Intelligent rice fertilizer efficiency regulation and control system based on multi-source data

Abstract

The invention discloses an intelligent rice fertilizer efficiency regulation and control system based on multi-source data, and particularly relates to the technical field of intelligent agriculture. A temperature gradient value is generated by continuously collecting field temperature data to calculate a time change rate, and a predicted deviation value is generated by synchronously comparing a meteorological predicted temperature with a real-time temperature; environment temperature and humidity parameters are separated and processed, and independent weight coefficients are applied to generate thermal index values; comparing the gradient value, the deviation value and the thermal index value with a preset critical threshold value, a tolerance range and a thermal threshold value in real time to generate a three-level dynamic adjustment amount, executing weighted average operation by adopting a weight coefficient, and acting a result on a basic fertilization threshold value to generate a self-adaptive threshold value; and finally, performing proportional fine adjustment on the fertilizing amount. According to the method, quick response of environment sudden change is realized, decision timeliness of a temperature sudden change scene is improved, a high-temperature and high-humidity composite stress effect is effectively quantified, fertilization precision is optimized through a multi-factor dynamic compensation mechanism, and the problem that environment parameter coupling influence of a traditional system is not quantified is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent agriculture, and in particular to a rice fertilizer efficiency intelligent regulation system based on multi-source data. BACKGROUND

[0002] The field of intelligent agriculture technology involves the practice of optimizing agricultural production processes using modern information technology. The core content involves the integration of data collection, analysis, and decision support systems, which are applied to crop management, resource allocation, and efficiency improvement. Overall, it covers sensor deployment, Internet of Things connection, artificial intelligence algorithm processing, and automated execution equipment, achieving intelligent and precise agricultural operations.

[0003] Among them, the rice fertilizer efficiency intelligent regulation system based on multi-source data refers to the problem of inaccurate fertilizer effect regulation in rice planting. This topic covers information acquisition from soil moisture, weather parameters, crop growth status and other multi-source data, automatic adjustment of fertilization operations based on input data through data integration and rule-driven mechanisms. The means include real-time data collection, preset threshold comparison and fertilization equipment linkage control.

[0004] The existing technology relies on isolated data source processing mode, meteorological data, soil data and crop data are analyzed independently, and there is a lack of multi-parameter spatio-temporal correlation mechanism. Therefore, we propose a rice fertilizer efficiency intelligent regulation system based on multi-source data. SUMMARY

[0005] The main purpose of the present application is to provide a rice fertilizer efficiency intelligent regulation system based on multi-source data, which solves the technical problems of environmental mutation response lag, multi-factor coupling influence unquantized, and large deviation between meteorological data and field measurement leading to inaccurate fertilization decision-making in the prior art by continuously calculating the temperature change rate, quantifying the meteorological prediction deviation, separating and weighting the temperature and humidity parameters, implementing three-level dynamic compensation and soil humidity closed-loop correction.

[0006] To achieve the above purpose, the technical scheme adopted by the present application is: A rice fertilizer efficiency intelligent regulation system based on multi-source data, comprising: A real-time gradient calculation module: calling rice field temperature sensor data stream, selecting temperature readings at two consecutive time points, performing subtraction operation to obtain difference, dividing by sampling interval time unit, calculating change rate, generating temperature gradient value; A prediction deviation processing module: calling meteorological application program interface future temperature prediction data, calling real-time temperature sensor readings in rice field, performing subtraction operation to obtain prediction value minus real-time value result, applying absolute value operation processing, generating temperature prediction deviation value; The heat index generation module: calls the real-time data of the ambient temperature sensor, calls the real-time data of the ambient humidity sensor, applies a multiplication operation to process the temperature value and the temperature weight coefficient, applies a multiplication operation to process the humidity value and the humidity weight coefficient, performs an addition operation to integrate the weighted results, and generates a heat index value; The threshold dynamic decision module: calls the preset threshold parameter of the rice growth stage, calls the temperature gradient value, compares the value with the critical gradient threshold parameter, sets the gradient adjustment amount according to the comparison result, calls the temperature prediction deviation value, compares the value with the tolerance range parameter, sets the prediction adjustment amount according to the comparison result, calls the heat index value, compares the value with the heat threshold parameter, sets the heat adjustment amount according to the comparison result, applies a weighted average operation to integrate each adjustment amount, calculates the average value based on the weight coefficient parameter, applies the average value to the basic fertilization threshold parameter, and generates an adaptive fertilization threshold; The fertilization control output module: calls the adaptive fertilization threshold, calls the real-time data of the soil humidity sensor, compares the value with the standard humidity range parameter, adjusts the fertilization amount proportion according to the comparison result, performs a proportional operation to calculate the final fertilization amount, and generates a fertilization control amount.

[0007] Preferably, the temperature gradient value is specifically a temperature change rate value, the temperature prediction deviation value is specifically a predicted and actual temperature difference value, the heat index value includes a temperature weighted component and a humidity weighted component, the adaptive fertilization threshold includes a gradient adjustment amount, a prediction adjustment amount, and a heat adjustment amount, and the fertilization control amount is specifically a fertilization amount proportion.

[0008] Preferably, the real-time gradient calculation module includes a data acquisition sub-module, a difference calculation sub-module, and a change rate generation sub-module. The data acquisition sub-module acquires a real-time data stream of a rice field temperature sensor, selects temperature readings at two consecutive time points, records the readings, and generates time point temperatures. The difference calculation sub-module calls the time point temperature values, acquires the first time point temperature value and the second time point temperature value, performs a subtraction operation to calculate a difference value, and generates a temperature difference value. The change rate generation sub-module calls the temperature difference value, calls a sampling interval parameter, performs a division operation to calculate a change rate, and generates a temperature gradient value.

[0009] Preferably, the prediction deviation processing module includes a data acquisition sub-module, a difference operation sub-module, and a deviation generation sub-module. The data acquisition sub-module acquires a future three-hour temperature prediction value of a meteorological application program interface and acquires a real-time temperature sensor reading of a rice field, records the value, and generates a temperature value pair. The difference operation sub-module calls the temperature value pair, extracts the predicted temperature value and the real-time temperature value, performs a subtraction operation to calculate a difference value, and generates a temperature difference value. The deviation generation submodule calls the temperature difference value, applies absolute value calculation to process the difference value, and generates the temperature prediction deviation value.

[0010] Preferably, the heat index generation module includes a temperature-weighted submodule, a humidity-weighted submodule, and a heat index integration submodule. The temperature weighting submodule acquires real-time data from the ambient temperature sensor, processes the temperature value and the temperature weighting coefficient 0.7 using multiplication operations, and generates a temperature weighting value. The humidity weighting submodule acquires real-time data from the ambient humidity sensor, processes the humidity value and the humidity weighting coefficient 0.3 using multiplication operations, and generates a humidity weighting value. The heat index integration submodule calls the temperature weighted value and the humidity weighted value, performs an addition operation to integrate the two weighted results, and generates the heat index value.

[0011] Preferably, the threshold dynamic decision-making module includes a gradient adjustment calculation submodule, a prediction adjustment calculation submodule, a hot adjustment calculation submodule, and an adjustment integration submodule: The gradient adjustment calculation submodule calls the temperature gradient value, calls the critical gradient threshold parameter, compares the temperature gradient value with the critical gradient threshold parameter, if the temperature gradient value is greater than the critical gradient threshold parameter, sets the gradient adjustment amount to -1%, otherwise sets the gradient adjustment amount to zero, and generates the gradient adjustment amount value. The prediction adjustment calculation submodule calls the temperature prediction deviation value and the tolerance range parameter, compares the temperature prediction deviation value with the tolerance range parameter. If the temperature prediction deviation value is greater than the tolerance range parameter, the prediction adjustment amount is set to the temperature prediction deviation value multiplied by negative zero point two; otherwise, the prediction adjustment amount is set to zero, and the prediction adjustment amount value is generated. The heat adjustment calculation submodule calls the heat index value and the heat threshold parameter, compares the heat index value with the heat threshold parameter, and sets the heat adjustment amount to negative two percentage points if the heat index value is greater than the heat threshold parameter; otherwise, it sets the heat adjustment amount to zero and generates the heat adjustment amount value. The adjustment and integration submodule calls the gradient adjustment value, the prediction adjustment value, the thermal adjustment value, and the weight coefficient parameters gradient zero point four, prediction zero point three, and thermal zero point three. It calculates the gradient adjustment value multiplied by the gradient weight coefficient parameter, adds the prediction adjustment value multiplied by the prediction weight coefficient parameter, adds the thermal adjustment value multiplied by the thermal weight coefficient parameter, and obtains the weighted average value. It then calls the basic fertilization threshold parameter, applies the weighted average value to the basic fertilization threshold parameter, and generates the adaptive fertilization threshold.

[0012] Preferably, the fertilization control output module includes a parameter calling submodule, an adjustment calculation submodule, and a control quantity generation submodule: The parameter call submodule obtains real-time data from the soil moisture sensor, calls the adaptive fertilization threshold parameter, records the data and the threshold, and generates control input values. The adjustment calculation submodule calls the control input value, extracts the soil moisture value and fertilization threshold, compares the soil moisture value with the standard moisture range parameter, sets the adjustment ratio parameter based on the comparison result, and generates the fertilization adjustment value. The control quantity generation submodule calls the control input value, calls the fertilization adjustment value, applies proportional calculation to calculate the final fertilization amount, and generates the fertilization control quantity.

[0013] Compared with the prior art, the present invention has the following beneficial effects: This invention generates a temperature gradient value by continuously monitoring the rate of temperature change and combining the field temperature difference collected at adjacent time points with a fixed sampling interval. This processing logic breaks through the limitations of traditional static temperature monitoring, capturing the instantaneous impact of environmental abrupt changes on rice metabolism. It introduces the calculation of the absolute deviation between meteorological forecast data and real-time sensor data, establishing a quantitative mechanism for prediction reliability and reducing decision-making errors caused by the lag in meteorological data. A separate weighting strategy for temperature and humidity parameters is adopted, multiplying each parameter by independent coefficients and then superimposing them to generate a heat index value, accurately quantifying the combined stress effect of high temperature and high humidity, and avoiding distortion in the assessment of a single environmental parameter. A three-level dynamic compensation mechanism is set up, including gradient adjustment, prediction adjustment, and heat adjustment. Independent compensation items are generated based on real-time comparison results of critical gradient thresholds, tolerance ranges, and heat thresholds. Weighted average calculations are performed using preset weight coefficients, and the results are applied to the basic fertilization threshold, achieving quantitative fusion of the coupled effects of multiple environmental factors. Finally, by combining real-time comparison of soil moisture with the standard range, the fertilization amount is proportionally fine-tuned to form a closed-loop control. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the working module of the intelligent rice fertilizer efficiency regulation system based on multi-source data in an embodiment of the present invention; Figure 2 This is a schematic diagram of the workflow of the threshold dynamic decision module in some embodiments of the present invention; Detailed Implementation

[0015] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the linguistic context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0016] It should be understood that the terms "system," "device," "unit," and / or "module" as used in this specification are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0017] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0018] The following is a detailed description of the intelligent rice fertilizer efficiency regulation system based on multi-source data provided in the embodiments of this specification, with reference to the accompanying drawings.

[0019] like Figure 1 As shown, the intelligent rice fertilizer efficiency regulation system based on multi-source data provided in this embodiment includes: a real-time gradient calculation module, a prediction deviation processing module, a heat index generation module, a threshold dynamic decision-making module, and a fertilizer control output module.

[0020] Specifically, the real-time gradient calculation module calls the data stream from the rice field temperature sensor, selects the temperature readings at two consecutive time points, performs a subtraction operation to obtain the difference, divides it by the sampling interval time unit, calculates the rate of change, and generates the temperature gradient value. The prediction deviation processing module calls the meteorological application interface to obtain future temperature prediction data, calls the real-time temperature sensor readings in the rice field, performs a subtraction operation to obtain the result of subtracting the real-time value from the predicted value, applies absolute value operation to process the result, and generates the temperature prediction deviation value. The heat index generation module calls real-time data from the ambient temperature sensor and real-time data from the ambient humidity sensor. It applies multiplication to process the temperature value and temperature weighting coefficient, applies multiplication to process the humidity value and humidity weighting coefficient, performs addition to integrate the weighted results, and generates the heat index value. The threshold dynamic decision-making module calls the preset threshold parameters for the rice growth stage, calls the temperature gradient value, compares the value with the critical gradient threshold parameter, sets the gradient adjustment amount based on the comparison result, calls the temperature prediction deviation value, compares the value with the tolerance range parameter, sets the prediction adjustment amount based on the comparison result, calls the heat index value, compares the value with the heat threshold parameter, sets the heat adjustment amount based on the comparison result, applies a weighted average calculation to integrate the adjustment amounts, calculates the average value based on the weight coefficient parameter, applies the average value to the basic fertilization threshold parameter, and generates an adaptive fertilization threshold. The fertilization control output module calls the adaptive fertilization threshold, calls the real-time data of the soil moisture sensor, compares the value with the standard moisture range parameter, adjusts the fertilization ratio according to the comparison result, performs proportional calculation to calculate the final fertilization amount, and generates the fertilization control amount.

[0021] Specifically, the temperature gradient value is the temperature change rate value, the temperature prediction deviation value is the difference between the predicted and actual temperatures, the heat index value includes a temperature-weighted component and a humidity-weighted component, the adaptive fertilization threshold includes a gradient adjustment amount, a prediction adjustment amount, and a heat adjustment amount, and the fertilization control amount is the fertilization ratio.

[0022] Specifically, in some embodiments of the invention, the real-time gradient calculation module includes a data acquisition submodule, a difference calculation submodule, and a rate of change generation submodule; The data acquisition submodule acquires real-time data streams from temperature sensors in the rice field, selects two consecutive time points for temperature readings, records these readings, and generates the time point temperature. The difference calculation submodule calls the time point temperature values, acquires the temperature values ​​of the first and second time points, performs a subtraction operation to calculate the difference, and generates the temperature difference. The rate of change generation submodule calls the temperature difference, calls the sampling interval parameter, performs a division operation to calculate the rate of change, and generates the temperature gradient value.

[0023] For example, the real-time gradient calculation module reads the raw values ​​of the temperature sensor at two consecutive time points (e.g., T1=28.5℃, T2=29.1℃) from the data acquisition submodule, generating a temperature value pair at each time point. The difference calculation submodule calls this value pair to perform a subtraction operation (29.1-28.5=0.6℃), outputting the temperature difference. The rate of change generation submodule, combined with a preset sampling interval parameter (Δt=10 minutes), divides the temperature difference by the time interval (0.6℃ / 10min), ultimately generating a temperature gradient value of 0.06℃ / min. When a sensor failure is detected, the system automatically switches to a backup node and triggers an alarm signal.

[0024] Furthermore, in some embodiments of this application, the prediction deviation processing module includes a data acquisition submodule, a difference calculation submodule, and a deviation generation submodule; The data acquisition submodule acquires the temperature forecast value for the next three hours from the meteorological application interface, obtains the real-time temperature sensor readings in the rice field, records the values, and generates temperature value pairs. The difference calculation submodule calls the temperature value pairs, extracts the predicted temperature value and the real-time temperature value, performs a subtraction operation to calculate the difference, and generates a temperature difference. The deviation generation submodule calls the temperature difference, applies absolute value operation to process the difference, and generates a temperature prediction deviation value.

[0025] The data acquisition submodule synchronously calls the meteorological API to obtain the predicted temperature for the next 3 hours (e.g., 30.2℃) and the real-time field temperature (e.g., 28.7℃), packaging them into temperature value pairs. The difference calculation submodule unpacks the data and performs algebraic subtraction (30.2-28.7=1.5℃). The deviation generation submodule applies an absolute value operation to the difference to generate a temperature prediction deviation value of 1.5℃. If the meteorological API does not respond, the locally stored ARIMA prediction model is used for data compensation.

[0026] In some embodiments of the present invention, the heat index generation module includes a temperature-weighted submodule, a humidity-weighted submodule, and a heat index integration submodule: The temperature weighting submodule acquires real-time data from the ambient temperature sensor, applies a multiplication operation to process the temperature value and the temperature weighting coefficient 0.7, and generates a temperature weighted value. The humidity weighting submodule acquires real-time data from the ambient humidity sensor, applies a multiplication operation to process the humidity value and the humidity weighting coefficient 0.3, and generates a humidity weighted value. The thermal index integration submodule calls the temperature weighted value and the humidity weighted value, performs an addition operation to integrate the two weighted results, and generates a thermal index value.

[0027] For example, the temperature-weighted submodule reads the raw temperature value (e.g., 30℃), multiplies it by a temperature weighting coefficient of 0.7 stored in the EEPROM, and outputs a temperature-weighted value of 21. The humidity-weighted submodule simultaneously reads the relative humidity value (e.g., 80%), multiplies it by a humidity weighting coefficient of 0.3, and outputs a humidity-weighted value of 24. The heat index integration submodule performs algebraic addition (21+24) on the two weighted values ​​to generate a heat index value of 45. The weighting coefficients are dynamically adjusted according to the rice growth stage: 0.6 (temperature) / 0.4 (humidity) is used during the tillering stage, and 0.7 / 0.3 is used during the heading stage.

[0028] like Figure 2 As shown, in some embodiments of the present invention, the threshold dynamic decision-making module includes a gradient adjustment calculation submodule, a prediction adjustment calculation submodule, a hot adjustment calculation submodule, and an adjustment integration submodule: The gradient adjustment calculation submodule calls the temperature gradient value, calls the critical gradient threshold parameter, compares the temperature gradient value with the critical gradient threshold parameter, if the temperature gradient value is greater than the critical gradient threshold parameter, sets the gradient adjustment amount to -1%, otherwise sets the gradient adjustment amount to zero, and generates the gradient adjustment amount value. The prediction adjustment calculation submodule calls the temperature prediction deviation value and the tolerance range parameter, compares the temperature prediction deviation value with the tolerance range parameter. If the temperature prediction deviation value is greater than the tolerance range parameter, the prediction adjustment amount is set to the temperature prediction deviation value multiplied by negative zero point two; otherwise, the prediction adjustment amount is set to zero, and the prediction adjustment amount value is generated. The heat adjustment calculation submodule calls the heat index value and the heat threshold parameter, compares the heat index value with the heat threshold parameter, and sets the heat adjustment amount to negative two percentage points if the heat index value is greater than the heat threshold parameter; otherwise, it sets the heat adjustment amount to zero and generates the heat adjustment amount value. The adjustment and integration submodule calls the gradient adjustment value, the prediction adjustment value, the thermal adjustment value, and the weight coefficient parameters gradient zero point four, prediction zero point three, and thermal zero point three. It calculates the gradient adjustment value multiplied by the gradient weight coefficient parameter, adds the prediction adjustment value multiplied by the prediction weight coefficient parameter, adds the thermal adjustment value multiplied by the thermal weight coefficient parameter, and obtains the weighted average value. It then calls the basic fertilization threshold parameter, applies the weighted average value to the basic fertilization threshold parameter, and generates the adaptive fertilization threshold.

[0029] It should be noted that the gradient adjustment calculation submodule compares the temperature gradient value (e.g., 0.6℃ / min) with the critical gradient threshold (0.5℃ / min for Nanjing 46) in the variety parameter library. When the gradient value exceeds the threshold, the gradient adjustment amount is set to -1%. The prediction adjustment calculation submodule compares the predicted temperature deviation value (e.g., 6℃) of the treatment with the tolerance range parameter (5℃). When the deviation exceeds the limit, the prediction adjustment amount is set to -1.2% (calculation formula: -0.2×6). The heat adjustment calculation submodule compares the heat index value of the treatment (e.g., 29) with the heat threshold parameter (28). When the threshold is exceeded, the heat adjustment amount is set to -2%. The adjustment integration submodule performs a three-level weighted operation, calls the basic fertilization threshold of 20g / ㎡ at the jointing stage, and generates an adaptive fertilization threshold of 19.73g / ㎡ after applying the weighted adjustment amount. The weighting coefficients are automatically switched according to day and night: 0.4 (gradient) / 0.3 (prediction) / 0.3 (heat index) during the day and 0.3 / 0.4 / 0.3 at night.

[0030] Furthermore, in this embodiment of the invention, the fertilization control output module includes a parameter calling submodule, an adjustment calculation submodule, and a control quantity generation submodule: The parameter call submodule acquires real-time data from the soil moisture sensor, calls the adaptive fertilization threshold parameter, records the data and the threshold, and generates the control input value. The adjustment calculation submodule calls the control input value, extracts the soil moisture value and the fertilization threshold, compares the soil moisture value with the standard moisture range parameter, sets the adjustment ratio parameter based on the comparison result, and generates the fertilization adjustment value. The control quantity generation submodule calls the control input value, calls the fertilization adjustment value, applies the ratio calculation to calculate the final fertilization amount, and generates the fertilization control quantity.

[0031] The parameter call submodule reads soil moisture sensor data (e.g., 45%) and adaptive fertilization threshold (19.73 g / m²), and packages it as control input values. The adjustment calculation submodule compares the soil moisture value to be processed with the standard moisture range [40%, 60%]: when the moisture is within the range, the fertilization adjustment value is set to 0 (if 38% moisture is detected, it is set to -5%). The control quantity generation submodule calculates the final fertilization amount (19.73 g / m² × (1 + 0) = 19.73 g / m²), converts it into a PWM signal with a 63% duty cycle to drive the fertilization valve. When the soil moisture > 70%, the hardware protection circuit is triggered to forcibly interrupt fertilization.

[0032] For example, the parameter call submodule reads soil moisture sensor data (e.g., 45%) and the adaptive fertilization threshold (19.73 g / m²), and packages it as a control input value. The adjustment calculation submodule compares the soil moisture value to be processed with the standard moisture range [40%, 60%]: when the moisture is within the range, the fertilization adjustment value is set to 0 (if 38% moisture is detected, it is set to -5%). The control quantity generation submodule calculates the final fertilization amount (19.73 g / m² × (1 + 0) = 19.73 g / m²), and converts it into a PWM signal with a 63% duty cycle to drive the fertilization valve. When the soil moisture > 70%, the hardware protection circuit is triggered to forcibly interrupt fertilization.

[0033] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A rice fertilizer efficiency intelligent regulation system based on multi-source data, characterized in that, Comprise: Real-time gradient calculation module: call rice field temperature sensor data stream, select two consecutive time point temperature readings, perform subtraction operation to obtain difference value, divide by sampling interval time unit, calculate rate of change, generate temperature gradient value; Forecast bias processing module: call weather application program interface future temperature prediction data, call real-time rice field temperature sensor readings, perform subtraction operation to obtain predicted value minus real-time value result, apply absolute value operation processing, generate temperature prediction bias value; Heat index generation module: call environmental temperature sensor real-time data, call environmental humidity sensor real-time data, apply multiplication operation processing temperature value and temperature weight coefficient, apply multiplication operation processing humidity value and humidity weight coefficient, perform addition operation to integrate weighted results, generate heat index value; Threshold dynamic decision module: call rice growth stage preset threshold parameter, call temperature gradient value, compare the value with critical gradient threshold parameter, set gradient adjustment amount according to comparison result, call temperature prediction bias value, compare the value with tolerance range parameter, set prediction adjustment amount according to comparison result, call heat index value, compare the value with heat threshold parameter, set heat adjustment amount according to comparison result, apply weighted average operation to integrate each adjustment amount, calculate average value based on weight coefficient parameter, apply average value to basic fertilization threshold parameter, generate adaptive fertilization threshold; Fertilization control output module: call adaptive fertilization threshold, call soil humidity sensor real-time data, compare the value with standard humidity range parameter, adjust fertilization amount proportion according to comparison result, perform proportional operation to calculate final fertilization amount, generate fertilization control amount. 2.The system according to claim 1, wherein, The temperature gradient value is specifically a temperature rate of change value, the temperature prediction bias value is specifically a predicted and actual temperature difference value, the heat index value includes temperature weighted component and humidity weighted component, the adaptive fertilization threshold includes gradient adjustment amount, prediction adjustment amount and heat adjustment amount, and the fertilization control amount is specifically a fertilization amount proportion. 3.The rice fertility intelligent regulation and control system based on multi-source data according to claim 1, characterized in that, The real-time gradient calculation module comprises a data acquisition submodule, a difference calculation submodule and a rate generation submodule; The data acquisition submodule acquires real-time rice field temperature sensor data stream, selects two consecutive time point temperature readings, records the readings, and generates time point temperatures; The difference calculation submodule calls the time point temperature values, obtains a first time point temperature value and a second time point temperature value, performs subtraction operation to calculate difference value, and generates temperature difference value; The rate generation submodule calls the temperature difference value, calls the sampling interval parameter, performs division operation to calculate rate of change, and generates temperature gradient value. 4.The system according to claim 1, wherein, The prediction bias processing module comprises a data acquisition submodule, a difference operation submodule and a bias generation submodule; The data acquisition submodule acquires weather application program interface future three-hour temperature prediction value and real-time rice field temperature sensor readings, records the values, and generates temperature value pair; The difference operation submodule calls the temperature value pair, extracts predicted temperature value and real-time temperature value, performs subtraction operation to calculate difference value, and generates temperature difference value; The bias generation submodule calls the temperature difference value, applies absolute value operation processing to the difference value, and generates temperature prediction bias value.

5. The multi-source data-based rice fertilizer efficiency intelligent regulation system according to claim 1, characterized in that, The heat index generation module comprises a temperature weighting submodule, a humidity weighting submodule, and a heat index integration submodule: The temperature weighting submodule acquires real-time data of an ambient temperature sensor, applies a multiplication operation to process a temperature value and a temperature weight coefficient 0.7, and generates a temperature weighting value; The humidity weighting submodule acquires real-time data of an ambient humidity sensor, applies a multiplication operation to process a humidity value and a humidity weight coefficient 0.3, and generates a humidity weighting value; The heat index integration submodule calls the temperature weighting value, calls the humidity weighting value, performs an addition operation to integrate the two weighting results, and generates a heat index value. 6.The system according to claim 1, wherein, The threshold dynamic decision module comprises a gradient adjustment calculation submodule, a prediction adjustment calculation submodule, a heat adjustment calculation submodule, and an adjustment integration submodule: The gradient adjustment calculation submodule calls the temperature gradient value, calls the critical gradient threshold parameter, compares the temperature gradient value with the critical gradient threshold parameter, sets the gradient adjustment amount to -1% if the temperature gradient value is greater than the critical gradient threshold parameter, otherwise sets the gradient adjustment amount to zero, and generates a gradient adjustment value; The prediction adjustment calculation submodule calls the temperature prediction deviation value, calls the tolerance range parameter, compares the temperature prediction deviation value with the tolerance range parameter, sets the prediction adjustment amount to the temperature prediction deviation value multiplied by -0.2 if the temperature prediction deviation value is greater than the tolerance range parameter, otherwise sets the prediction adjustment amount to zero, and generates a prediction adjustment value; The heat adjustment calculation submodule calls the heat index value, calls the heat threshold parameter, compares the heat index value with the heat threshold parameter, sets the heat adjustment amount to -2% if the heat index value is greater than the heat threshold parameter, otherwise sets the heat adjustment amount to zero, and generates a heat adjustment value; The adjustment integration submodule calls the gradient adjustment value, calls the prediction adjustment value, calls the heat adjustment value, calls the weight coefficient parameters gradient 0.4, prediction 0.3, and heat 0.3, calculates the weighted average value of the gradient adjustment value multiplied by the gradient weight coefficient parameter, plus the prediction adjustment value multiplied by the prediction weight coefficient parameter, plus the heat adjustment value multiplied by the heat weight coefficient parameter, calls the basic fertilization threshold parameter, and applies the weighted average value to the basic fertilization threshold parameter to generate an adaptive fertilization threshold.

7. The multi-source data-based rice fertilizer efficiency intelligent regulation system according to claim 1, characterized in that, The fertilization control output module comprises a parameter calling submodule, an adjustment calculation submodule, and a control amount generation submodule: The parameter calling submodule acquires real-time data of a soil humidity sensor, calls the adaptive fertilization threshold parameter, records the data and the threshold, and generates a control input value; The adjustment calculation submodule calls the control input value, extracts the soil humidity value and the fertilization threshold, compares the soil humidity value with the standard humidity range parameter, sets the adjustment proportion parameter according to the comparison result, and generates a fertilization adjustment value; The control amount generation submodule calls the control input value, calls the fertilization adjustment value, applies a proportional operation to calculate the final fertilization amount, and generates a fertilization control amount.

Images