Intelligent accounting platform for agricultural water consumption

Abstract

The invention discloses an intelligent accounting platform for agricultural water consumption, and relates to the technical field of agricultural Internet of Things. The agricultural water consumption intelligent accounting platform comprises a multi-sensor data acquisition module which is used for acquiring multi-dimensional data of an agricultural field environment in real time through multiple sensors, comprises at least one flowmeter, at least one soil humidity sensor and at least one meteorological sensor, and ensures that the acquired data is comprehensive and accurate; interference of external factors on data is eliminated, data reliability is ensured, normal operation and immediate response of the system can be ensured by edge calculation under the condition that a network is unstable, efficient closed loop from data acquisition to decision execution is realized, high efficiency and accuracy of agricultural irrigation in a complex environment are ensured, and the method is suitable for popularization and application. The problems that an irrigation plan cannot be matched with the actual water demand of crops and multi-dimensional changes of a field environment cannot be reflected in real time due to the fact that single flowmeter equipment cannot provide finer data are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural Internet of Things, and particularly to an intelligent accounting platform for agricultural water consumption. Background Art

[0002] In current agricultural water accounting platforms, flow meters are often used as the main data acquisition devices. The flow meters are installed on irrigation pipelines to measure the water flow rate per unit time to estimate the total water consumption. However, this method faces many limitations in practical applications. In diverse farmland environments, the applicable environment of flow meters is limited. In irrigation water environments with a relatively high sediment content, the flow meters may be blocked or worn, resulting in inaccurate data. The parameter acquisition is single, only reflecting the water flow rate and unable to comprehensively consider other key factors such as soil humidity, weather conditions, and crop water demand status. Moreover, in traditional systems, the accuracy of flow meters is easily affected by external factors such as water pressure fluctuations and temperature changes. Such single devices cannot provide more refined data, resulting in the inability of irrigation plans to match the actual water demand of crops and the inability to reflect the multi-dimensional changes in the field environment in real time. Summary of the Invention

[0003] (1) Technical Problems to be Solved

[0004] In view of the deficiencies of the prior art, the present invention provides an intelligent accounting platform for agricultural water consumption, which solves the problem that a single flow meter device cannot provide more refined data, resulting in the inability of irrigation plans to match the actual water demand of crops and the inability to reflect the multi-dimensional changes in the field environment in real time.

[0005] (2) Technical Solutions

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent accounting platform for agricultural water consumption, comprising:

[0007] A multi-sensor data acquisition module for real-time acquisition of multi-dimensional data of the agricultural field environment through a variety of sensors, including at least one flow meter, at least one soil humidity sensor, and at least one meteorological sensor, and the various sensors are respectively used for monitoring irrigation water flow, soil moisture, and meteorological environment;

[0008] A data processing and fusion module for receiving the multi-dimensional data from the multi-sensor data acquisition module and fusing the data of the multi-sensors based on a data fusion algorithm to generate a unified water consumption estimation value, wherein the data fusion algorithm includes weighted average, Kalman filter, or Bayesian inference algorithm for correcting data deviation caused by sensor errors or environmental factors;

[0009] An adaptive calibration module, which is used to calibrate the multi-sensor data according to real-time environmental changes and historical data, and automatically adjust the weights of each sensor, so that the system can maintain data accuracy in a changing agricultural environment;

[0010] An edge computing module, which is used to perform real-time processing and decision-making on the data locally at the sensor node or data acquisition device, and generate irrigation control instructions, and the irrigation control instructions are used to dynamically adjust the irrigation amount, irrigation time or permeability according to real-time data.

[0011] Preferably, the flowmeter is an ultrasonic flowmeter, which is used to monitor the irrigation water flow in real time through the ultrasonic principle. The flowmeter can still maintain high accuracy in the presence of sediment or other pollutants, and is not affected by the change of water flow direction and flow rate;

[0012] The soil moisture sensor includes a multi-point sensor array, which is used to monitor the change of soil moisture in real time at different soil levels, and monitor the soil moisture at different levels, so as to accurately evaluate the water demand of crops;

[0013] The meteorological sensor includes a temperature and humidity sensor, a precipitation sensor, and a wind speed sensor, which are used to monitor the field humidity and evaporation amount under different meteorological environments. Among them, the wind speed sensor is used to monitor the micro-climate change in the field and predict the evaporation rate, so as to adjust the irrigation strategy.

[0014] Preferably, the process of the data processing and fusion module generating a unified water consumption estimation value by fusing the data of multiple sensors based on the data fusion algorithm includes: collecting multi-dimensional data of the agricultural field environment in real time through multiple sensors. The data collection process includes: the flowmeter monitors the irrigation water flow, the soil moisture sensor monitors the soil moisture content, and the meteorological sensor monitors the meteorological data, and the meteorological data includes temperature and humidity, precipitation, and wind speed; then preprocess the collected multi-dimensional data, including data cleaning, denoising, filling missing values, and standardization, so that the data output by different sensors can be fused on the same scale; to ensure that the data from different sensors can be fused within the same time window, align the data through timestamps, and perform time interpolation and synchronization processing for sensors with different sampling frequencies; for the data collected by multiple sensors at different locations, use the Geographic Information System (GIS) to align the spatial positions to ensure that geographical location differences are considered when fusing data; then select a data fusion algorithm for comprehensive processing of multi-sensor data according to the agricultural water consumption estimation requirements and sensor data characteristics;

[0015] During comprehensive processing, the data from each sensor is first preliminarily estimated to obtain the estimated water consumption value for each individual sensor. Then, based on the reliability, accuracy, historical performance, and applicability to the current environment of each sensor, a weight is assigned to each sensor. For example, during the rainy season, the prediction of the meteorological sensor may be more important, while during drought, the weight of the soil moisture sensor may increase. Finally, a corresponding data fusion algorithm is used to combine the data from different sensors and output a comprehensive and unified estimated water consumption value. The fused estimated water consumption value is used as input and passed to the subsequent adaptive calibration module, edge computing module, and irrigation control system. The system dynamically adjusts the irrigation volume, irrigation time, and irrigation frequency based on this estimated value to achieve precise and real-time irrigation control.

[0016] Preferably, the adaptive calibration module includes:

[0017] A calibration rule library that stores various calibration algorithms and parameters, and selects the corresponding calibration algorithm according to environmental variables, where the environmental variables include different crops, soil types, and climate conditions;

[0018] An automatic learning mechanism that, based on historical data and real-time sensor data, automatically optimizes the weights of the sensors and the calibration model through deep learning algorithms.

[0019] Preferably, the adaptive calibration module calibrates the multi-sensor data according to real-time environmental changes and historical data, and automatically adjusts the weights of each sensor. The process is as follows:

[0020] Real-time data from different sensors is received. At the same time, historical data is obtained, and the sensor data is denoised and standardized to unify the format. Based on the calibration rule library, according to real-time environmental variables (including soil type, crop type, and climate conditions), the corresponding calibration method is automatically selected. The automatic learning mechanism trains the model through deep learning algorithms (such as deep neural networks, convolutional neural networks, long short-term memory networks, etc.) to make it learn the relationship between real-time sensor data and irrigation effects, where the irrigation effects include soil moisture changes and crop growth. Through continuous training and learning, the model automatically adjusts the weight of each sensor. For more accurate sensors, the weight will be automatically increased, while for sensors with poor data quality, the weight will be automatically decreased. Through the error feedback mechanism, based on real-time data and historical data, the readings and calibration parameters of the sensors are automatically corrected. Especially when the environment changes, such as during rainfall or temperature changes, the model will automatically adjust the weights of the sensors dynamically. After adaptive calibration, the data between sensors is fused, and the output of each sensor is adjusted according to the optimization results of its weight and calibration model to generate a unified estimated water consumption value.

[0021] Preferably, the edge computing module includes:

[0022] A real-time data processing unit, configured to receive data collected by sensors and adjust the irrigation strategy according to the set decision rules and real-time analysis results, where the decision rules are updated in real time according to the dynamic changes of the sensor data;

[0023] A local decision control unit, which can independently execute irrigation regulation operations when the data transmission is interrupted or the network is unstable, ensure the normal operation of the irrigation equipment, and synchronize the results to the cloud after the network is restored.

[0024] Preferably, the edge computing module processes data at the local node, generates irrigation control instructions according to the fused data, and the edge computing module can continue to execute irrigation control operations when the data transmission is interrupted or the network is unstable, ensuring the continuous operation of the system.

[0025] Preferably, the specific process of the edge computing module for real-time processing and decision-making of data at the sensor node or the local data acquisition device and generating irrigation control instructions is as follows:

[0026] The edge computing module connects to various sensors in the agricultural field, including flow meters, soil moisture sensors, and meteorological sensors, and real-time receives and collects data from each sensor, including key indicators such as soil moisture, temperature, precipitation, and flow rate; before the data is transmitted to the edge computing module, basic preprocessing operations are first performed, such as denoising, outlier removal, missing value filling, and data standardization;

[0027] The real-time data processing unit receives data streams from each sensor node in real time, analyzes the data in real time according to the set decision rules, and obtains real-time analysis results; the decision rules include water demand models based on different crops, dynamic responses to soil moisture changes, and predictions of meteorological data, including predictions of precipitation, temperature, and humidity; the real-time data processing unit also fuses data from multiple sensors to obtain a comprehensive estimate of agricultural water demand; the data fusion algorithms used include weighted average, Kalman filtering, Bayesian inference, etc., which can provide optimal estimates in case of data inconsistency or sensor errors;

[0028] The edge computing module updates the decision rules in real time according to the dynamic changes of sensor data, including: if the readings of a certain sensor fluctuate violently, including a sudden decrease in soil humidity, adjust the irrigation strategy to respond to the urgent water needs of the soil; if the weather sensor predicts an increase in precipitation, automatically reduce the irrigation volume; under drought or heatwave weather conditions, the system will increase the irrigation frequency and volume to ensure that the crops receive sufficient water; the local decision control unit calculates the appropriate irrigation volume, irrigation time or infiltration rate based on real-time data and the updated decision rules; this decision sends instructions to the control system or irrigation equipment to adjust the irrigation parameters in real time to ensure that the crops receive appropriate water supply according to the current environmental conditions; in the case of unstable network or data transmission interruption, the local decision control unit works independently and continues to execute the irrigation control operation to ensure that the irrigation equipment operates normally without cloud support; if a communication interruption occurs, the local decision control unit continues to execute the irrigation operation according to the last synchronized decision rules until the network is restored; after the network connection is restored, all locally executed control operations and data results are synchronized to the cloud to ensure the latest status and data update of the system;

[0029] After real-time processing and analysis of sensor data, the edge computing module generates specific irrigation control instructions; according to the current environmental conditions, including soil humidity and meteorological conditions, the irrigation instructions include irrigation volume, irrigation time and infiltration rate; the irrigation volume automatically calculates the most suitable irrigation volume according to the needs of the crops and the soil to prevent over-irrigation or under-irrigation; the irrigation time dynamically adjusts the start time and duration of irrigation according to sensor feedback and weather prediction; the infiltration rate adjusts the infiltration rate of irrigation water according to soil type and humidity changes to ensure that the water penetrates evenly into the soil; the edge computing module sends the generated irrigation instructions to the irrigation equipment through the local communication channel to ensure that the irrigation equipment operates according to the parameters calculated in real time; after the irrigation equipment executes the irrigation operation, the irrigation effect is fed back to the edge computing module as the input for the next decision, and the irrigation effect includes soil humidity changes and crop growth conditions; based on the feedback data, the edge computing module will fine-tune the irrigation strategy to cope with the changes in the real-time environment and the needs of different crops; if the soil humidity does not reach the expected target, the system will automatically increase the amount of the next irrigation; if the crops grow well, the system may reduce the irrigation frequency; after the network is restored, the edge computing module will synchronize all locally executed decision results, irrigation control instructions and feedback data to the cloud data center to ensure that the overall system status is consistent with the cloud system; by synchronizing historical data, the cloud system can update the overall irrigation plan for the farmland and conduct further analysis and optimization.

[0030] Preferably, the irrigation frequency adjustment formula in the irrigation strategy dynamically adjusted according to real-time sensor data in the decision rule is:

[0031] firrigation = f base ·(1 + β·ΔT);

[0032] Wherein, f irrigation is the adjusted irrigation frequency, f base is the basic irrigation frequency, set according to crop requirements, β is the irrigation frequency adjustment coefficient, and ΔT is the change in temperature or humidity, used to determine the adjustment of the irrigation frequency.

[0033] A control method for an intelligent accounting platform for agricultural water consumption includes the following steps:

[0034] a: Real-time collect multi-dimensional data of the agricultural field environment through a multi-sensor data acquisition module;

[0035] b: Use a data processing and fusion module to fuse the data to generate a unified water consumption estimation value;

[0036] c: Real-time calibrate the fused data through an adaptive calibration module and automatically adjust the weights of the data sources;

[0037] d: Analyze and make decisions on the real-time data based on an edge computing module, generate irrigation control instructions, and adjust the irrigation strategy according to changes in the farmland environment.

[0038] (III) Beneficial Effects

[0039] The present invention provides an intelligent accounting platform for agricultural water consumption. It has the following beneficial effects:

[0040] This intelligent accounting platform for agricultural water consumption provides data inputs from different sensors through multi-sensor fusion, ensuring that the collected data is comprehensive and accurate. It combines an adaptive data calibration algorithm to precisely correct the fused data, eliminating interference from external factors and ensuring data reliability. At the same time, using the edge computing module combined with the calibrated data, real-time data processing and decision-making are carried out locally, generating irrigation adjustment instructions and directly controlling irrigation equipment. Even in the case of unstable network, edge computing can ensure the normal operation and instant response of the system, comprehensively realizing an efficient closed-loop from data collection to decision execution, ensuring the efficiency and precision of agricultural irrigation in complex environments, and solving the problem that a single flow meter device cannot provide more refined data, resulting in the inability of the irrigation plan to match the actual water demand of the crops and the inability to reflect the multi-dimensional changes in the field environment in real time. Description of the Drawings

[0041] Figure 1 is a schematic framework diagram of the whole of the present invention;

[0042] Figure 2 is a schematic flow diagram of the present invention. Detailed Embodiments

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0044] Please refer to Figure 1 and Figure 2 , the present invention provides a technical solution: an intelligent accounting platform for agricultural water consumption, including:

[0045] A multi-sensor data acquisition module for real-time collecting multi-dimensional data of the agricultural field environment through a variety of sensors, including at least one flow meter, at least one soil humidity sensor, and at least one meteorological sensor. The various sensors are respectively used to monitor the irrigation water flow, soil moisture, and meteorological environment;

[0046] A data processing and fusion module for receiving the multi-dimensional data from the multi-sensor data acquisition module and fusing the data of the multi-sensors based on a data fusion algorithm to generate a unified water consumption estimation value. Among them, the data fusion algorithm includes weighted average, Kalman filter, or Bayesian inference algorithm, which is used to correct the data deviation caused by sensor errors or environmental factors;

[0047] An adaptive calibration module for calibrating the multi-sensor data according to real-time environmental changes and historical data, and automatically adjusting the weights of each sensor, so that the system can maintain data accuracy in a changing agricultural environment;

[0048] An edge computing module for performing real-time processing and decision-making on data at the sensor node or local data acquisition device, and generating irrigation control instructions. The irrigation control instructions are used to dynamically adjust the irrigation amount, irrigation time, or permeability according to real-time data.

[0049] The flow meter is an ultrasonic flow meter for real-time monitoring of the irrigation water flow through the ultrasonic principle. The flow meter can still maintain high accuracy in the presence of sediment or other pollutants and is not affected by the change of water flow direction and velocity;

[0050] The soil humidity sensor includes a multi-point sensor array for real-time monitoring of soil moisture changes at different soil levels, and monitoring soil humidity at different levels to accurately evaluate the water requirements of crops;

[0051] The meteorological sensor includes a temperature and humidity sensor, a precipitation sensor, and a wind speed sensor, which are used to monitor the field humidity and evaporation amount under different meteorological environments. Among them, the wind speed sensor is used to monitor the microclimate change in the field and predict the evaporation rate, so as to adjust the irrigation strategy;

[0052] The embodiments of the flow meter, soil humidity sensor, and meteorological sensor are embodied. In particular, the flow meter adopts the ultrasonic principle, which can improve the accuracy of data collection in an environment with sediment and pollutants; the soil humidity sensor introduces a multi-point sensor array to achieve precise monitoring of different soil layers; the meteorological sensor adds the function of the wind speed sensor, which can further enhance the perception of the microclimate.

[0053] The process of the data processing and fusion module fusing the data of multiple sensors based on the data fusion algorithm to generate a unified water consumption estimation value includes: real-time collecting multi-dimensional data of the agricultural field environment through multiple sensors. The data collection process includes: the flow meter monitors the irrigation water flow, the soil humidity sensor monitors the soil moisture content, and the meteorological sensor monitors the meteorological data, and the meteorological data includes temperature and humidity, precipitation, and wind speed; subsequently, preprocessing the collected multi-dimensional data, including data cleaning, denoising, filling missing values, and standardization, so that the data output by different sensors can be fused at the same scale; to ensure that the data from different sensors can be fused within the same time window, align the data through timestamps, and for sensors with different sampling frequencies, perform time interpolation and synchronization processing; for the data collected by multiple sensors at different positions, use the Geographic Information System (GIS) to align the spatial positions to ensure that geographical location differences are considered during data fusion; subsequently, according to the agricultural water consumption estimation requirements and the characteristics of sensor data, select a data fusion algorithm to comprehensively process the multi-sensor data;

[0054] The data fusion algorithms include:

[0055] The weighted average method assigns different weights to each sensor according to its reliability or accuracy, and then calculates the weighted average value of the data of each sensor. The calculation formula is as follows:

[0056]

[0057] Where, W est is the final water consumption estimation value; w i is the weight of the i-th sensor, which is assigned according to the accuracy or reliability of the sensor; x i is the measurement value of the i-th sensor, and n is the number of sensors;

[0058] Kalman filtering algorithm, which processes the data from flow meters and soil moisture sensors with noise. Kalman filtering combines the multi-dimensional data of multiple sensors through a state space model, filters the noise and obtains the optimal estimation result. The process of Kalman filtering includes prediction, update, and correction steps to eliminate the measurement errors of sensors. The calculation formula is as follows:

[0059]

[0060] P k|k =(I-K k ·H k )·P k|k-1 ;

[0061] where, is the state estimate at the current moment, such as the water consumption estimate; K k is the Kalman gain; z k is the sensor measurement value at the current moment, such as the output of the flow meter and the humidity sensor; H k is the observation matrix, indicating how to predict the observation value from the state estimate, and P k|k is the error covariance matrix, indicating the accuracy of the current estimate;

[0062] Bayesian inference algorithm, which converts the measurement data of each sensor into probability values through the calculation of prior distribution and posterior distribution, and makes inferences by combining the confidence of sensor data; Bayesian inference is used to handle the uncertainty of data and adjust the final estimation result according to the observed sensor data. The calculation formula is as follows:

[0063]

[0064] where, P(W|D) is the posterior probability, representing the probability distribution of water consumption W after the given data D; P(D|W) is the likelihood function, representing the probability of observing data D when the water consumption W is given; P(W) is the prior distribution, representing the prior belief of water consumption W; P(D) is the evidence, which is the marginalization of all possible observed data;

[0065] In the comprehensive processing, first, the data from each sensor are preliminarily estimated to obtain the individual water consumption estimation value of each sensor; then, according to the reliability, accuracy, historical performance, and applicability of the current environment of each sensor, weights are assigned to each sensor; for example, in the rainy season, the prediction of the meteorological sensor may be more important, while in the drought, the weight of the soil moisture sensor may increase; finally, the corresponding data fusion algorithm is used to combine the data of different sensors and output a comprehensive and unified water consumption estimation value;

[0066] The estimated water consumption value after fusion is used as input and transmitted to the subsequent adaptive calibration module, edge computing module, and irrigation control system. The system dynamically adjusts the irrigation amount, irrigation time, and irrigation frequency based on this estimated value to achieve precise and real-time irrigation control.

[0067] The adaptive calibration module includes a calibration rule library and an automatic learning mechanism. The rule library stores various calibration algorithms and parameters, and selects the corresponding calibration algorithm according to environmental variables, which include different crops, soil types, and climate conditions. The automatic learning mechanism automatically optimizes the weights of the sensors and the calibration model based on historical data and real-time sensor data through deep learning algorithms.

[0068] The function of the adaptive calibration module is further improved by introducing a calibration rule library and an automatic learning mechanism. The rule library can store calibration strategies suitable for different agricultural environments and can automatically adjust the calibration algorithm according to factors such as crops, soil, and climate. The automatic learning mechanism continuously optimizes the calibration process through deep learning, making the system more adaptable and accurate in long-term use.

[0069] The adaptive calibration module calibrates the multi-sensor data according to real-time environmental changes and historical data, and automatically adjusts the weights of each sensor. The process is as follows: receive real-time data from different sensors in real time. At the same time, obtain historical data, and perform denoising and standardization processing on the sensor data to unify the format. Based on the calibration rule library, automatically select the corresponding calibration method according to real-time environmental variables, which include soil type, crop type, and climate conditions.

[0070] Select the corresponding calibration method. The specific steps are as follows:

[0071] For the soil type, according to the soil type, including sandy soil and clay, select a suitable calibration algorithm.

[0072] For the crop type, according to the crop type, including rice, wheat, and corn, automatically adjust the calibration algorithm. Rice has a high demand for water, while corn may require less water.

[0073] For the climate conditions, according to different climate conditions, including drought, rainfall, and temperature, adjust the calibration rules. In the rainy season, meteorological sensors are more important, while in the dry season, the calibration of soil moisture sensors is more critical.

[0074] The automatic learning mechanism trains a model through deep learning algorithms (such as deep neural networks, convolutional neural networks, long short-term memory networks, etc.) to enable it to learn the relationship between real-time sensor data and irrigation effects, where the irrigation effects include soil moisture changes and crop growth. Through continuous training and learning, the model automatically adjusts the weights of each sensor. For more accurate sensors, the weights will be automatically increased, while for sensors with poor data quality, the weights will be automatically decreased. Through the error feedback mechanism, based on real-time data and historical data, the readings and calibration parameters of the sensors are automatically corrected. Especially when the environment changes, such as during rainfall or temperature changes, the model will automatically adjust the weights of the sensors dynamically. After adaptive calibration, the data between sensors is fused, and the output of each sensor is adjusted according to its weight and the optimization results of the calibration model to generate a unified water consumption estimation value.

[0075] According to environmental changes and real-time data, the weights of each sensor are dynamically adjusted to ensure that the system can always generate accurate water consumption estimation results from accurate data. The soil moisture changes and crop growth conditions in real-time irrigation effects are fed back to the adaptive calibration module, and then the calibration parameters and weights are updated according to the feedback to continuously optimize the calibration process of sensor data. As new data is continuously input, the automatic learning mechanism will continuously optimize the deep learning model and adjust the calibration algorithm through historical data, enabling the adaptive calibration module to maintain optimal performance in a changing agricultural environment.

[0076] In the dry season, when the climate conditions change, the calibration rule base will automatically select an algorithm suitable for the arid environment and enhance the weight of the soil moisture sensor. Through the automatic optimization of the deep learning model, the system will dynamically adjust the irrigation strategy according to sensor data and crop requirements during real-time irrigation.

[0077] The adaptive calibration module continuously adjusts the weights of each sensor through dynamically selecting calibration algorithms adapted to environmental changes, an automatic optimization mechanism based on deep learning, and real-time data feedback, thereby ensuring that the system can always maintain high-precision calibration of sensor data under different agricultural environmental conditions. This module can improve the accuracy and robustness of agricultural water consumption estimation and ultimately achieve precise irrigation management.

[0078] For the sensor calibration model in the adaptive calibration module, for each sensor i, under the given environmental variables E (such as soil type, crop type, climate conditions, etc.), calibration is performed using a calibration function:

[0079]

[0080] where is the calibrated sensor data; f i (x i, E) is the calibration function of sensor i, which performs calibration based on the environmental variable E and the raw measurement value x i for calibration;

[0081] For weight automatic adjustment, the weight w i is automatically adjusted as the environmental variable and historical data change. An error feedback mechanism is used to optimize the weight of each sensor. The calculation formula is as follows:

[0082]

[0083] where, w i (t) is the weight of the i-th sensor at time t; α is the learning rate, which controls the sensitivity of weight adjustment; errori(t) is the error of the i-th sensor at time t, such as the deviation between prediction and actual data; error i threshold is the error threshold of the sensor. When the error exceeds this threshold, the weight will be significantly adjusted.

[0084] The edge computing module includes a real-time data processing unit and a local decision control unit. The real-time data processing unit is used to receive the data collected by the sensors and adjust the irrigation strategy according to the set decision rules and real-time analysis results. The decision rules are updated in real time according to the dynamic changes of the sensor data; the local decision control unit can independently execute the irrigation regulation operation when the data transmission is interrupted or the network is unstable, ensure the normal operation of the irrigation equipment, and synchronize the results to the cloud after the network is restored;

[0085] The edge computing module processes data at the local node and generates irrigation control instructions based on the fused data. Moreover, the edge computing module can continue to execute the irrigation control operation when the data transmission is interrupted or the network is unstable, ensuring the continuous operation of the system.

[0086] The specific process of the edge computing module for real-time data processing and decision-making at the sensor node or local data acquisition device and generating irrigation control instructions is as follows:

[0087] The edge computing module is connected to various sensors in the agricultural field, including flow meters, soil moisture sensors, and meteorological sensors, and receives and collects data from each sensor in real time, including key indicators such as soil moisture, air temperature, precipitation, and flow rate; before the data is transmitted to the edge computing module, basic preprocessing operations are first performed, such as denoising, outlier removal, missing value filling, and data standardization;

[0088] The real-time data processing unit receives data streams from each sensor node in real time, and analyzes the data in real time according to the set decision rules to obtain real-time analysis results. The decision rules include moisture demand models based on different crops, dynamic responses to soil moisture changes, and predictions of meteorological data, including predictions of precipitation, temperature, and humidity. The real-time data processing unit also fuses data from multiple sensors to obtain a comprehensive estimate of agricultural water demand. The data fusion algorithms used include weighted average, Kalman filtering, Bayesian inference, etc., which can provide optimal estimates in case of data inconsistency or sensor errors.

[0089] The edge computing module updates the decision rules in real time according to the dynamic changes of sensor data, including: if the readings of a certain sensor fluctuate violently, including a sudden decrease in soil moisture, the irrigation strategy is adjusted to respond to the urgent need for water in the soil; if the meteorological sensor predicts an increase in precipitation, the irrigation amount is automatically reduced; under drought or heatwave weather conditions, the system increases the irrigation frequency and amount to ensure that the crops receive sufficient water. The local decision control unit calculates the appropriate irrigation amount, irrigation time, or infiltration rate based on the real-time data and the updated decision rules. This decision sends instructions to the control system or irrigation equipment to adjust the irrigation parameters in real time to ensure that the crops receive appropriate water supply according to the current environmental conditions. In the case of unstable network or data transmission interruption, the local decision control unit works independently and continues to execute the irrigation control operation to ensure that the irrigation equipment operates normally without cloud support. If a communication interruption occurs, the local decision control unit continues to execute the irrigation operation according to the last synchronized decision rules until the network is restored. After the network connection is restored, all locally executed control operations and data results are synchronized to the cloud to ensure the latest status and data update of the system.

[0090] After real-time processing and analysis of sensor data, the edge computing module generates specific irrigation control instructions; according to the current environmental conditions, including soil moisture and meteorological conditions, the irrigation instructions include irrigation volume, irrigation time, and infiltration rate; the irrigation volume automatically calculates the most suitable irrigation volume according to the needs of the crops and soil to prevent over-irrigation or under-irrigation; the irrigation time dynamically adjusts the start time and duration of irrigation based on sensor feedback and weather prediction; the infiltration rate adjusts the infiltration rate of irrigation water according to soil type and moisture changes to ensure uniform infiltration of water into the soil; the edge computing module sends the generated irrigation instructions to the irrigation equipment through the local communication channel to ensure that the irrigation equipment operates according to the parameters calculated in real time; after the irrigation equipment executes the irrigation operation, the irrigation effect is fed back to the edge computing module as the input for the next decision-making, and the irrigation effect includes soil moisture changes and crop growth conditions; based on the feedback data, the edge computing module will fine-tune the irrigation strategy to cope with the changes in the real-time environment and the needs of different crops; if the soil moisture does not reach the expected target, the system will automatically increase the amount of the next irrigation; if the crops grow well, the system may reduce the irrigation frequency;

[0091] After the network is restored, the edge computing module will synchronize all decision results, irrigation control instructions, and feedback data executed locally to the cloud data center to ensure that the overall system state is consistent with the cloud system; by synchronizing historical data, the cloud system can update the overall irrigation plan for the farmland and conduct further analysis and optimization;

[0092] Summary: The edge computing module generates irrigation control instructions by performing real-time data processing and decision-making locally at the sensor nodes or data acquisition devices, realizing the dynamic adjustment of agricultural water consumption; this module has real-time performance, independence, and adaptability, and can adjust the irrigation strategy in real time according to environmental changes. At the same time, it can also keep the irrigation equipment running normally in case of unstable network, ensuring the continuity of agricultural production and irrigation efficiency;

[0093] In the edge computing module, irrigation control instructions are generated through real-time data processing and decision-making; based on sensor data and preset decision rules, the irrigation volume, time, and infiltration rate are adjusted; the calculation formula for the irrigation volume is as follows:

[0094] I amount =(W target -W current )·C crop ·C soil ·(1-C weather );

[0095] Where, I amount is the amount of water to be irrigated (unit: cubic meters); W target is the target soil moisture (or water content); W currentis the current soil humidity (or moisture content); C rop is the moisture demand coefficient of the crop, adjusted according to the crop type; C soil is the influence coefficient of the soil type, adjusted according to the soil structure, such as clay, sandy soil; C weather is the influence coefficient of meteorological conditions, such as precipitation, temperature, which affects the water evaporation rate;

[0096] The irrigation time calculation formula is as follows:

[0097]

[0098] where, T irrigation is the irrigation time required (unit: hour); I amount is the irrigation amount; Q flow is the water flow rate of the irrigation system (unit: cubic meters per hour).

[0099] The decision rule dynamically adjusts the irrigation strategy according to real-time sensor data. The irrigation frequency adjustment formula in the irrigation strategy is:

[0100] f irrigation = f base ·(1 + β·ΔT);

[0101] where, f irrigation is the adjusted irrigation frequency, f base is the basic irrigation frequency, set according to the crop demand, β is the irrigation frequency adjustment coefficient, and ΔT is the change in temperature or humidity, used to determine the adjustment of the irrigation frequency.

[0102] As Figure 2 shown, a control method for an intelligent accounting platform for agricultural water consumption includes the following steps:

[0103] a: Real-time collect multi-dimensional data of the agricultural field environment through a multi-sensor data acquisition module;

[0104] b: Use a data processing and fusion module to fuse the data and generate a unified water consumption estimation value;

[0105] c: Through an adaptive calibration module, perform real-time calibration on the fused data and automatically adjust the weights of the data sources;

[0106] d: Based on an edge computing module, analyze and make decisions on real-time data, generate irrigation control instructions, and adjust the irrigation strategy according to changes in the farmland environment.

[0107] It should be further noted that in the specific implementation process, multi-sensor fusion provides data input from different sensors to ensure comprehensive and accurate data collection. The fused data is precisely corrected by combining an adaptive data calibration algorithm to eliminate the interference of external factors on the data and ensure data reliability. At the same time, the edge computing module is used to combine the calibrated data to perform real-time data processing and decision-making locally, generate irrigation adjustment instructions and directly control irrigation equipment. Even in the case of unstable network, edge computing can ensure the normal operation and instant response of the system, comprehensively realizing an efficient closed-loop from data collection to decision execution, ensuring the efficiency and precision of agricultural irrigation in complex environments, and solving the problem that a single flowmeter device cannot provide more refined data, resulting in the inability of the irrigation plan to match the actual water demand of crops and the inability to reflect the multi-dimensional changes in the field environment in real time.

[0108] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0109] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent agricultural water consumption calculation platform, characterized in that: include: A multi-sensor data acquisition module, used to collect multi-dimensional data of the agricultural field environment in real time through a variety of sensors, including at least one flow meter, at least one soil moisture sensor and at least one meteorological sensor, wherein the multiple sensors are used to monitor irrigation water flow, soil moisture and meteorological environment respectively; A data processing and fusion module, used for receiving the multi-dimensional data from the multi-sensor data acquisition module, and fusing the data of the multi-sensors based on a data fusion algorithm to generate a unified water consumption estimation value, wherein the data fusion algorithm includes a weighted average, a Kalman filter or a Bayesian reasoning algorithm, which is used for correcting data deviations caused by sensor errors or environmental factors; An adaptive calibration module, used to calibrate the multi-sensor data according to real-time environmental changes and historical data, and automatically adjust the weight of each sensor; The edge computing module is used to process and make decisions on the data in real time locally at the sensor node or data acquisition device, and generate irrigation control instructions, which are used to dynamically adjust the irrigation amount, irrigation time or infiltration rate according to the real-time data.

2. The intelligent agricultural water consumption calculation platform according to claim 1 is characterized by: The flow meter is an ultrasonic flow meter, which is used to monitor the irrigation water flow in real time through the ultrasonic principle; The soil moisture sensor includes a multi-point sensor array for real-time monitoring of soil moisture changes at different soil levels and monitoring soil moisture at different levels; The meteorological sensors include a temperature and humidity sensor, a precipitation sensor, and a wind speed sensor, which are used to monitor field humidity and evaporation under different meteorological environments. The wind speed sensor is used to monitor field microclimate changes and predict evaporation rate.

3. The intelligent agricultural water consumption calculation platform according to claim 2 is characterized by: The data processing and fusion module fuses the data of multiple sensors based on the data fusion algorithm to generate a unified water consumption estimation value. The process includes: real-time collection of multi-dimensional data of agricultural field environment by multiple sensors, followed by pre-processing of the collected multi-dimensional data, including data cleaning, denoising, filling missing values ​​and standardization, aligning the data by timestamps, and performing time interpolation and synchronization processing for sensors with different sampling frequencies; aligning the spatial positions of the data collected by multiple sensors at different locations using the geographic information system (GIS); then, according to the agricultural water consumption estimation requirements and sensor data characteristics, selecting the data fusion algorithm for comprehensive processing of the multi-sensor data; during the comprehensive processing, firstly making a preliminary estimate of the data from each sensor to obtain a separate water consumption estimation value for each sensor; then, assigning a weight to each sensor according to the reliability, accuracy, historical performance and applicability of each sensor to the current environment; finally, using the corresponding data fusion algorithm, combining the data from different sensors to output the water consumption estimation value.

4. The intelligent agricultural water consumption calculation platform according to claim 3 is characterized by: The adaptive calibration module comprises: A calibration rule base storing a variety of calibration algorithms and parameters, and selecting a corresponding calibration algorithm according to environmental variables, including different crops, soil types, and climate conditions; An automatic learning mechanism that automatically optimizes sensor weights and calibration models through a deep learning algorithm based on historical data and real-time sensor data.

5. The intelligent agricultural water consumption calculation platform according to claim 4 is characterized by: The adaptive calibration module calibrates the multi-sensor data according to real-time environmental changes and historical data, and automatically adjusts the weight of each sensor. The process is as follows: Receive real-time data from different sensors in real time, obtain historical data, and perform denoising, standardization, and unified formatting on sensor data; automatically select the corresponding calibration method based on the calibration rule base according to real-time environmental variables, including soil type, crop type, and climate conditions; the automatic learning mechanism trains the model through a deep learning algorithm to learn the relationship between real-time sensor data and irrigation effects, including soil moisture changes and crop growth; the model automatically adjusts the weight of each sensor through continuous training and learning; Through the error feedback mechanism, the sensor readings and calibration parameters are automatically corrected based on real-time and historical data; after adaptive calibration, the data between sensors are fused, and the output of each sensor is adjusted according to its weight and the optimization result of the calibration model to generate a unified water consumption estimate.

6. The intelligent agricultural water consumption calculation platform according to claim 5 is characterized by: The edge computing module includes: A real-time data processing unit, which is used to receive data collected by sensors and adjust irrigation strategies according to set decision rules and real-time analysis results, wherein the decision rules are updated in real time according to dynamic changes in sensor data; The local decision control unit can independently perform irrigation control operations when data transmission is interrupted or the network is unstable, ensuring the normal operation of the irrigation equipment and synchronizing the results to the cloud after the network is restored.

7. The intelligent agricultural water consumption calculation platform according to claim 6 is characterized by: The edge computing module processes data at a local node and generates irrigation control instructions based on the fused data. The edge computing module can continue to perform irrigation control operations when data transmission is interrupted or the network is unstable, thereby ensuring the continuous operation of the system.

8. The intelligent agricultural water consumption calculation platform according to claim 7 is characterized by: The specific process of the edge computing module processing and making decisions on data in real time at the sensor node or data acquisition device and generating irrigation control instructions is as follows: the edge computing module receives and collects data from various sensors in real time by connecting to various sensors at the agricultural site; the real-time data processing unit receives data streams from each sensor node in real time, and analyzes the data in real time according to the set decision rules to obtain real-time analysis results; the real-time data processing unit also fuses data from multiple sensors to obtain an estimated value of agricultural water demand; The edge computing module updates the decision rules in real time according to the dynamic changes of sensor data; the local decision control unit calculates the irrigation amount, irrigation time or penetration rate based on the real-time data and the updated decision rules; in the case of network instability or data transmission interruption, the local decision control unit works independently and continues to perform irrigation control operations to ensure that the irrigation equipment maintains normal operation without cloud support; if there is a communication interruption, the local decision control unit continues to perform irrigation operations according to the last synchronized decision rules until the network is restored; after the network connection is restored, all locally executed control operations and data results are synchronized to the cloud; after real-time processing and analysis of sensor data, the edge computing module generates irrigation control instructions; The edge computing module sends the generated irrigation instructions to the irrigation equipment through the local communication channel.

9. The intelligent agricultural water consumption calculation platform according to claim 8 is characterized by: The decision rule dynamically adjusts the irrigation frequency in the irrigation strategy according to the real-time sensor data, and the adjustment formula is: f irrigation =f base ·(1+β·ΔT); Among them, f irrigation is the adjusted irrigation frequency, f base is the basic irrigation frequency, which is set according to crop demand, β is the irrigation frequency adjustment coefficient, and ΔT is the temperature or humidity change, which is used to determine the adjustment of irrigation frequency.

10. A control method for the intelligent agricultural water consumption accounting platform according to any one of claims 1 to 9, characterized in that: The steps include: a: Real-time collection of multi-dimensional data of agricultural field environment through multi-sensor data acquisition module; b: using a data processing and fusion module to fuse the data and generate a unified water consumption estimation value; c: The fused data is calibrated in real time through the adaptive calibration module, and the weight of the data source is automatically adjusted; d: Analyze and make decisions based on real-time data based on the edge computing module, generate irrigation control instructions, and adjust irrigation strategies according to changes in the farmland environment.

Images