Dynamic plant spacing adjusting method and system for cotton close planting and high-yield cultivation

Abstract

The invention provides a dynamic plant spacing adjustment method and system for cotton close-planting high-yield cultivation. The method comprises the steps that plant morphology, field microenvironment and target cotton plant close-planting photosynthetic rate initial data of cotton in different growth stages are collected; based on the data, calculating the optimal plant spacing range at the current stage by using a pre-constructed plant spacing adaptation model which takes the cotton close planting yield maximization as an objective function and takes the plant canopy overlapping rate, the inter-row illumination transmittance and the soil humidity as constraint conditions; obtaining the actual plant spacing of the cotton field, comparing the actual plant spacing with the optimal range, judging whether to adjust, generating an adjusting instruction if the plant spacing exceeds the range, controlling the automatic transplanting device to operate, and collecting data again after adjustment to verify whether the plant spacing accords with the optimal range. According to the method, the plant spacing adaptation model is constructed, and the problem of multi-constraint synchronous coordination is solved by combining dynamic weight coefficient and growth stage characteristic matching; data are regularly collected to update model parameters, the cotton close planting yield is improved, and cultivation management intelligence is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cotton cultivation, in particular to a plant spacing dynamic adjustment method and system for high-yield cotton cultivation. BACKGROUND

[0002] In the field of high-yield cotton cultivation, traditional plant spacing management usually adopts fixed parameters or experience-based stage adjustment strategies. Farmers usually set the initial plant spacing according to historical planting guidelines or regional experience, and make limited plant spacing corrections through manual inspection during the growth cycle. Although this method can meet the basic cultivation needs, it has significant dynamic response defects. The plant spacing setting cannot adapt to the morphological variation and microenvironment fluctuations of different growth stages of cotton (such as the vegetative growth in the seedling stage and the reproductive growth in the flowering and bolling stage), often leading to problems such as excessive canopy overlap rate, poor ventilation, aggravated diseases, insufficient inter-row light transmittance inhibiting photosynthesis, and soil humidity imbalance affecting the photosynthetic rate of dense planting. In addition, the traditional method lacks data-driven optimization model support, and the adjustment decision highly depends on manual experience, which has strong subjectivity, low quantitative precision, and cannot synchronize and coordinate multiple constraint conditions (such as yield-canopy-light-humidity balance), making it difficult to meet the needs of modern agriculture for precision and intelligent cultivation. SUMMARY

[0003] The present application aims to at least solve the technical problems of strong subjectivity, low quantitative precision, and inability to synchronize and coordinate multiple constraint conditions in the prior art. It particularly innovatively proposes a plant spacing dynamic adjustment method and system for high-yield cotton cultivation.

[0004] To achieve the above-mentioned purposes of the present application, the present application provides a plant spacing dynamic adjustment method for high-yield cotton cultivation, characterized in that the method comprises: S1, collecting plant morphological parameters, field microenvironment parameters, and initial data of the target cotton plant's dense planting photosynthetic rate at different growth stages of cotton; S2, based on the plant morphological parameters, field microenvironment parameters, and initial data of the target cotton plant's dense planting photosynthetic rate, using a pre-constructed plant spacing adaptation model to calculate the optimal plant spacing range at the current growth stage; the plant spacing adaptation model takes the maximum yield of cotton dense planting as the objective function, and takes the canopy overlap rate, inter-row light transmittance, and soil humidity as the constraint conditions; S3, obtaining the current actual plant spacing data of the cotton field, comparing the actual plant spacing with the optimal plant spacing range, and determining whether adjustment is needed; if the actual plant spacing exceeds the optimal plant spacing range, generating a plant spacing adjustment instruction; S4, controlling the field automatic transplanting device to perform adjustment operations according to the plant spacing adjustment instruction; S5, after the adjustment operation is completed, the plant form parameters, the field microenvironment parameters and the dense planting photosynthetic rate of the target cotton plant in the adjustment area are collected again to verify whether the adjusted plant spacing meets the optimal plant spacing range; if not, steps S2 to S5 are repeated until the optimal plant spacing range is met.

[0005] In another aspect, the present application also provides a plant spacing dynamic adjustment system for cotton dense planting and high yield cultivation, the system comprising: a processor; a memory for storing processor executable instructions; wherein the processor is configured to implement the plant spacing dynamic adjustment method for cotton dense planting and high yield cultivation when executing the executable instructions.

[0006] The beneficial effects of the present application are: the present application constructs a plant spacing adaptation model integrating plant canopy overlap rate, inter-row light transmittance and soil humidity multi-constraint conditions with the goal of maximizing dense planting yield, and uses a genetic algorithm to solve the optimal plant spacing range, thereby realizing quantitative and accurate decision-making for plant spacing adjustment and overcoming the problems of strong subjectivity and low precision of traditional methods; the dynamic weight coefficient is matched with the growth stage characteristics, the multi-environment constraints and yield targets are coordinated synchronously, and the problem of synchronous coordination of multi-constraint conditions is solved; the model parameters are updated by regularly collecting measured data to ensure that the plant spacing adaptation model continuously adapts to environmental changes during the growth cycle, forming a closed-loop mechanism of "data collection-model optimization-dynamic adjustment-effect verification", and finally realizing accurate improvement of cotton dense planting yield and intelligentization of cultivation management.

[0007] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter in the description of the application. BRIEF DESCRIPTION OF DRAWINGS

[0008] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the description of the embodiments, given by reference to the following drawings: Figure 1 is a flowchart of a plant spacing dynamic adjustment method for cotton dense planting and high yield cultivation according to the present application. DETAILED DESCRIPTION

[0009] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, in which the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0010] Example 1 As shown in Figure 1 , a plant spacing dynamic adjustment method for cotton dense planting and high yield cultivation, characterized in that the method comprises: S1. Collect initial data on plant morphology parameters, field microenvironment parameters, and densely planted photosynthetic rate of target cotton plants at different growth stages. In step S1, it should be noted that plant morphological parameters include plant height, canopy diameter, number of main stem nodes, and functional leaf area; field microenvironment parameters include photosynthetically active radiation between rows, soil volumetric water content, relative humidity, and soil temperature. Initial data on the dense-planted photosynthetic rate of the target cotton plant is acquired using a dense-planting sensor at the base of the cotton boll. The specific operation is as follows: After the cotton boll has set (5-7 days after flowering), select target cotton bolls with uniform growth in the upper part of the cotton plant. From 9:00 to 11:00 daily (the peak period of natural photosynthesis in dense planting), slowly absorb photosynthetically active cotton from the nectary at the base of the cotton boll using a sterilized disposable microcapillary tube. Record the absorbed volume (accurate to 0.1 μL) and collection time, and calculate the photosynthetic rate per unit time (μL / h). For large-scale monitoring, a micro-capacitive dense-planting sensor installed at the base of the cotton boll can be used. The sensor senses the change in capacitance value after contact with dense planting, converts it into dense-planting volume data in real time, and synchronously transmits it to the data acquisition terminal. The initial data on the dense-planted photosynthetic rate of each target cotton boll is automatically calculated and stored.

[0011] S2. Based on plant morphological parameters, field microenvironment parameters and initial data of the photosynthetic rate of the target cotton plants in dense planting, the optimal plant spacing range for the current growth stage is calculated using a pre-constructed plant spacing adaptation model. The plant spacing adaptation model takes maximizing cotton yield in dense planting as the objective function and plant canopy overlap rate, inter-row light transmittance, and soil moisture as constraints. S3. Obtain the current actual plant spacing data of the cotton field, compare the actual plant spacing with the optimal plant spacing range, and determine whether adjustment is needed; if the actual plant spacing exceeds the optimal plant spacing range, generate a plant spacing adjustment command. S4. Control the automatic transplanting device in the field to perform adjustment operations according to the plant spacing adjustment command; In step S4, it is important to explain in detail that the automatic field transplanting device in this embodiment is existing technology. The specific operation process for adjusting plant spacing using an existing transplanting device is as follows: First, based on the generated plant spacing adjustment command, the target area and target plant spacing value to be adjusted are determined. Second, the command is transmitted to the control system of the automatic field transplanting device. After parsing the command, the control system drives the robotic arm of the transplanting device for precise positioning. Subsequently, the robotic arm rearranges the plants in the target area to ensure that the adjusted plant spacing meets the requirements of the optimal plant spacing range. Finally, the adjusted plant spacing data is monitored in real time by built-in sensors and fed back to the control system to verify the adjustment effect. If the monitoring results show that the adjustment does not achieve the expected accuracy, a secondary fine-tuning process is triggered until the plant spacing meets the set range.

[0012] S5. After the adjustment operation is completed, collect the plant morphology parameters, field microenvironment parameters and dense planting photosynthetic rate of the target cotton plants in the adjusted area again to verify whether the adjusted plant spacing meets the optimal plant spacing range; if not, repeat steps S2 to S5 until it meets the optimal plant spacing range.

[0013] In step S5, it is necessary to explain in detail that within 24-48 hours after the adjustment operation is completed (after the plant root system has been initially fixed and the canopy microenvironment has tended to stabilize), the plant morphology parameters of the target cotton plants in the adjustment area should be re-collected according to the parameter collection method in step S1 (focusing on measuring canopy diameter and functional leaf area; using a laser rangefinder to accurately obtain canopy diameter data, with the error controlled within ±0.5cm; functional leaf area is measured by scanning with a leaf area meter, recording the area of ​​a single leaf and the total area of ​​functional leaves of the whole plant), and field microenvironment parameters (prioritizing monitoring of photosynthetically active radiation between rows; using a portable photosynthetically active radiation meter to measure at 50cm above the ground in the middle of the cotton plant between rows, repeating 3 times at each point and taking the average value; soil volumetric water content). Data were collected using a time-domain reflectometer (TDR) in the root zone of cotton plants (20-30 cm deep), with soil moisture measured at three locations at each monitoring point to ensure data representativeness. The photosynthetic rate of the target cotton bolls was also measured (target cotton bolls with uniform growth in the upper and middle parts of the plant were selected; during the peak photosynthetic period of dense planting from 9:00 to 11:00 daily, photosynthetic data from the nectaries at the base of the boll were collected using a sterilized microcapillary tube; the collected volume (accurate to 0.1 μL) and collection time were recorded, and the photosynthetic rate per unit time was calculated; for large-scale monitoring, the dense planting volume data was acquired in real time using a microcapacitive dense planting sensor installed at the base of the cotton boll, and synchronously transmitted to the data acquisition terminal to automatically generate a photosynthetic rate curve for each target cotton boll). After data collection, the measured plant morphology parameters and field microenvironment parameters are input into the plant spacing adaptation model. The canopy overlap rate (the canopy coverage area of ​​a single plant is calculated by the canopy diameter and plant spacing, and then the proportion of canopy overlap area between rows is statistically analyzed), the light transmittance between rows (the measured photosynthetically active radiation value between rows is divided by the total photosynthetically active radiation value in the field at the same time), and the soil moisture (the average value of soil volumetric water content in the root zone) are then determined to see if the constraints are met (canopy overlap rate ≤ maximum allowable canopy overlap rate, light transmittance between rows ≥ minimum allowable light transmittance, soil moisture ≥ lower limit of soil moisture and ≤ optimal soil moisture). At the same time, the photosynthetic rate of the target cotton boll is compared with the photosynthetic rate of the same batch of cotton bolls before adjustment to assess whether the improvement in density has reached the preset threshold (e.g., ≥10%, which can be adjusted according to variety characteristics and growth stage).If all the above constraints are met and the photosynthetic rate of dense planting reaches the expected improvement target, then the adjusted plant spacing is determined to be within the optimal plant spacing range. If not met (e.g., the canopy overlap rate still exceeds the maximum allowable value, the light transmittance between rows does not meet the standard, or the photosynthetic rate of dense planting does not increase or even decreases), then the measured data and adjustment operation records of this verification need to be added to the model update dataset (for the dynamic update of the plant spacing adaptation model in step S6), and return to step S2, re-enter the verification data to calculate the new optimal plant spacing range, and then execute step S3 to generate plant spacing instructions for the secondary adjustment (clarifying the layer and area of ​​the secondary adjustment; for example, if the original plant spacing is too large and the canopy overlap rate exceeds the standard, the instruction adjustment direction is to reduce the plant spacing, and the adjustment layer is 50% of the original deviation), step S4 to control the automatic transplanting device to perform precise secondary adjustment (the positioning accuracy of the robotic arm is improved to ±1cm to avoid secondary damage to the plant roots), until step S5 verification is passed (constraints are met and the photosynthetic rate of dense planting reaches the expected value).

[0014] The principle of the method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation in this embodiment is as follows: First, by collecting plant morphological parameters (such as plant height, canopy diameter, and functional leaf area), field microenvironment parameters (such as photosynthetically active radiation between rows and soil volumetric water content), and initial data on the photosynthetic rate of the target cotton plants at different growth stages, multi-dimensional inputs are provided for the construction of the plant spacing adaptation model. Then, with maximizing cotton yield through dense planting as the core objective, the model incorporates plant canopy overlap rate (to avoid excessive canopy density leading to poor ventilation and light penetration), light transmittance between rows (to ensure the photosynthetic needs of the lower leaves), and soil moisture (to maintain root absorption function and support dense planting). Using photosynthetic water supply as a constraint, a dynamic weighting coefficient (matching the characteristics of different growth stages, such as canopy-light coordination during the vegetative growth stage and soil moisture-photosynthetic synergy during the reproductive growth stage) is integrated to construct a plant spacing adaptation model. Subsequently, a genetic algorithm is used to solve the model. Through encoding plant spacing variables, setting a fitness function (a comprehensive evaluation of the objective function and constraints), and performing operations such as selection (preserving optimal solutions), crossover (combining optimal solutions), and mutation (exploring new solutions), the optimal plant spacing range that satisfies all constraints is searched, achieving optimal plant spacing. The system makes precise decisions regarding spacing adjustments. It then compares the current actual plant spacing with the optimal range. If the spacing exceeds the optimal range, it generates instructions including the target plant spacing, adjustment direction (expansion or reduction), and adjustment area. This controls an automatic field transplanting device (using a robotic arm to precisely position and rearrange plants) to perform the adjustment. After adjustment, parameters of the adjusted area are re-collected within 24-48 hours to verify if the plant spacing meets the optimal range (constraints are met and the photosynthetic rate of dense planting reaches a preset threshold). If not, the adjustment process is repeated. Finally, measured data (plant morphology, environmental parameters, photosynthetic rate of dense planting, and adjustment records) are collected every n days to retrain and optimize the dynamic weight coefficients and plant spacing-photosynthetic rate function parameters of the plant spacing adaptation model. After verifying performance (matching rate ≥ r% and relative error ≤ v%), the original model is replaced. This forms a closed-loop mechanism of "data collection - model optimization - dynamic adjustment - effect verification," ensuring the model continuously adapts to environmental changes during the growth cycle, maintaining the accuracy and effectiveness of plant spacing adjustments, and ultimately achieving a precise increase in cotton yield through dense planting.

[0015] As an optional embodiment of the present invention, optionally, calculating the optimal plant spacing range for the current growth stage using a pre-constructed plant spacing adaptation model in step S2 includes: S201. Obtain standardized input data from initial data on plant morphology parameters, field microenvironment parameters, and the densely planted photosynthetic rate of the target cotton plants. In step S201, it should be noted that the process of acquiring standardized input data is as follows: First, the plant morphological parameters are normalized by mapping plant height, canopy diameter, number of main stem nodes, and functional leaf area to the [0,1] interval to eliminate the influence of dimensional differences on model calculation. Second, the field microenvironment parameters are standardized, including inter-row photosynthetically active radiation, soil volumetric water content, relative humidity, and soil temperature, using the Z-score standardization method to convert them into standard normal distribution data with a mean of 0 and a standard deviation of 1. Finally, for the initial data of the densely planted photosynthetic rate of the target cotton plants, a linear transformation is used to scale it to a uniform range, and outliers (such as data points exceeding the mean ± 3 standard deviations) are removed to ensure the stability and reliability of the input data. All standardized data are integrated into an input matrix that can be directly read by the model according to a preset format.

[0016] S202. Based on standardized input data, construct a plant spacing adaptation model with the objective function of maximizing yield in dense planting, and introduce dynamic weight coefficients to adjust the priority of each element in the objective function of the plant spacing adaptation model according to the characteristics of the growth stage. In step S202, it should be noted that the introduction of dynamic weighting coefficients aims to match the characteristic requirements of cotton at different growth stages, ensuring that the plant spacing adaptation model can flexibly adjust the priority of each element within the objective function under multiple constraints. For example, during the vegetative growth stage, canopy overlap rate and inter-row light transmittance have a more significant impact on plant health and development; therefore, the dynamic weighting coefficients will assign higher priority to these two factors. During the reproductive growth stage, the correlation between soil moisture and dense-plant photosynthetic rate increases, at which point the model will automatically adjust the weights, focusing on optimizing the synergistic relationship between soil moisture and dense-plant photosynthesis. In this way, the model can dynamically balance the contradictions between multiple objectives at different growth stages, thereby achieving more accurate plant spacing decisions. Furthermore, the specific values ​​of the dynamic weighting coefficients are derived from historical data training and periodically updated in conjunction with real-time collected field microenvironment parameters to ensure that they always reflect the actual needs of the current growth stage. S203. Based on the objective function adjusted by dynamic weight coefficients, plant canopy overlap rate, inter-row light transmittance and suitable soil moisture range are introduced as constraints to construct a mathematical optimization model containing the objective function and constraints. The mathematical optimization model is solved by genetic algorithm to obtain the optimal plant spacing range that satisfies the constraints and maximizes the yield of dense planting at the current growth stage.

[0017] In step S203, it should be noted that the solution process of the genetic algorithm specifically includes the following steps: First, the plant spacing variable is encoded by discretizing the plant spacing range into chromosome form that can be operated on by the algorithm using binary or real number encoding. Second, a fitness function is set, which comprehensively considers the objective function (maximizing yield in dense planting) and constraints (canopy overlap rate, inter-row light transmittance, and soil moisture). The constraints are incorporated into the objective function through weighted summation or penalty function methods to ensure the feasibility and optimality of the solution. Next, a selection operation is performed, selecting better solutions from the current population based on individual fitness values ​​and retaining them for the next generation. At the same time, a crossover operation is introduced, randomly exchanging some gene fragments between two parent individuals with a certain probability to generate new offspring individuals. Finally, a mutation operation is performed, randomly changing the values ​​of certain gene loci with a small probability to increase the diversity of the population and explore the potential space of better solutions. During each generation of evolution, the optimal solution in the current population is recorded, and it is determined whether the termination condition is met (such as reaching the preset maximum number of iterations or fitness value convergence). If the termination condition is met, the range of plant spacing corresponding to the optimal solution is output as the optimal plant spacing for the current growth stage; otherwise, the iteration continues until the condition is met. In addition, the following points should be noted when using genetic algorithms to solve the problem: First, the initial population should cover as much of the feasible solution space as possible to improve global search capability; second, the crossover and mutation probabilities should be set reasonably to avoid excessively high crossover rates leading to the loss of superior genes, or excessively low mutation rates causing the algorithm to get stuck in local optima; third, the weights of each element in the fitness function should be dynamically adjusted according to the characteristics of different growth stages to ensure that the algorithm can flexibly respond to environmental changes and differences in demand. Ultimately, the optimal plant spacing range obtained through the above optimization process not only satisfies all constraints but also maximizes the yield of densely planted cotton.

[0018] As an optional embodiment of the present invention, optionally, the expression of the objective function in step S202 is: in, This indicates the target yield value for densely planted crops. express The direct impact of plant spacing at any given time on yield has a weighting. The optimal photosynthetic rate for dense planting at this stage is represented. Indicates the current plant spacing. express The plant spacing-dense photosynthetic rate function at time point. express The compensation weight for canopy overlap at time step. Indicates the canopy overlap rate. Indicates the maximum permissible canopy overlap. express Light transmittance weight at any given time Indicates the light transmittance between rows. This represents the minimum permissible light transmittance. express Soil moisture weight at any given time Indicates soil moisture. Indicates the lower limit of soil moisture. This indicates the optimal soil moisture level.

[0019] As an optional embodiment of the present invention, optionally, the expression of the mathematical optimization model in step S203 is: ; , ; , ; in, This represents the overall optimization value for the stage. express Priority weighting of yield based on planting density at specific times. This represents the plant spacing-dense planting photosynthetic function. express Priority weight of canopy overlap rate at time step. Represents the utility function of canopy overlap rate. Indicates the priority weight of light transmittance. This represents the light transmittance utility function. Indicates the priority weight of soil moisture. Represents the soil moisture utility function. Indicates the canopy overlap rate. Indicates the maximum permissible canopy overlap. Indicates the light transmittance between rows. This represents the minimum permissible light transmittance. This represents the optimal light transmittance. Indicates soil moisture. Indicates the lower limit of soil moisture. Indicates the optimal soil moisture. Indicates the minimum plant spacing. This indicates the maximum plant spacing.

[0020] As an optional embodiment of the present invention, optionally, in step S3, the current actual plant spacing data of the cotton field is obtained, the actual plant spacing is compared with the optimal plant spacing range, and it is determined whether adjustment is needed; if the actual plant spacing exceeds the optimal plant spacing range, a plant spacing adjustment instruction is generated, including: S301. Collect actual plant spacing data in cotton fields; In step S301, it should be noted that the actual plant spacing data collection process is completed collaboratively by sensors and an automatic field inspection device. The sensor array is deployed in the cotton field to monitor changes in plant distance in real time, while the automatic inspection device moves along a preset path, using a laser ranging module and visual recognition technology to accurately measure the plant spacing. The collected data is filtered to remove noise interference and uploaded to the central control system via a wireless communication module to form a complete plant spacing distribution map. Furthermore, to ensure data accuracy, the system performs cross-validation after each collection, comparing the sensor measurements with the image analysis results. If the deviation exceeds a preset threshold, a manual review mechanism is triggered.

[0021] S302. Based on the optimal plant spacing range, extract the minimum and maximum plant spacing corresponding to the current growth stage as the judgment threshold. In step S302, it should be noted that the extraction process of the judgment threshold needs to be adjusted in conjunction with the dynamic weight coefficient of the current growth stage. Specifically, based on the optimal plant spacing range obtained in step S203, the minimum and maximum plant spacing are mapped to the lower and upper limits of the judgment threshold, respectively. Simultaneously, considering the different sensitivity of plant spacing at different growth stages, the system introduces a buffer zone, the width of which is determined by historical data and real-time environmental parameters. For example, during the vegetative growth stage, due to the rapid expansion of the plant canopy, the system will appropriately expand the buffer zone to avoid frequent adjustments; while during the reproductive growth stage, to ensure maximum photosynthetic rate in dense planting, the buffer zone will be reduced to improve adjustment accuracy. Furthermore, the setting of the judgment threshold also needs to comprehensively consider the changing trends of field microenvironment parameters, such as fluctuations in soil moisture or short-term anomalies in light conditions, thereby improving the robustness of the decision-making.

[0022] S303. Calculate the absolute deviation between the actual plant spacing data and the median of the optimal plant spacing range. And determine whether the absolute deviation satisfies ,in, Indicates the minimum plant spacing. Indicates the maximum plant spacing; In step S303, it should be noted that if the absolute deviation does not exceed the optimal plant spacing range, the current plant spacing is determined to be within a reasonable range and no adjustment is needed; if the absolute deviation exceeds the range, the system proceeds to the next step of generating adjustment instructions. Specifically, the system determines the specific adjustment strategy based on the direction of deviation between the actual plant spacing and the median of the optimal plant spacing. For example, when the actual plant spacing is less than the minimum plant spacing, the system will prioritize increasing the plant spacing to avoid excessive competition for resources among plants; while when the actual plant spacing is greater than the maximum plant spacing, the system will reduce the plant spacing to increase the yield potential of dense planting per unit area. In addition, the real-time changes in field microenvironment parameters, such as soil moisture and light intensity, must be comprehensively considered during the judgment process to ensure that the adjustment plan can adapt to the current actual growth conditions.

[0023] S304, if If the threshold is exceeded, a plant spacing adjustment instruction is generated based on the deviation direction, which includes the target plant spacing, adjustment direction, and adjustment area.

[0024] In step S304, it should be noted that the generated plant spacing adjustment command must include detailed execution information to ensure the accuracy and efficiency of field operations. The target plant spacing is calculated by the system based on the deviation between the actual plant spacing and the optimal plant spacing range, and fine-tuned in conjunction with the dynamic weighting coefficient of the current growth stage to adapt to the specific needs of different stages. The adjustment direction is determined based on the specific deviation; for example, when the actual plant spacing is less than the minimum plant spacing, the adjustment direction is to increase the plant spacing; conversely, it is to decrease the plant spacing. The delineation of the adjustment area is based on the location of abnormal areas in the plant spacing distribution map, and the range of plants requiring adjustment is accurately identified through sensor data and image analysis results.

[0025] In addition, plant spacing adjustment instructions must include priority indicators to prioritize areas with the greatest impact on yield from dense planting, given limited resources. Priority settings comprehensively consider the degree of deviation, trends in field microenvironment parameters, and historical adjustment records to achieve global optimization. After the instruction is issued, the central control system monitors the adjustment process in real time, verifies the adjustment effect through a feedback mechanism, and makes dynamic corrections based on actual conditions. If the adjusted plant spacing still does not reach the optimal range, the adjustment instruction is regenerated until the requirements are met.

[0026] As an optional embodiment of the present invention, the method may further include: S6. During the cotton growth cycle, the plant spacing adaptation model is dynamically updated every n days.

[0027] The dynamic update in step S6 includes: S601. Collect plant morphology parameters, field microenvironment parameters, and measured data of densely planted photosynthetic rate of target cotton plants at each monitoring point in the cotton field during the update cycle, and collect the execution records of plant spacing adjustment operations during this cycle. In step S601, it should be noted that the collected data must undergo preprocessing to ensure its quality and consistency. Specifically, plant morphological parameters, including plant height, canopy width, and leaf coverage, are extracted using image recognition technology combined with sensor data. Field microenvironment parameters, encompassing soil moisture, light intensity, temperature, and carbon dioxide concentration, are monitored and recorded in real time by multiple types of sensors. Measured data on the photosynthetic rate of densely planted crops are obtained using dedicated sampling equipment and calibrated in conjunction with laboratory analysis results. For the execution records of plant spacing adjustments, the system automatically summarizes the time, area, adjustment level, and post-adjustment effect evaluation data for each adjustment. All collected information will be integrated into a unified database, and data cleaning methods will be used to remove outliers and noise interference, while missing values ​​will be filled to ensure the completeness of subsequent analysis.

[0028] Furthermore, after data preprocessing, the system analyzes the overall trend within the update cycle, extracting key characteristic variables and their changing patterns. For example, if a significant decrease in photosynthetic rate at a monitoring point due to dense planting is observed, accompanied by limited canopy expansion, it may indicate intensified resource competition or environmental stress in that area. These analytical results serve as crucial inputs for the dynamic updating of the plant spacing adaptation model, used to recalibrate model parameters and optimize weight coefficient configuration. In this way, the model can better reflect the actual needs of the current growth stage, thereby continuously maximizing the yield potential of densely planted cotton.

[0029] S602. Preprocess the plant morphology parameters, field microenvironment parameters, measured data of the photosynthetic rate of the target cotton plants under dense planting, and execution records to form a model update dataset. In step S602, it should be noted that the preprocessing mainly includes data cleaning, feature extraction, and normalization. Data cleaning aims to remove outliers and noise interference, such as eliminating unreasonable data points caused by sensor malfunctions or sudden environmental changes, while filling in missing values ​​to ensure data integrity. Feature extraction involves reducing the dimensionality and transforming the original data to extract key variables that significantly affect model updates, such as plant canopy expansion rate and soil moisture fluctuation level. Normalization unifies data of different dimensions to the same scale to eliminate the interference of numerical range differences on model training and improve computational efficiency. In addition, when forming the model update dataset, the system also performs stratified sampling to ensure that the sample distribution can comprehensively cover different growth stages and field microenvironment conditions, thereby improving the model's generalization ability.

[0030] After preprocessing, the system calibrates the parameters and optimizes the weights of the plant spacing adaptation model based on the updated dataset. Specifically, by introducing incremental learning algorithms (such as online learning or mini-batch gradient descent), the model can quickly adapt to the changing trends of new data while retaining historical knowledge. For example, when it is found that the photosynthetic rate of dense planting at a certain monitoring point is continuously decreasing and is associated with plant spacing adjustment records, the system will reassess the optimal plant spacing range for that area and dynamically adjust the weights of each element in the fitness function. This dynamic update mechanism not only reflects the actual changes in the field environment in a timely manner but also effectively avoids model performance degradation caused by long-term fixed parameters. Ultimately, the updated model will guide subsequent plant spacing adjustment operations with higher accuracy.

[0031] S603. Divide the model update dataset into a training set and a test set. Use the training set to retrain the dynamic weight coefficients and parameters of the plant spacing-dense planting photosynthetic rate function in the plant spacing adaptation model, and use the optimizer to optimize them. At the same time, take maximizing cotton dense planting yield as the objective function, and plant canopy overlap rate, inter-row light transmittance, and soil moisture as constraints. In step S603, it should be noted that the ratio of the training set to the test set needs to be dynamically adjusted based on the amount and distribution characteristics of the data to ensure the accuracy and stability of the model update. Typically, the system uses cross-validation to split the dataset, ensuring the representativeness of the training and test sets under different growth stages and environmental conditions. During retraining, the dynamic weight coefficients in the plant spacing adaptation model will be fine-tuned based on the latest collected data, especially parameters related to dense planting photosynthetic rate, canopy overlap, light transmittance, and soil moisture. The degree of adjustment needs to be weighed in conjunction with the priority requirements of the current growth stage.

[0032] The choice of optimizer is based on limitations of model complexity and computational resources. For example, adaptive learning rate algorithms (such as Adam or RMSProp) are used to improve convergence speed and avoid getting trapped in local optima. Meanwhile, the objective function design needs further refinement, decomposing the maximization of cotton yield through dense planting into multiple sub-objectives, including improving photosynthetic rate per unit area, optimizing plant health index, and improving resource utilization efficiency. Regarding constraints, in addition to existing indicators, new boundary constraints need to be introduced, such as a quantitative assessment of inter-plant competition pressure and tolerance for field microenvironment fluctuations, to ensure that the model output adjustment scheme better reflects actual needs.

[0033] Furthermore, after model training is complete, the system will comprehensively evaluate the performance of the updated model using a test set. Evaluation metrics include prediction accuracy, generalization ability, and robustness, especially performance in previously unseen scenarios. If the test results do not meet the expected standards, secondary optimization will be performed by adjusting training parameters or expanding the dataset. Finally, the validated updated model will be deployed to the central control system and synchronized to all execution terminals.

[0034] S604. Use the test set to verify the performance of the updated plant spacing adaptation model. Calculate the matching rate between the optimal plant spacing range predicted by the plant spacing adaptation model and the actual adjusted plant spacing. If the matching rate is ≥ r% (90%) and the relative error is ≤ v% (4%), then replace the original plant spacing adaptation model with the updated model. If not, adjust the learning rate of the optimizer and return to S603 for retraining.

[0035] In step S6044, it should be noted that during the verification process, the matching rate is calculated based on sample data in the test set. The model's practical guiding ability is evaluated by comparing the optimal plant spacing range predicted by the model with the actual adjusted plant spacing distribution. The relative error reflects the degree of deviation between the predicted and actual values, and its calculation formula is the ratio of the absolute value of the difference between the predicted and actual values ​​to the actual value. If both the matching rate and the relative error meet the preset standards, it indicates that the updated plant spacing adaptation model has high practicality and reliability, and can directly replace the original model and be put into operation. If the expected performance is not met, the system will automatically trigger an optimization mechanism to adjust the optimizer's learning rate to improve model performance. The learning rate adjustment strategy typically employs dynamic decay methods, such as exponential decay or piecewise constant decay, to balance the model's convergence speed and accuracy. Furthermore, during retraining, the system will further analyze the data distribution of the training and test sets to identify potential sample imbalance issues and address them through oversampling or undersampling. It's worth noting that when the target performance is not achieved after multiple iterations, the system generates an anomaly report, detailing the key parameters, optimization process, and reasons for failure for each training iteration. This information is fed back to the manual monitoring platform for technical personnel to reference and intervene manually. Simultaneously, the system suggests expanding the data collection scope or extending the update cycle to obtain more high-quality training data, thereby improving the model's generalization ability and robustness. Ultimately, this closed-loop optimization process ensures that the plant spacing adaptation model is always in optimal condition.

[0036] As an optional embodiment of the present invention, optionally, the expression for calculating the matching rate between the optimal plant spacing range predicted by the plant spacing adaptation model and the actual adjusted plant spacing in step S604 is as follows: in, Indicates the plant spacing matching rate. Indicates the total number of adjustments. Indicates an indicator function, Indicates the first The actual plant spacing after the second adjustment Indicates the first The optimal plant spacing is predicted in the next step. This represents the matching tolerance coefficient. This indicates a predicted plant spacing range of half width.

[0037] Example 2 A dynamic plant spacing adjustment system for high-density, high-yield cotton cultivation includes: processor; Memory used to store processor-executable instructions; The processor is configured to implement a method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation when executing executable instructions.

[0038] It should be noted that the computer device includes a processor, a memory, and may also include one or more of a multimedia component, an input / output (I / O) interface, and a communication component.

[0039] The processor controls the overall operation of the computer device to complete all or part of the steps in the above-mentioned method for dynamically adjusting plant spacing in high-density cotton cultivation.

[0040] Memory is used to store various types of data to support the operation of the computer device. This data may include, for example, instructions for any application or method used to operate on the computer device, as well as application-related data. Memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0041] The multimedia component may include a screen and an audio component, wherein the screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals; for example, the audio component may include a microphone for receiving external audio signals, the received audio signals may be further stored in memory or transmitted via a communication component; the audio component may also include at least one speaker for outputting audio signals.

[0042] I / O interfaces provide interfaces between the processor and other interface modules, such as keyboards, mice, buttons, etc.; these buttons can be virtual buttons or physical buttons.

[0043] The communication component is used for wired or wireless communication between the computer device and other devices; wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G or 5G, or one or more combinations thereof, and the corresponding communication component may include: Wi-Fi module, Bluetooth module, NFC module, mobile communication module.

[0044] As a preferred embodiment, the computer device may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the above-described method for dynamically adjusting plant spacing in high-density cotton cultivation.

[0045] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation, characterized in that the method... include: S1. Collect initial data on plant morphology parameters, field microenvironment parameters, and densely planted photosynthetic rate of target cotton plants at different growth stages. S2. Based on plant morphological parameters, field microenvironment parameters and initial data of photosynthetic rate of densely planted target cotton plants, the optimal plant spacing range for the current growth stage is calculated using a pre-constructed plant spacing adaptation model. The plant spacing adaptation model takes maximizing cotton yield in dense planting as the objective function and plant canopy overlap rate, inter-row light transmittance, and soil moisture as constraints. S3. Obtain the current actual plant spacing data of the cotton field, compare the actual plant spacing with the optimal plant spacing range, and determine whether adjustment is needed; if the actual plant spacing exceeds the optimal plant spacing range, generate a plant spacing adjustment command. S4. Control the automatic transplanting device in the field to perform adjustment operations according to the plant spacing adjustment command; S5. After the adjustment operation is completed, collect the plant morphology parameters, field microenvironment parameters and dense planting photosynthetic rate of the target cotton plants in the adjusted area again to verify whether the adjusted plant spacing meets the optimal plant spacing range; if not, repeat steps S2 to S5 until it meets the optimal plant spacing range.

2. The method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation as described in claim 1, characterized in that, In step S2, the optimal plant spacing range for the current growth stage is calculated using a pre-built plant spacing adaptation model, including: S201. Obtain standardized input data from initial data on plant morphology parameters, field microenvironment parameters, and the densely planted photosynthetic rate of the target cotton plants. S202. Based on standardized input data, construct a plant spacing adaptation model with the objective function of maximizing yield in dense planting, and introduce dynamic weight coefficients to adjust the priority of each element in the objective function of the plant spacing adaptation model according to the characteristics of the growth stage. S203. Based on the objective function adjusted by dynamic weight coefficients, plant canopy overlap rate, inter-row light transmittance and suitable soil moisture range are introduced as constraints to construct a mathematical optimization model containing the objective function and constraints. The mathematical optimization model is solved by genetic algorithm to obtain the optimal plant spacing range that satisfies the constraints and maximizes the yield of dense planting at the current growth stage.

3. The method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation as described in claim 2, characterized in that, The expression for the objective function in step S202 is: in, This indicates the target yield value for densely planted crops. express The direct impact of plant spacing at any given time on yield has a weighting. The optimal photosynthetic rate for dense planting at this stage is represented. Indicates the current plant spacing. express The plant spacing-dense photosynthetic rate function at time point. express The compensation weight for canopy overlap at time step. Indicates the canopy overlap rate. Indicates the maximum permissible canopy overlap. express Light transmittance weight at any given time Indicates the light transmittance between rows. This represents the minimum permissible light transmittance. express Soil moisture weight at any given time Indicates soil moisture. Indicates the lower limit of soil moisture. This indicates the optimal soil moisture level.

4. The method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation as described in claim 2, characterized in that, The expression for the mathematical optimization model in step S203 is: ； ， ； ， ； in, This represents the overall optimization value for the stage. express Priority weighting of yield based on planting density at specific times. This represents the plant spacing-dense planting photosynthetic function. express Priority weight of canopy overlap rate at time step. Represents the utility function of canopy overlap rate. Indicates the priority weight of light transmittance. This represents the light transmittance utility function. Indicates the priority weight of soil moisture. Represents the soil moisture utility function. Indicates the canopy overlap rate. Indicates the maximum permissible canopy overlap. Indicates the light transmittance between rows. This represents the minimum permissible light transmittance. This represents the optimal light transmittance. Indicates soil moisture. Indicates the lower limit of soil moisture. Indicates the optimal soil moisture. Indicates the minimum plant spacing. This indicates the maximum plant spacing.

5. The method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation as described in claim 1, characterized in that, In step S3, the current actual plant spacing data of the cotton field is obtained, and the actual plant spacing is compared with the optimal plant spacing range to determine whether adjustment is needed. If the actual plant spacing exceeds the optimal plant spacing range, the plant spacing adjustment instructions generated include: S301. Collect actual plant spacing data in cotton fields; S302. Based on the optimal plant spacing range, extract the minimum and maximum plant spacing corresponding to the current growth stage as the judgment threshold. S303. Calculate the absolute deviation between the actual plant spacing data and the median of the optimal plant spacing range. And determine whether the absolute deviation satisfies ,in, Indicates the minimum plant spacing. Indicates the maximum plant spacing; S304, if If the threshold is exceeded, a plant spacing adjustment instruction is generated based on the deviation direction, which includes the target plant spacing, adjustment direction, and adjustment area.

6. The method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation as described in claim 1, characterized in that, The method also includes: S6. During the cotton growth cycle, the plant spacing adaptation model is dynamically updated every n days.

7. The method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation as described in claim 6, characterized in that, The dynamic update in step S6 includes: S601. Collect plant morphology parameters, field microenvironment parameters, and measured data of densely planted photosynthetic rate of target cotton plants at each monitoring point in the cotton field during the update cycle, and collect the execution records of plant spacing adjustment operations during this cycle. S602. Preprocess the plant morphology parameters, field microenvironment parameters, measured data of the photosynthetic rate of the target cotton plants under dense planting, and execution records to form a model update dataset. S603. Divide the model update dataset into a training set and a test set. Use the training set to retrain the dynamic weight coefficients and parameters of the plant spacing-dense planting photosynthetic rate function in the plant spacing adaptation model, and use the optimizer to optimize them. At the same time, take maximizing cotton dense planting yield as the objective function, and plant canopy overlap rate, inter-row light transmittance, and soil moisture as constraints. S604. Use the test set to verify the performance of the updated plant spacing adaptation model. Calculate the matching rate between the optimal plant spacing range predicted by the plant spacing adaptation model and the actual adjusted plant spacing. If the matching rate is ≥r% and the relative error is ≤v%, then replace the original plant spacing adaptation model with the updated model. If not, adjust the learner rate of the optimizer and return to S603 for retraining.

8. The method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation as described in claim 7, characterized in that, In step S604, the expression for calculating the matching rate between the optimal plant spacing range predicted by the plant spacing adaptation model and the actual adjusted plant spacing is as follows: in, Indicates the plant spacing matching rate. Indicates the total number of adjustments. Indicates an indicator function, Indicates the first The actual plant spacing after the second adjustment Indicates the first The optimal plant spacing is predicted in the next step. This represents the matching tolerance coefficient. This indicates a predicted plant spacing range of half width.

9. The method for dynamically adjusting plant spacing in high-density, high-yield cotton cultivation as described in claim 1, characterized in that, Plant morphological parameters include plant height, canopy diameter, number of main stem nodes, and functional leaf area; Field microenvironment parameters include inter-row photosynthetically active radiation, soil volumetric water content, relative humidity, and soil temperature.

10. A dynamic plant spacing adjustment system for high-density, high-yield cotton cultivation, characterized in that the system... include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method for dynamic adjustment of plant spacing in high-yield cotton dense planting according to any one of claims 1 to 9 when executing executable instructions.

Images