematic diagram of the environmental monitoring station. The specific operation method is as follows: When farmers come to the field and want to understand the current moisture and nutrient status of field crops, start the system, first confirm that the black rubber cover is covering the probe, and collect the dark current information at this time (that is, the reflectance when there is no light, theoretically 0). Then remove the black rubber cover of the probe and place the standard reflection white board ( Figure 4 H) is placed about 10cm below the probe to collect the photocurrent information of the system (i.e. the reflectivity of the standard whiteboard, which is theoretically 1). Check whether the standard reflectivity is normal at this time through the screen of the integrated water and fertilizer control platform. If it is normal, proceed to the next step. The bracket connected to the whiteboard is retractable and foldable, and the whiteboard is also detachable. When not in use, fold the whiteboard and place it under the awning or take it away.

[0056] Among them, J and K represent fixed frames, which can be disassembled, extended and rotated. Their function is to ensure that the spectral probe can always be located about 1m above the crop canopy according to the needs of different field configurations and different crops, so that the probe can completely cover the crop canopy according to the changes in the crop growth process, so as to facilitate the collection of effective spectral reflectance information. It should be noted that Figure 4 The diagram of the installation of two spectral probes on the left and right is based on the consideration that the moisture and nutrient status of corn and soybean can be determined simultaneously based on the corn-soybean strip compound planting. If farmers use it for monocropping corn, monocropping soybean or other monocropping crops, they only need to install one spectral probe.

[0057] The available spectral reflectance is transmitted to the water and nutrient model that has been built in the system through wireless transmission to calculate the water and nutrient status of the current crop. The minimum water and nutrient parameter thresholds of the crop in each key growth period are set in advance in the program, and the parameters obtained by the model are compared with the set parameters. If any type of nutrients and water calculated by the spectral model is lower than the set threshold, the system will control the valve to open so that water and nutrients flow to the plant. Note that whether it is only lacking water or fertilizer or both, the system will control the valve to open, but the pipes that control the valve to open are different, which has been explained in the previous article. The minimum thresholds for crops in different growth periods are set in advance in the system by agricultural experts or experienced farmers.

[0058] The minimum threshold reference values ​​of moisture and nutrients of corn and soybeans provided by the present invention (Table 1 and Table 2) are as follows:

[0059]

[0060] Table 1. System setting thresholds for moisture and nitrogen fertilizer content in corn leaves

[0061]

[0062] Table 2. System setting thresholds for moisture and nitrogen fertilizer content in soybean leaves

[0063] When the system is fully opened, the system starts timing and sets the irrigation time to 90 minutes. When the time is up, the system responds and the solenoid valve closes. After the plant fully absorbs (every 24 hours), the spectral information of the crop canopy is used to monitor the water and nutrient status again. If it is still lower than the set value after irrigation, the cycle will be repeated until the value calculated by the model is higher than the set value, and the program terminates.

[0064] When the value calculated by the model is lower than the set value, it means that the crops do not need to supplement nutrients and water at this time, and there is no need for irrigation and fertilization, and the program terminates.

[0065] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A precise fertilization and irrigation method based on corn and soybean ontology information, characterized in that: The following steps are involved: S1: Based on the big data of the full growth period of the two crops, a leaf nitrogen content estimation model and a water estimation model based on spectral information were established, and the feasibility of the model was verified on the data set of independent samples; S2: Based on the sensitive spectral band information contained in the model constructed in step S1, an innovatively developed spectral probe sensor is installed; S3: Install an environmental monitoring station in the field, which includes temperature and humidity, wind speed and wind direction sensors and a sensor that can sense external irradiance, and import the collected spectral reflectance data into the intelligent water-fertilizer integrated control platform; S4: Substituting the available canopy spectral information of corn and soybean leaves collected by the spectral equipment into the model in step S1 to obtain the predicted nitrogen content and water content; S5: The minimum thresholds of nitrogen content and water content corresponding to nutrient and water deficit are set in advance in the intelligent water-fertilizer integrated control platform. When the predicted value calculated by the model is lower than the set minimum threshold for a period of time, the system automatically makes a judgment through the built-in program, opens the dual-channel solenoid valve, and then the water and fertilizer enter the dripper belt through tube A and are finally replenished to the crops. Here, a fixed drip irrigation time is set. After the time is up, the system automatically terminates and closes the valve at the same time. After 24 hours, the spectral information of the crop canopy is collected again and substituted into the model to calculate the current nutrient and water content. If it is still lower than the set minimum threshold, the above steps are repeated. If it is higher than the set minimum threshold, the water-fertilizer integrated drip irrigation task is terminated.

2. The method for precise fertilization and irrigation based on corn and soybean ontology information according to claim 1, characterized in that: In step S1, the nitrogen content model of corn and soybean is: Corn: Y(corn leaf nitrogen content) = 0.1988*(RVI(696,531))2+18.756*RVI(696,531)+124.31, R2 = 0.868, R2 of independent sample test = 0.915, RMSE = 5.022; Soybean: Y (nitrogen content of soybean leaves) = -0.0387 (RVI (723, 531))2 + 6.318 * RVI (723, 531) + 0.7848, R2 = 0.813, R2 of independent sample data set test = 0.852, RMSE = 3.

962.

3. The method for precise fertilization and irrigation based on corn and soybean ontology information according to claim 1, characterized in that: In step S1, the general model of moisture content of corn and soybean leaves is: Y=-31.439*(NDVI(1498,1551))2+8.6383*NDVI(1498,1551)+0.2405, R2=0.565; The independent sample test has R2=0.344 and RMSE=0.

153.

4. The method for precise fertilization and irrigation based on corn and soybean ontology information according to claim 1, characterized in that: In step S2, the spectrum probe sensor has spectrum information of five independent sensitive bands (531nm, 696nm, 723nm, 1498nm, 1551nm).

5. The method for precise fertilization and irrigation based on corn and soybean ontology information according to claim 1, characterized in that: In step S3, if the light intensity is lower than 90,000 Lux, it is determined to be cloudy / overcast, and the collected spectral sensor data is automatically filtered out; if the light intensity is higher than 90,000 Lux, it is determined to be sunny.

6. A precision fertilization and irrigation system based on corn and soybean ontology information, comprising a precision fertilization and irrigation method based on corn and soybean ontology information according to any one of claims 1 to 5, characterized in that: It includes a water source, a constant pressure controller, a debris filter 1, a debris filter 2, a fertilizer storage tank, a water-fertilizer integrated control platform module and an environmental monitoring station module.

Images