Crop growth environment data dynamic monitoring and regulation method based on nano sensing

Abstract

The application discloses a crop growth environment data dynamic monitoring and regulation method based on nano sensing, relates to the technical field of crop environment monitoring and management, and synchronously acquires crop growth environment data and leaf inclination angle data through a deployed nano sensing network; an environment variation vector reflecting environment mutation intensity is constructed based on the environment data, combined with leaf posture abnormal characteristic values processed by a trigonometric function, and a comprehensive risk value is calculated; a spatial risk distribution matrix reflecting the continuous distribution of the risk space is generated by using an inverse distance weighted interpolation method. By analyzing the characteristics of the matrix, transient interference caused by wind gust and the like and sustained stress caused by environment drastic change are intelligently distinguished, and two differentiated response management strategies, namely, inhibitory regulation and hierarchical regulation, are triggered. Early, accurate and spatially resolved perception and preventive intervention on the flash in the facility are realized, and the problems of high false alarm rate, response lag and extensive regulation of traditional methods are effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of crop environment monitoring and management, and more particularly to a crop growth environment data dynamic monitoring and regulation method based on nanosensing. BACKGROUND

[0002] In intelligent agricultural production, crops in the seedling stage are extremely sensitive to dramatic changes in the growth environment, especially in early spring or autumn and winter. The environment in agricultural production facilities is prone to rapid fluctuations in temperature, light, and humidity due to sudden weather changes. Such dramatic changes in a short period of time often lead to an instantaneous transpiration rate of crops that exceeds the water supply capacity of the root system, causing acute physiological water loss and wilting of leaves, i.e., flash seedling phenomenon. This phenomenon occurs rapidly and is highly harmful. Once visible symptoms appear, irreversible damage to the plant tissue often occurs, severely affecting the survival rate of seedlings and subsequent growth. It is a long-standing and difficult problem in facility seedling and early cultivation management.

[0003] Traditionally, monitoring and prevention of flash seedling mainly rely on threshold monitoring of a single or a few environmental parameters in the facility. For example, when the temperature or light exceeds a certain fixed limit, global ventilation or shading measures are initiated. However, this response mode based on point monitoring and fixed thresholds has inherent limitations. First, the environment inside the facility is not uniform. Due to the influence of structure, orientation, external airflow, and internal equipment, factors such as temperature, light, and humidity often exhibit complex gradient or patchy distribution in space. Near the ventilation port, door and window, or areas directly exposed to sunlight, environmental parameter changes are often much earlier and more dramatic than in the internal areas. Traditional point monitoring is difficult to capture this spatial heterogeneity, which can easily lead to missed judgment of local high-risk areas or misjudgment and over-intervention of overall low-risk areas. Second, the essence of flash seedling is the collapse of crop physiological water balance, and traditional methods only rely on environmental physical parameters as indirect proxy indicators, which cannot directly sense the physiological stress state of crops.

[0004] There is a time difference between the time when the environmental parameters meet the standard and when the crops show visible wilting. This makes the traditional early warning lagging, and it cannot achieve real prevention. Furthermore, even if an early warning is issued, the traditional regulation method is to treat the entire facility space without discrimination and with uniform intensity, such as opening all ventilation windows to the same opening degree. This one-size-fits-all regulation mode not only has high energy consumption and low efficiency, but also may cause secondary stress to crops due to unnecessary environmental disturbance in non-risk areas.

[0005] To address the aforementioned issues, there is an urgent need in this field for an intelligent early warning and control method capable of accurately identifying the risk of seedling flashes within facilities, particularly their spatial distribution characteristics. This method must overcome the limitations of traditional point-based monitoring, enabling the perception of uneven environmental changes at a global spatial level; it must move beyond reliance on simple environmental parameters, integrating direct or indirect characteristics reflecting early physiological responses in crops; and finally, it must be able to generate and execute differentiated, locally precise control strategies based on the spatial distribution map of the risk, thereby implementing appropriate measures in areas where intervention is truly needed, achieving a shift from lagging and extensive homogeneous control to proactive and precise targeted intervention.

[0006] Existing technologies have not yet been able to effectively solve the specific technical challenge of early, accurate, and spatially differentiated perception and control of seedling flash risk in complex spatial heterogeneous environments. Summary of the Invention

[0007] To address the aforementioned technical problems, this technical solution provides a method for dynamic monitoring and control of crop growth environment data based on nanosensors, thus resolving the issues raised in the background section.

[0008] In a first aspect, embodiments of this application provide a method for dynamic monitoring and control of crop growth environment data based on nanosensors, comprising the following steps: acquiring environmental data streams and leaf tilt angles of target grid points; constructing an environmental variation vector for the target grid points based on the environmental data streams; processing the leaf tilt angles using a sine function to obtain leaf posture anomaly characteristic values; arbitrarily selecting several discrete points within a preset radius of the target grid points, denoted as first-level grid points; calculating the comprehensive risk value of the first-level grid points based on the Euclidean norm of the environmental variation vectors and the leaf posture anomaly characteristic values; and calculating the comprehensive risk value of each first-level grid point based on its relative spatial position and comprehensive risk value with respect to the target grid point, using inverse distance... A spatial risk distribution matrix is ​​constructed using a weighted interpolation method. The mean and standard deviation of the comprehensive risk values ​​of all first-level grid points in the spatial risk distribution matrix are calculated. If the comprehensive risk value of a first-level grid point in the spatial risk distribution matrix exceeds a preset intensity threshold and the standard deviation exceeds a preset standard deviation threshold, then the first-level grid point is marked as a second-level grid point. The average Euclidean distance between the second-level grid point and all other second-level grid points within a preset radius is calculated and recorded as the risk dispersion. When the mean exceeds a preset risk threshold, if the risk dispersion exceeds a preset dispersion threshold, then inhibitory control is implemented. If the risk dispersion does not exceed a preset dispersion threshold, then tiered control is implemented.

[0009] Secondly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for dynamic monitoring and control of crop growth environment data based on nanosensors.

[0010] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0011] 1. By constructing a spatial risk distribution matrix, discrete sensor information is fused into a continuous spatial risk field, and the risk intensity and risk dispersion characteristics are analyzed, which can clearly identify whether the risk is global, locally aggregated or randomly dispersed. This not only determines whether there is a risk, but also determines where the risk is and how it is distributed, thereby providing a reliable spatial basis for subsequent precise intervention, achieving precise perception and positioning of spatial heterogeneity of risk, and solving the problem of missed / incorrect reporting in traditional point monitoring.

[0012] 2. Not only the mutation rate of light intensity, air temperature and air humidity is monitored, but also the early stress signals of crops are captured simultaneously. By calculating the comprehensive risk value, both are fused, so that the early warning signal is derived from the cross verification of environmental driving and crop physiological response, which can identify the stress risk earlier and more reliably before the visible wilting symptoms appear in crops, and fuse the environmental mutation and early physiological response signals of crops, achieving truly advanced warning and overcoming the lag of traditional fixed environmental threshold warning.

[0013] 3. According to the risk dispersion and other characteristics, gust disturbance and sustained stress are automatically distinguished, avoiding unnecessary intervention. For real sustained stress, the maximum risk area position coordinates and spatial direction vector are identified by spatial gradient analysis, and then hierarchical regulation along the risk gradient is implemented, and different regions are matched with different intensity of ventilation, atomization and shading measures. This strategy greatly optimizes resource allocation, effectively controls the core risk while minimizing interference and energy consumption in non-risk areas, proposes an intelligent decision mechanism of inhibitory regulation and hierarchical regulation, and realizes precise and efficient regulation according to demand. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 A step schematic diagram of the crop growth environment data dynamic monitoring and regulation method based on nano sensing provided by the embodiments of the present application is provided.

[0015] Figure 2 A logic flow schematic diagram of hierarchical regulation provided by the embodiments of the present application is provided. DETAILED DESCRIPTION

[0016] The crop growth environment data dynamic monitoring and regulation method based on nano sensing provided by the embodiments of the present application solves the technical problem in the prior art that the flash seedling risk of facility seedlings in a complex spatial heterogeneity environment cannot be perceived and prevented early, accurately and spatially.

[0017] The existing technology is difficult to effectively perceive and cope with the flash risk caused by environmental sudden change in facility agriculture. The core problem lies in the neglect of complex spatial differences in the facility, the lack of physiological signals of crops, and the extensive regulation means. To solve this technical problem, our design idea focuses on how to realize the overall upgrade from local to global, from environment to crops, and from extensive to precise.

[0018] To overcome the limitations of traditional point monitoring, it is necessary to establish continuous perception ability of the entire facility space environment. We simultaneously collect environmental data and crop leaf posture data distributed at multiple locations in the facility. The change amount of environmental data at different times of each monitoring point is constructed as an environmental variation vector to represent the intensity and direction of environmental change at that point. At the same time, the change of leaf inclination is converted into an abnormal feature value of leaf posture through trigonometric function, which serves as direct evidence of early physiological water deficit in crops. These two types of data together constitute the bottom signal of risk assessment.

[0019] To solve the problem of spatial heterogeneity, it is necessary to convert the risk information of discrete points into a continuous spatial risk field. We define an analysis range around each monitoring point and select a series of calculation points within this range. By integrating the environmental variation intensity and leaf abnormal features of the monitoring point, we calculate the comprehensive risk value of these calculation points. Then, using spatial interpolation algorithm, we weight and integrate the comprehensive risk values of each calculation point according to its spatial distance from all monitoring points, and finally generate a spatial risk distribution matrix that reflects the continuous distribution of risks in the facility. This step realizes the key leap from isolated point data to overall risk situation map.

[0020] After obtaining the global risk situation, it is necessary to intelligently diagnose its pattern to distinguish different types of threats. We analyze the overall intensity and spatial dispersion of the spatial risk distribution matrix. When the overall risk is high and the high-risk points are highly dispersed in space, it is consistent with the characteristics of instantaneous disturbance such as gust, i.e., it is determined as a transient disturbance pattern. Conversely, if the high-risk points are spatially aggregated, it is consistent with the characteristics of environmental sudden change-induced stress, and it is determined as a sustained stress pattern. This automatic discrimination based on spatial statistical characteristics realizes the preliminary classification of the source of risk.

[0021] Based on different risk patterns, different regulation strategies are implemented. For transient disturbance, only short-term and mild local ventilation is started to buffer, avoiding overreaction. For sustained stress, further precise positioning is required. We analyze the internal gradient of the high-risk aggregation area to identify the dominant direction of risk spread and the most core position of risk. Taking this core as the starting point, we divide the space into multiple zonal partitions along the direction of decreasing risk, and implement hierarchical regulation from strong to weak, such as strong ventilation and shading in the risk core area, and moderate intensity intervention in the peripheral area. This realizes the optimal allocation of regulation resources in space.

[0022] It possesses self-optimization capabilities based on feedback from its effects. After the implementation of control measures, changes in risk intensity and spatial distribution are reassessed. If the risk significantly decreases and its distribution becomes more concentrated, it indicates that the control is effective and precise, and the scope of subsequent monitoring analysis will be appropriately narrowed to improve local sensitivity. If the risk distribution spreads or its intensity does not decrease, the scope of analysis will be expanded and the control efforts will be strengthened. By continuously comparing actions and effects, its perception and decision-making parameters can be dynamically adjusted, thereby continuously adapting to the unique characteristics of specific facility environments, forming a complete closed loop from perception, diagnosis, decision-making, execution to optimization.

[0023] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0024] like Figure 1 The diagram shown is a structural schematic of the crop growth environment data dynamic monitoring and control method based on nanosensors provided in this application embodiment. It includes the following steps: acquiring the environmental data stream and leaf tilt angle of the target grid point; constructing an environmental variation vector for the target grid point based on the environmental data stream; processing the leaf tilt angle using a sine function to obtain leaf posture anomaly characteristic values; arbitrarily selecting several discrete points within a preset radius of the target grid point, denoted as first-level grid points; calculating the comprehensive risk value of the first-level grid points based on the Euclidean norm of the environmental variation vector and the leaf posture anomaly characteristic values; and, based on the relative spatial position and comprehensive risk value of each first-level grid point relative to the target grid point, through inverse... A spatial risk distribution matrix is ​​constructed using distance-weighted interpolation. The mean and standard deviation of the comprehensive risk values ​​of all first-level grid points in the spatial risk distribution matrix are calculated. If the comprehensive risk value of a first-level grid point in the spatial risk distribution matrix exceeds a preset intensity threshold and the standard deviation exceeds a preset standard deviation threshold, then the first-level grid point is marked as a second-level grid point. The average Euclidean distance between the second-level grid point and all other second-level grid points within a preset radius is calculated and recorded as the risk dispersion. When the mean exceeds a preset risk threshold, if the risk dispersion exceeds a preset dispersion threshold, then inhibitory control is implemented. If the risk dispersion does not exceed a preset dispersion threshold, then tiered control is implemented.

[0025] Nanosensing refers to the technology of using nanomaterials or nanostructures as sensing elements to monitor physical, chemical, or biological parameters in the crop growth environment with high precision and in real time. Its key feature is its ability to achieve microscopic-scale sensing, providing a foundation for acquiring refined environmental data streams and crop physiological response data.

[0026] Target grid points refer to specific spatial locations within a crop growing area that are selected as monitoring and analysis centers. These points are typically determined through pre-planning or dynamic selection and represent the environmental and crop status of a local area.

[0027] The environmental data stream refers to a series of data collected continuously by sensors (such as nanosensors) reflecting the state of the crop growing environment. This data stream can include environmental physical parameters such as light intensity, air temperature, air humidity, etc.

[0028] The leaf angle refers to the angle of the crop leaf relative to the horizontal plane or vertical plane. Changes in this angle can serve as an indirect indicator of crop physiological state, particularly water stress or growth posture changes.

[0029] The environmental variation vector refers to a vector that quantifies the degree and direction of change in the environmental parameters of the target grid point at different times or in different spaces by processing the environmental data stream. This vector can reflect the severity of environmental changes.

[0030] The leaf posture abnormality feature value refers to a value obtained by a specific mathematical processing (such as a sine function processing) of the leaf angle data, which is used to represent the abnormality degree of the crop leaf posture, and thus reflects the possible physiological stress state of the crop.

[0031] The first-level grid point refers to a number of discrete monitoring points arbitrarily selected within a predetermined radius around the target grid point. These points are used to expand the monitoring range and obtain environmental and crop information in the surrounding area of the target grid point.

[0032] The comprehensive risk value refers to a value calculated by combining environmental factors such as the Euclidean norm of the environmental variation vector and crop physiological factors such as the leaf posture abnormality feature value, which is used to quantify the comprehensive index of the flash risk faced by the crop. The higher the value, the greater the risk faced by the crop.

[0033] The inverse distance weighted interpolation method is a spatial interpolation technique that estimates the value of an unknown point by considering the distance relationship between known points and unknown points. The closer the distance between points, the greater the weight of the unknown point.

[0034] The spatial risk distribution matrix refers to a data structure that visualizes and quantifies the comprehensive risk values of each grid point in two-dimensional or three-dimensional space through spatial interpolation methods. This matrix can intuitively show the spatial distribution of crop flash risk.

[0035] The preset intensity threshold refers to a reference limit set for the comprehensive risk value when determining whether a first-level grid point is a second-level grid point. When the comprehensive risk value exceeds this threshold, it indicates that the risk intensity has reached a certain level.

[0036] The preset standard deviation threshold refers to a reference limit set for the standard deviation of the comprehensive risk value when determining whether a first-level grid point is a second-level grid point. When the standard deviation exceeds this threshold, it indicates that there is significant spatial variability in the risk distribution.

[0037] The secondary grid points refer to the primary grid points whose comprehensive risk value and standard deviation both exceed the preset threshold. These points are considered as representatives of high-risk areas and need further attention and regulation.

[0038] The risk dispersion refers to the average Euclidean distance between the secondary grid points and all other secondary grid points within their preset radius. This value measures the degree of aggregation or dispersion of high-risk areas in space.

[0039] The preset risk threshold refers to a reference limit set for the average value of the comprehensive risk when deciding whether to implement regulatory measures. When the average value exceeds this threshold, it indicates that the overall risk level is high.

[0040] The preset dispersion threshold refers to a reference limit set for the risk dispersion when deciding which regulatory strategy to implement. This threshold is used to distinguish whether the risk is concentrated or dispersed.

[0041] The suppressive regulation refers to a preliminary regulatory measure aimed at quickly reducing the overall risk intensity, but possibly with a wider range, when the overall risk level is high and the risk dispersion is large.

[0042] The hierarchical regulation refers to a more refined and differentiated intervention strategy targeting the spatial distribution characteristics of the risk when the overall risk level is high but the risk dispersion is small, or after the suppressive regulation, there is still risk.

[0043] The present application realizes early perception of crop physiological stress state by fusing environmental data stream obtained by nano-sensing and leaf inclination angle, constructing environmental variation vector and leaf posture abnormality characteristic value. Through spatial interpolation to construct risk distribution matrix, the limitations of traditional point monitoring are overcome, and the spatial heterogeneity of flash seedling risk in facilities can be accurately identified. According to the average value and dispersion of risk, the suppressive or hierarchical regulation strategy is intelligently selected, avoiding the high energy consumption and secondary stress problem of extensive regulation, so as to realize early, accurate and spatially resolved perception and prevention and control of flash seedling risk of facility seedlings in complex spatial heterogeneity environment.

[0044] Further, the specific construction process of the environmental variation vector is as follows: the environmental data stream includes light intensity, air temperature, and air humidity; the target grid point is taken as the origin, and the light intensity, air temperature, and air humidity collected by the target grid point at the current monitoring time are normalized and taken as components in three orthogonal dimensions to construct the agricultural environment vector at the current time; based on the environmental data stream collected by the same target grid point at the last monitoring time, the agricultural environment vector of the target grid point at the last monitoring time is constructed; the agricultural environment vector at the current time and the agricultural environment vector at the last monitoring time are subjected to vector subtraction operation to obtain the environmental variation vector of the target grid point.

[0045] In the present embodiment, if the current time is the initial time, that is, there is no last monitoring time, the preset vector is directly selected as the environmental variation vector for operation.

[0046] The environmental data stream is a series of real-time or quasi-real-time data for describing the state of the crop growth environment. The light intensity, air temperature, and air humidity are selected as the key components of the environmental data stream because the three are the core environmental factors that affect the physiological and ecological processes of crops, such as photosynthesis, transpiration, respiration, and the occurrence and development of diseases and pests. The light intensity directly affects the photosynthetic rate; the air temperature affects the enzyme activity and metabolic rate; and the air humidity is closely related to the water balance of crops and the opening and closing of stomata. Real-time collection of these data through nanosensors can provide basic and comprehensive information for subsequent environmental state analysis.

[0047] In order to quantify and compare the effects of different environmental factors on crop growth, it is necessary to unify them to a comparable scale. Normalization is to convert environmental data of different dimensions and different numerical ranges into a unified dimensionless interval, such as [0, 1] or [-1, 1], to eliminate the effects of dimension and order of magnitude differences, and to ensure that each environmental factor has equal importance or comparability in vector construction.

[0048] In order to evaluate the variation or change trend of the environment, a reference point is needed. The agricultural environment vector of the last monitoring time is used as the benchmark for the current environmental change. By collecting environmental data and constructing agricultural environment vectors for the same target grid point at consecutive monitoring times, environmental state snapshots can be formed over time, providing necessary comparison data for subsequent change calculation.

[0049] Through the above technical solution, the key environmental factors such as light intensity, air temperature, and air humidity are normalized and constructed into a multi-dimensional agricultural environment vector, and further through vector subtraction operation, the environmental variation vector is obtained, which can realize the fine and quantitative capture of the dynamic change of the crop growth environment. This method not only eliminates the differences in dimensions and numerical ranges of different environmental factors, ensuring the fairness of each factor in evaluation, but also through the introduction of vector difference in time dimension, it can accurately reflect the change trend and amplitude of environmental factors in a short time. This makes it possible to more sensitively identify small fluctuations or potential stress in the environment, providing more accurate and reliable environmental change data for subsequent comprehensive risk value calculation, thereby significantly improving the accuracy and timeliness of crop growth environment dynamic monitoring and control.

[0050] Further, the specific obtaining process of the comprehensive risk value is: calculating the blade pitch angle change amount of the same target grid point at the current monitoring moment and the last monitoring moment; inputting the blade pitch angle change amount into a preset sine function to output a characteristic scalar representing the physiological influence of blade drooping as a blade posture abnormality characteristic value; calculating the Euclidean norm of the environmental variation vector and multiplying it by a preset environmental weight coefficient to obtain an environmental stress contribution value; multiplying the blade posture abnormality characteristic value by a preset physiological weight coefficient to obtain a physiological stress contribution value; and averaging and summing the environmental stress contribution value and the physiological stress contribution value to obtain the comprehensive risk value of the target grid point.

[0051] In the embodiment, the application can effectively integrate external environmental changes and physiological responses of crops. Specifically, by calculating the blade pitch angle change amount and processing it through a sine function, the physiological influence of blade drooping can be more accurately quantified, overcoming the limitations that may exist with only raw inclination data. At the same time, by introducing environmental weight coefficients and physiological weight coefficients, environmental stress and physiological stress can be reasonably weighted in comprehensive risk assessment according to their actual importance to crops, avoiding one-sidedness of single index assessment. Finally, through averaging and summing, a comprehensive risk value reflecting the degree of stress suffered by crops is obtained, which not only considers environmental causes, but also takes into account the physiological performance of crops, thereby providing more accurate and reliable basis for subsequent risk identification and precise regulation, significantly improving the fine level and effectiveness of monitoring and regulation.

[0052] Further, the specific construction process of the spatial risk distribution matrix is: calculating the Euclidean distance from the primary grid point to the preset target grid point; calculating the interpolation weight of the primary grid point relative to each target grid point based on the power function inverse of the Euclidean distance; weighting and averaging the comprehensive risk values of each target grid point to obtain the comprehensive risk value of the primary grid point; and repeating the above calculation process for all primary grid points to generate the spatial risk distribution matrix.

[0053] In the embodiment, the preset target grid point represents that for a specific primary grid point, all target grid points within the pre-defined surrounding spatial area can be selected as the preset target grid point.

[0054] The interpolation weight of the first-level grid point relative to each target grid point is calculated based on the power function inverse proportional to the Euclidean distance. This step utilizes the core idea of inverse distance weighted interpolation, that is, the target grid point closer in distance has a greater impact on the first-level grid point. By introducing the power function inverse proportional relationship, the influence degree of distance on the weight can be flexibly adjusted, for example, when the power index is larger, the target grid point closer in distance will obtain a higher weight, so that its comprehensive risk value has a more significant impact on the first-level grid point. Specifically, the interpolation weight can be expressed as the negative p-th power of the distance, that is, 1 / d^p, where d is the Euclidean distance and p is a preset power index. The calculated weight is usually normalized to ensure that the total influence weight of all target grid points on the first-level grid point is 1.

[0055] Through the above technical solutions, the spatial risk distribution matrix can be accurately and efficiently constructed, so that the comprehensive risk value of the first-level grid point can more accurately reflect the actual risk level of its location. This detailed interpolation process ensures the standardization and repeatability of risk assessment, avoiding evaluation deviation caused by unclear interpolation method. On this basis, a more detailed and reliable data basis is provided for subsequent secondary grid point identification, so that the high-risk area can be more accurately located. Therefore, the application supports more detailed risk control decisions, so that suppressive control and hierarchical control can be based on more accurate spatial risk information, significantly improving the pertinence and effectiveness of the control.

[0056] Further, the specific execution process of the suppressive control is as follows: a dynamic duration is calculated according to the risk intensity, and the specific calculation formula of the dynamic duration is as follows: wherein, represents the dynamic duration, is a basic duration, is a preset proportion coefficient, is the risk intensity, is a preset maximum duration; the partition ventilation window within the preset distance of the secondary grid point is opened to a first preset opening degree and runs at a preset power, and is automatically closed after the dynamic duration; after the partition ventilation window is automatically closed, the risk intensity is calculated again and recorded as a first-level risk intensity, and the risk dispersion is calculated again and recorded as a first-level risk dispersion; when the first-level risk intensity does not exceed the preset risk threshold, no treatment is performed; when the first-level risk intensity exceeds the preset risk threshold and the first-level risk dispersion exceeds the preset dispersion threshold, hierarchical control is performed.

[0057] In the embodiment, after determining the dynamic duration, the sub-area ventilation window within the preset distance of the secondary grid point is opened to the first preset opening degree and runs at a preset power for the dynamic duration and then automatically closes. The sub-area ventilation window can be an independently controllable ventilation opening in the greenhouse or the big shed, such as a side window or a top window, and the first preset opening degree can be set to 20% or 50% to achieve the preliminary ventilation cooling or dehumidification effect. The preset power refers to the running power of the ventilation equipment, such as the rotating speed of the fan or the power of the air extractor. After the dynamic duration ends, the control automatically closes the sub-area ventilation window, and completes a round of suppression control. In addition to the ventilation window, local spraying or local shading can also be opened to achieve the purpose of suppressing the risk.

[0058] The application can intelligently adjust the duration of the suppression control according to the actual risk degree of the crop growth environment, avoid the insufficient or excessive control caused by the fixed duration, and thus improve the accuracy of the control and the resource utilization efficiency. In addition, after the preliminary suppression control is completed, the environmental risk is immediately re-evaluated, the first risk intensity and the first risk dispersion are calculated, and the control effect can be timely judged. If the risk has been effectively controlled, further operation is stopped to avoid unnecessary energy consumption; if the risk still exceeds the threshold value and there is a diffusion trend, more refined hierarchical control can be timely started to ensure the rapid response and effective treatment of the complex risk condition, and the intelligence and adaptability of the crop growth environment control are significantly improved.

[0059] Further, the specific execution process of the hierarchical control is as follows: the environmental variation vector of the secondary grid point is obtained and the module length thereof is calculated, if the module length exceeds the preset module length threshold value, the direction of the environmental variation vector is taken as the judgment space direction; at least one verification path is generated along the direction within the preset angle range of the judgment space direction with the secondary grid point as the starting point; on the verification path, the comprehensive risk value is extracted at a preset interval to form a comprehensive risk value sequence; the Spearman correlation coefficient of the comprehensive risk value sequence and the preset comprehensive risk value sequence is calculated, if the Spearman correlation coefficient of the verification path exceeding the preset number threshold value is greater than the preset monotonic threshold value, the corresponding verification path is marked as an effective path, and in all effective paths, the spatial coordinates of the secondary grid point with the maximum comprehensive risk value are taken as the maximum risk region position coordinates.

[0060] In this embodiment, at least one verification path is generated from the secondary grid point as the starting point, along the direction within the preset angle range of the judgment spatial direction as the ray direction. Once the judgment spatial direction is determined, in order to more comprehensively evaluate the potential propagation path of the risk, it is necessary to generate multiple verification paths from the secondary grid point along the judgment direction and within a certain angle range near the judgment direction. These paths are potential trajectories simulating the possible diffusion or influence of the risk, which are used for subsequent risk assessment. The preset angle range can be a sector area extending to both sides with the judgment spatial direction as the center, for example, ±30 degrees or ±45 degrees. Within the angle range, multiple directions can be selected as the ray direction at equal intervals, and each ray extends outward from the secondary grid point to form a verification path. The length of the path can be preset to a certain fixed value, or dynamically adjusted according to the module length of the environmental variation vector. The larger the module length, the longer the path is likely to be.

[0061] The Spearman correlation coefficient is a non-parametric statistical indicator used to measure the direction and strength of monotonic relationship between two variables. Here, it is used to evaluate the similarity between the sequence of comprehensive risk values on the verification path and a preset ideal risk diffusion pattern. If the correlation coefficient is high and exceeds the preset monotonicity threshold, it means that the risk distribution on the path is consistent with the expected risk diffusion pattern, and therefore it is considered to be an effective path.

[0062] Among all the effective paths, the spatial coordinates of the secondary grid point with the maximum comprehensive risk value are determined as the maximum risk area position coordinates, which represent the area where the risk is most concentrated or most severe. The preset sequence of comprehensive risk values can be an idealized risk decreasing model, such as a linearly decreasing or exponentially decreasing sequence from high to low, reflecting the trend of gradually decreasing intensity as the risk source spreads outward. The calculation of the Spearman correlation coefficient is a standard statistical method. The preset quantity threshold and the preset monotonicity threshold can be set according to actual experience or model training results.

[0063] Through the above technical solution, when the preliminary inhibitory control is not enough to solve the problem, the application can perform more in-depth and fine analysis on the risk area. By obtaining the environmental variation vector of the secondary grid point and judging its module length and direction, the potential propagation trend of the risk can be preliminarily identified.

[0064] On this basis, multiple verification paths are generated from the secondary grid point as the starting point, along the direction within the preset angle range of the judgment spatial direction, and the comprehensive risk values are extracted at preset intervals on these paths to form a sequence of comprehensive risk values. Subsequently, the Spearman correlation coefficient of these sequences of comprehensive risk values and the preset sequence of comprehensive risk values is calculated, and the effective paths that highly match the actual risk propagation pattern are selected according to the preset quantity threshold and the preset monotonicity threshold.

[0065] Finally, by determining the spatial coordinates of the secondary grid point with the maximum integrated risk value in all effective paths as the maximum risk area position coordinates, the application can accurately locate the area with the most concentrated or most serious risk, thereby providing accurate risk source positioning for subsequent hierarchical regulation, avoiding blind or large-scale regulation, significantly improving the targeting and efficiency of regulation, and ensuring that resources are concentrated in the areas that need intervention the most, thereby more effectively controlling the crop growth environment risk.

[0066] Further, the specific execution process of hierarchical regulation further includes: taking the maximum risk area position coordinates as the starting point, dividing the preset space into K consecutive regulation partitions in the descending direction of the integrated risk value according to a preset rule; executing a first regulation intensity for the first regulation partition, and executing a regulation intensity instruction of a regulation intensity that decreases by stages for the second to Kth regulation partitions in turn; wherein K is an integer greater than 1, and the regulation intensity of the Kth regulation partition is lower than that of the K-1th regulation partition; for the Kth regulation partition, the regulation intensity instruction of the corresponding regulation intensity includes: opening the corresponding regulation partition to the Kth preset opening degree, starting the directional atomizing nozzle to operate at the Kth preset atomizing amount, and adjusting the local sunshade net to the Kth preset shading rate; after the execution of the regulation intensity instruction of the Kth regulation partition is completed, the risk intensity and risk dispersion are obtained again, denoted as secondary risk intensity and secondary risk dispersion, and the preset radius range is adjusted accordingly.

[0067] In the present embodiment, Figure 2 The logical flow diagram of hierarchical regulation provided by the present embodiment takes the maximum risk area position coordinates as the starting point, which is the point with the most concentrated risk determined according to the foregoing steps. Taking this point as the starting point means that the focus and intensity of regulation will be centered around this high-risk area. The descending direction of the integrated risk value means that the regulation partitions are divided according to the spatial distribution of risk intensity from high to low.

[0068] This is achieved by analyzing the gradient of the integrated risk value in the spatial risk distribution matrix, for example, from the maximum risk area position coordinates, whenever the integrated risk value drops by a preset threshold or reaches a certain preset percentage, a new regulation partition boundary is drawn. The preset rule can include various division strategies, such as distance-based division, risk value gradient-based division, or dynamic division based on factors such as crop sensitivity to environmental changes, regional microclimate characteristics, etc. The K consecutive regulation partitions mean that the entire preset space that needs to be regulated is subdivided into K adjacent sub-regions, each corresponding to a regulation level. K is an integer greater than 1, ensuring the hierarchical nature of the regulation.

[0069] For the Kth regulation subzone, the regulation intensity instruction corresponding to the regulation intensity includes: opening the corresponding regulation subzone ventilation window to the Kth preset opening degree, starting the directional atomizing nozzle to operate at the Kth preset atomizing amount, and adjusting the local sunshade net to the Kth preset shading rate. Wherein, the subzone ventilation window, the directional atomizing nozzle and the local sunshade net are specific environmental regulation execution devices. The Kth preset opening degree, the Kth preset atomizing amount and the Kth preset shading rate are specific parameters preset according to the regulation intensity level corresponding to the current regulation subzone. For example, for the first subzone with the highest risk, the ventilation window can be opened to the maximum opening degree, the atomizing nozzle operates at the maximum atomizing amount, and the sunshade net is adjusted to the maximum shading rate; and for the Kth subzone with lower risk, these parameters are correspondingly reduced to provide more moderate regulation. The fine setting of these parameters is the key to realize differentiated regulation.

[0070] The fine and differentiated regulation of the crop growth environment risk can be realized. Taking the maximum risk area position coordinate as the center, the continuous regulation subzones are divided in the descending direction of the comprehensive risk value, and the regulation intensity is gradually decreased for different subzones, which effectively avoids extensive regulation, thereby accurately matching the risk level of different areas and improving the pertinence and efficiency of regulation. For example, strong regulation is applied in the area with the highest risk, and weak regulation is applied in the area with lower risk, which can quickly suppress the core risk and avoid excessive intervention in the low-risk area, thereby reducing energy consumption and the negative impact on the normal growth of crops. In addition, after the regulation is completed, the risk intensity and the risk dispersion are obtained again, and the preset radius range is dynamically adjusted according to the risk intensity and the risk dispersion, thereby forming a closed-loop feedback regulation mechanism, which makes the whole regulation process more adaptive and intelligent, can continuously optimize the regulation strategy according to the actual effect, and ensures that the crop growth environment is always in the best state.

[0071] Further, the specific process of adjusting the preset radius range is as follows: calculating the risk intensity change rate and the risk dispersion change rate according to the secondary risk intensity and the secondary risk dispersion; calculating the difference between the primary risk intensity and the secondary risk intensity, and dividing the difference by the primary risk intensity, and the quotient obtained is the risk intensity change rate; calculating the difference between the primary risk dispersion and the secondary risk dispersion, and dividing the difference by the primary risk dispersion, and the quotient obtained is the risk dispersion change rate; when the risk intensity change rate is greater than zero and the risk dispersion change rate is greater than zero, it is determined that the effective concentration regulation is performed, and the preset radius range is adjusted according to the first adjustment formula to obtain the first adjustment radius, and the specific first adjustment formula is as follows: wherein, the first adjustment radius is represented by R1, the preset radius range is represented by R, the preset minimum radius range is represented by Rmin, the preset first adjustment coefficient is represented by a1. represents the risk dispersion change rate; when the risk intensity change rate is greater than zero and the risk dispersion change rate is less than or equal to zero, it is determined that the risk diffusion is regulated, and the preset radius range is adjusted according to a second adjustment formula to obtain a second adjustment radius, and the specific second adjustment formula is as follows: , wherein, represents the second adjustment radius, is a preset radius range, is a preset maximum radius range, is a preset second adjustment coefficient; when the risk intensity change rate is less than or equal to zero, it is determined that the regulation is invalid, and a warning is issued.

[0072] In this embodiment, the present application provides an adaptive monitoring radius adjustment mechanism. This mechanism can dynamically adjust the preset radius range of subsequent monitoring and regulation according to the actual effect of the regulation measures. When the regulation is effective and the risk is concentrated, the monitoring range is reduced to improve the resource utilization efficiency; when the risk intensity is reduced but has a diffusion trend, the monitoring range is expanded to ensure comprehensive coverage of the potential risk area; when the regulation is invalid, a warning is issued in time to avoid the problem from getting worse. This dynamic adjustment capability significantly enhances the intelligence, flexibility and response speed of the entire monitoring and regulation, enabling it to more accurately and efficiently cope with dynamic risks in the crop growing environment, thereby improving the management level and production efficiency of crop growth.

[0073] Further, the specific process of issuing a warning is: controlling the movable industrial vision acquisition device to move to the maximum risk area position coordinates and acquiring images of the corresponding crops to obtain crop canopy vision images; sending the crop canopy vision images and the maximum risk area position coordinates to the user terminal.

[0074] In this embodiment, the movable industrial vision acquisition device is a device that can move autonomously or remotely controlled, and is equipped with an industrial-level vision sensor. This device can be mounted on a track, unmanned aerial vehicle, automatic guided vehicle, etc. The mobility of the device enables it to accurately reach a specific area for data acquisition. The industrial-level vision sensor ensures the quality and stability of image acquisition, which is suitable for the complexity of agricultural environment. Controlling the movable industrial vision acquisition device to move to the maximum risk area position coordinates and acquiring images of the corresponding crops means that after the mobile control receives the maximum risk area position coordinates, it plans a path according to the coordinates and drives the movable industrial vision acquisition device to move to the target position.

[0075] After reaching the target position, the visual sensor is started to take photos or videos of the crops, collecting multi-angle, multi-spectral or high-resolution image data to obtain more detailed visual information of the crops. The crop canopy visual image refers to the image data obtained by the visual collection device, reflecting the information of the crop canopy shape, color, texture, health status, etc. These images can be RGB images, multi-spectral images, hyperspectral images or thermal infrared images, etc., which can be used to analyze the growth potential, disease and pest, nutrition status, water stress and other physiological indicators of the crops. The crop canopy visual image and the maximum risk area position coordinates are sent to the user terminal, and the data can be uploaded to the cloud platform or server through the wireless communication module, and then accessed and displayed by the user terminal through the application program or web interface. The user terminal can receive and display the image, and at the same time mark the maximum risk area position coordinates on the map to realize intuitive risk visualization.

[0076] When it is determined that the regulation is invalid and a warning is issued, it is no longer just to provide an abstract warning, but to drive the movable industrial visual collection device to accurately move to the maximum risk area position coordinates and collect images of the crops in the area, so as to obtain intuitive and detailed crop canopy visual images. These images are sent to the user terminal together with the maximum risk area position coordinates, so that the user can directly observe the specific performance of the crops in the risk area, such as leaf discoloration, wilting, disease spots or signs of insect pests, etc. This greatly enhances the practicality and operability of the warning information, and the user can make more accurate judgment and diagnosis on the abnormal condition of the crops based on the visual image and accurate position information, so as to take targeted and fine intervention measures in time to avoid the risk from further expanding, and improve the intelligent level and decision-making efficiency of the crop growth environment regulation.

[0077] The embodiment of the present application also provides a computer readable storage medium for storing a computer program, which is executed by a processor to realize the steps of the crop growth environment data dynamic monitoring and regulation method based on nanosensing.

[0078] Those skilled in the art will appreciate that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CDROM, optical storage, etc.) containing computer-usable program code.

[0079] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0080] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0082] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0083] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dynamic monitoring and control of crop growth environment data based on nanosensors, characterized in that, Includes the following steps: The environmental data stream and blade tilt angle of the target grid point are obtained. Based on the environmental data stream, the environmental variation vector of the target grid point is constructed. The blade tilt angle is processed by a sine function to obtain the blade attitude anomaly feature value. Within the preset radius of the target grid point, select any number of discrete points and denot them as first-level grid points; The comprehensive risk value of the first-level grid points is calculated based on the Euclidean norm of the environmental variation vector and the characteristic value of the blade attitude anomaly. Based on the relative spatial location and comprehensive risk value of each first-level grid point and the target grid point, a spatial risk distribution matrix is ​​constructed using the inverse distance weighted interpolation method. Calculate the average and standard deviation of the comprehensive risk values ​​of all first-level grid points in the spatial risk distribution matrix. If the comprehensive risk value of a first-level grid point in the spatial risk distribution matrix exceeds a preset intensity threshold and the standard deviation exceeds a preset standard deviation threshold, then mark the first-level grid point as a second-level grid point. Calculate the average Euclidean distance between a secondary grid point and all other secondary grid points within a preset radius, and record it as the risk dispersion. When the average value exceeds the preset risk threshold, if the risk dispersion exceeds the preset dispersion threshold, then inhibitory control is implemented; if the risk dispersion does not exceed the preset dispersion threshold, then tiered control is implemented.

2. The method for dynamic monitoring and control of crop growth environment data based on nanosensors according to claim 1, characterized in that, The specific construction process of the environmental variation vector is as follows: Environmental data streams include light intensity, air temperature, and air humidity; Using the target grid point as the origin, the light intensity, air temperature, and air humidity collected at the target grid point at the current monitoring time are normalized and then used as components in three orthogonal dimensions to construct the agricultural environment vector at the current time. Based on the environmental data stream collected at the previous monitoring time for the same target grid point, construct the agricultural environment vector of that target grid point at the previous monitoring time; The environmental variation vector of the target grid point is obtained by subtracting the agricultural environment vector at the current time from the agricultural environment vector at the previous monitoring time.

3. The method for dynamic monitoring and control of crop growth environment data based on nanosensors according to claim 1, characterized in that, The specific process for obtaining the comprehensive risk value is as follows: Calculate the change in blade tilt angle at the same target grid point between the current monitoring time and the previous monitoring time; The change in leaf tilt angle is input into a preset sine function, and a characteristic scalar that characterizes the physiological impact of leaf drooping is output as a characteristic value of abnormal leaf posture. Calculate the Euclidean norm of the environmental variation vector and multiply it by the preset environmental weight coefficient to obtain the environmental stress contribution value; The physiological stress contribution value is obtained by multiplying the abnormal leaf posture feature value with the preset physiological weight coefficient. The comprehensive risk value of the target grid point is obtained by averaging the contribution values ​​of environmental stress and physiological stress.

4. The method for dynamic monitoring and control of crop growth environment data based on nanosensors according to claim 1, characterized in that, The specific construction process of the spatial risk distribution matrix is ​​as follows: Calculate the Euclidean distance from the first-level grid point to the preset target grid point; The interpolation weights of the first-level grid points relative to each target grid point are calculated based on the inverse power function of Euclidean distance. The comprehensive risk value of each target grid point is obtained by weighting the comprehensive risk value of the first-level grid point by taking the weighted average of the comprehensive risk values ​​of each target grid point. Repeat the above calculation process for all first-level grid points to generate a spatial risk distribution matrix.

5. The method for dynamic monitoring and control of crop growth environment data based on nanosensors according to claim 1, characterized in that, The specific execution process of the inhibitory regulation is as follows: The dynamic duration is calculated based on the risk intensity. The specific formula for calculating the dynamic duration is as follows: ,in, Indicates dynamic duration. Based on duration, This is a preset proportional coefficient. The risk intensity is... The preset maximum duration; Open the ventilation windows of the partitions within a preset distance of the secondary grid points to the first preset opening degree and operate at a preset power for the specified dynamic duration, then automatically close them. After the zoned ventilation windows automatically close, the risk intensity is recalculated and recorded as Level 1 risk intensity, and the risk dispersion is recalculated and recorded as Level 1 risk dispersion. If the level 1 risk intensity does not exceed the preset risk threshold, no action will be taken; When the intensity of Level 1 risk exceeds the preset risk threshold and the dispersion of Level 1 risk exceeds the preset dispersion threshold, tiered regulation will be implemented.

6. The method for dynamic monitoring and control of crop growth environment data based on nanosensors according to claim 5, characterized in that, The specific implementation process of the hierarchical regulation is as follows: Obtain the environmental variation vector of the secondary grid point and calculate its magnitude. If the magnitude exceeds the preset magnitude threshold, the direction of the environmental variation vector is used as the spatial direction for judgment. Starting from the secondary grid point, at least one verification path is generated by taking the direction within the preset angle range of the spatial direction as the ray direction. Along the verification path, comprehensive risk values ​​are extracted at preset intervals to form a comprehensive risk value sequence; Calculate the Spearman correlation coefficient between the comprehensive risk value sequence and the preset comprehensive risk value sequence. If the Spearman correlation coefficient of the verification path that exceeds the preset number threshold is greater than the preset monotonic threshold, then mark the corresponding verification path as a valid path. Among all valid paths, the spatial coordinates of the secondary grid point with the largest comprehensive risk value are taken as the location coordinates of the maximum risk area.

7. The method for dynamic monitoring and control of crop growth environment data based on nanosensors according to claim 6, characterized in that, The specific implementation process of the tiered regulation also includes: Starting from the coordinates of the area with the greatest risk, the preset space is divided into K consecutive control zones along the decreasing direction of the comprehensive risk value according to preset rules. The first level of control intensity is applied to the first control zone, and control intensity instructions with progressively decreasing intensity are applied to the second to the Kth control zones. Where K is an integer greater than 1, and the control intensity of the Kth level control zone is lower than that of the (K-1)th level control zone. For the Kth level control zone, the corresponding control intensity instructions include: Open the corresponding control zone ventilation window to the K-level preset opening, start the directional atomizing nozzle to operate at the K-level preset atomization amount, and adjust the local shading net to the K-level preset shading rate; After the control intensity command for the Kth level control zone is executed, the risk intensity and risk dispersion are obtained again and recorded as the secondary risk intensity and secondary risk dispersion, and the preset radius range is adjusted accordingly.

8. The method for dynamic monitoring and control of crop growth environment data based on nanosensors according to claim 7, characterized in that, The specific process for adjusting the preset radius range is as follows: Calculate the rate of change of risk intensity and the rate of change of risk dispersion based on the secondary risk intensity and the secondary risk dispersion. Calculate the difference between the primary risk intensity and the secondary risk intensity, and divide this difference by the primary risk intensity. The quotient is the rate of change of risk intensity. Calculate the difference between the primary risk dispersion and the secondary risk dispersion, and divide this difference by the primary risk dispersion. The resulting quotient is the risk dispersion change rate. When both the rate of change of risk intensity and the rate of change of risk dispersion are greater than zero, it is determined to be effective centralized control, and the preset radius range is adjusted according to the first adjustment formula to obtain the first adjustment radius. The specific first adjustment formula is as follows: ,in, Indicates the first adjustment radius. For the preset radius range, The preset minimum radius range, The preset first adjustment coefficient, This represents the rate of change of risk dispersion; When the rate of change of risk intensity is greater than zero and the rate of change of risk dispersion is less than or equal to zero, it is determined to be a risk diffusion control measure, and the preset radius range is adjusted according to the second adjustment formula to obtain the second adjustment radius. The specific second adjustment formula is as follows: ,in, Indicates the second adjustment radius. For the preset radius range, The preset maximum radius range, This is the preset second adjustment coefficient; When the rate of change of risk intensity is less than or equal to zero, it is determined that the regulation is ineffective and an early warning is issued.

9. The method for dynamic monitoring and control of crop growth environment data based on nanosensors according to claim 8, characterized in that, The specific process for issuing the warning is as follows: The mobile industrial vision acquisition device is controlled to move to the coordinates of the maximum risk area and acquire images of the corresponding crops to obtain visual images of the crop canopy. The visual image of the crop canopy and the coordinates of the location of the area of ​​greatest risk are sent to the user terminal.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-9.

Images