Intelligent rice fertilization method and system based on SOD (superoxide dismutase)

Abstract

The invention relates to the technical field of rice fertilization, in particular to an intelligent rice fertilization method and system based on SOD enzyme. Comprising the steps of setting a plurality of monitoring sub-regions based on initial data of a paddy field; generating an activity abnormal value of each monitoring sub-region according to a preset activity diagnosis model, and judging whether a risk early warning instruction is generated or not according to all the activity abnormal values; obtaining a feedback data packet according to the risk early warning instruction, and setting a fertilization compensation strategy according to the feedback data packet and a preset fertilization adjustment model; the SOD enzyme activity monitoring device is used for periodically collecting SOD enzyme activity data in each monitoring sub-region, so that whether the nutrient state of rice in each monitoring sub-region is abnormal or not is judged, early warning is carried out on the abnormal nutrient state of the rice in time, fertilization parameters are dynamically corrected, accurate monitoring of the state of the rice is achieved, and the quality of the rice is improved. The fertilization precision and the rice cultivation efficiency are improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of rice fertilization, in particular to a rice intelligent fertilization method and system based on SOD enzyme. BACKGROUND

[0002] In rice production, a long-term experience-based timed and quantitative fertilization mode is adopted, which has problems of neglecting field differences, feedback lag and low fertilizer utilization rate, and is easy to cause yield loss and environmental pollution.

[0003] Although the existing precision agriculture technology, such as soil testing or unmanned aerial vehicle remote sensing, has certain progress, there are still fundamental defects. Soil nutrient monitoring cannot reflect the crop absorption condition in real time; and remote sensing means based on leaf color, vegetation index and the like are indirect monitoring, which is easy to be disturbed by the environment, and when the crop phenotype appears visible change, the physiological stress has occurred for a long time, the diagnosis lags, and the real "predictive" management cannot be achieved. SUMMARY

[0004] The purpose of the application is to solve the above technical problems, and the application provides a rice intelligent fertilization method and system based on SOD enzyme, which aims to realize accurate monitoring of the state of rice and improve the fertilization accuracy and rice cultivation efficiency.

[0005] In some embodiments of the application, the SOD enzyme activity monitoring device is used to periodically collect SOD enzyme activity data in each monitoring sub-region, so as to judge whether the nutrient state of the rice in each monitoring sub-region is abnormal, timely alarm the abnormal state of the nutrient of the rice, dynamically correct the fertilization parameters, and realize accurate monitoring of the state of the rice.

[0006] In some embodiments of the application, by constructing a fertilization adjustment model, multiple fertilization scenarios are constructed, the fertilization parameters are accurately adjusted according to the different growth environments of the rice combined with the real-time SOD enzyme activity curve, the diagnosis interference of environmental factors on the nutrient state of the rice is avoided, and the fertilization accuracy and rice cultivation efficiency are improved.

[0007] In some embodiments of the application, a rice intelligent fertilization method based on SOD enzyme is provided, which comprises: Based on the initial data of the rice field, multiple monitoring sub-regions are set; According to the preset activity diagnosis model, the activity abnormal value of each monitoring sub-region is generated, and whether a risk warning instruction is generated is judged according to all the activity abnormal values; According to the risk warning instruction, a feedback data packet is obtained, and a fertilization compensation strategy is set according to the feedback data packet and the preset fertilization adjustment model; Among them, multiple activity points and multiple auxiliary points are arranged in a single monitoring sub-region.

[0008] In some embodiments of the present application, the preset active diagnosis model comprises: A sequence of monitoring sub-regions A is established, A=(a1, a2…an), wherein ai is the ith monitoring sub-region, and n is the number of monitoring sub-regions; i …an n ), wherein ai is the ith monitoring sub-region, and n is the number of monitoring sub-regions; i is the ith monitoring sub-region, and n is the number of monitoring sub-regions; According to the sequence of monitoring sub-regions A, ai is sequentially set as a target sub-region; i According to the sequence of monitoring sub-regions A, ai is sequentially set as a target sub-region; An active base value curve is generated according to the rice category of the target sub-region; A monitoring time axis of the target sub-region is set, and the monitoring time axis comprises a plurality of monitoring time nodes; A diagnosis sub-model of the target sub-region is set according to the active base value curve and the monitoring time axis; The diagnosis sub-models of the respective monitoring sub-regions are sequentially set; An active diagnosis model is generated according to all the diagnosis sub-models.

[0009] In some embodiments of the present application, whether to generate a risk warning instruction is determined, comprising: According to the sequence of monitoring sub-regions A, ai is sequentially set as a target sub-region; i According to the sequence of monitoring sub-regions A, ai is sequentially set as a target sub-region; The diagnosis sub-model corresponding to the to-be-diagnosed sub-region is set as a first diagnosis model; An active data packet and an auxiliary data packet of the to-be-diagnosed sub-region at a current monitoring time node are obtained; An active abnormal value b of the to-be-diagnosed sub-region at the current monitoring time node is generated; A active abnormal value threshold B1 is preset; If b>B1, a risk warning instruction of the to-be-diagnosed sub-region is generated at the current monitoring time node; Whether to generate a risk warning instruction of each monitoring sub-region is sequentially determined.

[0010] In some embodiments of the present application, the active abnormal value b of the to-be-diagnosed sub-region at the current monitoring time node is generated, comprising: b=e*[ (s i -s') 2 ]; e=U1*[ η i *v i ]; Wherein e is an abnormal compensation coefficient; θ1 is the number of active points of the to-be-diagnosed sub-region; s is' represents the SOD enzyme activity value of the i-th active site in the sub-region to be diagnosed at the current monitoring time point; s' represents the standard activity value generated based on the primary diagnostic model at the current monitoring time point; U1 is the preset first conversion coefficient; θ2 is the preset number of environmental disturbance indicators; η i v is the influencing factor of the i-th environmental disturbance index; i This represents the deviation value of the i-th environmental disturbance index in the sub-region to be diagnosed at the current monitoring time point.

[0011] In some embodiments of this application, the pre-fertilizer adjustment model includes: Multiple feature indicators are set based on historical data; Multiple fertilization scenarios were established based on all characteristic indicators; Establish a fertilization scenario sequence W, W=(w1, w2…w i …w m ), where w i Let m be the i-th fertilization scenario; m is the number of fertilization scenarios. Based on the fertilization scenario sequence W, w is set sequentially. i For targeted fertilization scenarios; Filter relevant data packages for the target fertilization scenario based on historical data; Multiple activity comparison curves for the target fertilization scenario are generated based on the associated data packets; Compensation sub-strategies are generated for each activity comparison curve, and a sub-strategy library for the target fertilization scenario is generated based on all compensation sub-strategies. Based on all activity comparison curves and the sub-strategy library, a regulatory sub-model for the target fertilization scenario is generated; The regulation sub-models for each fertilization scenario are generated sequentially, and the fertilization regulation model is constructed based on all the regulation sub-models.

[0012] In some embodiments of this application, a fertilization compensation strategy is set, including: Receive risk warning instructions; Define the monitoring sub-region corresponding to the risk warning instruction as the abnormal sub-region; Set collection instructions for abnormal sub-regions based on risk warning instructions; According to the acquisition command, obtain the feedback data packets of the sequential sub-regions. Generate first-level activity curves and first-level scenarios for abnormal sub-regions based on feedback data packets; Generate similarity values ​​between the primary scene and each fertilization scene; The regulation sub-model corresponding to the fertilization scenario with the maximum value among all similar values ​​is set as the first-level regulation model; Fertilization compensation strategies for abnormal sub-regions are set based on the first-level regulation model.

[0013] In some embodiments of the present application, the fertilization compensation strategy is set according to the primary regulation model, including: According to the primary regulation model, a series of activity contrast curves T is established; T=(t1,t2…t i …t r ), wherein t i is the i th activity contrast curve in the primary regulation model; r is the number of activity contrast curves; The fitting values between the primary activity curve and each activity contrast curve are generated in sequence; A series of fitting values H is established, H=(h1,h2…h i …h r ), wherein h i is the fitting value of the primary activity curve and the i th activity contrast curve; r is the number of activity contrast curves; A fitting value threshold H1 is preset; The maximum value h max in the series of fitting values is obtained; If h max >H1, the compensation sub-strategy of the activity contrast curve corresponding to h max is set as the fertilization compensation strategy; If h max <H1, a primary fusion instruction is generated, and the fertilization compensation strategy is generated according to the primary fusion instruction; According to the preset correction time node, it is judged whether a primary correction instruction of the fertilization compensation strategy is generated; An execution record package of the fertilization compensation strategy is generated.

[0014] In some embodiments of the present application, it also includes: According to the preset update time node, all execution record packages are obtained; According to all execution record packages, it is judged whether an update instruction of the fertilization regulation model is generated.

[0015] In some embodiments of the present application, a rice intelligent fertilization system based on SOD enzyme is provided, including: A central control unit is configured to set a plurality of monitoring sub-regions according to initial data of a rice field; A plurality of activity points and a plurality of auxiliary points are arranged in the monitoring sub-regions; An activity monitoring unit includes a plurality of activity sub-modules, and the activity sub-modules are arranged at the activity points; The activity sub-modules are configured to collect SOD enzyme activity data of the activity points; An auxiliary monitoring unit includes a plurality of auxiliary sub-modules, and the auxiliary sub-modules are arranged at the auxiliary points; The central control unit includes: The first processing module is configured to construct an activity diagnosis model; The first processing module is further configured to generate an activity abnormal value of each monitoring sub-region, and determine whether to generate a risk warning instruction according to all the activity abnormal values; The second processing module is configured to construct a fertilization adjustment model, and determine whether to generate an update instruction of the fertilization adjustment model according to a preset update time node; The third processing module is configured to obtain a feedback data packet according to the risk warning instruction; The third processing module is further configured to set a fertilization compensation strategy according to the feedback data packet and the preset fertilization adjustment model.

[0016] In some embodiments of the present application, the first processing module is further configured to: establish a monitoring sub-region sequence A, A=(a1, a2…ai…an), wherein ai is the i th monitoring sub-region, and n is the number of monitoring sub-regions; set ai as a target sub-region in sequence according to the monitoring sub-region sequence A; generate an activity base value curve according to the rice category of the target sub-region; set a monitoring time axis of the target sub-region, and the monitoring time axis includes a plurality of monitoring time nodes; set a diagnosis sub-model of the target sub-region according to the activity base value curve and the monitoring time axis; set diagnosis sub-models of each monitoring sub-region in sequence; generate an activity diagnosis model according to all the diagnosis sub-models.

[0017] Compared with the prior art, the SOD enzyme-based intelligent rice fertilization method and system has the following beneficial effects: The SOD enzyme activity monitoring device is used to periodically collect SOD enzyme activity data in each monitoring sub-region, so as to determine whether the nutrient state of the rice in each monitoring sub-region is abnormal, timely warn the abnormal nutrient state of the rice, dynamically correct the fertilization parameters, and realize accurate monitoring of the state of the rice.

[0018] By constructing the fertilization adjustment model, a plurality of fertilization scenarios are constructed, the fertilization parameters are accurately adjusted according to the real-time SOD enzyme activity curve combined with different growth environments of the rice, the diagnosis interference of environmental factors on the nutrient state of the rice is avoided, and the fertilization accuracy and the rice cultivation efficiency are improved. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a flowchart of a SOD enzyme-based intelligent rice fertilization method in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0020] The specific embodiments of the present application will be further described in conjunction with the drawings and examples. The following examples are used to illustrate the present application, but not to limit the scope of the present application.

[0021] In the description of the present application, it should be understood that the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0022] The terms "first", "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0023] In the description of the present application, it should be noted that unless otherwise specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0024] As Figure 1 shown, a rice intelligent fertilization method based on SOD enzyme according to the preferred embodiment of the present application, characterized in that, comprising: S101: setting a plurality of monitoring sub-regions based on the initial data of the rice field; S102: generating activity anomaly values of each monitoring sub-region according to a preset activity diagnosis model, and determining whether to generate a risk warning instruction according to all activity anomaly values; S103: obtaining a feedback data packet according to the risk warning instruction, and setting a fertilization compensation strategy according to the feedback data packet and a preset fertilization adjustment model; Among them, a plurality of activity points and a plurality of auxiliary points are arranged in a single monitoring sub-region.

[0025] Specifically, the initial data includes parameters affecting rice SOD enzyme activity, such as the area of ​​the paddy field, the number of rice types, soil uniformity within the paddy field, and the variability in rice sowing time. These initial data are quantified to generate corresponding initial reference values. A segmentation evaluation value is generated based on the sum of all initial reference values. A larger segmentation evaluation value indicates a greater likelihood of growth differentiation within the paddy field (i.e., differences in SOD enzyme activity curves at different locations at the same time point), and the corresponding monitoring area is smaller. The paddy field is then divided according to the monitoring area, generating multiple monitoring sub-regions, where each sub-region contains the same rice type and growth cycle. The actual area of ​​each monitoring sub-region is less than or equal to the preset monitoring area.

[0026] Specifically, the greater the influence of each parameter in the initial data on growth differentiation, the larger the corresponding initial reference value, and the same range of values ​​for the initial reference values ​​corresponding to each parameter after quantification.

[0027] Specifically, the active sites are equipped with SOD enzyme activity monitoring devices to periodically collect SOD enzyme activity data. The auxiliary sites are used to obtain soil nutrient data (content of available nitrogen, phosphorus, and potassium in the soil), meteorological data (light intensity, temperature, precipitation, and humidity) and other relevant data affecting rice growth in the monitored sub-areas. The corresponding data acquisition devices are set according to the types of data to be collected at each auxiliary monitoring site.

[0028] Specifically, the pre-defined activity diagnostic model includes: Establish a monitoring sub-region sequence A, A=(a1,a2…a3) i …a n ), where a i Let be the i-th monitoring sub-region, and n be the number of monitoring sub-regions; Based on the monitoring sub-region sequence A, a is set sequentially. i For the target sub-region; Generate an active base value curve based on the rice category of the target sub-region; Set a monitoring timeline for the target sub-region, which includes multiple monitoring time nodes; A diagnostic sub-model for the target sub-region is set based on the active base value curve and the monitoring time axis; Set up the diagnostic sub-models for each monitoring sub-region in sequence; An active diagnostic model is generated based on all diagnostic sub-models.

[0029] Specifically, the rice category of the target sub-region is acquired, the historical SOD enzyme activity data of the same category of rice is acquired according to the rice category, and the SOD enzyme normal activity level of the current rice category at different growth stages (tillering stage, jointing stage, booting stage, etc.) is generated through screening analysis, so as to construct the corresponding activity base value curve.

[0030] Specifically, the time interval corresponding to the adjacent monitoring nodes is set according to the fluctuation degree of the activity base value curve corresponding to each rice category. The greater the fluctuation degree, the shorter the corresponding time interval. The specific mapping relationship can be set according to the historical parameters.

[0031] Specifically, each point in the activity base value curve represents the normal SOD enzyme activity value of the rice in the target sub-region at the corresponding generation time node.

[0032] It can be understood that in the above embodiment, the SOD enzyme activity monitoring device is used to periodically collect the SOD enzyme activity data in each monitoring sub-region, so as to judge whether the nutrient state of the rice in each monitoring sub-region is abnormal, timely alarm the abnormal nutrient state of the rice, dynamically correct the fertilization parameters, and realize the accurate monitoring of the state of the rice.

[0033] In the preferred embodiment of the present application, whether to generate a risk warning instruction includes: According to the monitoring sub-region sequence A, a is set in turn i is a to-be-diagnosed sub-region; The diagnosis sub-model corresponding to the to-be-diagnosed sub-region is set as a first diagnosis model; The activity data packet and the auxiliary data packet of the to-be-diagnosed sub-region at the current monitoring time node are acquired; The activity abnormal value b of the to-be-diagnosed sub-region at the current monitoring time node is generated; The activity abnormal value threshold B1 is pre-set; If b>B1, the risk warning instruction of the to-be-diagnosed sub-region is generated at the current monitoring time node; Whether to generate the risk warning instruction of each monitoring sub-region is judged in turn.

[0034] Specifically, the greater the activity abnormal value, the greater the deviation of the SOD enzyme activity level of the rice in the to-be-diagnosed sub-region from the normal level, and the greater the possibility that the nutrient state of the rice is abnormal.

[0035] Specifically, the activity abnormal value b of the to-be-diagnosed sub-region at the current monitoring time node is generated, including: b=e*[ (s i -s') 2 ]; e=U1*[ η i *v i ]; wherein e is an anomaly compensation coefficient; θ1 is the number of active points of the sub-region to be diagnosed; s i is the SOD enzyme activity value of the i-th active point of the sub-region to be diagnosed at the current monitoring time node; s' is a standard activity value generated based on the first diagnosis model at the current monitoring time node; U1 is a preset first conversion coefficient; θ2 is the number of preset environmental disturbance indicators; η i is the influence factor of the i-th environmental disturbance indicator; v i is the deviation value of the i-th environmental disturbance indicator in the sub-region to be diagnosed at the current monitoring time node.

[0036] Specifically, the value of the anomaly compensation coefficient e is within a preset value range through the preset first conversion coefficient, and η i *v i The greater the value of e, the greater the value of e, and the mapping relationship between the two can be set according to historical parameters, and the value of e is always greater than 1.

[0037] Specifically, the growth cycle of the rice in the sub-region to be diagnosed at the current monitoring time node is determined, and the activity value in the activity base value curve in the first diagnosis model corresponding to the growth cycle is set, that is, the optimal SOD enzyme activity value of the rice in the sub-region to be diagnosed at the current monitoring time node.

[0038] Specifically, the environmental disturbance indicators include but are not limited to soil nutrient data (the content of available nitrogen, phosphorus and potassium in the soil), meteorological data (light intensity, temperature, precipitation, humidity) and other parameters affecting the growth state of rice, and each environmental disturbance indicator is quantitatively processed so that each environmental disturbance indicator is within the same value range, and the greater the data value of each environmental disturbance indicator (for example, the greater the nutrient content, the higher the temperature, the greater the humidity), the greater the reference value of each environmental disturbance indicator.

[0039] Specifically, the growth cycle of the rice in the sub-region to be diagnosed at the current monitoring time node is determined, and the optimal reference value of each environmental disturbance indicator (i.e. the reference value most suitable for the current rice development) of the growth cycle is obtained, and the deviation value is set according to the difference between the real-time reference values of each environmental disturbance indicator of the collection vehicle in the sub-region to be diagnosed. The greater the difference, the greater the corresponding deviation value, and the mapping relationship between the two can be set according to historical parameters.

[0040] Specifically, the influence factor of each environmental disturbance indicator is set according to its influence on the growth state of rice, and the greater the influence, the greater the value of the corresponding influence factor.

[0041] In a preferred embodiment of this application, the pre-fertilizer adjustment model includes: Multiple feature indicators are set based on historical data; Multiple fertilization scenarios were established based on all characteristic indicators; Establish a fertilization scenario sequence W, W=(w1, w2…w i …w m ), where w i Let m be the i-th fertilization scenario; m is the number of fertilization scenarios. Based on the fertilization scenario sequence W, w is set sequentially. i For targeted fertilization scenarios; Filter relevant data packages for the target fertilization scenario based on historical data; Multiple activity comparison curves for the target fertilization scenario are generated based on the associated data packets; Compensation sub-strategies are generated for each activity comparison curve, and a sub-strategy library for the target fertilization scenario is generated based on all compensation sub-strategies. Based on all activity comparison curves and the sub-strategy library, a regulatory sub-model for the target fertilization scenario is generated; The regulation sub-models for each fertilization scenario are generated sequentially, and the fertilization regulation model is constructed based on all the regulation sub-models.

[0042] Specifically, the historical data includes all fertilization records (including regular fertilization records and data on compensatory fertilization based on analysis of rice nutrient status). Specifically, the characteristic indicators include all the environmental disturbance indicators mentioned above, and also add two parameters: growth period and rice type. By quantifying all the characteristic indicators, each characteristic indicator is made to be within the same value range, and multiple value intervals are established for each characteristic indicator. Based on the random combination of all value intervals, multiple fertilization scenarios are established.

[0043] Specifically, the value ranges of all characteristic indicators in any two fertilization scenarios are not exactly the same.

[0044] Specifically, based on the value ranges of various characteristic indicators in the target fertilization scenario, relevant data matching these ranges are filtered from historical data to generate a training data package for the target fertilization scenario. Multiple activity comparison curves are then extracted from this training data package, where each curve represents an abnormal nutrient state in rice. By filtering the data in the training package, optimal fertilization remedial parameters corresponding to each activity comparison curve are generated. This allows for the setting of compensation sub-strategies for each activity comparison curve.

[0045] Specifically, the compensation sub-strategy includes a fertilizer compensation amount (which can be an increase or a decrease) and a fertilization period (i.e., how long the fertilization is completed) and environmental adjustment data (e.g., increasing or decreasing soil moisture, pest control, etc.).

[0046] It can be understood that in the above embodiments, by constructing a fertilization adjustment model, a plurality of fertilization scenarios are constructed, and the fertilization parameters are accurately adjusted according to the real-time SOD enzyme activity curve in combination with different growth environments of the rice, so as to avoid the interference of environmental factors on the diagnosis of the nutrient state of the rice and improve the fertilization accuracy and the rice cultivation efficiency.

[0047] In the preferred embodiments of the present application, the fertilization compensation strategy is set, including: obtaining a risk early warning instruction; setting a monitoring sub-area corresponding to the risk early warning instruction as an abnormal sub-area; setting a collection instruction of the abnormal sub-area according to the risk early warning instruction; obtaining a feedback data packet of the abnormal sub-area according to the collection instruction generating a first-level activity curve and a first-level scenario of the abnormal sub-area according to the feedback data packet; generating a similarity value of the first-level scenario and each fertilization scenario; setting a fertilization compensation strategy of the abnormal sub-area according to a first-level adjustment model corresponding to a maximum value in all similarity values;

[0048] Specifically, the real-time value of each feature index is obtained through each auxiliary point set in the abnormal sub-area, and the corresponding similarity value is set according to the difference between the real-time value of each feature index and the reference value of the feature index corresponding to each fertilization scenario. The smaller the difference is, the larger the corresponding similarity value is, and the mapping relationship between the two can be set according to historical parameters.

[0049] Specifically, the fertilization compensation strategy is set according to the first-level adjustment model, including: establishing an activity comparison curve sequence T according to the first-level adjustment model; T=(t1,t2…t i …t r ), wherein t i is the i th activity comparison curve in the first-level adjustment model; r is the number of activity comparison curves; generating a fitting value between the first-level activity curve and each activity comparison curve in turn; establishing a fitting value sequence H, H=(h1,h2…h i …h r ), wherein h i ​is the fitting value of the first activity curve and the i-th activity contrast curve; r is the number of activity contrast curves; a preset fitting value threshold H1 is set; obtaining the maximum value h of the fitting value sequence max ; if h max > H1, h max is set, and the compensation sub-strategy of the corresponding activity contrast curve is the fertilization compensation strategy; if h max < H1, a first fusion instruction is generated, and the fertilization compensation strategy is generated according to the first fusion instruction; a preset correction time node is used to determine whether a first correction instruction for generating the fertilization compensation strategy is generated; an execution record package of the fertilization compensation strategy is generated.

[0050] Specifically, the greater the overlap degree of the first activity curve and the current activity contrast curve, the higher the corresponding fitting value, indicating that the possibility of the abnormal problem of the nutrient state of the rice in the abnormal sub-region being the abnormal problem corresponding to the current activity contrast curve is greater.

[0051] Specifically, the first fusion instruction refers to selecting the compensation sub-strategies corresponding to the maximum value and the second maximum value of the fitting value, and weighting the fertilizer compensation amounts in the two compensation sub-strategies, the weight coefficients of which are the ratios of the fitting value of each to the sum of the fitting values of the two, so as to generate the fertilizer compensation amount corresponding to the abnormal sub-region, and generate the corresponding fertilization compensation strategy.

[0052] Specifically, the time interval between adjacent correction time nodes is set according to the fitting value, the greater the fitting value, the longer the corresponding time interval, the real-time SOD enzyme activity curve is obtained according to each correction time node, and it is determined whether the recovery speed of the SOD enzyme activity reaches the expectation, if so, the fertilization compensation strategy is not corrected, if not, the corresponding fertilizer compensation amount is adjusted appropriately.

[0053] Specifically, the adjustment parameters (i.e. the correction of the fertilizer compensation amount) of each correction time node are recorded, and the corresponding execution record package is generated after the fertilization compensation strategy is executed.

[0054] In the preferred embodiment of the present application, the following is further included: all execution record packages are obtained according to a preset update time node; an update instruction for generating the fertilization adjustment model is determined according to all the execution record packages.

[0055] Specifically, the sub-strategy library in each fertilization scene in the fertilization adjustment model is optimized in turn according to all the execution record packages, and the adjustment accuracy of the fertilization parameters is improved.

[0056] According to the SOD enzyme-based rice intelligent fertilization method in any one of the above preferred embodiments, the preferred embodiment provides a SOD enzyme-based rice intelligent fertilization system, comprising: A central control unit is configured to set a plurality of monitoring sub-regions according to initial data of the rice field; A plurality of active points and a plurality of auxiliary points are arranged in the monitoring sub-regions; An active monitoring unit comprises a plurality of active sub-modules, and each active sub-module is arranged at each active point; Each active sub-module is configured to collect SOD enzyme activity data of each active point; An auxiliary monitoring unit comprises a plurality of auxiliary sub-modules, and each auxiliary sub-module is arranged at each auxiliary point; The central control unit comprises: A first processing module is configured to construct an activity diagnosis model; The first processing module is further configured to generate activity outliers of each monitoring sub-region, and determine whether to generate a risk warning instruction according to all activity outliers; A second processing module is configured to construct a fertilization adjustment model, and determine whether to generate an update instruction of the fertilization adjustment model according to a preset update time node; A third processing module is configured to obtain a feedback data packet according to the risk warning instruction; The third processing module is further configured to set a fertilization compensation strategy according to the feedback data packet and the preset fertilization adjustment model.

[0057] In the preferred embodiment of the present application, the first processing module is further configured to: establish a monitoring sub-region sequence A, A=(a1, a2…ai…an), wherein ai is the i-th monitoring sub-region, and n is the number of monitoring sub-regions; set ai as a target sub-region in sequence according to the monitoring sub-region sequence A; generate an activity base value curve according to the rice category of the target sub-region; set a monitoring time axis of the target sub-region, and the monitoring time axis comprises a plurality of monitoring time nodes; set a diagnosis sub-model of the target sub-region according to the activity base value curve and the monitoring time axis; set diagnosis sub-models of each monitoring sub-region in sequence; generate an activity diagnosis model according to all diagnosis sub-models.

[0058] According to the first concept of the present application, the SOD enzyme activity monitoring device is used to periodically collect SOD enzyme activity data in each monitoring sub-region, so as to determine whether the nutrient state of the rice in each monitoring sub-region is abnormal, timely warn the nutrient abnormal state of the rice, dynamically correct the fertilization parameters, and realize accurate monitoring of the state of the rice.

[0059] According to the second concept of the application, by constructing a fertilization adjustment model, various fertilization scenarios are constructed, and the fertilization parameters are accurately adjusted according to the different growth environments of the rice and the real-time SOD enzyme activity curve, so as to avoid the interference of environmental factors on the diagnosis of the nutrient state of the rice and improve the fertilization accuracy and the rice cultivation efficiency.

[0060] The above is only the preferred embodiment of the application. It should be pointed out that, for ordinary skilled persons in the technical field, several improvements and replacements can be made without departing from the technical principles of the application, and these improvements and replacements should also be regarded as the protection scope of the application.

Claims

1. A smart fertilization method for rice based on SOD enzyme, characterized in that, include: Multiple monitoring sub-regions were set up based on initial data from paddy fields; Based on the preset activity diagnostic model, generate activity anomaly values ​​for each monitoring sub-region, and determine whether to generate a risk warning instruction based on all activity anomaly values. Obtain feedback data packets based on risk warning instructions, and set fertilization compensation strategies based on feedback data packets and pre-fertilization adjustment models; Within each monitoring sub-region, there are multiple active points and multiple auxiliary points.

2. The intelligent rice fertilization method based on SOD enzyme as described in claim 1, characterized in that, Pre-defined activity diagnostic models, including: Establish a monitoring sub-region sequence A, A=(a1,a2…a3) i …a n ), where a i Let be the i-th monitoring sub-region, and n be the number of monitoring sub-regions; Based on the monitoring sub-region sequence A, a is set sequentially. i For the target sub-region; Generate an active base value curve based on the rice category of the target sub-region; A monitoring timeline is set for the target sub-region, and the monitoring timeline includes multiple monitoring time nodes; A diagnostic sub-model for the target sub-region is set based on the active base value curve and the monitoring time axis; Set up the diagnostic sub-models for each monitoring sub-region in sequence; An active diagnostic model is generated based on all diagnostic sub-models.

3. The intelligent rice fertilization method based on SOD enzyme as described in claim 2, characterized in that, Determining whether to generate a risk warning instruction includes: Based on the monitoring sub-region sequence A, a is set sequentially. i The sub-region to be diagnosed; The diagnostic sub-model corresponding to the sub-region to be diagnosed is set as a first-level diagnostic model; Obtain the active data packets and auxiliary data packets of the sub-region to be diagnosed at the current monitoring time point; Generate the abnormal activity value b of the sub-region to be diagnosed at the current monitoring time point; Pretreatment activity outlier threshold B1; If b > B1, a risk warning instruction for the sub-region to be diagnosed is generated at the current monitoring time point; The system sequentially determines whether to generate risk warning instructions for each monitoring sub-area.

4. The intelligent rice fertilization method based on SOD enzyme as described in claim 3, characterized in that, Generate the abnormal activity value b of the sub-region to be diagnosed at the current monitoring time point, including: b=e*[ (s i -s') 2 ]; e=U1*[ or i *v i ]; Where e is the anomaly compensation coefficient; θ1 is the number of active points in the sub-region to be diagnosed; s i s' represents the SOD enzyme activity value of the i-th active site in the sub-region to be diagnosed at the current monitoring time point; s' represents the standard activity value generated based on the primary diagnostic model at the current monitoring time point; U1 is the preset first conversion coefficient; θ2 is the preset number of environmental disturbance indicators; η i v is the influencing factor of the i-th environmental disturbance index; i This represents the deviation value of the i-th environmental disturbance index in the sub-region to be diagnosed at the current monitoring time point.

5. The intelligent rice fertilization method based on SOD enzyme as described in claim 4, characterized in that, Pre-fertilizer regulation model, including: Multiple feature indicators are set based on historical data; Multiple fertilization scenarios were established based on all characteristic indicators; Establish a fertilization scenario sequence W, W=(w1, w2…w i …w m ), where w i Let m be the i-th fertilization scenario; m is the number of fertilization scenarios. Based on the fertilization scenario sequence W, w is set sequentially. i For targeted fertilization scenarios; Filter relevant data packages for the target fertilization scenario based on historical data; Multiple activity comparison curves for the target fertilization scenario are generated based on the associated data packets; Compensation sub-strategies are generated for each activity comparison curve, and a sub-strategy library for the target fertilization scenario is generated based on all compensation sub-strategies. Based on all activity comparison curves and the sub-strategy library, a regulatory sub-model for the target fertilization scenario is generated; The regulation sub-models for each fertilization scenario are generated sequentially, and the fertilization regulation model is constructed based on all the regulation sub-models.

6. The intelligent rice fertilization method based on SOD enzyme as described in claim 5, characterized in that, Establish a fertilizer compensation strategy, including: Receive risk warning instructions; Define the monitoring sub-region corresponding to the risk warning instruction as the abnormal sub-region; Set collection instructions for abnormal sub-regions based on risk warning instructions; According to the acquisition command, obtain the feedback data packets of the sequential sub-regions. Generate first-level activity curves and first-level scenarios for abnormal sub-regions based on feedback data packets; Generate similarity values ​​between the primary scene and each fertilization scene; The regulation sub-model corresponding to the fertilization scenario with the maximum value among all similar values ​​is set as the first-level regulation model; Fertilization compensation strategies for abnormal sub-regions are set based on the first-level regulation model.

7. The intelligent rice fertilization method based on SOD enzyme as described in claim 6, characterized in that, Fertilization compensation strategies are set according to the first-level regulation model, including: Based on the first-order regulation model, an activity comparison curve series T was established; T=(t1,t2…t i …t r ), where t i represents the i-th activity comparison curve in the first-order regulation model; r represents the number of activity comparison curves. The first-level activity curve and the fitting values ​​between each activity comparison curve are generated sequentially. Establish a sequence of fitted values ​​H, where H = (h1, h2…h…) i …h r ), where h i is the fitted value between the first-order activity curve and the i-th activity comparison curve; r is the number of activity comparison curves; Preset the fitting value threshold H1; Get the maximum value h in the fitted value series max ; If h max H1, set h max The compensator strategy for the corresponding activity comparison curve is the fertilization compensation strategy. If h max <H1, generate a first-level fusion instruction, and generate a fertilization compensation strategy according to the first-level fusion instruction; Determine whether to generate a first-level correction instruction for the fertilization compensation strategy based on the preset correction time node; Generate an execution record package for the fertilization compensation strategy.

8. The intelligent rice fertilization method based on SOD enzyme as described in claim 7, characterized in that, Also includes: Retrieve all execution record packages based on preset update time nodes; Determine whether to generate an update instruction for the fertilization adjustment model based on all execution record packages.

9. A smart rice fertilization system based on SOD enzyme, employing the smart rice fertilization method based on SOD enzyme as described in any one of claims 1-8, characterized in that, include: The central control unit is used to set up multiple monitoring sub-areas based on the initial data from the paddy field; Multiple active points and multiple auxiliary points are set up within the monitoring sub-area; An activity monitoring unit includes multiple activity sub-modules, each of which is located at a specific activity point. The active submodule is used to collect SOD enzyme activity data at each active site; The auxiliary monitoring unit includes multiple auxiliary sub-modules, each of which is located at a specific auxiliary point. The central control unit includes: The first processing module is used to construct the activity diagnostic model; The first processing module is also used to generate activity anomaly values ​​for each monitoring sub-region, and to determine whether to generate a risk warning instruction based on all activity anomaly values; The second processing module constructs a fertilization adjustment model and determines whether to generate an update instruction for the fertilization adjustment model based on a preset update time node. The third processing module is used to obtain feedback data packets based on risk warning instructions; The third processing module is also used to set a fertilization compensation strategy based on the feedback data packet and the pre-fertilization adjustment model.

10. The intelligent rice fertilization system based on SOD enzyme as described in claim 9, characterized in that, The first processing module is also used for: Establish a sequence of monitoring sub-regions A, A=(a1,a2…ai…an), where ai is the i-th monitoring sub-region and n is the number of monitoring sub-regions; Based on the monitoring sub-region sequence A, ai is sequentially set as the target sub-region; Generate an active base value curve based on the rice category of the target sub-region; A monitoring timeline is set for the target sub-region, and the monitoring timeline includes multiple monitoring time nodes; A diagnostic sub-model for the target sub-region is set based on the active base value curve and the monitoring time axis; Set up the diagnostic sub-models for each monitoring sub-region in sequence; An active diagnostic model is generated based on all diagnostic sub-models.

Images