Intelligent summer corn transpiration measuring and calculating method and system based on silicon accumulation data

Abstract

The invention relates to the technical field of agricultural water management, in particular to an intelligent summer corn transpiration measuring and calculating method and system based on silicon accumulation data, and the method comprises the steps: collecting canopy spectrum and environmental data, inverting layered silicon accumulation, screening steady-state parameters, constructing silicon gradient mapping, and predicting transpiration and future water consumption. According to the method, silicon element distribution characteristics in the vertical direction of a canopy are analyzed by constructing physical layered mapping logic, double calibration is performed on a remote sensing inversion result in combination with laboratory chemical analysis data, a physiological stability interval is delimited by using a statistical method, abnormal environmental noise is eliminated, and a variable step size iteration solving mechanism based on a cumulative gradient is constructed. The calculation precision is adaptively adjusted according to the metabolic activity degree, fine analysis of the moisture transmission process in different physiological states is realized, accumulative errors caused by environmental fluctuation and spatial heterogeneity are eliminated, and it is ensured that the prediction result of the moisture consumption rate of the whole summer corn plant can truly reflect the actual water consumption demand of crops.

Description

Technical Field

[0001] This invention relates to the field of agricultural water management technology, and in particular to an intelligent method and system for calculating the transpiration of summer maize based on silicon accumulation data. Background Technology

[0002] The field of agricultural water management technology refers to a collection of technologies related to the acquisition, allocation, utilization, and regulation of water resources in agricultural production. These technologies include crop water requirement analysis, field water status monitoring, irrigation system formulation and implementation, and quantitative calculation of crop water consumption. By combining crop physiological characteristics, meteorological factors, soil moisture parameters, and field observation data, the dynamics of farmland water are systematically managed to support irrigation decisions and water resource allocation. In grain crop cultivation, particular attention is paid to the quantitative description of crop transpiration and evapotranspiration processes and their relationship with environmental and crop growth indicators.

[0003] Among them, the intelligent measurement method of summer maize transpiration refers to collecting plant growth indicators and environmental data based on the division of crop growth stages, and combining the accumulation of silicon in the plant as a physiological characteristic parameter. It usually calculates and estimates the transpiration of summer maize per unit time by recording the changes in leaf silicon content at different growth stages, combined with meteorological data such as temperature, sunshine, wind speed and air humidity, as well as morphological indicators such as plant height and leaf area index, thereby completing a quantitative description of the crop water consumption process.

[0004] Existing technologies rely solely on recording changes in leaf silicon content and combining it with conventional meteorological data for estimation. This ignores the spatial variability in silicon accumulation along the vertical direction of the crop canopy. Using only a single or averaged indicator makes it difficult to accurately characterize the actual physiological activity of different parts of the plant. Furthermore, directly using unfiltered fluctuating environmental data in the calculation can easily introduce noise interference from unsteady states. It also ignores the dynamic changes in metabolic intensity at different accumulation stages, resulting in a lack of adaptive adjustment mechanisms for physiological states in the calculation process. Consequently, the final quantitative description of water consumption deviates significantly from the actual physiological process. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an intelligent method and system for calculating the transpiration of summer maize based on silicon accumulation data.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent method for calculating the transpiration of summer maize based on silicon accumulation data, comprising the following steps: S1: Collect spectral images of the summer maize canopy, divide the summer maize canopy into upper, middle and lower regions, and simultaneously collect environmental parameters within the monitoring area to generate a multi-source monitoring dataset; S2: Calculate the reflectance of characteristic bands in the upper, middle and lower parts of the summer maize canopy based on the multi-source monitoring dataset, determine the silicon accumulation in the upper, middle and lower parts of the summer maize canopy, and generate the multi-dimensional calibrated silicon accumulation of the entire summer maize plant. S3: Define the physiological stability range of silicon in summer maize, extract environmental parameters within the monitoring area from the multi-source monitoring dataset, screen each environmental parameter corresponding to the moment when the multi-dimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of silicon in summer maize, and generate a steady-state environmental parameter set. S4: Obtain the global range of changes in silicon accumulation during the summer maize growth cycle and divide the silicon accumulation gradient intervals of summer maize. Construct a silicon accumulation gradient step size mapping table based on the intervals. S5: Referring to the steady-state environmental parameter set, perform inversion calculation of transpiration in the upper, middle and lower parts of the summer maize canopy, match it with the silicon cumulative gradient step size mapping table, predict the future water consumption rate of summer maize, and obtain the total transpiration calculation result.

[0007] The present invention is improved in that the multi-source monitoring dataset includes spectral images of the upper canopy, middle canopy, and lower canopy of summer maize, as well as temperature, humidity, wind speed, and light intensity of the monitoring area. The multi-dimensional calibrated silicon accumulation of the entire summer maize plant includes silicon accumulation in the upper, middle, and lower canopy of summer maize after calibration based on comparison deviation. The steady-state environmental parameter set includes temperature, humidity, wind speed, and light intensity of the monitoring area corresponding to the selected values ​​falling within the physiologically stable silicon range of summer maize. The silicon accumulation gradient step size mapping table specifically represents the mapping relationship data between the silicon accumulation gradient interval of summer maize within its growth cycle and the step size for solving summer maize transpiration. The total transpiration calculation result includes the sum of transpiration from the upper, middle, and lower canopy parts of summer maize, the total transpiration of summer maize, and the predicted future water consumption rate of summer maize.

[0008] The present invention is improved in that step S1 is specifically as follows: S101: Control a drone equipped with remote sensing equipment to collect spectral images of the upper part of the summer maize canopy from a vertical perspective at the top of the canopy, and control a ground mobile robot with sensor lenses to collect spectral images of the middle part and the lower part of the summer maize canopy. Analyze the spatial location information of the spectral images of the upper, middle and lower parts of the summer maize canopy and map them to the corresponding physical layering regions to generate layered spectral images. S102: Synchronously collect environmental parameters within the monitoring area, including air temperature, air humidity, wind speed, and light intensity values, verify the timestamp of each value collection, and generate synchronous environmental monitoring parameters; S103: Using the timestamp of data acquisition as the reference index, perform association matching, map the matched synchronous environmental monitoring parameters to the corresponding image entries in the layered spectral image, remove redundant data with misaligned timestamps, and obtain a multi-source monitoring dataset.

[0009] The present invention is improved in that step S2 is specifically as follows: S201: Extract the layered spectral image from the multi-source monitoring dataset, extract the grayscale value of the layered spectral image and perform spectral radiometric calibration operation with reference to the standard whiteboard data, determine the reflection intensity value of the red band and the reflection intensity value of the near-infrared band respectively, and generate the regional characteristic band reflectance. S202: Combining the regional characteristic band reflectivity, the pre-measured summer maize leaf area and summer maize plant height and standardizing them, calculate the silicon element precipitation values ​​in the upper, middle and lower parts of the summer maize canopy, respectively, as the estimated silicon accumulation in the region; S203: Collect leaf samples of summer maize under multiple growth conditions and test them to obtain the measured value of silicon content. Calculate the numerical deviation between the estimated silicon accumulation in the region and the measured value of silicon content. Use the numerical deviation to perform a weighted correction on the estimated silicon accumulation in the region to obtain the multidimensional calibrated silicon accumulation of the whole summer maize plant.

[0010] The present invention is improved in that step S3 is specifically as follows: S301: Obtain a set of summer maize leaf silicon accumulation samples under a continuous time series, calculate the arithmetic mean and standard deviation of the summer maize leaf silicon accumulation samples, delineate the upper and lower boundaries of the numerical distribution of the summer maize leaf silicon accumulation sample data, and generate the physiological stability interval of summer maize silicon. S302: Traverse the time index in the multi-source monitoring dataset, extract the monitoring area environmental parameters that match the calculation time of the multi-dimensional calibration silicon accumulation of the whole summer maize plant, establish the mapping sequence of the multi-dimensional calibration silicon accumulation of the whole summer maize plant and each environmental parameter in the monitoring area in the time dimension, and generate time-series corresponding environmental parameters. S303: Filter each environmental parameter corresponding to the moment when the multidimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of silicon in the summer maize, determine whether the multidimensional calibration silicon accumulation of the whole summer maize plant at each moment is within the numerical range of the physiological stability range of silicon in the summer maize, retain the time-series corresponding environmental parameters that meet the judgment conditions, and generate a steady-state environmental parameter set.

[0011] The present invention is improved in that step S4 is specifically as follows: S401: Obtain the full range of changes in silicon accumulation during the summer maize growth cycle and divide the summer maize silicon accumulation gradient interval. Statistically calculate the maximum and minimum values ​​of silicon accumulation in summer maize leaves during the summer maize growth cycle. Calculate the difference between the maximum and minimum values ​​to obtain the full range of changes. Cut the full range of changes into multiple non-overlapping continuous numerical segments according to equal spacing or non-linear rules to generate the summer maize silicon accumulation gradient interval. S402: For each of the summer maize silicon accumulation gradient intervals, set the corresponding numerical solution step size according to the metabolic activity of silicon element in the corresponding accumulation stage, and establish a key-value pair relationship between the interval index of the summer maize silicon accumulation gradient interval and the numerical solution step size as the gradient interval solution step size. S403: Using the divided summer maize silicon cumulative gradient interval as the lookup key and the gradient interval solution step size as the corresponding value, store all the associated entries of the summer maize silicon cumulative gradient interval and the gradient interval solution step size to obtain the silicon cumulative gradient step size mapping table.

[0012] The present invention is improved in that step S5 is specifically as follows: S501: Call the environmental parameters of the monitoring area in the steady-state environmental parameter set, iteratively calculate the theoretical silicon accumulation of each layer of summer maize canopy based on the coupling relationship between water flow and silicon transfer, calculate the difference between the theoretical silicon accumulation and the multidimensional calibration silicon accumulation of the whole summer maize plant, retrieve the silicon accumulation gradient step size mapping table according to the silicon accumulation gradient interval of the summer maize where the difference is located, obtain the solution step size of the gradient interval and adjust the transpiration estimation value to obtain the transpiration inversion value of each layer; S502: Read the transpiration values ​​of the upper, middle and lower parts of the summer maize canopy from the transpiration inversion values ​​of each level, perform a summation operation to summarize the total water evaporation of the whole plant, and use the total water evaporation as a key indicator to characterize the current water metabolism intensity of the plant to generate the total transpiration of summer maize. S503: Import the total transpiration of summer maize, the multidimensional calibration silica accumulation of the whole summer maize plant, the leaf area of ​​summer maize, and the set of steady-state environmental parameters into a multivariate regression analysis model, fit the functional relationship between the total transpiration of summer maize, the multidimensional calibration silica accumulation of the whole summer maize plant, the leaf area of ​​summer maize, and the set of steady-state environmental parameters and the water consumption rate, deduce the water consumption value at the next time step, and generate the total transpiration measurement result.

[0013] A smart system for measuring the transpiration of summer maize based on silicon accumulation data, the system comprising: The multi-source data acquisition module collects spectral images of the summer maize canopy, divides the summer maize canopy into upper, middle and lower regions, and simultaneously collects environmental parameters within the monitoring area to generate a multi-source monitoring dataset. The silicon accumulation inversion module calculates the reflectance of characteristic bands in the upper, middle and lower parts of the summer maize canopy based on the multi-source monitoring dataset, determines the silicon accumulation in the upper, middle and lower parts of the summer maize canopy, and generates the multi-dimensional calibrated silicon accumulation of the entire summer maize plant. The steady-state parameter screening module defines the physiological stability range of silicon in summer maize, extracts environmental parameters within the monitoring area from the multi-source monitoring dataset, and screens each environmental parameter corresponding to the moment when the multi-dimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of silicon in summer maize, thereby generating a steady-state environmental parameter set. The silicon gradient mapping module obtains the global range of changes in silicon accumulation during the summer maize growth cycle and divides the silicon accumulation gradient intervals of summer maize. Based on the intervals, it constructs a silicon accumulation gradient step size mapping table. The transpiration prediction module, referring to the steady-state environmental parameter set, performs inversion calculations of the transpiration of the upper, middle and lower parts of the summer maize canopy, matches it with the silicon cumulative gradient step size mapping table, predicts the future water consumption rate of summer maize, and obtains the total transpiration measurement result.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, the distribution characteristics of silicon elements in the vertical direction of the canopy are analyzed by constructing a physical layer mapping logic. The remote sensing inversion results are double-calibrated by combining laboratory chemical analysis data. The physiological stability interval is defined by statistical methods and abnormal environmental noise is eliminated. A variable step-size iterative solution mechanism based on cumulative gradient is constructed. The calculation accuracy is adaptively adjusted according to the metabolic activity level to achieve a refined analysis of the water transport process under different physiological states. The cumulative error caused by environmental fluctuations and spatial heterogeneity is eliminated, ensuring that the prediction results of the whole plant water consumption rate of summer maize can truly reflect the actual water consumption demand of the crop. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a detailed flowchart of step S1 of the present invention; Figure 3 This is a detailed flowchart of step S2 of the present invention; Figure 4 This is a detailed flowchart of step S3 of the present invention; Figure 5 This is a detailed flowchart of step S4 of the present invention; Figure 6 This is a detailed flowchart of step S5 of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] Please see Figure 1 This invention provides a technical solution: an intelligent method for calculating the transpiration of summer maize based on silicon accumulation data, comprising the following steps: S1: Collect spectral images of the summer maize canopy, divide the summer maize canopy into upper, middle and lower regions, and simultaneously collect environmental parameters within the monitoring area to generate a multi-source monitoring dataset; S2: Calculate the reflectance of characteristic bands in the upper, middle and lower parts of the summer maize canopy based on the multi-source monitoring dataset, determine the silicon accumulation in the upper, middle and lower parts of the summer maize canopy, and generate the multi-dimensional calibrated silicon accumulation of the whole summer maize plant. S3: Define the physiological stability range of silicon in summer maize, extract environmental parameters within the monitoring area from the multi-source monitoring dataset, screen each environmental parameter corresponding to the moment when the multi-dimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of silicon in summer maize, and generate a set of steady-state environmental parameters. S4: Obtain the global range of changes in silicon accumulation during the summer maize growth cycle and divide the silicon accumulation gradient intervals of summer maize. Construct a silicon accumulation gradient step size mapping table based on the intervals. S5: Referring to the steady-state environmental parameter set, the transpiration of the upper, middle and lower parts of the summer maize canopy is inverted and calculated. It is matched with the silicon cumulative gradient step size mapping table to predict the future water consumption rate of summer maize and obtain the total transpiration measurement results.

[0018] The multi-source monitoring dataset includes spectral images of the upper, middle, and lower canopies of summer maize, as well as temperature, humidity, wind speed, and light intensity in the monitoring area. The multi-dimensional calibrated silicon accumulation of the entire summer maize plant includes silicon accumulation in the upper, middle, and lower canopies of summer maize after calibration based on comparison bias. The steady-state environmental parameter set includes temperature, humidity, wind speed, and light intensity in the monitoring area corresponding to the selected values ​​falling within the physiologically stable silicon range of summer maize. The silicon accumulation gradient step size mapping table specifically shows the mapping relationship between the silicon accumulation gradient interval of summer maize and the step size for solving summer maize transpiration during the summer maize growth cycle. The total transpiration measurement results include the sum of transpiration in the upper, middle, and lower parts of the summer maize canopy, the total summer maize transpiration, and the predicted future water consumption rate of summer maize.

[0019] Please see Figure 2 Step S1 is as follows: S101: Control a drone equipped with remote sensing equipment to collect spectral images of the upper part of the summer maize canopy from a vertical perspective at the top of the canopy, and control a ground mobile robot with sensor lenses to collect spectral images of the middle part and the lower part of the summer maize canopy. Analyze the spatial location information of the spectral images of the upper, middle and lower parts of the summer maize canopy and map them to the corresponding physical layering regions to generate layered spectral images. Flight control commands are sent to a drone equipped with remote sensing devices. The drone retrieves data from its GPS module, hovers directly above the center of the summer maize planting area, and adjusts its multispectral camera to a vertically downward angle. It then performs a photographic task at a preset height of 20 meters above the top of the canopy, acquiring spectral images covering the upper part of the summer maize canopy. Simultaneously, path planning commands are sent to a ground-based mobile robot, controlling its parallel movement along the rows and extending the sensor lenses on its robotic arm to different heights. The average plant height of summer maize at this growth stage is set at 220 cm, obtained by the arithmetic mean of historical measured height data from 100 summer maize plants within the monitoring area. Based on a preset physical stratification logic, the boundary height is calculated: the average plant height is multiplied by two-thirds to obtain the boundary height between the upper and middle sections. Centimeters; multiply the average plant height by one-third to obtain the dividing height between the middle and lower parts, i.e. Centimeters. The robotic arm is controlled to position the sensor lens at 110 cm above the ground to collect spectral images of the middle part of the summer maize canopy, and at 40 cm above the ground to collect spectral images of the lower part of the summer maize canopy. The spatial depth information and height coordinate data of each pixel in the collected images are analyzed. Pixels with height coordinate values ​​greater than 146.7 cm are classified as the upper region, pixels with height coordinate values ​​between 73.3 cm and 146.7 cm are classified as the middle region, and pixels with height coordinate values ​​less than 73.3 cm are classified as the lower region. Based on this spatial mapping relationship, the original image data is reconstructed to generate a layered spectral image.

[0020] S102: Synchronously collect environmental parameters within the monitoring area, including air temperature, air humidity, wind speed, and light intensity values, verify the timestamp of each value collection, and generate synchronous environmental monitoring parameters; At the same moment the spectral images are acquired, meteorological monitoring stations deployed within the monitoring area are triggered to perform synchronous data sampling. Temperature sensors utilize the response characteristics of thermistors to changes in ambient heat, converting resistance changes into voltage signals and resolving them into air temperature values. Humidity sensors use the dielectric constant changes of water molecule adsorption by humidity-sensitive capacitors to obtain relative humidity values. Wind speed sensors measure the rotation frequency of the wind turbine cups to obtain wind speed values. Light intensity sensors use photodiodes to convert light energy into current signals to obtain light intensity values. An internal high-precision clock module timestamps each received data stream to milliseconds. For example, at a certain sampling instant, the timestamp is recorded as "10:00:00:50 milliseconds," simultaneously recording an air temperature of 30.5 degrees Celsius, air humidity of 60%, wind speed of 2.5 meters per second, and light intensity of 50,000 lux. This set of parameters with the same timestamp is packaged and stored to generate synchronous environmental monitoring parameters, ensuring a unified time dimension reference for subsequent data fusion.

[0021] S103: Using the timestamp of data acquisition as the reference index, perform correlation matching, map the matched synchronous environmental monitoring parameters to the corresponding image entries in the layered spectral image, remove redundant data with misaligned timestamps, and obtain a multi-source monitoring dataset; The frame generation timestamp from the layered spectral image metadata is used as the baseline index. A time alignment tolerance window is set based on the frame rate characteristics of the image acquisition device. The size of this window is determined by calculating half the frame interval; for example, if the device frame rate is 10 frames per second, the frame interval is 100 milliseconds. The tolerance radius is then calculated. Milliseconds. Iterate through the timestamps in the synchronized environmental monitoring parameters and calculate the absolute value of the difference between each timestamp and the image baseline index. For example, if the difference between the timestamp of a certain set of environmental parameters and the image baseline index is 10 milliseconds, perform the following judgment: If the set of environmental parameters falls within the tolerance window, perform data binding to map the environmental parameters to the layered spectral image of that frame. If the difference calculation result of another set of redundant data is 70 milliseconds, perform the following judgment: The system determines that the data is misaligned and performs a removal operation. This logic filters out all unmatched data, retaining only data pairs that strictly correspond in the time dimension, thus obtaining a multi-source monitoring dataset.

[0022] Please see Figure 3 Step S2 is as follows: S201: Extract layered spectral images from the multi-source monitoring dataset, extract the gray values ​​of the layered spectral images, and perform spectral radiometric calibration calculations with reference to the standard whiteboard data to determine the reflection intensity values ​​of the red band and the near-infrared band respectively, and generate the regional characteristic band reflectance. The raw digital quantization values ​​of the layered spectral images are extracted from the multi-source monitoring dataset. The preset standard whiteboard reflectance parameters and the blackboard noise digital quantization value acquired at the time of acquisition are retrieved from memory. The standard whiteboard reflectance parameter is obtained by measuring the ratio of reflected light intensity to incident light intensity under a standard laboratory light source. The blackboard noise value is the dark current noise acquired with the lens cap closed. The reflectance of the standard whiteboard in the red band is set to 0.98, the digital quantization value of the current pixel in the red band is 120, the digital quantization value of the standard whiteboard in the red band is 200, and the digital quantization value of the blackboard noise is 5. Spectral radiometric calibration calculations are performed. Similarly, for the near-infrared band, with a standard whiteboard reflectivity of 0.99, a pixel digital quantization value of 150, a whiteboard digital quantization value of 210, and a blackboard digital quantization value of 5, the following calculation is performed: The above operation is performed on all pixels within the image region to obtain the red band reflectance intensity value and the near-infrared band reflectance intensity value, thereby generating the regional characteristic band reflectance.

[0023] S202: Combining the regional characteristic band reflectance, as well as the pre-measured summer maize leaf area and summer maize plant height and standardizing them, the silicon element precipitation values ​​of the upper, middle and lower parts of the summer maize canopy are calculated respectively as the estimated silicon accumulation in the region. A pre-built silicon element inversion model was invoked. This model, based on partial least squares regression, was trained as follows: A large amount of summer maize canopy spectral data was collected as the input matrix, and silicon content data of the corresponding plants, measured by laboratory chemical analysis, was simultaneously collected as the response matrix. By finding a multi-dimensional direction vector in the feature space to maximize the covariance between spectral features and silicon content, the regression coefficients were determined. Historical databases were used to retrieve the mean red reflectance for this growth stage, which was 0.5 with a standard deviation of 0.1; and the mean leaf area, which was 4500 cm² with a standard deviation of 200 cm². The current red light band reflectance value of 0.578 was standardized. Standardization processing is applied to the current leaf area of ​​4600 square centimeters: The standardized values ​​are substituted into the model regression equation, with the equation coefficients determined during the training process (e.g., red light coefficient of -2.5, leaf area coefficient of 1.2, and intercept of 10.0). A weighted summation operation is then performed to estimate the silicon content in the middle of the canopy. Similarly, calculate the upper and lower values, assuming the upper value is 7.5 and the lower value is 9.8. Summarize these values ​​as the estimated silicon accumulation for the region.

[0024] S203: Collect leaf samples of summer maize under multiple growth conditions and test them to obtain the measured value of silicon content. Calculate the numerical deviation between the estimated silicon accumulation in the region and the measured value of silicon content. Use the numerical deviation to make a weighted correction to the estimated silicon accumulation in the region and obtain the multidimensional calibrated silicon accumulation of the whole summer maize plant. The laboratory uploaded measured values ​​of silicon content in summer maize leaves. These values ​​were obtained through chemical analysis using the molybdenum blue colorimetric method, based on randomly collected summer maize leaf samples within the monitoring area. The measured silicon content for the corresponding sample was set at 8.8 g / kg, and the estimated silicon accumulation in the aforementioned area was 8.65 g / kg. The absolute value of the numerical deviation between the two values ​​was calculated. Calculate the proportion of this deviation to the measured value: Based on this ratio, a weighting correction coefficient is constructed: This coefficient is used to correct the estimated silicon accumulation in the region: This correction process was applied to the estimated values ​​for the upper, middle, and lower parts of the whole plant. By introducing high-precision laboratory data to correct the remote sensing inversion results, systematic errors caused by environmental interference were eliminated, and the multidimensional calibrated silicon accumulation of the whole summer maize plant after calibration with laboratory data was obtained.

[0025] Please see Figure 4 Step S3 is as follows: S301: Obtain a set of continuous time series samples of silicon accumulation in summer maize leaves, calculate the arithmetic mean and standard deviation of the silicon accumulation samples of summer maize leaves, delineate the upper and lower boundaries of the numerical distribution of the silicon accumulation sample data of summer maize leaves, and generate the physiological stability interval of silicon in summer maize. The process of defining the upper and lower boundaries of the numerical distribution of silicon accumulation sample data in summer maize leaves is as follows: A confidence level adjustment coefficient is preset as a quantitative benchmark for measuring the tolerance of physiological data fluctuations; The product of the standard deviation of silicon accumulation in summer maize leaves and the confidence adjustment coefficient was calculated, and the product was defined as the one-sided discrete amplitude that characterizes the allowable deviation from the central trend under normal physiological conditions. The arithmetic mean and the sum of the one-sided discrete amplitudes of the silicon accumulation samples in summer maize leaves were calculated. The calculation results were determined as the upper limit of the physiological stability range of silicon in summer maize, and the critical point of silicon accumulation during positive fluctuations was defined. The difference between the arithmetic mean and the one-sided discrete amplitude is calculated, and the calculation result is determined as the lower limit of the physiological stability range of silicon in summer maize, defining the critical point of silicon accumulation when it fluctuates negatively. Construct a closed numerical range with the lower limit as the starting boundary and the upper limit as the ending boundary, and mark the closed numerical range as the physiological stability interval of summer maize silicon. Data on silicon accumulation in summer maize leaves from the same period over the past five growth cycles were extracted from historical databases, with a total sample size of 1000 groups. The arithmetic mean of this sample set was calculated to be 15.0 g / kg, and the standard deviation was 0.5 g / kg. A confidence level adjustment coefficient of 1.96 was set, derived from the standard critical value corresponding to the 95% confidence level in the normal distribution statistical table. The one-sided discrete amplitude was calculated. Calculate the upper limit of the physiological homeostasis range for silicon in summer maize: ;Calculate the lower limit of the value: Construct a closed numerical range This range characterizes the normal silicon metabolism fluctuation range of summer maize under non-stress conditions. It is used to identify and eliminate abnormal physiological data caused by pests, diseases, or extreme environments, ensuring the purity of subsequent model inputs and generating the stable silicon physiological range of summer maize.

[0026] S302: Traverse the time index in the multi-source monitoring dataset, extract the environmental parameters of the monitoring area that match the calculation time of the multi-dimensional calibration silicon accumulation of the whole summer maize plant, establish the mapping sequence of the multi-dimensional calibration silicon accumulation of the whole summer maize plant and each environmental parameter in the monitoring area in the time dimension, and generate time-series corresponding environmental parameters. Using timestamps from the multi-source monitoring dataset as keys, environmental parameters consistent with the calculation time of the multidimensional calibration silicon accumulation of summer maize are retrieved. Assuming the calculation time corresponding to the calibration silicon accumulation is a specific point in time, the monitoring database is used to locate the recorded temperature of 30 degrees Celsius, humidity of 60%, wind speed of 2.5 meters per second, and light intensity of 50,000 lux. A data association operation is performed, linking these four environmental parameters as an array vector to the silicon accumulation data entry for that time. This process is repeated for all discrete time points throughout the monitoring period to establish a time-dimensional mapping sequence, ensuring that each set of silicon accumulation data is associated with the instantaneous meteorological conditions affecting it, generating time-series corresponding environmental parameters.

[0027] S303: Screen each environmental parameter corresponding to the moment when the multidimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of summer maize silicon, determine whether the multidimensional calibration silicon accumulation of the whole summer maize plant at each moment is within the numerical range of the physiological stability range of summer maize silicon, retain the time-series corresponding environmental parameters that meet the judgment conditions, and generate a steady-state environmental parameter set. Read the multidimensional calibration silicon accumulation values ​​of the whole summer maize plant from the environmental parameters corresponding to the time series, and compare them with the physiological stability range of silicon in summer maize. Perform a comparison. If the calibration silicon accumulation value at a certain moment is 15.5, perform a judgment: If the result is true, it indicates that the plant is in a normal physiological homeostasis at that moment, and the corresponding parameters such as temperature and humidity are stored in the homeostasis set. If the value at a certain moment is 16.5, perform the following judgment: If the result is true, it means the value exceeds the upper limit, and the plant may be in an abnormal metabolic state; if the value is 13.5, perform the following judgment: If the result is true, it means the value is below the lower limit. For the two cases exceeding the range mentioned above, the data is judged as non-steady-state data and is removed from subsequent calculations. Only records meeting the steady-state conditions are output, generating a steady-state environment parameter set.

[0028] Please see Figure 5 Step S4 is as follows: S401: Obtain the full range of changes in silicon accumulation during the summer maize growth cycle and divide the summer maize silicon accumulation gradient interval. Statistically calculate the maximum and minimum values ​​of silicon accumulation in summer maize leaves during the summer maize growth cycle. Calculate the difference between the maximum and minimum values ​​to obtain the full range of changes. Cut the full range of changes into multiple non-overlapping continuous numerical segments according to equal spacing or non-linear rules to generate the summer maize silicon accumulation gradient interval. By traversing historical silicon accumulation data throughout the entire growth cycle of summer maize, the historical maximum value was identified as 25.0 g / kg, and the minimum value as 5.0 g / kg. The entire range of changes was calculated. Based on the required data processing accuracy, the segmentation interval is set to 2.0 grams per kilogram. The number of intervals is calculated as follows: Perform interval partitioning to generate 10 consecutive numerical segments: , until Each numerical segment is defined as an independent classification unit. Through this discretization process, the continuously changing amount of silicon accumulation can be divided into different physiological levels, thus providing a classification basis for matching the optimal calculation parameters for different metabolic intensities and generating the silicon accumulation gradient interval for summer maize.

[0029] S402: For each summer maize silicon accumulation gradient interval, set the corresponding numerical solution step size according to the metabolic activity of silicon element in the corresponding accumulation stage, and establish the key-value pair relationship between the interval index of summer maize silicon accumulation gradient interval and the numerical solution step size as the gradient interval solution step size. The process of setting the corresponding numerical solution step size based on the metabolic activity of silicon in the corresponding accumulation stage is as follows: Calculate the slope of silicon accumulation within each silicon accumulation gradient interval of summer maize, and take the absolute value of the slope of silicon accumulation as an evaluation index characterizing the degree of metabolic activity. Preset a base step size value and a slope threshold; Compare the evaluation metrics with the slope threshold; When the evaluation index is greater than the slope threshold, a preset first adjustment coefficient with a value less than one is selected, and the multiplication operation between the base step size value and the preset first adjustment coefficient is performed. The result of the operation is determined as the numerical solution step size. When the evaluation index is less than or equal to the slope threshold, a preset second adjustment coefficient with a value greater than one is selected, and the multiplication operation between the base step size value and the preset second adjustment coefficient is performed. The result of the operation is determined as the numerical solution step size. For each gradient interval, the absolute value of the rate of change of historical data within the interval is calculated as an indicator of metabolic activity. For example, in the interval... Within the range, the calculated slope of silicon accumulation per unit time is 0.8; within the interval... Within this range, the slope is 0.2. A base step size of 0.01 and a slope threshold of 0.5 are set; this threshold is obtained by calculating the arithmetic mean of the slopes across all gradient intervals. For the interval... Compare and evaluate the indicators and thresholds: The study determined that metabolism was active and data fluctuations were large during this stage. To improve simulation accuracy and prevent iteration overshoot, the step size needed to be reduced. A preset first adjustment coefficient of 0.5 (less than 1) was selected, and the numerical solution step size was calculated as follows: For intervals Compare and evaluate the indicators and thresholds: The metabolism is determined to be slow during this stage, so the step size can be increased to improve computational efficiency. A preset second adjustment coefficient of 1.5 (greater than 1) is selected, and the numerical solution step size is calculated as follows: After completing the calculations for all intervals, establish the correspondence between the interval index and the step size, which serves as the step size for solving the gradient interval.

[0030] S403: Use the divided summer maize silicon cumulative gradient interval as the lookup key and the gradient interval solution step size as the corresponding value to store all the associated entries of summer maize silicon cumulative gradient interval and gradient interval solution step size to obtain the silicon cumulative gradient step size mapping table. Use the cumulative gradient interval of summer maize silicon divided in S401 as the lookup key, and the solution step size of the gradient interval calculated in S402 as the corresponding value. For example, write the key-value pair "key: Value: 0.005 and key: The value is 0.015. All related entries for all intervals are stored sequentially in the database. This mapping table realizes a fast index from physiological state to calculation parameters, ensuring that in subsequent inversion calculations, the optimal iteration step size that balances accuracy and efficiency can be quickly matched based on the current silicon accumulation value, thus obtaining the silicon accumulation gradient step size mapping table.

[0031] Please see Figure 6 Step S5 is as follows: S501: Call the environmental parameters of the monitoring area in the steady-state environmental parameter set, iteratively calculate the theoretical silicon accumulation of each layer of summer maize canopy based on the coupling relationship between water flow and silicon transfer, calculate the difference between the theoretical silicon accumulation and the multidimensional calibration silicon accumulation of the whole summer maize plant, retrieve the silicon accumulation gradient step size mapping table according to the silicon accumulation gradient interval of summer maize where the difference is located, obtain the gradient interval solution step size and adjust the transpiration estimation value, and obtain the transpiration inversion value of each layer; Using current data from the steady-state environmental parameter set, the theoretical silicon accumulation is calculated using the modified Penman-Montes equation combined with a silicon-water coupled transport model. It is assumed that the model outputs a theoretical silicon accumulation of 8.40 g / kg in the mid-canopy under initial evaporation parameters, while the calibrated silicon accumulation obtained in S203 is 8.50 g / kg. The difference is calculated as follows: Based on the current calibration value of 8.50, it falls within the range. The silicon cumulative gradient step size mapping table was retrieved to obtain a step size of 0.01. Since the theoretical value is less than the calibration value ( The transpiration rate parameter in the model is increased step by step, and the theoretical value is recalculated. An iterative loop is executed, continuously adjusting the model input parameters to approximate the measured value until the difference is less than a preset allowable error (e.g., 0.01). At this point, the transpiration rate in the model is the inversion result. Assume that upon final convergence, the inverted transpiration rate is 50 ml / h in the upper canopy, 35 ml / h in the middle, and 15 ml / h in the lower canopy.

[0032] S502: Read the transpiration values ​​of the upper, middle and lower parts of the summer maize canopy from the transpiration inversion values ​​of each level, perform a summation operation to summarize the total water evaporation of the whole plant, and use the total water evaporation as a key indicator to characterize the current water metabolism intensity of the plant to generate the total transpiration of summer maize. Read the transpiration rate values ​​for each layer obtained after iterative convergence in S501. Perform a whole-plant summation operation to calculate the sum of the transpiration rates retrieved from the upper, middle, and lower parts of the summer maize canopy: The calculated result of 100 ml per hour is labeled as the total transpiration of summer maize. This value quantifies the total flux of water released from the entire summer maize plant into the atmosphere at the current moment, reflecting the actual water consumption capacity of the crop under specific silicon accumulation and environmental conditions. It provides a core input variable for subsequent water consumption prediction and generates the total transpiration of summer maize.

[0033] S503: Import the total transpiration of summer maize, the multidimensional calibration silicon accumulation of the whole summer maize plant, the leaf area of ​​summer maize, and the set of steady-state environmental parameters into a multivariate regression analysis model, fit the functional relationship between the total transpiration of summer maize, the multidimensional calibration silicon accumulation of the whole summer maize plant, the leaf area of ​​summer maize, and the set of steady-state environmental parameters and the water consumption rate, deduce the water consumption value at the next time step, and generate the total transpiration measurement result; A nonlinear prediction model based on a generalized regression neural network was constructed. The model training process involved using historical data on total transpiration, silicon accumulation, and environmental parameters of summer maize from different periods as input samples, and the measured water consumption at the next time step as the target output. A Gaussian kernel function was used to establish the connection weights between samples. The model consists of an input layer, a pattern layer, a summation layer, and an output layer. The specific calculation formula for the prediction model is as follows: ,in, This indicates the predicted water consumption value at the next time step (e.g., after 1 hour). The input feature vector at the current moment includes the normalized total transpiration of summer maize, the multidimensional calibrated silica accumulation of the whole plant, air temperature, humidity, wind speed and light intensity; This indicates the number of historical training samples selected; Indicates the first Feature vectors of historical training samples; Indicates the first The measured water consumption values ​​at the next moment corresponding to each historical training sample; This represents the smoothing parameter, used to control the radial range of the kernel function. This parameter is set based on the dispersion of the historical data; the sparser the data distribution, the smoother the effect. The larger the set value, the more likely it is to be set using cross-validation in this embodiment. The calculation is illustrated using two historical samples as an example: Setting the smoothing parameter... The normalized input feature vector at the current time step. for This corresponds to the total transpiration of 100 ml / h calculated by S502 and the silicon accumulation of 8.5 g / kg (normalized) calculated by S203, respectively. The first historical sample was selected. The corresponding measured value of water consumption at the next moment Select the second historical sample The corresponding measured value of water consumption at the next moment First, calculate the squared Euclidean distance between the current vector and each historical vector. Squared distance from sample 1: Squared distance from sample 2: Calculate the kernel function activation value for each sample. Activation value of sample 1: Activation value of sample 2: Calculate the numerator (the sum of weighted objective values): Calculate the denominator (sum of activation values): Calculate the predicted value : Based on the predicted water supply of 105.00 ml / h, and considering the current soil moisture supply capacity (set at 90.00 ml / h), calculate the water deficit value: Based on a preset irrigation decision gradient (the threshold is determined through crop water stress experiments): if the deficit value is less than 5, it is considered that the water is sufficient and irrigation is not performed; if the deficit value is between 5 and 25, it is considered that there is a slight deficit and precision drip irrigation is performed; if the deficit value is greater than 25, it is considered that there is a severe deficit and sprinkler irrigation is performed. Since the calculated deficit value of 15.00 is between 5 and 25, the control decision of "perform precision drip irrigation" is output, generating the total transpiration measurement result and guiding the irrigation.

[0034] A smart system for measuring the transpiration of summer maize based on silicon accumulation data, the system comprising: The multi-source data acquisition module collects spectral images of the summer maize canopy, divides the summer maize canopy into upper, middle and lower regions, and simultaneously collects environmental parameters within the monitoring area to generate a multi-source monitoring dataset. The silicon accumulation inversion module calculates the reflectance of characteristic bands in the upper, middle and lower parts of the summer maize canopy based on the multi-source monitoring dataset, determines the silicon accumulation in the upper, middle and lower parts of the summer maize canopy, and generates the multi-dimensional calibrated silicon accumulation of the entire summer maize plant. The steady-state parameter screening module defines the physiological stability range of silicon in summer maize and extracts environmental parameters within the monitoring area from the multi-source monitoring dataset. It then screens each environmental parameter corresponding to the moment when the multi-dimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of silicon in summer maize, and generates a steady-state environmental parameter set. The silicon gradient mapping module obtains the global range of changes in silicon accumulation during the summer maize growth cycle and divides the silicon accumulation gradient intervals of summer maize. Based on the intervals, it constructs a silicon accumulation gradient step size mapping table. The transpiration prediction module, referring to the steady-state environmental parameter set, performs inversion calculations of transpiration in the upper, middle and lower parts of the summer maize canopy, matches it with the silicon cumulative gradient step size mapping table, predicts the future water consumption rate of summer maize, and obtains the total transpiration measurement results.

[0035] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for intelligently calculating the transpiration of summer maize based on silicon accumulation data, characterized in that, Includes the following steps: S1: Collect spectral images of the summer maize canopy, divide the summer maize canopy into upper, middle and lower regions, and simultaneously collect environmental parameters within the monitoring area to generate a multi-source monitoring dataset; S2: Calculate the reflectance of characteristic bands in the upper, middle and lower parts of the summer maize canopy based on the multi-source monitoring dataset, determine the silicon accumulation in the upper, middle and lower parts of the summer maize canopy, and generate the multi-dimensional calibrated silicon accumulation of the entire summer maize plant. S3: Define the physiological stability range of silicon in summer maize, extract environmental parameters within the monitoring area from the multi-source monitoring dataset, screen each environmental parameter corresponding to the moment when the multi-dimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of silicon in summer maize, and generate a steady-state environmental parameter set. S4: Obtain the global range of changes in silicon accumulation during the summer maize growth cycle and divide the silicon accumulation gradient intervals of summer maize. Construct a silicon accumulation gradient step size mapping table based on the intervals. S5: Referring to the steady-state environmental parameter set, perform inversion calculation of transpiration in the upper, middle and lower parts of the summer maize canopy, match it with the silicon cumulative gradient step size mapping table, predict the future water consumption rate of summer maize, and obtain the total transpiration calculation result.

2. The intelligent calculation method for summer maize transpiration based on silicon accumulation data according to claim 1, characterized in that, The multi-source monitoring dataset includes spectral images of the upper, middle, and lower canopies of summer maize, as well as temperature, humidity, wind speed, and light intensity in the monitoring area. The multi-dimensional calibrated silicon accumulation of the entire summer maize plant includes silicon accumulation in the upper, middle, and lower canopies of summer maize after calibration based on comparison bias. The steady-state environmental parameter set includes temperature, humidity, wind speed, and light intensity in the monitoring area corresponding to the selected values ​​falling within the physiologically stable silicon range of summer maize. The silicon accumulation gradient step size mapping table specifically represents the mapping relationship between the silicon accumulation gradient interval of summer maize during its growth cycle and the step size for solving summer maize transpiration. The total transpiration calculation result includes the sum of transpiration from the upper, middle, and lower parts of the summer maize canopy, the total summer maize transpiration, and the predicted future water consumption rate of summer maize.

3. The intelligent calculation method for summer maize transpiration based on silicon accumulation data according to claim 1, characterized in that, Step S1 is as follows: S101: Control a drone equipped with remote sensing equipment to collect spectral images of the upper part of the summer maize canopy from a vertical perspective at the top of the canopy, and control a ground mobile robot with sensor lenses to collect spectral images of the middle part and the lower part of the summer maize canopy. Analyze the spatial location information of the spectral images of the upper, middle and lower parts of the summer maize canopy and map them to the corresponding physical layering regions to generate layered spectral images. S102: Synchronously collect environmental parameters within the monitoring area, including air temperature, air humidity, wind speed, and light intensity values, verify the timestamp of each value collection, and generate synchronous environmental monitoring parameters; S103: Using the timestamp of data acquisition as the reference index, perform association matching, map the matched synchronous environmental monitoring parameters to the corresponding image entries in the layered spectral image, remove redundant data with misaligned timestamps, and obtain a multi-source monitoring dataset.

4. The intelligent calculation method for summer maize transpiration based on silicon accumulation data according to claim 3, characterized in that, Step S2 is as follows: S201: Extract the layered spectral image from the multi-source monitoring dataset, extract the grayscale value of the layered spectral image and perform spectral radiometric calibration operation with reference to the standard whiteboard data, determine the reflection intensity value of the red band and the reflection intensity value of the near-infrared band respectively, and generate the regional characteristic band reflectance. S202: Combining the regional characteristic band reflectivity, the pre-measured summer maize leaf area and summer maize plant height and standardizing them, calculate the silicon element precipitation values ​​in the upper, middle and lower parts of the summer maize canopy, respectively, as the estimated silicon accumulation in the region; S203: Collect leaf samples of summer maize under multiple growth conditions and test them to obtain the measured value of silicon content. Calculate the numerical deviation between the estimated silicon accumulation in the region and the measured value of silicon content. Use the numerical deviation to perform a weighted correction on the estimated silicon accumulation in the region to obtain the multidimensional calibrated silicon accumulation of the whole summer maize plant.

5. The intelligent calculation method for summer maize transpiration based on silicon accumulation data according to claim 1, characterized in that, Step S3 is as follows: S301: Obtain a set of summer maize leaf silicon accumulation samples under a continuous time series, calculate the arithmetic mean and standard deviation of the summer maize leaf silicon accumulation samples, delineate the upper and lower boundaries of the numerical distribution of the summer maize leaf silicon accumulation sample data, and generate the physiological stability interval of summer maize silicon. S302: Traverse the time index in the multi-source monitoring dataset, extract the monitoring area environmental parameters that match the calculation time of the multi-dimensional calibration silicon accumulation of the whole summer maize plant, establish the mapping sequence of the multi-dimensional calibration silicon accumulation of the whole summer maize plant and each environmental parameter in the monitoring area in the time dimension, and generate time-series corresponding environmental parameters. S303: Filter each environmental parameter corresponding to the moment when the multidimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of silicon in the summer maize, determine whether the multidimensional calibration silicon accumulation of the whole summer maize plant at each moment is within the numerical range of the physiological stability range of silicon in the summer maize, retain the time-series corresponding environmental parameters that meet the judgment conditions, and generate a steady-state environmental parameter set.

6. The intelligent calculation method for summer maize transpiration based on silicon accumulation data according to claim 1, characterized in that, Step S4 is as follows: S401: Obtain the full range of changes in silicon accumulation during the summer maize growth cycle and divide the summer maize silicon accumulation gradient interval. Statistically calculate the maximum and minimum values ​​of silicon accumulation in summer maize leaves during the summer maize growth cycle. Calculate the difference between the maximum and minimum values ​​to obtain the full range of changes. Cut the full range of changes into multiple non-overlapping continuous numerical segments according to equal spacing or non-linear rules to generate the summer maize silicon accumulation gradient interval. S402: For each of the summer maize silicon accumulation gradient intervals, set the corresponding numerical solution step size according to the metabolic activity of silicon element in the corresponding accumulation stage, and establish a key-value pair relationship between the interval index of the summer maize silicon accumulation gradient interval and the numerical solution step size as the gradient interval solution step size. S403: Using the divided summer maize silicon cumulative gradient interval as the lookup key and the gradient interval solution step size as the corresponding value, store all the associated entries of the summer maize silicon cumulative gradient interval and the gradient interval solution step size to obtain the silicon cumulative gradient step size mapping table.

7. The intelligent calculation method for summer maize transpiration based on silicon accumulation data according to claim 1, characterized in that, Step S5 is as follows: S501: Call the environmental parameters of the monitoring area in the steady-state environmental parameter set, iteratively calculate the theoretical silicon accumulation of each layer of summer maize canopy based on the coupling relationship between water flow and silicon transfer, calculate the difference between the theoretical silicon accumulation and the multidimensional calibration silicon accumulation of the whole summer maize plant, retrieve the silicon accumulation gradient step size mapping table according to the silicon accumulation gradient interval of the summer maize where the difference is located, obtain the solution step size of the gradient interval and adjust the transpiration estimation value to obtain the transpiration inversion value of each layer; S502: Read the transpiration values ​​of the upper, middle and lower parts of the summer maize canopy from the transpiration inversion values ​​of each level, perform a summation operation to summarize the total water evaporation of the whole plant, and use the total water evaporation as a key indicator to characterize the current water metabolism intensity of the plant to generate the total transpiration of summer maize. S503: Import the total transpiration of summer maize, the multidimensional calibration silica accumulation of the whole summer maize plant, the leaf area of ​​summer maize, and the set of steady-state environmental parameters into a multivariate regression analysis model, fit the functional relationship between the total transpiration of summer maize, the multidimensional calibration silica accumulation of the whole summer maize plant, the leaf area of ​​summer maize, and the set of steady-state environmental parameters and the water consumption rate, deduce the water consumption value at the next time step, and generate the total transpiration measurement result.

8. The intelligent calculation method for summer maize transpiration based on silicon accumulation data according to claim 5, characterized in that, The process of defining the upper and lower boundaries of the numerical distribution of silicon accumulation sample data in summer maize leaves is as follows: A confidence level adjustment coefficient is preset as a quantitative benchmark for measuring the tolerance of physiological data fluctuations; The product of the standard deviation of silicon accumulation in summer maize leaves and the confidence adjustment coefficient was calculated, and the product was defined as the one-sided discrete amplitude that characterizes the allowable deviation from the central trend under normal physiological conditions. The arithmetic mean and the sum of the one-sided discrete amplitudes of the silicon accumulation samples in summer maize leaves were calculated. The calculation results were determined as the upper limit of the physiological stability range of silicon in summer maize, and the critical point of silicon accumulation during positive fluctuations was defined. The difference between the arithmetic mean and the one-sided discrete amplitude is calculated, and the calculation result is determined as the lower limit of the physiological stability range of silicon in summer maize, defining the critical point of silicon accumulation when it fluctuates negatively. Construct a closed numerical range with the lower limit as the starting boundary and the upper limit as the ending boundary, and mark the closed numerical range as the physiological stability interval of summer maize silicon.

9. The intelligent calculation method for summer maize transpiration based on silicon accumulation data according to claim 6, characterized in that, The process of setting the corresponding numerical solution step size based on the metabolic activity of silicon in the corresponding accumulation stage is as follows: Calculate the slope of silicon accumulation within each silicon accumulation gradient interval of summer maize, and take the absolute value of the slope of silicon accumulation as an evaluation index characterizing the degree of metabolic activity. Preset a base step size value and a slope threshold; Compare the evaluation metrics with the slope threshold; When the evaluation index is greater than the slope threshold, a preset first adjustment coefficient with a value less than one is selected, and the multiplication operation between the base step size value and the preset first adjustment coefficient is performed. The result of the operation is determined as the numerical solution step size. When the evaluation index is less than or equal to the slope threshold, a preset second adjustment coefficient with a value greater than one is selected, and the multiplication operation between the base step size value and the preset second adjustment coefficient is performed. The result of the operation is determined as the numerical solution step size.

10. A smart system for measuring the transpiration of summer maize based on silicon accumulation data, characterized in that, The system comprises: (1) the intelligent calculation method for summer maize transpiration based on silicon accumulation data according to any one of claims 1-9; (2) the system comprising: The multi-source data acquisition module collects spectral images of the summer maize canopy, divides the summer maize canopy into upper, middle and lower regions, and simultaneously collects environmental parameters within the monitoring area to generate a multi-source monitoring dataset. The silicon accumulation inversion module calculates the reflectance of characteristic bands in the upper, middle and lower parts of the summer maize canopy based on the multi-source monitoring dataset, determines the silicon accumulation in the upper, middle and lower parts of the summer maize canopy, and generates the multi-dimensional calibrated silicon accumulation of the entire summer maize plant. The steady-state parameter screening module defines the physiological stability range of silicon in summer maize, extracts environmental parameters within the monitoring area from the multi-source monitoring dataset, and screens each environmental parameter corresponding to the moment when the multi-dimensional calibration silicon accumulation of the whole summer maize plant falls within the physiological stability range of silicon in summer maize, thereby generating a steady-state environmental parameter set. The silicon gradient mapping module obtains the global range of changes in silicon accumulation during the summer maize growth cycle and divides the silicon accumulation gradient intervals of summer maize. Based on the intervals, it constructs a silicon accumulation gradient step size mapping table. The transpiration prediction module, referring to the steady-state environmental parameter set, performs inversion calculations of the transpiration of the upper, middle and lower parts of the summer maize canopy, matches it with the silicon cumulative gradient step size mapping table, predicts the future water consumption rate of summer maize, and obtains the total transpiration measurement result.

Images